EFTA01098878

IWUROPSYCHOLOGIA ELSEVIER Neuropsychologia 44 (2006) 2037-2078 www elsevier comilocateineumpsychologia Development of cognitive control and executive functions from 4 to 13 years: Evidence from manipulations of memory, inhibition, and task switching Matthew C. Davidson a'b, Dima Amso a, Loren Cruess Anderson c, Adele Diamond d,* Sackler Institutefor Developmental Psychobiologx frill Medical College of Cornell University; New York. NY. USA b Department ofPsychologx University ofMassachusetts. Amherst. MA. USA Shrive, Center. University ofMassachusetts Medical School, Waltham. MA. USA d Department ofPsychiatrx University ofBritish Columbia & Division ofChild & Adolescent Psychiatty, BC Children's Hospital. Vancouver. Canada Received 20 November 2005: received in revised form 7 February 2006: accepted 10 February 2006 Available online 31 March 2006 Abstract Predictions concerning development, interrelations, and possible independence of working memory, inhibition, and cognitive flexibility were tested in 325 panicipants (roughly 30 per age from 4 to 13 years and young adults.. 50% female). All were tested on the same computerized battery. designed to manipulate memory and inhibition independently and together. in steady state (single-task blocks) and during task-switching. and to be appropriate over the lifespan and for neuroimaging (MARI). This is one of the first studies, in children or adults, to explore: (a) how memory requirements interact with spatial compatibility and (b) spatial incompatibility effects both with stimulus-specific rules (Simon task) and with higher-level, conceptual rules. Even the youngest children could hold information in mind, inhibit a dominant response. and combine those as long as the inhibition required was steady-state and the rules remained constant. Cognitive flexibility (switching between rules), even with memory demands minimized, showed a longer developmental progression. with 13-year-olds still not at adult levels. Effects elicited only in Mixed blocks with adults were found in young children even in single-task blocks: while young children could exercise inhibition in steady state it exacted a cost not seen in adults, who (unlike young children) seemed to re-set their default response when inhibition of the same tendency was required throughout a block. The costs associated with manipulations of inhibition were greater in young children while the costs associated with increasing memory demands were greater in adults. Effects seen only in RT in adults were seen primarily in accuracy in young children. Adults slowed down on difficult trials to preserve accuracy: but the youngest children were impulsive; their RT remained more constant but at an accuracy cost on difficult trials. Contrary to our predictions of independence between memory and inhibition, when matched for difficulty RT correlations between these were as high as 0.8. although accuracy correlations were less than half that. Spatial incompatibility effects and global and local switch costs were evident in children and adults, differing only in size. Other effects (e.g.. asymmetric switch costs and the interaction of switching rules and switching response-sites) differed fundamentally over age. O 2006 Elsevier Ltd. All rights reserved. Keywords: Task switching: Inhibition: Working memory: Simon effect: Asymmetric switch costs: Global and local switch costs: Stimulus—response compatibility: Development: Children: Frontal lobe Mature cognition is characterized by abilities that include appropriately, and (c) to quickly and flexibly adapt behavior being able: (a) to hold information in mind, including compli- to changing situations. These abilities are referred to respec- cated representational structures, to mentally manipulate that tively as working memory, inhibition, and cognitive flexibility. information, and to act on the basis of it, (b) to act on the basis Together they are key components of both "cognitive control" of choice rather than impulse, exercising self-control (or self- and "executive functions" and have been studied in a wide vari- regulation) by resisting inappropriate behaviors and responding ety of experimental paradigms with diverse subject groups. Our battery of interrelated tasks enabled us to indepen- dently and systematically vary demands on these abilities and to track their development across a wider age range than hereto- • Corresponding author at: Department of Psychiatry. University of British fore investigated using the same measures at all ages. Hav- Columbia. 2255 Wesbrook Mall. Vancouver. BC. Canada V6T 2AL Tel.: +1 604 822 7220: fax: +1 604 822 7232. ing measures that span a wide age range is important given E-mail address: adele.diamondaubc.ca (A. Diamond). the protracted developmental progressions of many executive 0028-3932(8 — see front matter O 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropsychologia.2006.02.006 EFTA01098878 2038 MC. Davidson e at /Neumps)rhologia 44 (2006)2037-2078 function and cognitive control skills. While some cognitive did not have to be held in mind and in another case by using abilities develop early, executive functions do not reach their stimuli (Arrows) that pointed to where to respond. We increased peak until early adulthood (DeLuca et al., 2003; Diamond, the working memory requirements by introducing conceptual 2002; Fischer, Biscaldi, & Gezeck, 1997; Harnishfeger & Pope, rules, where the correct response required mental manipulation. 1996; Kail, 1991abc; Kail & Salthouse, 1994; Luciana & Instead of a rule being "for A press left:' a rule was "for A Nelson, 2002; Luciana, Conklin, Hooper, & Yarger, 2005; Luna, press on the side opposite A:' Thus, in addition to activating Garver, Urban, Lazar, & Sweeney, 2004; Lyons-Warren, Lillie, the rules associated with the two stimuli (the memory require- & Hershey, 2004; Munoz, Broughton, Goldring, & Armstrong, ment in standard Simon tasks), participants had to instantiate the 1998; Zelazo, Craik, & Booth, 2004). Each test in our battery appropriate rule for the particular spatial location of the stimulus can be performed by children as young as 4 years; yet adults on each trial. still find many of them challenging. The entire battery takes Task-switching paradigms target the ability to flexibly shift less than 30 min to complete. These tests are also designed from one mindset to another, often times acting according to to be appropriate for testing nonhuman primates and for neu- rules that would be incompatible with the other mindset. This roimaging research using functional magnetic resonance imag- has been studied extensively in adults (e.g., Allport, Styles, & ing (fMRI) (Diamond, O'Craven, & Savoy, 1998; O'Craven, Hsieh. 1994; Jersild, 1927; Meiran, Gotler, & Perlman, 1996; Savoy, & Diamond, 1998). Monsell & Driver, 2000; Rogers & Monsell, 1995; Meiran Across this wide age span, our battery provides within- et al., 2000a,b; Meiran, 2005; Spector & Biederman, 1976; subject measures of two classic paradigms in cognitive psy- Sudevan & Taylor, 1987), including the elderly (e.g., Kramer, chology, the Simon task and task switching. In the Simon task Hahn, & Gopher, 1999; Mayr, 1996; Meimn, Gotler, & Perlman, paradigm, a non-spatial aspect of the stimulus (such as its 2001), and in various clinical groups (e.g., Aron, Sahakian, color or identity) is relevant and its spatial location is irrele- & Robbins, 2003; Brown & Marsden, 1988; Downes et al., vant. Nevertheless, the well-replicated finding in adults is that 1989; Flowers & Robertson, 1985; Hayes, Davidson, Rafal & responses are faster and more often correct when the stimu- Keele, 1998; Mecklinger, von Cramon, Springer, & Mantles- lus and response are on the same side than when they are on von Cramon, 1999; Rogers et al., 1998). However, to date, opposite sides (the Simon effect, also called spatial incompati- only a handful of studies have looked at task switching in chil- bility or stimulus—response compatibility; e.g., Craft & Simon, dren (Cepeda, Kramer, & Gonzalez de Sather, 2001; Cohen, 1970; Fitts and Seger, 1953; Hommel, 1995; Hommel, Proctor, Bixenman, Meiran, & Diamond, 2001; Crone, Bunge, Van der & Vu, 2004; Lu & Proctor, 1995; Simon & Small, 1969; Simon, Molen, & Ridderinkhof, in press; Crone, Ridderinkhof, Worm, 1990; Simon & Berbaum, 1990). This effect indicates: (a) the Somsen, & van der Molen, 2004; Reimers & Maylor, 2005; influence of an irrelevant stimulus attribute on performance and Zelazo, Craik, & Booth, 2004). (b) a prepotent tendency to respond on the same side as the Switching is fundamentally difficult and a paradigmatic stimulus (confirmed at the neuronal level [see Georgopoulos, instance of when top-down executive control is required 1994; Georgopoulos, Lurito, Petrides, Schwartz, & Massey, because generally it cannot be done "on automatic:' It taxes 19891 and with lateralized readiness potentials [Valle-Inclan, both working memory and inhibition (the newly-relevant 19961) which must be inhibited when the locations of stimu- rules and stimulus-response relations must be activated and lus and response are incompatible. It thus provides insight into the previously-relevant ones suppressed). One cannot get in the an aspect of inhibitory control. A finding that the Simon effect "groove" of repeatedly doing the same thing or staying in the decreases over a certain age range provides evidence for when same mindset because periodically one will have to change that. developmental improvement in that aspect of inhibition occurs A groove is a good analogy because it takes effort to climb over and insight into when maturational changes in the neural sys- the banks of the groove (the mindset) one is in and settle, however tem underlying that might be occurring. That neural system temporarily, into another grove. Neuroimaging studies confirm overlaps substantially with the neural system activated during that task-switching (as opposed to continuing to do the same Stroop interference and other cognitive control paradigms. It task) activates the neural system associated with executive func- includes the anterior cingulate, lateral prefrontal cortex (dorso- tion and top-down cognitive control, that is lateral prefrontal lateral and ventrolateral), pre-SMA, premotor cortex, posterior cortex (dorsolateral and ventrolateral), inferior frontal junction and superior parietal cortex, inferior temporal cortex, the insula, (IFJ) and premotor cortex, pre-SMA and the anterior cingu- and precuneus (Bush, Shin, Holmes. Rosen, & Vogt, 2003; late, and the insula and cerebellum (Brass et al., 2003; Brass, Dassonville et al., 2001; Fan, Flombaum, McCandliss, Thomas, Derrfuss,Forstmann, & von Cramon, 2005; Braver, Reynolds, & & Posner, 2003; lacoboni, Woods, & Mazziotta, 1998; Liu, Donaldson, 2003; Crone, Wendelken,Donohue, & Bunge, 2005; Banich, Jacobson, & Tanabe, 2004; Maclin Gratton, & Fabiani, DiGirolamo et al., 2001; Dove, Pollmann, Schubert, Wiggins, 2001; Peterson et al., 2002; Thomas et al., 1999; Wager & Smith, & von Cramon, 2000; Dreher & Berman, 2002; Dreher & 2003). Grafman, 2003; Kimberg, Aguirre, & D'Esposito, 2000; Meyer We investigated spatial incompatibility effects both decreas- et al., 1998; Omori et al., 1999; Pollmann, 2001; Sohn, Ursu, ing and increasing the working memory requirements tradition- Anderson, Stenger, & Carter, 2000; Sylvester et al., 2003; Wager, ally required for Simon tasks. We decreased it in one case by Reading, & Jonides, 2004). Consistent with this, patients with providing icons depicting stimuli A and B over their respective frontal cortex damage are impaired at switching between tasks response-sites so that which response goes with which stimulus (Mon,Monsell, Sahakian, & Robbins, 2004; Diedrichsen, Mayr, EFTA01098879 M.C. Davidson es nl. /Neuropsychologia 44 (2006)2037-2078 2039 Dhaliwal, Keele. & Ivry, 2000; Keele & Rafal, 2000; Owen et Finally, Diamond (1991, 2002) and others (Anderson & al., 1993; Rogers et al., 1998; Shallice & Burgess, 1991). Spellman, 1995; Gemsbacher & Faust, 1991; Hasher, Stoltzfus, We report here on the developmental progression in almost Zacks, & Rypma, 1991) have hypothesized that working mem- 300 children from 4 to 13 years of age and the performance ory and inhibition are separable functions. This is consistent of young adults for comparison, all tested on the same test with the results of the factor analyses of Miyake et al. (2000) that battery. Various manipulations exploited task switching and spa- found working memory and inhibition to be moderately corre- tial incompatibility effects, with and without an added memory lated but clearly separable. Many scholars, however, have argued component, or taxed memory without taxing inhibition or task that there is no need to postulate an inhibitory function separate switching, enabling us to test predictions concerning interre- from working memory and have produced neural network mod- lations, independence, and the developmental progressions of els consistent with that (Cohen, Dunbar, & McClelland, 1990; working memory (how much information you must hold in Kimberg & Farah, 1993; Miller & Cohen, 2001; Morton & mind and how many steps must be mentally executed using that Munakata, 2002; Munakata, 2000). Given that we hypothesized information), inhibition (resisting an incorrect response you are that working memory and inhibition are independent, we pre- inclined to make in order to make the correct response), and dicted that performance on tasks that tax primarily memory or cognitive flexibility (switching between tasks or rules). The pre- primarily inhibition would not be highly correlated, and tested dictions we tested were generated from hypotheses concerning this for relatively easy tasks and for relatively difficult tasks inhibition and working memory and hypotheses concerning cog- requiring primarily memory or primarily inhibition, matched on nitive flexibility and task switching. difficulty. 1. Hypotheses relevant to inhibition and working 2. Hypotheses relevant to cognitive flexibility and task memory switching We hypothesized that inhibition would exact a greater relative Diamond (1990, 1991, 2002) has long maintained that it is cost for young children than for older children or young adults, the conjunction of simultaneous demands on holding informa- and thus predicted that inhibitory demands would account for a tion in mind and inhibition that is truly difficult, especially if greater proportion of the variance in children's performance than one's mental settings have to be continually re-set because the in adults, and the more so the younger the child. In young adults, task changes. We thus predicted that the most difficult condition in whom inhibitory control is more mature, we hypothesized at all ages would be the one that taxes inhibition and memory in that memory demands would exact a greater cost than inhibitory a switching context, where top-down executive control is con- demands. tinually required, and that that would be even more difficult than Because we hypothesized that inhibitory control is extremely having to hold three times as much information in mind but with problematic for very young children, we predicted they would no inhibition or switching component. Further, since we hypoth- perform poorly on all trials requiring inhibition (Incongruent esized that switching is so difficult, we predicted that having to trials and switch trials) and that those effects would be addi- switch between task sets would show a long developmental pro- tive. We predicted that older children and adults would show gression even when memory demands are minimized. the same "asymmetric switch costs" (a greater relative switch Diamond has recently theorized that several seemingly inde- cost for switching to the easier [Congruent] condition) previ- pendent findings in cognitive psychology can be integrated under ously reported in adults (Allport & Wylie, 2000; Allport et al., the hypothesis that the brain and mind tend to work at a grosser 1994; De long, 1995; Kleinsorge & Heuer, 1999; Los, 19%; level, and only with effort, or more optimal functioning, work Stoffels, 1996; Wylie & Allport, 2000). Further, for slightly in a more selective manner (a theory Diamond has called "all older children, who are beginning to exercise better inhibitory or none" (Diamond, 2005, in preparation)). For example, it is control, doing so should require greater effort than in older easier to take into account all salient aspects of a stimulus than participants. Hence, we predicted that undoing that inhibition only some of its properties. Indeed, it is difficult to ignore irrele- (switching back to making a dominant response) should exact vant properties of an attended stimulus, as the Simon effect and a greater cost in those children than in adults. Thus, we pre- children's difficulties on card sorting tasks so amply demon- dicted that beginning after 6 or 7 years, asymmetric switch strate (Diamond, Carlson, & Beck, 2005; Kirkham, Cruess, & costs would be larger in younger than older participants, but Diamond, 2003). that the youngest children would show an opposite pattern of Another finding that fits under the all or none rubric is that asymmetry. it is easier to inhibit a dominant response all the time than only The ability to simply hold items in mind (without any added some of the time. One of the most demanding cognitive require- requirement to manipulate that information or exercise inhibi- ments is to switch back and forth, to overcome inertial tendencies tion) develops early, is robust even in preschoolers, and shows favoring staying in the "groove" one is in (Kirkham et al., 2003). little improvement with age (Brown, 1975; Dempster, 1985; Once in a "groove," even if it was a difficult one to settle into Diamond, 1995). Given the early maturation of the ability to (because it required resisting a tendency to act otherwise, for hold items in mind, we predicted that although it would be harder example) it is not that difficult to continue along that path. It for everyone to hold more items in mind than fewer, the relative is re-mapping stimulus—response associations, changing mind- difficulty of that would not change over age. sets, that is quite difficult (Brass et al., 2003; Fagot 1994; Los, EFTA01098880 2040 M.C. Davidson et al. /Neumpsychologia 44 (2006)2037-2078 1996, 1999; Schuch & Koch, 2003, 2004; Wannk, Hommel, & when they are presented in the context of having to periodically Allport, 2003). We thus predicted that performance at all ages switch between rules than in a block of all nonswitch trials. would be better in Incongruent-only blocks (where inhibition is Such global switch costs (the difference in performance on non- consistently required on all trials) than in Mixed blocks (where switch trials in a Mixed block versus in a single-task block; Fagot inhibition is only required on the 50% of trials that are Incon- 1994, Mar, 2000) have been shown to be greater for elders than gruent), and that this difference would be greater the younger for younger adults (Kray, Eber, & Lindenberger, 2004; Kray & the children. This might seem obvious, but most studies of the Lindenberger, 2000; Mayr, 2000; van Asselen & Ridderinkhof, classic Stroop effect still tend to administer the conditions in 2000) and higher for children than for young adults (Cepeda single-task blocks (read all the words or state the ink color of et al., 2001; Cohen et al., 2001; Reimers & Maylor, 2005), all the words), missing the most difficult condition (switching though this has not been investigated in children as young as the between having to read the words and having to state the ink youngest tested here and though some studies have not found an color). age difference in global switch costs (Crone et al., in press; Kray, A further seemingly independent finding that provides Li, & Lindenberger, 2002). We predicted that global switch costs another example of the all or none principle is that it is eas- would not only be found in our youngest participants but would ier to switch everything, or nothing, than to switch one thing be more exaggerated the younger the child. (e.g., the rule or the response) but not the other (Hommel et al., Because of floor effects (subjects should already be slower 2001; Kleinsorge, 1999; Meiran, 2000a,b; Rogers & Monsell, and more error-prone in the Incongruent-only block), the effect 1995; Schuch & Koch, 2004). Similarly, if you are supposed to of context (the Mixed block versus single-task block) should press the color opposite to a stimulus it is easier to also press the be greater on Congruent than Incongruent trials. We predicted button on the side opposite to the stimulus (rather than the typ- that this would be more evident the younger the child. Thus, ical bias to respond on the same side as the stimulus; Hedge & performance on "easy" (Congruent, nonswitch) trials should fall Marsh, 1975). Issuing a global "change" or "opposite" command closer and closer to the level of "harder" trials in the context of to all systems appears to be preferred by our neural machinery sometimes having to switch back and forth the younger the child. over a more selective command to just the action system or to just one aspect of cognition. This has been demonstrated not 3. Methods only in young adults, but also in older adults (Mayr, 2001) and children (Crone et al., in press). We predicted that we would 3.1. Participants demonstrate these effects, heretofore documented only in adults A total of 325 individuals participated. ranging in age from 4 to 45 years. and older children, even in young children. Thus, we predicted Of these. I 1 children were excluded from the analyses for failing to press any that throughout our age span, participants would do better at button or consistently pressing both. Of the remaining 314 participants. 50% switching tasks if the response-site also changed and would be were female (157 female. 157 male). Table 1 shows the number and gender of slower and less accurate on switch trials when the response-site participants in each of the age groups. Children were recruited through local preschool and elementary school programs in the suburban Boston area. Adults remained the same as on the previous trial. Another way of putting some of the above points is that con- Table 1 text matters. For example, even "easy" trials do not seem so Number of participants within each age and gender group easy when they are presented in the context of switching between those and "harder trials. Knowing that sometimes you will have Age group' Mean age S.U. N Gender (years) (years) to switch can cause you to slow down (and perhaps err more) on Female Male trials where you do not have to switch. Local context matters; it 4 4.43 0.25 30 14 16 matters what trials came before a particular trial. For example, 5 5.19 0.17 30 14 16 was the rule on the preceding trial the same as on the current 6h 6.01 0.40 30 15 15 trial? Performance is better on nonswitch than on switch tri- 6h 6.22 0.35 30 12 IS als. Was the response-site on the preceding trial the same as on 7 7.12 0.20 30 13 17 8 7.97 0.28 30 10 10 the current trial? Studies in adults have shown that performance 9 9.07 0.30 30 17 13 is better on nonswitch, same-response-site trials than on non- 10 9.92 0.30 30 13 17 switch, response-site-switch trials and on rule-switch, response- II 11.01 0.32 28 II 17 site-switch trials than on rule-switch, same-response-site trials. 13 12.89 1.21 26 17 9 We predicted a different pattern in the youngest children and a Adults 26.30 5.40 20 14 6 more exaggerated version of the adult pattern in slightly older Total number of 314 157 157 children (see above). participants Global context also matters; it matters what kind of trial block 4 The age groups were used for illustrative purposes when preparing graphs. a given trial occurs in. Performance on the same type of trial All regression analyses used the actual ages of participants and treated age as a (e.g., Congruent, Incongruent) in the same type of local con- continuous variable. b TWo groups of 6-year-old children were tested to study the effects of short text (e.g., nonswitch) varies depending on the larger context vs. long presentation time at this intermediate age. For one group. stimulus (e.g., a single-task block or Mixed block). Performance even on presentation time was 2501ms. the slower version used with younger children. "easy" nonswitch trials (where the rule on the present trial is the For the second group. stimulus presentation time was 750 nu, the faster version same as on the previous trial) is usually slower and less accurate used with older children and adults. EFTA01098881 ACC. Davidson et al. /Neumpsychologia 44 (2006) 2037-2078 2011 were recruited from within the Eunice Kennedy Shrivcr Center in Waltham. MA. pointing down like this IE demonstrated] to this button. press this button. If the The majority of participants were Caucasian and from middle to upper middle arrow is on this side. pointing down across the screen like this IE demonstrated] class families. Informed consent was obtained from all adult participants and to this button. press this button. If the arrow is on the other side. pointing down from a parent of each child participant: assent was obtained from the younger across the screen like this IE demonstrated] to this button. press this button:' children and consent from the olderones. All participants received a small, token Congruent and Incongruent trials were presented in a randomized Mixed block present for their participation. of 20 trials. This requires inhibiting the tendency to respond on the same side as the stimulus when a diagonal arrow appears. but it requires little or no working memory. as the arrow points directly to the correct response button on all trials. 3.2. Procedures common to all tests in our batten• 3.3.3. Dols All tasks were presented on a Macintosh computer using MacStim to The Dots test was designed to tax both working memory and inhibition. present the stimuli and record responses. Participants held a button box while the other tests were designed to tax primarily either working memory (10cm x 14 cm x 3 cm) with both hands and used their thumbs to press the two or inhibition, not both. Here, a large dot (diameter = 2_5 cm). was presented response buttons. For each task a horizontal rectangle (6 cm x 18 cm) with a cen- either at the left or right on each trial (sec Fig. I). Two types of Dots (striped tral fixation cross was presented on the computer screen (25cm x 33 cm). Only or solid) were used. Striped Dots contained vertical black and white stripes. one stimulus was presented per trial and participants were positioned approxi- while solid Dots were a uniform gray color. These Dots were equated for visual mately 50cm from the screen. characteristics such as size and luminance. For half of the participants a striped Participants completed a set of four related tests designed to manipulate dot indicated they should make a response on the same side as the dot while demands on working memory and inhibitor)• control (see Fig. I ). For adults and a gray dot indicated they should respond on the side opposite the dot. These older children (>7 years). stimulus presentation time was 750ms. For younger rules were reversed for the other half of the participants. An initial block of 20 children (4-6 years). stimulus presentation time was 2500ms. In all cases the Congruent trials (with all responses on the same side as the dot) was followed interstimulus interval was 500 ms. resulting in total trial durations of 1250 and by a block of 20 Incongruent trials (with all responses on the side opposite the 3000 ms. respectively. An additional group of 6-year-oldchildren was tested with dot), and then by a Mixed block of 20 trials where Congruent and opposite the short (adult) presentation time (750ms) to study the effects of presentation trials were randomly intermixed. Instructions and practice were given before time at this intermediate age. the Congruent and Incongruent blocks. Instructions alone were given before the Each task began with condition-specific instructions and a short practice Mixed block. e.g.. "Remember. gray same side: striped opposite: Memory is block consisting of four or six trials. Different numbers of trials were used to required on all trials of the Dots test to remember the rules (respond on the same allow presentation of all relevant trial types within each practice block. Partic- or opposite side as the dot I. Inhibition is required on Incongruent trials to inhibit ipants could repeat the practice trials if needed to demonstrate learning of the the prepotent response to respond on the same side as the visual stimulus. This requirements for a given task. Most children learned the task requirements with task is similar to one used by Shaffer (1965) though there each subject received one practice block and no participant needed more than two practice blocks. only one type of trial block (Congruent. Incongruent. or Mixed) and therefore The criterion for demonstrating learning was 75-80% correct on the practice subjects did not have the benefit of testing with the two easier trial blocks before trials and to be able to verbally tell the experimenter the rules. Testing blocks receiving the Mixed block. The Dots task is also similar to a task used by Vu and contained 20 trials and each participant completed 1 block for each condition Proctor (2004) but the rules for their single-task blocks did not refer to stimulus of each task. except for the 2 conditions of the Abstract Shapes task. each of appearance and so the memory demand in their Mixed condition might have which contained 2 blocks (with a shoo break in between) for a total of 40 trials been greater than in ours. for each condition. The set of tests was administered with Arrows first. then Dots. Abstract Shapes (two then six shapes). and finally Pictures. A subset of 3.3.4. Abstract Shapes participants were tested with Arrows presented last and Pictures presented first In the Abstract Shapes test. unlike all other tests.each stimulus was presented to check for order effects. but this did not affect performance. so results for both in the center of the rectangle. Participants were taught a rule for each stimulus orders of presentation are collapsed together in the results reported here. ("for this one press the left button": "for this one press right") during short prac- tice blocks before each testing condition. There were two conditions involving 3.3. Procedures specifie le individual tests two- or six-Abstract-Shapes. Participants first completed the two-shapes condi- tion (2 blocks of 20 trials) and were then taught 4 additional rules. for a total 3.3.1. Pictures of 6 shapes. and were then tested on another two blocks of 20 trials. The six- This test is a classic Simon task. Here, a color picture of either a frog or Abstract-Shapes condition taxes memory heavily (participants must hold six butterfly was presented on the left or right side of the computer screen(see Fig. I). rules in mind). but it requires little or no inhibition (as the stimuli appear at the Each stimulus had an associated right or left response. The exact instructions center of the screen and do not preferentially activate the right or left hand). given participants were: "If you sec a butterfly, press the button on the left. whether the butterfly appears on the left or right: if you see a frog. press the 4. Results: general comments button on to the right. whether the frog appears on the left of right:' Small versions of the stimuli were attached next to the correct buttons on the response box to minimize the need to remember which stimulus was associated with which The three dependent measures were percentage of correct button. The stimuli were presented randomly on the left or right of the screen responses (accuracy), speed (reaction time MTH, and percent- over the block of 20 trials, yielding Congruent (compatible) and Incongruent age of anticipatory responses (AR). Linear regressions were used (incompatible) trials. for all analyses involving age and each subject's exact age was entered, not simply the person's age grouping. Within-subject 3.3.2. Arrows ANOVAs were used for analyses comparing tasks, conditions Here. a single large arrow was presented at the left or right of the computer screen. The arrow pointed either straight down (toward the response button on within task, or trial types. All binary comparisons included the same side as the arrow) or toward the opposite side at a 45' angle (toward the Tukey corrections for multiple comparisons. Whenever the vari- response button on the opposite side: see Fig. 0.0n Congruent trials. the arrow ance structure didnot conform to the requirements for parametric pointed straight down and participants were to respond on the same side as the analyses, logarithmic or arc sine transformations of the data were arrow. On Incongruent trials. the arrow pointed diagonally toward the opposite used to obtain the required conformity. All tables and figures side and participants were to respond on the side opposite the arrow. The precise instructions participants were told were. "1want you to push the button the arrow present the raw, untransfonned data. is pointing toward. If the arrow is on the side of the box pointing down like this A response time faster than 200 ms was considered antic- IE demonstrated] to this button. press this button.If the arrow is on the other side ipatory (too fast to be in response to the stimulus). Those EFTA01098882 2012 MC. Davidson et at /Neumpsychologia 44 (2006)2037-2078 CONGRUENT TRIALS INCONGRUENT TRIALS DOTS TEST: A SPATIAL INCOMPATIBILITY TASK with ARBITRARY STIMULI ABSTRACT SHAPES TEST: • A MEMORY LOAD TASK Press Left Press Right • Press Left Press Right Press Left ARROWS TEST: A SPATIAL INCOMPATIBILITY TASK Press Right with ICONIC STIMULI Press Right Press -elf Press Left 4] Press Right Press Right Press Right PICTURES TEST: A SPATIAL INCOMPATIBILITY TASK that is a CLASSICAL SIMON TASK Press Left 4 e Press Left Press Right Press Left 4 Press Rignt Press Left INHIBITORY CONTROL DEMAND MEMORY LOAD Low Medium High Low Dots-Congruent Dots-Incongruent. Arrows Pictures Medium 2-Abstract-Shapes Dots-Mixed These 2 cells are logically possible. e.g . 6-Abstract Side (a High 6-Abstract-Shapes 'Simon" task with 3 stimuli per response button) would be High Memory! Medium Inhib. However such tasks are too difficult Fig. I. Illustration of the tasks in our battery with a table summarizing the demands of each on memory and inhibition. EFTA01098883 M.C. Davidson et at. /Neuroptychologia 44 (2006)2037-2078 2043 responses were excluded from analyses of accuracy or speed, accuracy (RT: F(1,88)=4.58, p < 0.04; anticipatory responses: but were included in analyses of anticipatory responses (ARs). F(1,88) = 6.07, p <0.02). ARs occurred when a participant was either too eager and failed to wait for the stimulus on the current trial or failed to 5.2. Arrows release the button following the previous trial. These anticipa- tory responses indicate inhibitory failures and are reported as a The Arrows test was designed to require inhibitory control percentage of all possible responses where appropriate. A trial when a response was required on the side opposite the stimulus was considered correct if: (a) the first response following a stim- but to require little or no working memory as the stimuli them• ulus was correct and (b) RT was >200 ms following stimulus selves point to the correct response button. Performance was onset. better as a function of age, with increased accuracy, increased The percentage of correct responses was calculated by divid- speed, and reduced anticipatory responses (Table 2). This was ing the number of correct responses by the sum of correct highly significant for accuracy and anticipatory responses when plus incorrect responses. Anticipatory responses were excluded all ages were included in the analyses but not significant for from that calculation. The median RI' for correct responses speed of responding (accuracy: F(1,312)=57.06; p<0.0001; only was calculated for each participant. The median value, AR: F(1,312) = 35.73, p <0.0001). When the youngest children, rather than the mean value, was used to reduce the effect of tested with a long presentation time, were removed from the outlying RI's. analyses, the age-related improvements in speed, as well as accu- The youngest children received a slower version of our tasks racy and reduced incidence of anticipatory responses, were sig- than the rest of the children and adults. The stimuli were pre- nificant at p <0.0001 (F(1,222) = 76.88 [%correct]; 36.07 MTh sented to the 4- and 5-year-olds and one group of 6-year-olds for 38.56 [AM). The youngest children (4-6 years of age) showed much longer than they were presented to the rest of the children a steady reduction in anticipatory responses, and 6-year-olds and adults (trial durations of 3000 and 1250 ms, respectively). responded correctly significantly more often than children of 4 Analyses over all ages might exaggerate RT differences over or 5 years, but there was no difference over the age range of 4-6 age (since children given longer to respond will naturally take years in response speed (accuracy: F( 1,88) = 10.69; p <0.005; longer) and might underestimate accuracy differences (since AR: F(I,88) = 6.5, p <0.02). children given longer to respond are likely to make fewer errors). Hence, analyses of age differences are reported separately for 5.3. Dots the youngest children tested with a presentation time of 2500 ms and for all other participants tested with a presentation time of In the Dots test there were three conditions (Congru- 750 ms. ent, Incongruent, and Mixed). Performance in each condition The effects of gender, and interactions of gender with age, improved significantly as a function of age, with increased accu- were tested in all analyses. Significant effects were not found. racy and speed, and reduced anticipatory responses the older the Independent age-related regressions for male and female partic- participants (see Fig. 2). Unless othenvise noted, all results in the ipants showed comparable R2 values across the three dependent next three paragraphs for improvement over age are significant measures for all tests. at p < 0.0001. For the Congruent condition, performance improved over age 5. Results: basic level results for the tasks that included in the percentage of correct responses, RT. and reduced antici- an inhibitory component (Pictures, Arrows, and Dots) patory responses (F(1,312) = 34.68. 116.97, and 8.42 (p < 0.05 for AR), respectively with all subjects in the analyses). The 5.1. Pictures corresponding results for only those tested with the 750-ms stimulus presentation time (6-year-olds through adults) are The Pictures test was designed to provide a measure of the F(1,222) = 14.33 (p < 0.001), 55.05, and 2.59 (NS for AR). The Simon effect in children. It tests the effect of an inhibitory correspondingresults for only those tested with the 2500-ms pre- demand (resisting the impulse to respond on the same side as sentation time (children of 4-6 years) are F(1,88) = 18.19 and the stimulus) with little or no working memory demand since 8.54 (p <0.005), and 18.52. small icons were attached above the appropriate response keys For the Incongruent condition, with all subjects included, to indicate the correct response for each stimulus. The older the performance improved over age in accuracy (F(1,312)=46.60), subjects, the better their performance (see Table 2). This was speed (F(1,312) = 110.76), and reduced anticipatory responses highly significant when all ages were included in the analyses (F(1,312)=39.77). The corresponding results for those ≥6 (p <0.0001 for each of the three dependent variables) and for years of age are F(1,222)=33.09, 47.21, and 24.33. The ages 6 years through adults tested with the briefpresentation time corresponding results for those 4-6 years of age are (accuracy: F(1,222)= 17.93, p < 0.0001; RT: F(1,222) = 35.36, F(1,88) = 7.76 (p < 0.005), <I (NS), and 15.07. p <0.0001; anticipatory responses: F(1,222) = 10.8, p <0.001), For the Mixed condition, the results for improvement the effect of age being particularly marked on speed of respond- over age with all subjects included in the analyses are ing. The youngest children (4-6 years of age) improved in speed F(1,312)=66.65 (%correct), 62.15 (RT), and 42.84 (AR). and reduced anticipatory responses on the task over age, but For only those ≥6 years of age, the corresponding results are given a long time to respond showed no difference over age in F(1,222)=61.95, 10.68, and 31.51. For only those 4-6 years of EFTA01098884 2044 MC. Davidson et at. /Neumpsychologia 44 (2006)2037-2078 Table 2 Table of means for each of the task conditions by age of the participants Task condition Age in years Average Tukey results 4 5 6 6 7 8 9 10 11 13 26 Accuracy (percentage of correct responses) Pictures 91.67 87.59 93.92 88.07 85.26 88.03 86.14 91.45 91.08 92.25 100.00 98.23 A Arrows 83.36 79.11 90.22 77.87 73.67 77.97 80.22 78.38 84.63 88.19 95.19 90.06 B Dots Congruent 96.33 94.96 98.61 96.00 96.86 99.65 99.30 99.67 99.62 99.55 98.50 91.47 B Incongruent 86.93 86.12 92.62 88.05 88.77 90.46 89.51 93.83 96.38 95.02 89.73 76.37 D Mixed 68.57 68.03 77.37 71.24 71.97 73.70 74.94 76.96 81.71 85.81 96.22 82.71 C Abstract Shapes Two-shapes 88.95 88.96 90.34 87.69 84.82 88.00 87.38 88.65 89.91 94.95 96.80 89.68 B Six-shapes 76.87 77.16 87.39 73.29 72.60 73.97 81.34 78.42 79.16 86.46 89.92 79.69 C Average 84.67 83.13 90.07 83.17 81.99 84.54 85.55 86.77 88.93 91.75 95.19 Reaction time (in ms) Pictures 1037.48 952.67 881.20 665.55 602.25 563.40 513.95 523.65 471.02 473.79 422.08 646.09 C Arrows 1121.28 1150.60 1090.42 797.73 725.57 683.07 613.12 651.77 578.04 555.46 465.25 766.57 BE Dots Congruent 775.37 684.58 677.37 474.53 412.12 395.05 356.87 341.13 331.46 323.87 271.30 458.51 A Incongruent 1023.12 905.75 875.02 619.87 546.27 501.63 444.02 451.83 398.09 402.87 321.28 589.98 CD Mixed 1172.32 1195.47 1177.00 787.10 728.18 725.98 644.72 654.15 597.36 593.85 562.98 803.55 E Abstract Shapes Two-shapes 892.80 853.88 795.17 608.15 552.03 520.58 478.38 463.13 436.23 434.56 371.40 582.39 D Six-shapes 1121.20 1038.10 987.53 726.72 694.15 662.98 640.55 61233 592.23 568.29 532.93 743.38 B Average 1020.51 968.72 926.24 668.52 608.65 578.96 527.37 528.31 486.35 478.95 421.03 Percentage of anticipatory responses Pictures 12.17 12.17 6.67 6.83 8.33 4.00 2.17 0.67 1.79 0.77 0.50 5.10 E Arrows 19.33 17.00 9.00 21.33 23.17 18.67 10.83 9.00 6.96 2.31 0.50 12.56 B Dots Congruent 13.83 10.67 2.67 6.33 7.17 6.83 5.50 6.83 6.25 5.38 2.75 6.75 A Incongruent 21.17 15.33 8.33 11.50 10.33 8.33 8.83 5.50 2.68 3.85 0.25 8.74 A Mixed 28.67 21.00 15.00 23.00 25.50 26.67 16.17 11.33 5.18 4.42 3.50 16.40 D Abstract Shapes Two-shapes 18.00 11.50 7.75 8.42 7.08 6.58 3.67 4.42 2.50 1.35 0.88 6.56 AE Six-shapes 18.00 17.83 11.58 15.33 15.33 13.83 6.08 5.83 4.73 2.98 2.75 10.39 BC Average 18.74 15.07 8.71 13.25 13.85 12.13 7.61 6.23 4.30 3.01 1.59 age, the corresponding results are F(1,88)= 6.24 (p< 0.05), <1 mance. This effect was present at all ages and particularly (NS), and 11.21(p< 0.005). pronounced in the younger children (t(89)=5.35 [accuracy], When the stimuli were presented for only 750 ms, 6-year-olds 4.49 IRT], both p<0.0001; ARs, NS). It was present, though performed at a level of accuracy roughly comparable to that of smaller, in older children and adults 0(223) = 8.55 [accuracy[, 4-5-year-old children shown each stimulus for 2500 ms. While 10.41 1RT1, both p <0.0001 ARs, NS) decreasing from the children of 4 or 5 years could perform well in the single-task age of 6 years onward (accuracy: F(1,222)=7.46, p < 0.01; blocks, even the Incongruent one, their average accuracy dipped speed: F(1,222) = 5.23, p<0.02; see Fig. 3). Children of 4-6 below 70% in the Mixed block, even on Congruent trials. At the years, allowed a long time to respond, showed no change in fast stimulus presentation rate (750ms), it was not until the age the absolute size of the effect over age. However taking into of II years that children began responding at >80% correct on account their baseline speed on Congruent trials, the percent- average in the Mixed block. Even our oldest children (13 years age increase in RT on Incongruent trials decreased signifi- old) were not yet correct on 90% of the items in the Mixed block. cantly over these ages (children 4-6 years old: t(89)=4.23, p <0.0001). 6. Results: spatial compatibility effects Inhibition was required on only half the trials in the Pictures task (the Incongruent ones). Although children of4-5 years were 6.1. Spatial compatibility effects: Pictures task able to perform correctly on 90% of the Congruent trials, they were correct on only 80% of the Incongruent trials. Only the The Pictures test contained two intermixed trial types. Con- older subjects, and the 6-year-olds given a long time to respond, gruent and Incongruent, with spatial incompatibility present were able to perform at ≥85% on Incongruent trials in the Pic- on the Incongruent trials. Participants made fewer errors tures task (88%, 88%, 89%, 94%, and 85%, at ages 10,11 and 13 and responded faster on Congruent than Incongruent trials years, young adult, and 6 years allowed a long time to respond, (0[313[=10.1 [accuracy], 8.38 [RTI, both p< 0.0001; antici- respectively). Accuracy at ages 6-9 years, given a short time patory responses NS; see Fig. 3). These comparisons indicate to respond, was comparable to that seen at 4-5 years with the that the presence of spatial incompatibility affected perfor- longer response window. EFTA01098885 M.C. Davidson et al. /Neap rhologia 44 (2006)2037-2078 100 es • — •• • & Congruent i incongnani t• - t • a. 9 90 I Mixed Percent Correct 80 70 60 (A) 4 5 6 6 7 8 9 10 I 13 20 1400 1200 11000 .5 800 iE 800 • ▪ --a. F. IS 400 cc▪ 200 j8) 0 4 6 7 8 9 10 11 13 26 40 Anticipatory Responses Errors %Anticipatory Responses 30 20 10 • CI • • 0 6 8 7 8 9 10 11 13 StimuS presented for 2500 ms Stimuli presented for 750 ms (C) Age In Years Fig. 2. Dots conditions: (A) accuracy. (B) reaction time and (C) anticipatory response errors. 6.2. Spatial compatibility effects: Arrows task p <0.001) on Congruent than Incongruent trials (showing inter- ference similar to the Simon effect). Similarly, participants 6 The Arrows test also presented Congruent and Incongruent years and older, tested with the 750-ms presentation time, were trials randomly intermixed. The youngest participants (4-6 years also more accurate and faster on Congruent than Incongru• of age, tested with the 2500.ms presentation time) were both ent trials (accuracy: 4223] = 8.76, p <0.0001; RT: 42231=7.91, more accurate (4891=7.25. p <0.0001) and faster 01891=3.44, p <0.0001). Among those ≥6 years, the difference in accuracy EFTA01098886 2046 MC. Davidson es at. /Neumps)rhologia 44 (2006)2037-2078 16 g ° 14 167 0 12 C • = E 10 0 c o. 8 ,c E 2 5 o 2 4 O go 0 2 (A) 4 5 6 6 7 B 9 10 11 13 26 f t 140 •1) 6. 6. 125 .S 80 •O c ai 65 36 (B) 4 6 b 7 6 9 10 I I 13 25 31. • y Ot 20 CAGE • Th .t- 1=CC o 2 15 ren • o 10 o o 2a 5 ix 3 cco €U 0 11 26 2 4 5 6 6 7 8 9 10 13 it I Staub presented for 2500 ms Staub presented for 750 ms (c) Age in Years Fig. 3. Simon effect on the Pictures task. (A) Difference in percent correct: Congruent minus Incongruent trials. (B)difference in reaction time: Incongruent minus Congruent trials and (C) percentage change in reaction time: (reaction time on Incongruent minus Congruent trials) divided by reaction time on Congruent trials. (but not speed) on Congruent versus Incongruent trials decreased p <0.01 (subjects ≥6 years old); NS for the youngest children. as a function of age (accuracy: F(1,222) = 13.51, p <0.0003). This effect of spatial incompatibility on speed did not change significantly over age. There was no significant effect of spatial 6.3. Spatial compatibility effects: Dots task incompatibility for accuracy or anticipatory responses on this task. There was a significant spatial incompatibility effect in the Mixed condition of the Dots task (where Congruent and Incon- 7. Discussion: compatibility effects gruent trials were again randomly intermixed). Participants were significantly faster on Congruent (spatially compatible) Based on our hypothesis that even very young children trials than on Incongruent (spatially incompatible) trials: can perform well when inhibition alone is taxed, we predicted $223) = 2.09, p < 0.04 (all subjects included); t(217)=2.49, they would perform well even on the Incongruent trials of EFTA01098887 ALC. Davidson a at. /Neuropsychologia 44 (2006)2037-2078 2047 the Pictures task, where memory demands were minimized. significant throughout, beginning at age 6, but did not change Since we hypothesized that inhibitory control shows a long over age and was not significant for accuracy. The lack of an developmental progression we predicted that the spatial accuracy cost on spatially incompatible (Incongruent) trials in incompatibility effect would decrease in size with age over an the Mixed block of the Dots task is in sharp contrast to the results extended period, despite some findings in the literature to the when comparing separate blocks of Congruent and Incongruent contrary. For example, Band, van der Molen, Overtoom, and trials on the task (see below where results for the different con- Verbaten (2000), using auditory stimuli and including neutral ditions of the Dots task are presented and discussed). trials as well as compatible and incompatible ones, found an inverse relation between the size of the Simon effect and age. 8. Results: local switch costs The effect on response speed was smaller in 5-year-olds than in subjects of 8, I I, and 21 years and the effect on accuracy 8.1. Local switch costs: Arrows task was smaller in children of 5 and 8 years than in the two older groups. They did, however, find that the effect of the stimulus's The Arrows test contained nonswitch and switch trials, irrelevant spatial location persisted longer for the younger depending on whether the rule on the present trial was the children. On the other hand, consistent with our prediction of a same as on the previous trial (nonswitch trials) or different reduced compatibility effect over age, Gerardi-Coulton (2000) (switch trials). The difference between performance on non- found evidence that even 2-year-old children show a propensity switch and switch trials administered in the same block is to respond on the same side as the stimulus, with the size of the known as the "local switch cost." Subjects were faster and more effect seeming to decrease over the next 6-12 months. Because accurate on nonswitch trials relative to switch trials (all sub- of problems with working with children so young, however, jects: 43131= 8.54 [%correct] and 8.33 [12'n: subjects <6 years: most of the 24-month-olds in that study did not provide useable 489] = 1.36, NS [%correct] and 5.92 [RT]; subjects ≥6 years: data, and the few who did may not have been representative. 42231= 3.91 [%correct] and 9.80 [RT]; all p <0.0001 except the The youngest children we tested (4-year-olds) showed evi- one place noted; no differences in AR). dence of being able to inhibit a dominant response. Certainly Among subjects 6 years old through young adults, tested they performed significantly better than chance even on Incon- with the briefer 750 ms stimulus presentation time, the accu- gruent trials on the Pictures test where memory demands were racy cost of switching showed a marked quadratic trend, with the minimized. Despite that, they still performed significantly bet- inverted U-shape peaking for accuracy switch cost at 9-10 years ter on Congruent than Incongruent trials. The Simon effect (F(1,222)=5.65, p < 0.02; see Fig. 4). The youngest children (faster and more accurate responses on spatially compatible than (4-6 years), given a longer time to respond (2500 ms stimu- incompatible trials) was evident on the Pictures task at all ages. lus presentation time), showed a significantly smaller switch However, age differences in the Simon effect (the cost of inhibit- cost than did the older children of 6-13 years given less time ing the pull to respond on the same side as the stimulus) on the to respond (F(1,292)=9.39, p <0.003). The youngest children Pictures task provide evidence that exercising this inhibition was achieved that small accuracy cost by using their allotted time to disproportionately harder for younger children. Consistent with slow down on the harder trials (i.e., the switch trials), and their our prediction, the Simon effect showed a decrease in size from RT switch costs were over twice those at any age from 6 years 6 years of age onward and a possible decrease in size between through young adults (F(1,312) = 16.52, p <0.0001). 4 and 6 years of age. We also looked at spatial incompatibility effects in the context 8.2. Local switch costs: Dots task of higher-order rules and different memory loads in the Arrows and Dots tests, where the rules were more abstract, no icons were Performance in the Mixed block of the Dots task was sig- provided to remind subjects of the stimulus—response mappings, nificantly slower and less accurate on switch than nonswitch and where both the identity and the spatial location of the stimu- trials. This local switch cost was significant for both accu- lus were relevant to determining the correct response. Although racy and speed (all subjects: t(3131=8.94, p<0.0001 [%cor- the rules for the Arrows task were more abstract, the memory rectl; 8.56, p < 0.0001 [RTI; subjects ≥6 years: 42221=9.27, demands were minimal because subjects needed only to look p <0.0001 [%correct]; 9.02, p <0.0001 MTh subjects of 4-6 at the stimulus to see where to respond. On the Dots task, the years: t[89] = 2.36, p <0.03 [%correct]; 4.31, p < 0.0001 [RT]; abstract rules were arbitrary and memory demands were greater. see Fig. 5). The local switch cost was evident on both Congruent On the Dots task, and to a lesser extent on the Arrows test, the and Incongruent trials in both accuracy and speed (Congruent tri- rules had to be instantiated on each trial by mentally integrating als: 43131= 8.22, p < 0.0001 [accuracy]; 6.60, p < 0.0001 [RT]; the rule for the appearance of the stimulus with the location of Incongruent trials: 43131=4.41, p < 0.0001 [accuracy]; 5.76, the stimulus (e.g., "since the dot is striped, I should press on the p <0.0001 [RT1). opposite side, and since the dot is on the left that means I should The magnitude of the local switch cost on accuracy in press on the right"). the Dots-Mixed condition was greatest at 6-13 years of age On the Arrows test, the spatial incompatibility effect in both and showed little change over that age range. The accuracy speed and accuracy was significant throughout our age range and switch cost at 6-13 years was greater than that for adults decreased from age 6 onward in accuracy but not in speed. On (F( I ,222) = 6.33,p < 0.01) and greater than that for the youngest the harder Dots test, the spatial incompatibility effect on RT was children (4-6 years old: F(1,292) = 9.39,p <0.003). Children of EFTA01098888 2048 MC Davidson a at /Neumps)rhologia 44 (2006)2037-2078 25 su is u 5 0 (a) 4 8 9 10 1.1 13 26 300 250 ‘1 c 200 s i= g 150 Z 1C c 100 0c my ss O 0 4 5 6 6 7 8 9 10 11 13 26 (b) Age In Years Fig. 4. Local switch costs on the Arrows task. (a) Local switch costs in Accuracy and (b) local switch costs in reaction time. 4 years and children of 7-8 years performed near chance on ciently tight for them that they had little room to show differential switch trials; the 4-year-olds showed a smaller accuracy switch RTs. cost because they also made many errors on nonswitch trials. As on the Arrows test, but to a lesser extent, the youngest 9. Discussion: local switch costs children benefited from the long time allotted to them for prepar- ing their responses and their RT switch costs were larger than As expected, performance was slower and less accurate on those for older children and adults (F(1,312) = 5.98, p <0.02). switch than nonswitch trials in both the Arrows task and the The difference in speed of responding on switch and non- Dots-Mixed condition. For both Arrows and Dots-Mixed, local switch trials tended to be smallest among subjects 6—8 years switch costs in accuracy were smaller in adults than in children of age, presumably because the response window was suffi- 6-13 years of age tested under the same conditions as adults. EFTA01098889 M.C. Davidson et al. /Neurops)rhologia 44 (2(X6)2037-207N 2049 -.- Rule Repeats (Nonswitch Trials) 100 1400 - 8-Rule Switches (Switch Trials) 1200 80 t •O 1000 Percent Correct 60 800 CE i= C)O 600 40 <4 Ct 400 20 200 0 0 4 5 8 8 7 8 9 10 11 13 26 4 5 7 9 9 10 11 13 20 Age in Years Fig. 5. Difference between switch and nonswitch trials in the Mixed block of Dots task. (A) Percent correct and (B) reaction time. Local switch costs on speed of responding, on the other hand, the studies that have used unpredictable switches report greater showed no differences between children of6-13 years and adults local switch costs in children than adults. Cohen et al. (2001) and remained fairly constant from 6 years through young adult- found greater local switch costs in accuracy, but not in RT, in hood on both Arrows and Dots-Mixed. children 5-11 years of age compared to adults using Meiran's Children of 4-6 years also showed smaller local switch costs task-switching paradigm adapted for children. Crone et al. (in in accuracy than did children of 6-13 years on both the Arrows press), using a paradigm similar to our Dots test, found that local task and the Dots-Mixed condition. Presumably children of 4-6 switch costs decreased with age from 8 to I I to 23 years. Cepeda years were able to use the considerable time allowed for them et al. (2001), who studied subjects aged 7 through 82 years of to respond to slow down on switch trials to preserve their accu- age, asking them the number of the digits displayed or the value racy. Their local switch costs in RT were greater than those for of the digits, report larger local switch costs for both young chil- participants at any older age, especially on the Arrows test. dren and older adults than for young adults. Similar results are Studies where the switches between tasks are unpredictable reported by Kray et al. (2004). have tended to find larger local switch costs in older ver- There are two differences between our results and those of sus younger adults, in contrast to the lack of difference in most studies. First, most studies find little or no difference in global switch costs over age (Kray et al., 2002; van Asselen & local switch costs in accuracy; the differences they find are Ridderinkhof, 2000). Studies with predictable switches, on the in RT. We found only accuracy differences between children other hand, have generally found that local switch costs either and adults and no RT differences. Second, we found that local did not change over age or are smaller in older adults, in con- switch costs in accuracy were greater and local switch costs trast to the larger global RT switch costs found in those studies in RT were smaller among children in our age range approxi- (Kray & Lindenberger, 2000; Mayr & Kliegl, 2000a,b; Mayr & mating the ages included in other studies (6-13 years of age) Liebscher, 2001; Salthouse, Firstoe, Lineweaver, & Coon, 1995; than in younger children rarely investigated previously in task- Verhaeghen & Salthouse, 1997; Verhaeghen & De Meersman, switching studies. Two differences in our design may account 1998). It is not because older adults are performing well that for our relatively large accuracy differences and small RI' dif- they show smaller, or equivalent, local switch costs compared ferences. One is the size of the window provided for subjects to young adults in predictable-switch studies. It is because their to compute their responses. Children find task-switching harder RT is elevated across the board in the Mixed block (on both than adults. When given a large enough response window so nonswitch and switch trials) that they show no further dispro- they can slow down on switch trials, and when that window does portionate increase in RT on switch trials. not exceed young children's ability to inhibit responding suffi- Results comparing children and young adults mirror those ciently long to compute the correct answer, children show larger comparing older versus younger adults. The one study that used RT switch costs than adults. When given a narrower response predictable switches found that local switch costs remained sta- window, or the time needed to compute the answer is longer ble from age 10 through middle adulthood, though global switch than young children are willing to delay their response, children costs were larger in children (Reimers & Maylor, 2005). All show larger switch costs in accuracy than adults. EFTA01098890 2050 MC. Davidson et at /Neumpsychologia 44 (2000 2037-2078 Two, unlike the vast majority of task switching studies, our ent and Congruent: 4313] = 18.51 [accuracy], 18.06 [RT], and stimuli were "univalent." Each stimulus was unique to a task or 5.45 [AR], all p <0.0001; see Fig. 2). rule. No stimulus had one meaning for the Congruent rule and Performance in the Incongruent and Mixed conditions can a different meaning for the Incongruent rule; different stimuli also be viewed as a percentage change from performance in the were used for the different rule sets. Meiran's model predicts, Congruent condition ((I or M minus C] divided by C), thus tak- and Meiran reports results showing, that switch costs largely dis- ing into account baseline performance. The percentage change appear if the stimuli are relevant to only one task (i.e., univalent; was far greater for performance in the Mixed condition than Meiran, 2000a.b). We clearly found robust switch costs with our the Incongruent one (see Fig. 6). The difference between accu- univalent stimuli, but some differences in what we found versus racy in the Congruent and Incongruent conditions decreased what others have reported might be due to this characteristic of significantly over age (F(1,312) = 14.95, p < 0.0001), but the our stimuli. decrease over age in the difference between how accurately par- ticipants performed the Congruent and Mixed conditions was far 10. Results: comparisons across the different conditions greater (F(1,312) =43.81,p <0.0001; see Fig. 6). Thus, the accu- of the Dots task (Congruent Single-Task Block, racy difference between the Mixed and Congruent conditions Incongruent Single-Task Block, and the Mixed block) decreased more sharply over age than did the accuracy differ- ence between the Incongruent and Congruent conditions (dif- Comparisons of performance among these three blocks show ference between accuracy difference scores: (F(1,312) = 12.02, significant differences in the percentage of correct responses, p <0.001)). Despite the marked improvement over age in accu- RTs, and number of anticipatory responses (F(2,939) = 278.03 racy in the Mixed condition, even for 13-year-olds the difference [%correctl, 134.55 [RT], 49.86 [AR], all p <0.0001). As can in accuracy in the Mixed condition versus theCongruent one was be seen in Fig. 2, performance was best in the Congruent condi- larger than for adults (F(1,88)=7.47, p < 0.01). tion, intermediate in the Incongruent one, and worst in the Mixed For neither the Incongruent nor Mixed conditions was there condition. Planned comparisons show that performance in each a significant linear trend for reduced percentage-change scores of the conditions was significantly different from performance in any dependent measure between the ages of 4-6 years, in the other two conditions in accuracy, speed, and anticipatory except for percentage-change in RT for the Mixed condition responses (with the single exception of percentage of anticipa- (F(1,312)=4.95, p <0.04). The percentage change in speed of tory responses in the Congruent and Incongruent blocks among responding in the Incongruent condition compared with the subjects >6 years; see Table 3). Although performance was bet- Congruent one remained quite stable over age. The percentage ter for Dots-Congruent than Dots-Incongruent, that difference change in RT in the Mixed condition compared with the Congru- pales in comparison with the difference between performance in ent condition was greater and increased significantly over age either of those conditions and Dots-Mixed (difference between (F(1,312) = 28.75,p <0.0001). Thus, with age participants were Mixed and Incongruent versus the difference between Incongru- better able to modulate their performance speed, slowing down in the more difficult Mixed condition to minimize any reduc- tion in accuracy; whereas younger subjects (even those given a Table 3 very long response window) tended to keep their response speed T values for planned comparisons between trial blocks within the Dots test more constant across conditions at the cost of accuracy in the Percentage of Response Anticipatory more difficult Mixed condition. correct responses speed responses The difference inresponse speed between the Mixed and Con- All subjects (d.f.= 313) gruent conditions increased over age while the RT difference Congruent vs. 13.78 14.46 7.46 between the Incongruent and Congruent conditions remained Incongruent blocks constant. Hence the difference between RT in the Mixed and Congruent vs. Mixed 28.55 24.49 14.58 blocks Congruent conditions showed a greater change over age than the Incongruent vs. Mixed 18.81 15.66 10.51 difference between the Incongruent and Congruent conditions blocks (difference between RT difference scores: F(1,312) = 42.09, Younger subjects (4-6 years old: cll.89) p <0.0001). Congruent vs. 7.1 7.12 4.49 We had predicted that cognitive flexibility would improve Incongruent blocks Congruent vs. Mixed 15.32 10.51 8.27 with age and that therefore the difference in performance blocks between Dots-Incongruent and Dots-Mixed would decrease over Incongruent vs. Mixed 8.83 5.99 4.82 age. That was strongly confirmed for subjects 6 years and blocks older tested with the 750 ms stimulus presentation time. The Older subjects (6-26 years old: d.f.=223) difference between their performance on the Incongruent and Congruent vs. 12.24 17.06 6.24 Incongruent blocks Mixed conditions steadily decreased in both speed and accuracy Congruent vs. Mixed 24.15 30.36 12.1 (F( I ,222) = 12.9, p < 0.0005 [%correct]; 3.72, p < 0.05 [RT]). blocks For children 4-6 years of age however, tested with the 2500 ms Incongruent vs. Mixed 16.91 18.74 9.43 stimulus presentation time, the difference between performance blocks in the Incongruent and Mixed conditions did not change consis- All significant at p <0.000!. tently over age in either speed or accuracy. EFTA01098891 M.C. Davidson a al. /Neap rhologia 44 (2006)2037-2078 2051 — Incongruent -60 —ID-- M xeil -50 .40 rA = -30 C) ca Z -20 U a. -10 - o (Al 4 5 a 9 10 II 13 te 120 0 E 100 R 80 0 1.1 IG 60 CC •_C h 40 .C £ C0 0 20 • - S 6 6 7 8 9 10 11 13 26 (B) Age in Years Fig. 6. The Incongruent and Mixed conditions as percentage change from the Congruent condition of the Dots task. (A) Percentage change in accuracy and (B) percentage change in reaction time. 11. Discussion: comparisons across the different memory demands minimized] is also consistent with this predic- conditions of the Dots task tion). Indeed, accuracy differences between these two conditions must continue to decrease after 13 years of age since the differ- We had predicted that inhibitory demands would account for ence in accuracy in Dots-Congruent and Dots-Incongruent was a greater proportion of the variance in children's performance still greater in 13-year-olds than in young adults. than in adults, and the more so the younger the child. The Con- Since it is harder to switch back and forth between inhibiting a gruent and Incongruent blocks of the Dots test each contained dominant response and making it, we predicted that performance the same memory load (one higher-order rule, with two embed- at all ages would be better in the Incongruent-only block of the ded rules). The two blocks differed only in that the Incongruent Dots test (where the tendency to respond on the same side as block required inhibition while the Congruent Block did not. the stimulus must be inhibited all the time) than in the Mixed The prepotent tendency to respond on the same side as the stim- block of the task (where that tendency must be inhibited on ulus had to be inhibited in the Incongruent block but should have only half the trials as the other half are Congruent trials), and facilitated performance in the Congruent Block. We predicted that this difference would be greater the younger the children. that the Dots-Incongruent block would be more difficult than Indeed, performance differences between the Dots-Incongruent Dots-Congruent Block, but more important, that the difference and Dots-Mixed conditions were large at all ages, and especially in performance between those two conditions would decrease large the younger the children. as predicted. over age as inhibitory control improved. That prediction was confirmed. Dots-Incongruent was more 12. Results: global switch costs difficult than Dots-Congruent, and the more so the younger the children. Accuracy and impulsivity differences between these The cost of knowing that on some trials you will have to two conditions decreased over age. (The larger spatial incom- switch rules can be evaluated by comparing (a) performance patibility effect we had found the younger the children [with on Congruent trials following Congruent trials within a block EFTA01098892 2052 N.C. Davidson a al. /Neumps)rhologia 44 (2006)2037-2078 50 El Congruent Trials: Nortswitch in Mixed BMA 45 minus Single-Task Block Inconenseol Nonsaith in Mixed Block 40 minus Single-Task Bloat 15 i mas c t' 10 O — C tft -c E 5 O1- 0 (A) 4 5 6 9 in 13 28 90 0 .12E° tm is; gga0 ▪ 70 c 2 2 o gt it' 03 o 0 x 60 E 2 0 r, E tz fr. tal— c g 50 7 00 etEF cexw 40 2 2 -- co c c c t& • s:" .tt 6 O 22c 20 —8s io 6 8 9 10 11 3 26 (8) Age In Years Fig. 7. Mixing costs on the Dots task: performance on trials in the single-task blocks compared with performance on comparable nonswitch trials in the Mixed-task block. (A) Difference in percentage of correct responses: trials in the Congruent and Incongruent blocks minus the corresponding nonswitch trials in the Mixed block and (B) difference in reaction time: nonswitch Congruent or Incongruent trials in the Mixed block minus corresponding trials in the single-task blocks. of only Congruent trials to (b) performance on Congruent trials The accuracy cost of this difference in global (i.e., trial- following Congruent trials within the Mixed block, and similarly block) context was roughly equal for Congruent and Incongru- by comparing Incongruent trials in the Incongruent block with ent trials especially among the younger subjects. From age 9 Incongruent nonswitch trials in the Mixed block. In both cases onwards there was a trend for the accuracy cost to be greater on all dependent measures the difference is clear. Although the for Congruent trials (see Fig. 7). The cost in speed of this local context of all these trials is similar (all follow a trial of the difference in global context was significantly greater for Congru- same type), when these occurred in the context of a Mixed block, ent than Incongruent trials (only nonswitch trials: all subjects: participants were significantly slower, less accurate, and more 4313] = 4A6,p <0.0001; subjects s6 years, 3000-ms trial dura- inclined to make anticipatory responses (see Fig. 7; nonswitch tion: 4891=2.73, p <0.01; subjects ≥6 years, 1250-ms trial Congruent Dots-Mixed trials versus Dots-Congruent single-task duration: 1[223]=4.18, p <0.0001; see Fig. 7). The difference block [43 I 3) = 13.48, p <0.0001 (accuracy); 18.21, p <0.0001 in this RT cost for Congruent versus Incongruent trials was (RI); 2.85, p <0.011; nonswitch Incongruent Dots-Mixed tri- greater for the children 4-6 years old than for the older sub- als versus Dots-Incongruent single-task block [1(313)=9.44, jects (F(1,312)=6.28, p < 0.01). p< 0.0001 (accuracy); 10.07, p< 0.0001 (RI); 3.74, p <0.0003 The mixing cost (the cost of Congruent [Incongruent] (ARM). trials being Mixed in with Incongruent [Congruent] ones) for EFTA01098893 M.C. Davidson a al. /Neuromrhologia 44 (2006)2037-2078 2053 accuracy was much greater for subjects under 10 years of age Hackley, & de Labra, 2002; Verbruggen, than for those 10 years and older (reduction of the cost in Liefooghe, Notebaert, & Vandierendonck, 2005; Wiihr, 2004, accuracy over age [all subjects]: F(1,312)=30.6, p <0.0001; 2005). see Fig. 7). The mixing cost as assessed by response speed, however, increased over age (F(1,312) = 60.10, p <0.0001; see 14. Results: interaction of local switch costs with Fig. 7), again showing that older subjects were better able (or prepotent response or its inhibition more likely) to modulate their speed to preserve their accuracy. These findings are true for Congruent and for Incongruent trials. 14.1. Arrows test: interaction ofrule switching with prepotent response or its inhibition 13. Discussion: global switch costs For younger children, there was barely any accuracy switch Global switch costs (worse performance on nonswitch tri- cost in the Arrows test. Their accuracy was much worse on als in a Mixed block versus in a single-task block) were found Incongruent trials whether or not they were switch trials. For here, as predicted. We had predicted they would be greater the 7—I0-year-olds, the cost to accuracy of switching was greater younger the children. Indeed, global switch costs on accuracy on Congruent trials (111191=6.41, p < 0.0001). The difference were greater for participants <10 years old than for those older in the accuracy cost of switching to Congruent versus Incongru- than 10 years. Global switch costs on accuracy declined from ent trials followed an inverted U-shaped function over age (see 9 to 13 years. However, global switch costs on RT showed the Fig. 8). It was negative at 6 and 11-13 years of age, showing a opposite pattern. They increased from age 6 to early adulthood. greater accuracy cost in switching to the Incongruent rule. It was Adults adjusted their speed to preserve their accuracy; younger largest at 8 years of age and intermediate at 7 and 9-10 years children did that less, resulting in a difference in the speed- of age. For adults, there was no effect of spatial incompatibility accuracy trade-off with age. on accuracy. Adults made more errors on switch than nonswitch We had also predicted that, because of floor effects for Incon- trials in the Arrows test and it made no difference whether a gruent trials, the effect of context (the Mixed block versus Congruent or Incongruent response was required. single-task block) would be greater on Congruent than Incon- The effect of switching on RT in the Arrows test, depending gruent trials, and that this would be more evident the younger on whether the rule on the switch trial was Congruent or Incon- the child. However, contrary to the portion of our prediction con- gruent, showed a different pattern. Switching took a greater toll cerning development, the size of the greater effect of context on on the speed with which the younger children responded when Congruent versus Incongruent trials did not change over age. the response rule on the switch trial was Incongruent rather than It may well be that difficulty undoing inhibition of the pre- Congruent (difference for children 4-6 years old: [RT on Incon- potent response accounts for why switching back to making a gruent switch minus nonswitch trials] versus [RT on Incongruent response consistent with that tendency shows a greater cost than switch minus nonswitch trials' (489] = 2.6,p < 0.03) with a sim- switching back to inhibiting that tendency, as Allport and others ilar difference for children 6-7 years old: (4591= 2.8, p< 0.01). have suggested (Allport et al., 1994; Allport & Wylie, 2000). For children 8-13 years of age, the RT cost of switching was However, it is also true that the easier condition provides more equivalent on Congruent and Incongruent trials. For young room to find an effect because performance is so good on that adults, the difference seen in the youngest children reversed condition on nonswitch trials. It is not that subjects are worse at and the RT cost of switching was greater on Congruent trials switching to the easier rule than to the harder rule. It is that the (1[191=2.75, p <0.01), consistent with reports in the literature floor is so much lower for the easier than the harder condition for adults (e.g., Allport et al., 1994; Allport & Wylie, 2000). on nonswitch trials that there is more room for an effect to be The progression over age was from an opposite pattern in the found for switching to the easier condition. youngest children to no difference in the older children to finally As noted above in the discussion of spatial incompatibility seeing a greater RT switch cost on Congruent than on Incongru- effects, the lack of an accuracy difference on spatially compati- ent trials for young adults. ble (Congruent) and spatially incompatible (Incongruent) trials in the Mixed block is in sharp contrast to the result of com- 14.2. Dots test: interaction of vile switching with prepotent paring separate blocks of Congruent and Incongruent trials on response or its inhibition the task. The latter shows a significant incompatibility effect for children of all ages in both speed and accuracy, though not for The difference between accuracy on switch and nonswitch adults. The cost in accuracy on the spatially incompatible block trials in the Mixed block of the Dots task was significantly compared to the compatible block of the Dots task was greater greater for Congruent than for Incongruent trials [all sub- than the cost in speed, and the accuracy cost decreased over jects: 4313) = 2.96, p < 0.003; children <6 years: 489) = 2.36, age from 8 years onward (see Fig. 6). Our results for adults are p <0.02; children >6 years: 1(223)=3.03, p <0.0041. The consistent with a wealth of studies where adults have shown no greater cost in accuracy of switching to the Congruent condi- cost (or greatly reduced cost) of inhibiting in steady-state the tion was evident at 7 through II years of age (see Fig. 8b). The urge to make the spatially incompatible response in single-task children for whom the Dots task was most difficult (those 4-5 blocks (Praamstra, Kleine, & Schnitzler, 1999; Ridderinkhof, years old even though given a large response window and those 2002; Sturmer, Leuthold, Soetens, Schroter, & Sommer. 2002; 6 years old given a shorter response window) showed no greater EFTA01098894 2054 MC Davidson a at /Neumps)rhologia 44 (2006)2037-2078 25 20 15 10 5 0 4 •10 co — -15 re o -20 -25 (a) 4 5 6 7 9 10 11 13 26 (b) Age in Years Fig. 8. Differential accuracy cost of switching to the Congruent rule rather than the Incongruent rule. (a) Arrows test and (b) Dots-Mixed condition. accuracy switch cost for Congruent or Incongruent trials, nor did difference in speed of responding comparing Incongruent those who found the task easiest, 13-year-olds and young adults. switch and nonswitch trials for children ≥8 years and for adults Beginning at 8 years of age there was also a greater switch (children 8-13 years old: 11143] = 2.18, p <0.001; young adults: cost in RT for Congruent than Incongruent trials, replicating t[19]=2.75, p <0.01) but not for children <8 years. Indeed, the pattern previously reported for adults (that the RT cost of for the youngest children (4-6 years of age) the opposite was switching to the rule consistent with one's prepotent inclination found: The RT cost of switching to an Incongruent trial was is greater than the cost of switching to the rule that requires greater for them than the RT cost of switching to a Congruent resisting that inclination [e.g., Allport & Wylie, 2000; Allport trial (4891=4.31, p <0.0001); minoring a similar finding on the et al., 1994]). The difference in speed of responding comparing Arrows task. For 6-year-olds performing the faster version of Congruent switch and nonswitch trials was greater than the the task, the RT cost of switching was equivalent on Congruent EFTA01098895 M.C. Davidson a al. /Neuromrhologia 44 (2006)2037-2078 2055 and Incongruent trials. Hence, the age progression was from a to the Incongruent rule was greater than the RT cost of switching greater RT switch cost for Incongruent trials (at 4-6 years), to to the Congruent one. For older children of 8-13 years on no difference, to a greater RI' switch cost for Congruent trials the Arrows test, the RT cost of switching was equivalent on (from 8 years onward; see Fig. 8). Congruent and Incongruent trials. Only for young adults did the difference seen in the youngest children reverse. The progres- 15. Discussion: interaction of local switch costs with sion over age on the Arrows test was from a greater KT switch prepotent response or its inhibition cost on Incongruent than Congruent trials for the younger chil- dren (4-7 years old) to no difference in the older children (8-13 "Asymmetric switch costs" refer to a greater relative cost in years old) to finally seeing a greater RT switch cost on Congru- switching to the rule consistent with your prepotent tendency ent trials for young adults. On the Dots test, beginning at 8 years (Congruent trials in our study) than in switching to the rule that of age and for older ages, the adult pattern was evident—greater requires inhibiting that tendency (Allport & Wylie, 2000; Allport RT switch costs on Congruent than Incongruent trials. et al., 1994; De Jong, 1995; Kleinsorge & Heuer, 1999; Los, The pattern we report in children also differs from previ- 1996; Stoffels, 19%; Wylie & Allport, 2000). One explanation ous reports for adults (and our own findings for adults) in that for this pattern is that greater inhibition is required of the easier children showed differences in the size of the switch costs on rule when responding according to the harder rule than vice accuracy as well as RT On the Arrows test, the youngest chil- versa, and that going back to responding according to easier rule dren made more errors on Incongruent trials, whether they were requires undoing that inhibition.Hence, for example, Allport and switch trials or not, and showed little difference in accuracy Wylie (2000) looked at switching between reading color words on switching to Congruent or Incongruent trials. Children 7-10 and saying the color of the ink in the Stroop task. To report the years of age, however, showed greater costs in accuracy when ink color requires inhibiting the tendency to read the word; to switching to Congruent than to Incongruent trials. The greater switch back to reading the word presumably requires undoing accuracy cost in switching to the Congruent versus Incongruent that inhibition. To read the word requires minimal inhibition rule was largest in magnitude at 8 years and next largest at 7 of reporting the ink color; hence there is minimal inhibition to and 9-10 years of age. For children 11-13 years of age, the cost undo when switching back to reporting the ink color (but see in accuracy of switching was slightly greater on Incongruent also Yeung & Monett, 2003). than Congruent trials. For adults there was no difference in the In the present experiment, to respond on the side opposite accuracy cost. the stimulus should require inhibiting the tendency to respond In the Dots-Mixed condition, the difference between accu- on the same side as the stimulus. Switching back to respond- racy on switch and nonswitch trials was greater for Congruent ing on the same side as the stimulus should require undo- than for Incongruent trials overall and at all individual ages ing that inhibition. We had predicted we would replicate the except 4 and 5 years of age, age 6 when the shorter presen- effect previously reported in adults (greater RT costs for switch- tation time was used, and of course young adults. The chil- ing to the rule consistent with subjects' inclinations than for dren for whom the task was most difficult (those 4-5 years switching to the rule requiring inhibition of that) but also pre- old even though given a large response window and those 6 dicted that the very youngest children, who have poor inhibitory years old given a shorter response window) showed no greater control, would perform poorly on all trials requiring inhibi- accuracy switch cost for Congruent or Incongruent trials. The tion (Incongruent trials and switch trials) and that the effects size of their accuracy cost for switching to the Incongruent rule would be additive. Thus, we predicted that the youngest chil- was limited by their relatively poor performance on even non- dren, unlike adults, would perform worse when switching to the switch Incongruent trials. At all other ages, children showed Incongruent rule rather than to the Congruent one. Further, we a greater accuracy cost when switching to the Congruent rule predicted that intermediate-age children, who are beginning to in the Dots task. This was largest at the intermediate ages of exercise better inhibitory control, would require greater effort 7-11 years. to do so than older participants. Hence, we predicted that undo- The only other study to examine this in children (Crone et ing that inhibition (switching back to making the Congruent al., in press), using a task similar to our Dots task, found greater response) should exact a greater cost in intermediate-age chil- costs in both speed and accuracy for switching to Congruent dren than in older participants. Thus, we predicted that beginning versus Incongruent trials, as did we. However, unlike us, they after 6 or 7 years, "asymmetric switch costs" would be larger did not find differences in this over age. in younger versus older participants. These predictions were It may be that the difficulty of undoing inhibition of the pre- confirmed. potent response accounts for why switching back to making a In both the Arrows test and the Dots-Mixed condition, adults response consistent with that tendency shows a greater cost than showed a greater RT cost (though no accuracy difference) for switching back to inhibiting that tendency, as Allport and oth- switching to the Congruent than the Incongruent rule. Those ers have suggested. However, it should be noted that the easier results replicate those of Allport and Wylie. However, a different condition also provides more room to find an effect because per- pattern was found in children. On the Arrows test, the youngest formance is so good on that condition on nonswitch trials. The children were slower to switch to the Incongruent rule than floor is so much lower for the easier than the harder condition the Congruent one. Similarly in the Dots-Mixed condition, for on nonswitch trials that there is more room for an effect to be the youngest children (4-6 years old) the KT cost of switching found for switching to the easier condition. EFTA01098896 2056 M.C. Davidson err at. /Neumps)rhologia 44 (2006)2037-2078 16. Results: interaction of local switch costs with Indeed, children of 6-9 years of age performed at or near chance response-site switching on rule-switching response-stay trials. In speed of responding the interaction of rule 16.1. Arrows test: interaction ofrule switching with switch x response change on the Arrows test began to be response-site switching evident at 9 years of age and was significant for adults and children of 9-13 years (RT: 11126)=3.55, p <0.0005). Among The correct response on Trial N might be in the same location children younger than 9 years, however, there was a tendency as on Trial N-1 or it might be at the opposite location. We had for the effects of a rule switch and of a response-site change hypothesized that when the rule switched there would be an to be additive. Children younger than 9 years were slower inclination to change where to respond as well, and that when the on rule-switch trials, whether or not the correct response-site rule remained the same, subjects would be faster when the same changed (main effect of a rule switch: t(I63)= 6.8 I, p < 0.0001). trial repeated (consistent with global commands to "change" Similarly, they tended to be slower on response-change trials. or "repeat"). We thus predicted an interaction between whether whether or not the rule switched. They were slowest if both the rule changed and whether the correct response-site changed. rule and response changed (F(1,178)=4.42, p < 0.0001: see On the Arrows test, our prediction was strongly confirmed for Fig. 9B). The adult pattern (of faster responses when neither accuracy (F(1,312)=41.89, p < 0.0001). The effect was most changed or both changed) was not seen until 9 years of age on marked at 6-9 years of age and smallest in adults (see Fig. 9A). the Arrows test. — Rule Repeals •••••-• Rule Swotches 4.; (O 8 •uo, a O ,C c :2 ci) 20 0 aE (A) 200 co Difference in RT (in msec): ct) 150 WO O _ ago O -100 6 7 8 9 26 (B) Age In Years Fig. 9. Cost of switching response locations in the Arrows task on switch trials and on nonswitch trials. (A) Difference in percent correct: opposite side minus same side and (B) difference in reaction time: opposite side minus same side. EFTA01098897 M.C. Davidson es a1. /Neuropsychologia 44 (2006)2037-2078 2057 The effect on switch trials is of most interest here because both rule and response-site switched (41331 = 3.58,p <0.0005). nonswitch, response-stay trials are simply repeat trials. Focusing Children of 4-8 years tended to be faster when the response- just on switch trials, the difference in error rates on response-stay site remained the same, even on switch trials (1(163)=1.77, and response-shift trials was smallest for the youngest children p =0.08). (4-6 years of age) and young adults (see Fig. 9A). Although this difference decreased from 9 to 13 years, the difference 16.2. Dots test: interaction of title switching with in accuracy at 13 years was still greater than seen in young response-site switching adults (F(1,42)=4.1, p <0.05). At all ages, this difference was for fewer errors to be made when both rule and response-site When both the rule and response-site remained the same, pre- changed than when just the rule changed. This accuracy differ- cisely the same trial was repeated. One would expect RT to be ence was most pronounced from 6 to 13 years. fast then, and faster than when the response location changed. The difference in response time on switch trials depending What is more interesting is that, as predicted, performance on on whether the response-site changed or not decreased from 4 switch trials was faster and more accurate when the correct to 6 years when given lots of time to respond, and was insignif- response location also switched than when it remained the same icant among children of 6, 8, and 10 years and young adults 0(308) = 8.03, p <0.0001 [accuracy]; 2.8, p < 0.005 [R11; see (see Fig. 9B). Not until 9 years of age were children faster when Fig. 10). The corresponding results for only the younger subjects Rule Repeats 40 Rule Switches 30 t; Ci g a 10 o I 0. g 0 • C • E •c -10 co y E •20 .30 (A) .40 250 o N 200 E g C e3 150 ta C 100 S 50 0 :g 0 5 0. O -° •100 -150 0 5 8 8 7 8 9 10 11 13 (B) Age In Years Fig. 10. Cost of switching response locations in the Dots task on switch trials and nonswitch trials. (A) Difference in percent correct: opposite side minus same side. (Across the age spectrum, and especially at 6-11 years. participants were correct on more switch trials when the response-site also switched from the previous trial.) and (B) difference in reaction time: opposite side minus same side. (The typical adult pattern of faster responding on switch trials if the response-site also switched from the previous trial, seen here and reported in numerous studies, was not evident until 13 years of age.) EFTA01098898 2058 MC Davidson a at /Neumpsychologia 44 (2006)2037-2078 are: (t[89[=4.25, p <0.0001 (accuracy); NS (RI). The corre- switch to be additive. Children of 4-8 years were slowest if both sponding results for just the older subjects are: (1[2231=8.25. the rule and correct response-site changed. They were faster p <0.0001 [accuracy'. 4.05, p <0.0001 [RTI). Indeed, from age when the response-site remained the same, even on switch tri- 6 to II (and especially 9-11 years) the effect on accuracy was als. Similarly, in the Dots-Mixed condition, children of 4 and 5 particularly large (see Fig. 10A). The adult pattern (of faster years given a long time to respond, and children of 6-8 years responses when neither changed or both changed) was not seen given less time to respond, were faster when the response-site until 13 years of age on the Dots (see Fig. 10B); 4 years later remained the same whether the rule changed or not. Especially than that pattern appears on the Arrows test. at 4 years, and at 6 years on the short-version, the RT effect On switch trials, there was little difference in children's error of task-switching and response-switching appeared to be addi- rates on response-stay and response-shift trials from 4 to 7 years tive. Thus, for RT we found the predicted all-or-none pattern of age. From 8 to 10 years accuracy was far better on switch in older children and young adults, but for younger children trials when the response-site also changed. The difference in we found worse RT performance on switch than nonswitch tri- accuracy was again small among young adults (see Fig. 10A). als and on response-switch than response-stay trials and those The difference in response time on switch trials depending on effects tended to be additive. whether the response-site switched or not decreased from 4 to 6 years when given lots of time to respond (as it did for Arrows 18. Results: comparing across the tests that required [compare Figs. 9B and 10B1), was large again from 6 to 8 years inhibition (Pictures, Arrows, and Dots) when given less time to respond, but not until 13 years was response time faster when both the rule and the response-site 18.1. Comparing performance in the Mixed block of each switched. Before then RTs were faster on switch trials when the of the tests response-site did not change (see Fig. 10B). The Pictures test (our classic Simon task with minimized 17. Discussion: interaction between rule switching and memory load) was substantially easier for children of all ages response-site switching than were the Arrows or Dots-Mixed tests (spatial incompati- bility tasks with higher level rules). Children showed far better When the rule and response remained the same, precisely the accuracy, faster response times, and markedly fewer instances of same trial was repeated. One would expect RTs to be faster on anticipatory reaching in the Pictures test compared with either such exact trial repetitions than on other trials. More interesting the Arrows or Dots tests (see Fig. II; Pictures versus Arrows: is what happens on switch trials. We predicted, consistent with t[293[= 13.3 [%correct], 9.42 [RTI, 9.06 [AR': Pictures versus the findings of others (Hommel et al., 2001; Kleinsorge, 1999; Dots-Mixed: t[293[= 18.53 [%correct], 8.59 [RT], 12.92 [AR]; Meiran, 2000a,b; Rogers & Monsell, 1995; Schuch & Koch. all six t-values significant at p <0.0001). By 9 years of age, 2004) and Diamond's all-or-none hypothesis (Diamond, 2005), anticipatory responses had all but disappeared on the Pictures that performance would be better when both the rule and the test. response changed than when the rule changed but the response Children tested in the faster presentation condition (children did not. ≥6 years of age) found the Arrows test to be almost as diffi- On both Arrows and Dots-Mixed, across the age spectrum, cult as the Dots-Mixed condition, judging by their comparable people were more accurate when both the rule and the response speed in the two conditions (see Fig. II), though other aspects changed than when just the rule switched. At all ages, fewer of their performance were still significantly worse on Dots- errors were made on the Arrows test when both the rule and Mixed than on Arrows (children 6-13 years of age: t[203] =4.05, response-site changed than when just the rule changed, with p<0.001 [accuracy]; NS [RT]; 2.72, p<0.01 [AR]). Without the difference in error rate being smallest for young adults and question however, the difference in performance of children largest for children 6-9 years of age. The difference in accuracy 6-13 years on the Pictures test compared with performance at 13 years was still greater than that seen in young adults. For on either the Arrows test or Dots-Mixed condition was far Dots-Mixed as well, fewer errors were made when both the rule greater than any difference in their performance on the Arrows and response-site changed than when just the rule changed, with test and the Dots-Mixed condition (Pictures minus Arrows ver- the difference in error rate being smallest for young adults and sus Dots-Mixed minus Arrows: t[2931= 15.15 [96correct], 8.92 the youngest children and largest for children 8-10 years of age. [RTI, 11.17 [AR], all significant at p <0.0001). Thus, we found the predicted all-or-none pattern for accuracy For adults and the youngest children, the results were dif- throughout our age spectrum. ferent. Adults found the Pictures and Arrows tests to be of For speed of responding, we replicated the pattern previ- comparable ease and found both of those conditions to be sig- ously reported for adults: faster responses when both the rule nificantly easier than the Dots-Mixed condition (see Fig. 11; and response switched than when only the rule switched but not Dots-Mixed versus Pictures: r(20) = 7.47 [KT]; Dots-Mixed ver- the response. In children, the RT interaction of rule-switch with sus Arrows: S20) = 7.21 [K1], both p <0.0001; no comparisons response-switch was not evident until the age of 9 on the Arrows between Pictures and Arrows yielded any significant results). task and the age of 13 on the more difficult Dots-Mixed condi- Thus, while performance on Arrows and Dots was roughly com- tion. On the Arrows test, children younger than 9 years of age parable for children of 6-13 years, performance on Arrows and showed a tendency for the effects of a rule switch and response Pictures was comparable for adults. EFTA01098899 M.C. Davidson er at /Neueopsyrhologia 44 (200612037-2078 2059 Percent Correct •••• Piclam, - ♦ Anon —II Dole (A) 4 5 0 0 7 8 9 10 11 13 28 1200• 0 re MO. to • 8O •0 600. 03 IX • 400 • 6 7 8 9 10 11 13 26 % Anticipatory Responses S 0 4 5 6 7 8 9 10 11 13 26 %mutt gaunter) for 2500 ms SEIM. crosemed log 750 ms (C) Age in Years Fig. II. Comparison or Mixed conditions of Dots. Arrows and Pictures. (A) Percent correct. (B) reaction time and (C) percentage of anticipatory responses. EFTA01098900 2060 MC. Davidson a al /Neummrhologia 44 (2006)2037-2078 While children of 4-6 years. tested with the longer stimulus a percentage of RT on Dots-Congruent remained constant (see presentation times, found the Pictures test to be easier than the Fig. 6). Arrows test or Dots-Mixed condition as did older children, chil- A third way to assess the relative cost of increasing inhibitory dren of 4-6 years (unlike older children) found the Arrows test demands is to compare performance on the Arrows task (which to be much easier than the Dots-Mixed condition (see Fig. II). required inhibition on the half the trials [the spatially incompati- All pair-wise comparisons between any two of the three tests ble ones) and required task-switching if encoded as two superor- were significant except for Arrows versus Dots-Mixed on RT, dinate rules, but required little memory as each stimulus pointed and all those were significant at p <0.0001, except for Pictures to its correct response-site) to performance (a) where inhibition versus Arrows on AR [459)= p < 0.01] and Arrows ver- of the spatially compatible response was required on all trials sus Dots-Mixed on AR [t(59)= p <0.0051. Thus, for the (rather than switching between spatially compatible and incom- youngest children no two tests were comparable in difficulty. The patible trials—Dots-Incongruent) or (b) where inhibition of that Dots-Mixed condition was significantly harder than the Arrows response was never required (Dots-Congruent). Inhibiting the test and the latter was significantly harder than the Pictures spatially-compatible response some of the time despite the min- test (except that in their RT data they showed the same pat- imal memory requirements (in the Arrows task) took a greater tern as older children [comparable performance on the Arrows toll on speed and accuracy at every age than did inhibiting the and Dots-Mixed tests with much faster responses in the Pictures spatially-compatible response on all trials (Dots-Incongruent, test]). see Table 2), though those differences were of course smaller than that between Arrows and Dots-Congruent (where no inhi- ./8.2. Comparing performance across conditions that bition was required). Accuracy differences between the Arrows differed in their demands on inhibition but required little or task and Dots-Incongruent condition were greatest at interme- no memory diate ages (children of 6-11 years tested in the faster condition) and smallest among the youngest children (4-6 years, given a Comparing performance in the Pictures test and the Dots- much longer time to respond) and among the two oldest groups Congruent condition enabled us to compare performance in the (13-year-olds and adults). Accuracy differences between Arrows presence versus absence of an inhibitory demand with mem- and Dots-Congruent were sizeable at all ages except among ory load held relatively constant. Both the Pictures test and young adults and decreased significantly from 7 to 26 years the Dots-Congruent condition required holding two rules in (F(I,192) = I5.73, p = 0.0001). Differences in response speed in mind. (In the Pictures test that memory demand could be min- the Arrows test and Dots-Incongruent condition were roughly imized by referring to the visible icons that showed which 200 ms at all ages, except among 4-year-olds, where the mean picture was mapped to the left or right. In the Dots-Congruent difference was only 100 ms (all subjects: /[3131= 14.76; sub- condition, the memory demand could be minimized by remem- jects ≥6 years: 4223] = 15.76; subjects <6 years: 1[891=6.27; bering the single higher-order rule.) While the memory demand all p <0.0001). Mean RT differences between Arrows and Dots- was roughly equivalent in Pictures and Dots-Congruent, the Congruent were roughly 350 ms or more for the youngest chil- former required inhibiting the tendency to make the spatially dren (4-6 years, given a long time to respond) and decreased compatible response on half the trials whereas the spatially linearly from 325 ms at 6 years (adult condition) to 180ms compatible response was always the correct response in the among young adults (except for a spike at 10 years; all Dots-Congruent condition (no inhibition required). Thus more subjects: x[313]=27.00; subjects >6 years: 42231=31.19; inhibitory control was required in the Pictures test than in subjects <6 years: 4891= 12.49; all p <0.0001; significant the Dots-Congruent condition. Across all ages, performance decrease from 6 years to young adulthood: F(1,222)=5.32, was consistently better in the Dots-Congruent condition than p <0.0001). in the Pictures test on all dependent measures (all subjects: Finally, comparing performance in the Pictures condition to 43B] = 23.16 [96 correct] and 24.0 [RT]; 2.31; subjects >6 the Dots-Incongruent condition, like the comparison of Arrows years: t[2231=13.9 [%correcti and 27.82 [RT1; subjects ≤6 to Dots-Incongruent, provides a measure of (a) inhibition in years: t[89] =4.16 [%correct] and 10.26 [RT1; all p < 0.0001; a switching context where it is only required on some trials see Table 2). The difference in accuracy on Dots-Congruent (with little or no memory requirement) versus (b) inhibition in versus Pictures decreased over the age range of 6-26 years a steady-state context where it is required on every trial (and (F(I,222) = 5.32, p <0.02). memory of a higher order rule and instantiating it on each trial Another way to assess inhibitory costs, and change in their are required). Accuracy was comparable in these two condi- size overage, is to look at the cost of steady-state inhibition (con- tions across all ages but when given only 1250 ms to respond sistently inhibiting the prepotent response in Dots-Incongruent) subjects consistently responded faster in Dots-Incongruent than versus consistently making the prepotent response in Dots- in Pictures (all subjects: 43131=7.2, p <0.0001; only those ≥6 Congruent. These results were presented above (see Fig. 2 and years: 42231= 8.57,p <0.0001; only those ≤6 years:11891= 1.9, Table 3). These costs (in both speed and accuracy) were sig- p = 0.06; see Table 2). The RT difference (faster in Dots- nificant for children at all ages, including the oldest children Incongruent than in Pictures) tended to increase over age from 6 (13 years old), but were not significant for adults. The accuracy years through young adulthood (F(1,222)=3.44, p = 0.065) and difference between the Dots-Congruent and -Incongruent con- was over twice as large among young adults as among 6-year- ditions decreased over age, while the RT on Dots-Incongruent as olds. EFTA01098901 M.C. Da rayon et at. /Neuromrholagia 44 (2006)2037-2078 2061 19. Discussion: comparing performance on the tests dition was significantly harder than the two-shape condition that required inhibition (Pictures, Arrows, and Dots) (all subjects: 43131=13.60 [accuracy], 20.36 [RT1, and 6.49 IARI, all significant at p <0.0001; subjects >6 years, 1250- When minimal memory was required and no task switching, ms ISI: t[2231= 12.27 [accuracy], 20.49 [Ref], and 6.39 [AR], very young children were reasonably successful at overcom- all significant at p <0.0001; subjects <6 years, 3000-ms ISI: ing the prepotent tendency to respond on the same side as the 489] = 6.16, p <0.0001 [accuracy], 9.89, p <0.0001 [RT1, and stimuli consistently on all trials (Dots-Incongruent; 86% cor- 2.48, p < 0.02 [ARI: see Fig. 12). Both conditions showed rect at 4-5 years) and on half the trials as long as the rules did age-related improvements in performance on all three depen- not change (Incongruent trials in the Pictures task; 80% cor- dent measures. For two shapes, age-related improvement was rect at 4-5 years). On the Dots-Incongruent task the youngest significant on all three dependent measures at p <0.0001 for children were able to perform at a relatively high level when subjects >6 years and for all the subjects together (subjects required to combine (a) holding a superordinate rule in mind ≥6 years: F(1,222)=21.87 [accuracy], 78.70 [RI], and 16.06 (mentally translating that into the appropriate embedded rule on [AR]; all subjects: F(1,312)=26.84 [accuracy], 140.13 IRTI, each trial) plus (b) inhibiting the dominant tendency to respond and 39.10 [A121). For children 4-6 years of age there was no on the same side as a stimulus, but importantly inhibition was difference in speed in the two-shapes condition over age, but required in steady-state and the rule remained constant. They the improvements in accuracy and impulsivity were signifi- had to exercise that inhibition on every trial, not switching cant (accuracy: F(I ,88)= 5.13, p < 0.03; AR: F(1,88) = 12.75, back and forth between sometimes exercising it and sometimes p <0.001). Results are similar for the six-shapes condition; age- not. In the Pictures task, there were no higher-order rules to related improvement was significant atp <0.0001 for all subjects mentally instantiate. Only two stimulus—response associations together and for those ≥6 years of age, except on RT which was were relevant and memory demands were minimized by hav- significant at p <0.0005 for those >6 years (subjects ≥6 years: ing a picture of each stimulus mounted immediately above its F(1,222)=30.48 [accuracy], 12.72 [RT], and 24.79 [AR]; all associated response button. Critically, the rules never changed. subjects: F(1,312)=29.64 [accuracy], 81.11 IRTI, and 39.56 However, inhibition rather than being continuously required, IARI). For the youngest participants there was no significant was needed on only half the trials. The performance of 4- to age difference in speed on the six-shapes condition, but accu- 5-year-olds on the Pictures task indicates that they could obey racy and impulsivity showed significant improvements over age two stimulus—response rules even though that meant switching (accuracy: F(1,88) = 11.32, p <0.001; AR: =4.88, p <0.03), as between sometimes responding on the Congruent side (the same was found for the two-shapes condition. side as the stimulus) and sometimes on the Incongruent side. To test whether there was more change in performance over Since it is harder to switch back and forth between inhibiting a age in the six-shapes condition than in the two-shapes con- dominant response and making it than to consistently inhibit that dition, difference scores were calculated for each participant response, we had predicted that performance at all ages would (performance in six-shapes minus two-shapes) for each of the be better in Dots-Incongruent than in Arrows (which required dependent measures. None of these difference scores (for accu- switching, but minimized memory demands). That prediction racy, speed, or anticipatory responses) varied significantly as a was confirmed. Accuracy was better and speed faster at every function of age when all subjects were included in the analyses. age in the Dots-Incongruent condition than on the Arrows test. The degree to which the six-shapes condition was more difficult Thus, inhibiting the spatially-compatible response some of the than the two-shapes condition generally did not change over time even when the stimuli pointed to the correct response (in age. This suggests that although participants of all ages were the Arrows task) took a greater toll on speed and accuracy at affected by the increased memory load (i.e.. all showed positive every age than did inhibiting the spatially-compatible response difference scores) the size of this effect changed little over age. all the time (Dots-Incongruent). We had also predicted that per- The magnitude of the difference in performance on the six- formance differences between Dots-Incongruent and Arrows versus two-shapes conditions showed no significant change over would decrease over age as cognitive flexibility improved. While age on any of the three dependent measures when all sub- the accuracy difference between these conditions was smaller jects were included or only the youngest children were used. in adults, otherwise the markedly better performance on Dots- However, for the 6-year-old through young-adult subjects, the Incongruent than Arrows was equally true across all ages, con- speed-accuracy tradeoff seemed to vary by age: Accuracy on trary to our prediction. the six-shapes condition more closely approximated that on the two-shape condition in older subjects [9 years through adults 20. Results: the Abstract Shapes test: conditions that versus 6-8 years old: F(1,162) = 7.39, p < 0.01] while the differ- differed in their demands on memory but required little ence in RT on the two conditions showed an opposite tendency, or no inhibition with a smaller RT difference between the two conditions in younger subjects [9 years through adults versus 6-8 years old: The Abstract Shapes test contained two conditions (two F(1,162)= II.73,p <0.0011. shapes and six shapes), designed to vary working memory Performance in the six-shape condition can also be viewed load (two arbitrary rules versus six). The inhibition requirement as a percentage change from performance in the two-shape con- was minimal, as all shapes were presented at central fixation dition ([six-shapes minus two-shapes] divided by two-shapes). (no spatial incompatibility). As predicted, the six-shape con- Overall, and for the older subjects, there was no significant EFTA01098902 2062 MC. Davidson er af. /Neumpsychologia 44 (2006)2037-2078 --- Dots - Mixed ma Abetted 6 0 60 O >. u to 0 50 u 4t 15 • 40 a re CO 2 30 1 20 t C)c 0.A 10 N 0 (A, 4 5 6 7 8 9 10 11 13 26 Dote - Incongruent 120 iiihn Mama 6 ott 0 E (In the background: Dots - .c — • too Incongruent as Percentage g !14 Change from Dots.Congment g co a+ f,, 60 ce 0 40 .c E • O rcipgto ..- a 20 g 2 (s) 4 5 6 6 7 8 Age in Years 9 10 11 13 26 Fig. 12. The six-Abstract-Shapes condition as percentage change from the two-Abstract-Shapes condition. (A) Percent change in accuracy-(two -shape minus six- shapes) divided by two-shapes (in the background: Dots-Mixed as percentage change front Dots-Congruent) and (B) percent change in reaction time-(two-shape minus six-shapes) divided by two-shapes (in the background: Dots-Incongruent as percentage change from Dots-Congruent). change over age in the accuracy difference between the six- For speed of responding, a different picture emerged. shape and two-shape conditions as a percentage of accuracy in Whereas change over age in the difference between accu- the two-shape condition. For the youngest children (4-6 years, racy in the six- and two-shape conditions as a percentage of tested with 2500-ms stimulus presentation time) the error rate two-shapes performance was significant only for the younger on six-shapes declined over age and so the percentage change children, the reverse was true for the RT difference. The RT in accuracy on the six-shape condition compared to the two- difference in the two conditions as a percentage of speed shape condition declined over this age range (F(1,89)=2.65, in the two-shapes condition changed significantly overall and p < 0.000 l)—showing the same pattern as seen for percent- among subjects >6 years, but not among the younger children age change in accuracy on Dots-Mixed compared with Dots- (all subjects: F(1,312)=29.39, p < 0.00011 subjects >6 years: Congruent though the difference for the latter was larger and F(1,222)=38.10, p <0.0001). Further, whereas the percentage remained larger (see Fig. I2A). Indeed, the size of the dif- change in accuracy in six-Abstract-Shapes over age resem- ferences, as well as the pattern, among the older subjects for bled that on Dots-Mixed, the percentage change in RT on six- six-shapes compared to two-shapes was similar to that for Dots- Abstract-Shapes over age resembled that on Dots-Incongruent Mixed compared to Dots-Congruent except that the difference (see Fig. 128). The change over age in speed on Dots-Mixed as a between six- and two-shapes was particularly small among 9- percentage of Dots-Congruent dwarfed the age-related change in year-olds and hence the linear trend did not reach significance six-Abstract-Shapes as a percentage of two-Abstract-Shapes or (see Fig. I2A). Dots-Incongruent as a percentage of Dots-Congruent. Finally, EFTA01098903 M.C. Davidson et al. /Neuropsychologia 44 (2006)2037-2078 2063 whereas the accuracy difference between six- and two-shapes 22. Results: comparison of performance on the Abstract decreased over age, the difference in response speed on the two Shapes test and the other tests conditions increased over age, as older subjects preserved their accuracy in the harder condition by sacrificing their speed (see 22.1. First -order comparisons among conditions Fig. 12). The easiest condition of all, across all ages, was Dots- Congruent (see Table 2). Accuracy was consistently highest 21. Discussion: the Abstract Shapes test: conditions that and RT consistently quickest in that condition at all ages. differed in their demands on memory but required little Indeed, accuracy and speed were significantly better (in all or no inhibition cases at p <0.0001) in the Dots-Congruent condition than on the three next easiest conditions, two-Abstract-Shapes, Pictures, We had predicted that even very young children would per- and Dots-Incongruent, with all participants included in the anal- form well at holding two rules in mind when inhibition is not yses, only the older subjects, or only the youngest subjects (see taxed. Consistent with that prediction, the performance of even Table 4). our youngest subjects was excellent in the two-Abstract-Shapes Across all ages, accuracy on the two-Abstract-Shapes condi- condition. tion, the Pictures test, and the Dots-Incongruent condition was Based on our hypothesis that the ability to hold items in excellent and fully comparable (see Fig. I3A and Table 2). mind matures early, we had predicted that although it would RT (as opposed to accuracy) was better on the two-Abstract- be harder for everyone to hold more items in mind than fewer, Shapes and Dots-Incongruent conditions than on the Pictures the relative difficulty of that would not change over age. Over- test (see Table 2; RT on two-Abstract-Shapes versus Pictures: all within-subject difference scores (six-Abstract-Shapes versus all subjects: O131= 10.55; only those ≥6 years: 4223] = 8.82; two) did not show any change over age in relative difficulty only those ≤6 years: 4891=6.81; all p <0.0001; RT on Dots- on any dependent measure. How to answer whether the rela- Incongruent versus Pictures: see above). tive difficulty changed over age is not straightforward, however. Table 4 because how the difference in difficulty was handled changed T values for planned comparisons between experimental conditions over age, i.e., the speed-accuracy tradeoff changed over age Accuracy in the six-shapes condition more closely approximated Percentage of Response speed correct responses that on the two-shape condition in older subjects (9-year-olds through young adults) while the difference in speed of respond- The three of the four easiest conditions' ing in the two conditions showed the opposite tendency, with a Dots-Congruent vs. two-Abstract-Shapes All subjects (d.f. =1.313) 13.52 —19.78 greater RT difference between the two conditions in the same Younger subjects (4-6 years: 12.54 —26.91 older subjects. This suggests that older subjects preserved their d.f. =1.89) accuracy in the harder six-shapes condition by sacrificing their Older subjects (6-26 years: 5.72 —6.29 speed. Hence the speed differential between the two conditions d.f. = 1.223) was largest for these subjects but the accuracy differential was Dots-Congruent vs. Pictures All subjects (di =1.313) 13.20 —24.00 smallest. Younger subjects (4-6 years: 13.91 —27.84 Another measure of whether the relative difficulty of these d.f. =1.89) two conditions changes over age might be performance on six- Older subjects (6-26 years; 4.23 —10.31 Abstract-Shapes as a percentage change from performance on d.f. = 1.223) two-Abstract-Shapes (thus correcting for differences in base- The three hardest conditions', Dots-Mixed vs. six-Abstract-Shapes line performance). In general, there was no significant change All subjects (d.f. = 1.313) —6.24 2.46 (p < 0.01) over age in the accuracy difference between the six-shape and Younger subjects (4-6 years: —4.22 1.30 NS two-shape conditions as a percentage of accuracy in the two- d.f. =1.89) shape condition. However, for the youngest children (4-6 years) Older subjects (6-26 years; —4.78 224 the percentage change in accuracy on the six-shape condi- d.f. =1223) tion relative to the two-shape condition declined over age (this All significant at p < 0.0001. unless otherwise noted. same pattern was seen for percentage change in accuracy on a Dots-Incongruent was the other very easy condition. For performance on Dots-Mixed compared with Dots-Congruent though the dif- Dots-Congruent vs. Dots-Incongruent. see Table 3. Accuracy on two-Abstract- Shapes. Pictures. and Dots-Incongruent was fully comparable. Response speed ference between the Dots conditions was larger and remained was faster on two-Abstracts-Shapes than on Pictures (all three comparisons sig- larger). nificant at p <0.0001) and on Dots-Incongruent than Pictures (for all subjects Whereas change over age in accuracy on six-Abstract-Shapes and older subjects. p < 0.(X)0 I: for younger subjects. p =0.06). Younger chil- as a percentage change from performance on two-Abstract- dren were faster on Dots-Incongruent than two-Abstract-Shapes (4891=3.54. Shapes was significant only for the youngest children, the p <0.001). while our older subjects were faster on two-Abstract-Shapes than on Dots-Incongruent (1(2231=-246. p <0.02). reverse was true for the RT difference. Here, again, the change b Arrows was theotherrelatively difficult task. For Dots-Mixed vs. Arrows. see over age in speed on Dots-Mixed as a percentage of Dots- the section comparing performance in the three Mixed conditions. There were Congruent dwarfed the age-related change in six-Abstract- no significant differences in either speed or accuracy on six-Abstract-Shapes Shapes as a percentage of two-Abstract-Shapes. and Arrows. EFTA01098904 206.1 MC. Davidson of al. /Neumps)rhologia 44 (2006)2037-2078 100 T T 90 T Percent Correct 80 70 - -a— Arc: 60 ye Ab..•'nct (A) 4 5 6 7 8 9 10 11 13 26 1200• Reaction Time (in msec) 1000• 800• 600 • I 400 (B1 4 5 6 6 7 8 9 10 11 13 26 35 30 %Of Anticipatory Responses 25 20 15 10 5 a 0 0 5 6 6 7 8 9 10 11 13 26 1 L 1 St mull PCOSeed00 be 2500 m5 Staub ',assented br 750 ms (C) Age In Yens Fig. 13. Comparisons of the two-Abstract-Shapes conditions with each other and with all the other tasks. Despite significant differences in performance among was intermediate on the Arrows test and the six-Abstract-Shapes these conditions, these four conditions (Dots-Congruent, condition between those four conditions, on the one hand, two-Abstract-Shapes, Dots-Incongruent, and Pictures) clearly and Dots-Mixed condition, on the other (significantly worse proved the easiest for participants. Across the age span, accuracy performance on the Arrows test than on the four easier tests: EFTA01098905 ALC. Davidson et al. /Nettropsychologia 44 (2006)2037-2078 2065 all subjects: x[3131=16.42 [accuracy] and 19.15 [RT]; only this comparison between difference scores was most highly those >6 years: :12231= 15.41 [accuracy] and 20.36 [RT]; significant was young adults (4191= 10.5, p <0.0001). only those <6 years: 489] = 6.68 [accuracy] and 9.30 [RT]; all The size of the differences in the percentage of correct p <0.0001; significantly better performance on Arrows than responses (Dots-Congruent minus Dots-Incongruent compared on Dots-Mixed, see above; significantly worse performance with two-Abstract-Shapes minus six-Abstract-Shapes) did not on the six-Abstract-Shapes condition than on the four easier differ significantly over the age range of 4-9 years. However, tests: all subjects: 43131= 18.57 [accuracy] and 23.73 [Rh]; beginning at 10 years, increased memory demands (six versus only those ≥6 years: 42231= 16.89 [accuracy] and 23.66 [RT]; two rules) took a greater toll on accuracy than did consistently only those ≤6 years: 489] = 8.26 [accuracy] and 10.15 IRTI; all inhibiting the tendency to respond on the same side as the stim- p <0.0001). ulus (subjects >10 years: 41031= 3.62, p <0.0005). Although no inhibitory or task-switching demands were Another way to look at the relative cost of increasing present in the six-Abstract-Shapes condition, holding six inhibitory demands is to compare performance on the Arrows arbitrary rules in mind for hard-to-name, abstract stimuli was task to that on Dots-Incongruent or Dots-Congruent. At all challenging at all ages. However, no test, not even six-Abstract- intermediate ages (6-I I years, all tested under adult condi- Shapes, was as difficult as the Dots-Mixed condition (where tions), having to switch between inhibiting and not (Arrows) subjects had to hold two superordinate rules in mind, instantiate versus settling in to exercising inhibition on all trials (Dots- them on each trial, inhibit a prepotent response tendency Incongruent) tended to take a greater toll on accuracy and speed on incompatible trials, and switch between same-side and than did having to hold six arbitrary rules in mind rather than two opposite-side rules) all subjects: 4313)=6.2 (accuracy) and 2.4 (subjects 6-11 years, adult conditions: t[2031=1.72, p=0.08 (p <0.04; RT); only those >6 years: 4223) = 4.2 (accuracy) and [%correct]; 1.88, p = 0.06 [RT]). For the youngest subjects (4-6 NS (RI'); only those <6 years: 489)=4.78 (accuracy) and 2.15 years) given much longer to respond and for the two oldest age (p < 0.01;RT); all p <0.0001 except where otherwise noted; see groups (13-year-olds and young adults), the accuracy and RT Fig. 13 and Tables 2 and 4). costs were comparable. At all ages, the RT costs for having to Performance on the Arrows test was roughly comparable to exercise inhibition in the Arrows task versus not having to do that in the six-Abstract-Shapes condition in both speed and accu- so in the Dots-Congruent condition took a much greater toll racy. Accuracy on both was significantly better than accuracy in on response speed than did increasing the memory load from the Dots-Mixed condition (see Fig. I 3A). RT on the Arrows test two to six arbitrary rules (all subjects: 43131= 11.37; only those was generally intermediate between that on the six-Abstract- >6 years: t[2231= 12.70; only those <6 years: t[891=5.12; all Shapes and Dots-Mixed conditions, especially between 4 and 8 p <0.0001). A larger accuracy difference between Arrows versus years of age (see Table 2), not significantly different from either. Dots-Congruent than between two- versus six-Abstract-Shapes, Thus, the task that taxed memory most heavily and included no however, was only true for the younger two-thirds of the subjects inhibitory or task-switching component (six-Abstract-Shapes) (subjects ≤9 years: 4210] = 6.65, p < 0.000 I ). This was not sig- proved approximately equivalent in difficulty at all ages to nificant for the older subjects; indeed for adults there was almost the task that taxed memory only minimally (since each stim- a trend in the reverse direction (see Table 2). ulus pointed to its correct response), but required inhibition on incompatible trials and task-switching when subjects used two 22.3. Correlations between performance on hierarchical rules (though one superordinate rule could be used memory-demanding and inhibition-demanding conditions instead). If working memory and inhibition are independent then one 22.2. Difference-score analyses of the relative costs of might expect little relation between performance on the two- increasing memory or inhibitory demands Abstracts-Shapes condition (that requires little or no inhibition) and the Pictures test (that requires little or no memory), despite For the youngest children (4-5 years old), inhibitory their relatively equivalent levels of difficulty judging by sub- demands even in steady-state, took a greater toll on RT than did jects' performance. Contrary to our predictions, performance memory demands. Their RT difference on Dots-Incongruent on the two was highly correlated, even after including age in versus Dots-Congruent (which differed only in their inhibitory the partial correlation analyses. This was especially true for requirements) was greater than their RT difference on two- speed of responding. The correlation between the two-Abstracts- versus six-Abstract-Shapes (which differed only in their Shapes condition and the Pictures test, controlling for age, was memory requirements; subjects 4-5 years old: 1[591=2.33, roughly twice as high for speed as it was for accuracy (all sub- p <0.03; see Table 2). That was very surprising because the jects: 4'3131=0.37 [accuracy] and 0.82 [RT]; only those ≥6 inhibitory demand in Dots-Incongruent feels rather minimal years: r[2231=0.32 [accuracy' and 0.67 IRT]; only those <6 to adults while the memory demand in six-Abstract-Shapes years: r[90] = 0.44 [accuracy] and 0.65 [RT]; all significant at feels quite substantial. For older children ≥8 years and adults' p <0.0001). memory took a greater toll on RT than did inhibition (subjects Similarly, if working memory and inhibition are independent ≥8 years: Dots-Incongruent minus Dots-Congruent versus six- then one might expect little relation between performance on Abstract-Shapes minus two-Abstract-Shapes, within-subject the six-Abstracts-Shapes condition (which heavily taxed mem- 4163] = 5.32, p <0.0001). The individual age group for which ory but required little or no inhibition) and the Arrows task EFTA01098906 2066 MC. Davidson a at /Neumpsychologia 44 (2006)2037-2078 (which taxed inhibition, but required little memory), even though Contrary to our prediction, they were correlated 0.74 for speed both presented relatively equivalent levels of difficulty judging and 0.30 for accuracy. Subjects who were better at exercising by subjects' performance. The partial correlations were high, inhibition also tended to be better at holding information in though not quite as high as those for two-Abstract-Shapes and mind and this was especially true for how fast they could Pictures, and were again higher for RT than for accuracy (all execute their responses. subjects: 4313] =0.30 [accuracy] and 0.74 MTh only those >6 We also predicted that differences in inhibitory demands years: r[2231=0.31 [accuracy] and 0.45 [RT]; all four signifi- would matter more for young children and differences in mem- cant at p <0.0001; only those ≤6 years: 489] = 0.24, p < 0.01 ory demands would matter more for young adults. Findings [accuracy, and 0.43. p <0.001 [M]). consistent with that prediction include that the difference in accuracy between Arrows versus Dots-Congruent (which differ 23. Discussion: comparison of performance on the in inhibitory demands) was larger for younger children than the Abstract Shapes test and the other tests accuracy difference between two- versus six-Abstract-Shapes (which differ in their memory demands), with a trend in the We predicted that the most difficult condition at all ages reverse direction being found for adults. Other evidence that would be the one that taxes inhibition and memory in a switch- inhibition appears to have been more difficult for younger chil- ing context (Dots-Mixed), and that at all ages that would be dren than holding information in mind can be seen in the greater even more difficult than holding more information in mind (six RT cost exacted by inhibitory demands even in steady-state than rules) but without an inhibition or switching component (six- memory demands for the youngest children (4-5 years old). For Abstract-Shapes). Consistent with our prediction, we found that example, the RT difference on Dots-Incongruent versus Dots- Dots-Mixed was indeed the hardest condition for participants Congruent (which differ only in their inhibitory demands) was of all ages and showed the longest developmental progression. greater for the youngest children than theirRT difference on two- At every age, even holding six arbitrary associations in mind versus six-Abstract-Shapes (which differ only in their memory between responses and stimuli that did not easily lend them- requirements). Consistent with our prediction that this would selves to verbal labels was easier than holding two superordinate reverse with age, beginning at age 10 years, increased memory rules in mind and switching randomly between the rule for mak- demands (six versus two rules) took a greater toll on accu- ing a prepotent response and the rule for inhibiting that to make racy than did consistently inhibiting the tendency to respond on the opposite response. Even young adults found Dots-Mixed to the same side as the stimulus (Dots-Incongruent versus Dots- be the most difficult condition and their performance was worse Congruent). there than on even six-Abstract-Shapes. Accuracy did not differ between Dots-Mixed and six-Abstract-Shapes, but beginning as 24. Results: effect of greater presentation time early as 5 years of age, participants could perform at that accu- racy level in six-Abstract-Shapes going at a faster pace than Comparison of the performance of the two groups of 6- they could in Dots-Mixed and at the cost of fewer anticipatory year-olds (tested with the two presentation times) showed that response errors. children of 6 years were able to get significantly more responses Because we hypothesized that memory and inhibition are correct, and made significantly fewer anticipatory responses, independent functions, we had predicted that performance on when they had more time to view the stimuli and compute their the memory-alone conditions of the Abstract Shapes task would responses. not be highly correlated with performance on conditions that Often, the 6-year-olds allowed 2.5 s to view each stimulus primarily taxed only inhibition. We had predicted this would and up to 3 s to compute their responses performed at the level be true both for quite easy and very difficult tasks matched on of children 3-4 years older (children 9-10 years of age) who difficulty. Our prediction of weak correlations between condi- were given only 0.75 s to view each stimulus and up to 1.25 s to tions that primarily taxed working memory and conditions that compute their responses. This was true for the size of the spatial primarily taxed inhibition, matching those conditions on diffi- incompatibility effect on accuracy in the Pictures task (Fig. 3A), culty, was not confirmed. Our basic Simon task (Pictures) with the size of the spatial incompatibility effect on accuracy aver- visible memory aids taxed inhibition while placing little or no aged over the Pictures, Arrows, and Dots tasks (Fig. I IA), the demand on memory. We predicted it would be roughly as easy as incidence of anticipatory responses in the Arrows test (Fig. I I C), remembering the rules for two-Abstract-Shapes when no inhibi- percentage of correct responses in the Incongruent and Mixed tion was required (our two-Abstract-Shapes condition) but that blocks of the Dots test (Fig. 3A), and the incidence of anticipa- performance on Pictures and two-Abstract-Shapes would not tory responses in the Mixed block of the Dots test (Fig. 3C). be highly correlated. As predicted, these two conditions were Indeed, on some measures the 6-year-olds given more time indeed roughly matched in difficulty and both relatively easy. to view the stimuli and compute their answers performed better However, contrary to our prediction, their correlation was 0.82 than children of even I 1 or 13 years tested with shorter stimu- for speed and 0.37 for accuracy. lus presentation times. The accuracy of 6-year-olds tested with Similarly, the six-Abstracts-Shapes condition (which the longer stimulus presentation time was comparable to that of required memory but little or no inhibition) and the Arrows I 3-year-olds given less time on the Pictures test (Fig. 11A), the test (which required inhibition but little or no memory) were Arrows test (Fig. 1 IA), and six-Abstract-Shapes test (Fig. I 3A). roughly matched in difficulty and both relatively difficult. In the interaction of rule switching and response switching on EFTA01098907 ALC. Davidson a al. /Neuropsychologia 44 (2006)2037-2078 2067 accuracy in the Arrows and in the Dots test, the 6-year-olds in six-Abstract-Shapes feels quite substantial. However, this is given more time showed a pattern that more closely approx- fully consistent with greater costs being exacted by inhibitory imated that seen in adults than did children of any other age demands even in steady-state than memory demands for the (Figs. 9A and 10A). The accuracy of these 6-year-olds in remem- youngest children. Also consistent with this prediction, we found bering six arbitrary stimulus-response associations compared that the spatial incompatibility effect (the cost of inhibiting the with their accuracy in remembering only two showed less of a pull to respond on the same side as the stimulus) was greater the cost than did that of even young adults (Fig. 12A). younger the children. This suggests that the younger the chil- In most cases the 6-year-olds tested at the shorter stimulus dren, the harder it was for them to muster inhibition at either presentation times performed roughly comparable to the chil- (a) the level of attention to disregard an irrelevant aspect of the dren of 4 and 5 years given more time. The one and only place stimulus (its spatial location) and/or (b) the level of response to that no advantage was found for giving 6-year-olds more time override the prepotent tendency to respond on the same side as and no cost was found for giving them less time was on the the stimulus. two-Abstract-Shapes task, where time to view the stimuli and We looked at the spatial incompatibility effect in the context determine their response had no effect on performance. of lower- and higher-order rules, different memory loads, and in the context of task-switching. While in the Pictures test, the typi- 25. General discussion cal low-level rules pertaining to individual stimuli were used and memory demands were minimized by the use of icons over the We investigated the development, and interactions overdevel- response-sites, we also investigated the spatial-incompatibility opment, of inhibitory control, memory, and task switching. effect in hybrid, conceptual tasks (Arrows and Dots) where the Our computerized battery included tasks designed to manip- rules were more abstract, spatial location had to be integrated ulate demands on retaining, and working with, information with stimulus identity, and no icons were provided to remind held in mind and/or inhibition, independently and together, in subjects of the stimulus—response mappings. The working mem- single-task and in task-switching contexts. The ability to inhibit ory requirements were greater for these later tasks because they attention to distractors makes possible selective and sustained required mental computation to determine the correct response. attention. The ability to inhibit a strong behavioral inclination Instead of the rule being "for A press left" (the typical rule on helps make change possible, as well as social politeness. Inhibi- Simon tasks, which requires attending only to the stimulus or a tion, thus, provides us a measure of control over our attention and particular property of the stimulus), the rule for the Arrows and actions. External stimuli and engrained behavioral tendencies Dots tasks was "for A press on the side opposite A:' Knowledge exert strong influences on our behavior, but inhibition permits of only which stimulus appeared or only where it appeared was us the possibility to act otherwise. The ability to hold and manip- insufficient on these tasks; those two pieces of information had ulate information in mind makes it possible for us to remember to be integrated. The Arrows and Dots tasks differed from each our plans and others' instructions, relate one thing to another, other in that the memory demands were minimal on the Arrows including relating the present to the future and the past, and to task because the stimulus pointed to the correct response on each act on the basis of information not perceptually present. Cogni- trial. tive flexibility is critical in a changing world. It is essential for In the Pictures task spatial incompatibility effects were signif- adaptability and for the creativity that comes from being able to icant for both RT and accuracy for children of all ages, produced see things in new or different ways. the greatest effect on RT, and decreased more in size as a function of age than on the Arrows or Dots tasks. Even at 13 years of age, 25.1. How our predictions concerning memory and children still showed a significantly greater Simon effect on the inhibitionfared Pictures task than did young adults. The spatial incompatibility effect was weakest and showed the least change over age in the We predicted that inhibitory demands would account for a Dots task, even though in the Dots and Arrows tasks the spatial greater proportion of the variance in children's performance location of the stimuli had to be explicitly taken into account than in adults, and the more so the younger the child. Con- and on the Pictures task it did not. The Dots-Mixed condition sistent with our prediction, Dots-Incongruent (where inhibition was the only task in which the spatial incompatibility effect was of the spatially-compatible response was required on all trials) not evident in both RT and accuracy. In Dots-Mixed, the spatial was more difficult than Dots-Congruent (where the spatially- incompatibility effect was evident only in RI', its effect on RT compatible response was correct on all trials) and the more was weaker for the youngest children (4-6 years old) and adults so the younger the children. Accuracy and impulsivity dif- than in the Arrows or Pictures tests, and the size of its effect on ferences between those two conditions decreased over age. RT did not change significantly over age. The Arrows condition The RI' difference on Dots-Incongruent versus Dots-Congruent produced a significant spatial incompatibility effect on both RT (which differed only in their inhibitory demands) was greater and accuracy, and a decrease in the size of the effect on accuracy for the youngest children than the RT difference on two- ver- (though not on 121) over age. These results — a stronger spatial sus six-Abstract-Shapes (which differed only in their memory incompatibility effect the easier the task — are consistent with requirements). This finding surprises adults who have taken our results in the literature showing that this effect decreases as a task battery because for adults the inhibitory demand in Dots- function of task difficulty (Hommel, 1993, 1994; Vu & Proctor, Incongruent feels rather minimal while the memory demand 2004). However, those results have previously been interpreted EFTA01098908 2068 MC. Davidson a at /Neumpsychologia 44 (2006)2037-2078 to mean that anything that increases response time will decrease rules are, for example, "If the stimulus is an animal, press right; the Simon effect (giving the automatic activation of the irrele- if the stimulus is a vehicle, press left"? Here, the same number vant stimulus location information time to decay). In contrast, of mental steps ([1] Is the stimulus an animal or a vehicle? [2] we found that although younger children took much longer to Where do I press for that?) would be required as in our Dots respond than older children, they showed a larger spatial incom- test ([ Which rule [same side or opposite side] pertains to this patibility effect. stimulus? [2] Which side is the stimulus on?). That would be a The larger Simon effect found for the younger children might conceptual task with higher-order rules (like our Dots task), but indicate that their ability to exercise inhibition of the pull to unlike our Dots task it would be a pure Simon task, not a hybrid. respond on the same side as the stimulus was weaker than that Similarly, Wascher and colleagues (Wascher, Schatz, Kuder, of older children. It may have also been affected by the greater & Verleger, 2001; Wascher & Wolber, 2004; Wiegand & likelihood of younger children to use verbal mediation. Though Wascher, 2005) provide evidence for the involvement of both the Pictures task could be solved by simple perceptual matching, visuomotor and cognitive mechanisms in Simon task perfor- some younger children named the stimulus out loud on some tri- mance. Modifying the task slightly (e.g., presenting vertical als. Similarly, on the Dots task, younger children often called out rather than horizontal stimuli) changed the distributions of RT the rule ("same" or "different") on trials in the Mixed condition. scores, and was taken as evidence for involvement of a cognitive In adults, the Simon effect is stronger, and does not diminish with component. An interesting developmental question is, "Does the consecutive incompatible trials, when the stimulus or response effect function across Simon task variants change over age and has a verbal property (Proctor & Vu, 2002). if so in what ways and why?" Our manipulations (increasing The lack of an accuracy cost on spatially incompatible (Incon- or decreasing the working memory load) might also change the gruent) trials in the Mixed block of the Dots task is in sharp con- distribution of RT scores but, unfortunately, the vincentization trast to the results when comparing separate blocks of Congruent procedure used by Wascher and colleagues involves splitting the and Incongruent trials on the task, which showed a significant RT distributions into quintiles (i.e., quartiles or even deciles) and spatial incompatibility effect for children of all ages in both would result in relatively few datapoints per bin in the current speed and accuracy, though not for adults. The lack of a spatial study. Future research with these tasks with more data per subject incompatibility effect for accuracy within the Dots-Mixed con- could allow more thorough investigation of the RT distributions dition, in conjunction with the increase in RT and reduction in across development. accuracy for all trials in that condition relative to the other two In young adults, in whom inhibitory control is more mature, Dots conditions, may indicate that participants exercised inhibi- we had predicted that memory demands would exact a greater tion on both Congruent and Incongruent trials when those trials cost than inhibitory demands. Beginning at 10 years, increased were intermixed in a difficult task-switching context. It may also memory demands (six versus two rules) took a greater toll be due to an order effect (see below). on accuracy than did consistently inhibiting the tendency to The results of Crone et al. (in press), with a task similar to respond on the same side as the stimulus (Dots-Incongruent our Dots task and with subjects ages 7-8, 10-I 2, and 20-25 versus Dots-Congruent)—the opposite of the pattern observed years, are similar to ours. Like us, they found a stronger Simon in the youngest children. Also in line with the greater role effect comparing across the single-task blocks than in the Mixed of memory in what is demanding for adults, taxing cognitive block, but unlike us they found that the Simon effect disappeared flexibility and inhibition in a switching context was not that altogether in the Mixed block. Like us, they found the single-task hard for adults if memory demands were minimal. Unlike block Simon effect to be significant for both accuracy and RT, the case for children of all ages in our study, the Arrows but unlike us they found no change in the size of the accuracy test was easy for adults. However, the Dots-Mixed condition, or RT Simon effects over age. which presented the same demands on inhibition and cognitive Our study of spatial compatibility effects with conceptual flexibility as did the Arrows test but in addition taxed working rules and switching between those rules, as well as stimulus- memory more, was difficult even for adults. The difference in level S—R associations, deserves additional follow-up. Our Dots accuracy on Dots-Mixed and Arrows, which differed only in and Arrows tests required integrating stimulus-appearance infor- the greater working memory demands in the Dots condition, mation with spatial location information. Since spatial location was greater for young adults than for children at any age from 7 was relevant to the correct response they were not pure Simon to 13 years. The greater memory demands in the Dots condition tasks, but hybrid spatial incompatibility tasks. There is evidence made a big difference for adults and the youngest children, but that the neural bases for working memory of object-appearance not for the majority of children (aged 7-13 years). and spatial-location information are somewhat different (Haxby, Since we are discussing performance on Dots-Incongruent Petit, Ungerleider, & Courtney, 2000; Levy & Goldman-Rakic, and Dots-Congruent here, it is appropriate to note that one of 2000; Mecklinger & Mueller, 19%) and spatial location is the striking differences between the results for children and exactly the stimulus property that feeds the spatial-compatibility adults was that effects elicited only in Mixed blocks (e.g., in bias. In standard Simon tasks, subjects would perform better if Dots-Mixed) with adults were found in children even in single- they could (theoretically) screen out the location of the stimulus, task blocks (e.g., Dots-Incongruent versus Dots-Congruent but on our Dots and Arrows tasks information about the location Blocks). For example, even though switching between rules of the stimulus is critical for determining the correct response. that require inhibiting a prepotent response or making it is what Will results be similar to those on our Dots test, if the conceptual was most difficult at all ages, even inhibition in steady state EFTA01098909 M.C. Davidson a al. /Neuromrhologia 44 (2006)2037-2078 2069 (Dots-Incongruent) was more difficult for children than going to the Incongruent (more difficult) condition; opposite to the with their prepotent response on every trial (Dots-Congruent). pattern typically reported in adults. We predicted that after that At every age, without exception, children were slower and less early period, we would see greater switch costs at all ages for accurate in the Dots-Incongruent block than the Dots-Congruent switching to the easier (Congruent) condition than to the harder Block. Thus, inhibition, even in steady state, was sufficiently dif- (Incongruent) condition (consistent with the asymmetric switch ficult for children to elicit a cost in their performance. This was costs previously reported in adults ]Allport & Wylie, 2000: not true for adults. Adults performed comparably in both speed Allport et al., 1994; De Jong, 1995; Kleinsorge & Heuer, 1999: and accuracy in the Dots-Congruent and Dots-Incongruent Los, 1996 and Stoffels, 1996; Wylie & Allport, 2000]). Further, conditions. They required having to switch between the two for intermediate-age children, who are beginning to exercise bet- conditions for a significant effect on their performance to be ter inhibitory control, we reasoned that doing so should require evident. greater effort than in older participants and so predicted that These results for adults are consistent with those reported in undoing that inhibition (switching back to making the dominant other studies of the spatial incompatibility effect in adults. As we response) should exact a greater cost in those children than in found here, adults tend not to show the spatial incompatibility adults. Thus, we predicted that beginning after 6 or 7 years of effect if Congruent and Incongruent trials are administered in age, asymmetric switch costs would be larger in younger than separate, single-task blocks (Stunner et al., 2002; Valle-Inclan older participants. These predictions were confirmed for RT On et al., 2002; Verbruggen et al., 2005; Wiihr, 2004,2005). Adults both Arrows and Dots-Mixed, children of 4-6 years showed a evidently re-set their default response if several trials in a row greater Kr cost for switching to the Incongruent rule than the are Incongruent and so are no slower on those Incongruent trials Congruent one. In both of those conditions, adults and older than on Congruent ones. Indeed, adults can show a reverse spa- children showed a greater RT switch cost for switching to the tial incompatibility effect when switching back to responding Congruent than the Incongruent rule (consistent with previous on the same side as a stimulus after several trials of respond- reports of asymmetric switch costs). In both Arrows and Dots- ing on the side opposite the stimulus (e.g., Logan & Zbrodoff, Mixed, the differential RT cost of switching to the Congruent 1979). Children from 4 to 13 years of age, on the other hand, rather than the Incongruent rule was largest at 7-10 years of evidently did not re-set their default response. They showed the age. For accuracy, on the other hand, across the age spectrum spatial incompatibility effect with Congruent and Incongruent on both Arrows and Dots-Mixed, people were more accurate trials administered in separate, single-task blocks. Inhibition in when both the rule and the response changed than when just the steady state took a toll on the performance of children even as rule changed. Thus the asymmetric switch costs reported in the old as 13 years, but not on that of adults. literature primarily for RT, were found here for accuracy at all Note that participants (including adults) in our study showed ages (even among the youngest children). a spatial compatibility effect in Dots-Mixed although all sub- Crone et al. (in press) found results for Kr that resemble jects were tested on Dots-Incongruent immediately before that. ours for accuracy and results for accuracy that resemble ours for Just as adults typically show little cost in making spatially- RT. Across their age spectrum (7-23 years), they found faster incompatible responses if they are tested on a block where they responses when both the rule and the response changed than need to do that on every trial or most trials, Tagliabue, Zorzi, when just the rule switched. Across our age spectrum (4-45 Umilta, and Bassignani (2000) found no Simon effect in Mixed years), we found a greater percentage of correct responses when blocks when adults were given a run of Incongruent trials before- both the rule and the response changed than when just the rule hand. Given those results it is possible that we might have found switched. We found this effect on accuracy to be largest at a much stronger spatial incompatibility effect in Dots-Mixed if 8-10 years; they found this effect on KI' to be largest at 7—8 it had been preceded by a block of Congruent trials. In order years. Mirroring in reverse these greater effects in young chil- to test participants at all ages under the same conditions we dren versus older children and adults, Mayr (2001) found the RT had not varied task order, so further study would be needed to effect (faster rule switching when the response also changed) to address that interesting possibility. It is also possible that other be greater in older versus younger adults. Crone et al. found effects observed here might appear different if the tasks were that accuracy costs did not show the pattern they expected; administered in a different order. For example, Dots-Incongruent more errors occurred on switch trials when the response-site was always tested after Dots-Congruent. We do not think that also changed than when it remained the same. We similarly caused a difference in the memory demands between the two found slower responding on rule-switch, response-switch tri- conditions because the rule for Dots-Incongruent was taught als until age 13 on our Dots-Mixed task which resembles Crone and practiced immediately before that block just as the rule for et al.'s (this encompasses two of Crone et al.'s three age groups), Dots-Congruent was taught and practiced immediately before but unlike Crone et al.'s results for accuracy, we found faster that block, but given that we did not vary task order we cannot responding when both rule and response switched for adults. prove that that is the case We had predicted that even the youngest children would find Based on our hypothesis that inhibitory control is extremely it easy to hold two rules in mind and that although it would be problematic for very young children, we predicted they would harder for everyone to hold more rules in mind than fewer, the perform poorly on all trials requiring inhibition (Incongruent relative difficulty of this would not change over age. Indeed, as trials and switch trials) and that those effects would be additive. predicted, holding two arbitrary rules in mind was easy even for That is, we predicted they would perform worst on switching our youngest participants. At all ages performance was excellent EFTA01098910 2070 MC. Davidson a at /Neumpsychologia 44 (2006)2037-2078 on the two-Abstracts-Shapes and Dots-Congruent conditions These findings concerning not taking the time they needed (the latter requiring holding a superordinate rule in mind and are fully consistent with results reported by Gerstadt, Diamond, mentally translating that into the appropriate embedded rule and Hong (1994) and Diamond, Kirkham, and Amso (2002) on on each trial). This is consistent with other evidence that chil- a different task, the Day—Night Stroop-like task, where chil- dren can hold two conditional rules in mind by 4(1/2 )-5 years of dren had to say "night" to a daytime image and "day" to a age (Campione & Brown, 1974; Doan & Cooper, 1971; Gollin, nighttime image. Gerstadt et al. (1994) found that: (a) those chil- 1964, 1965, 1966; Gollin & Liss, 1962; Osier & Kofsky, 1965; dren of 3(1/2 )-4(1/2 ) years who took more time to compute their Shepard, 1957). Although everyone found the increased memory answers were able to answer correctly on more trials than chil- load (six versus two Abstract Shapes) more difficult, the size of dren who answered more quickly and (b) within child, on those the effect changed little over age in either speed or accuracy trials where a child of 3(1/2 )-4(1/2 ) years took longer to respond, when all subjects were included in the analyses. More fine- the child was more likely to be correct. Diamond et al. (2002) grained analyses, however, showed that how the difference in manipulated time to view the stimulus and compute the response difficulty was handled differed over age. The speed-accuracy by chanting a ditty to the child either after the stimulus was pre- tradeoff changed over age. Accuracy on six-shapes more closely sented but before the child could respond or between trials before approximated that on two-shapes the older the subjects. RT the stimulus was presented. Diamond et al. found that 4- and in the two conditions, however, diverged more the older the 4(1/2 )-year-old children were correct on significantly more trials subjects. in the manipulation that gave them more stimulus-viewing and Across conditions, older participants slowed down to pre- more response-computation time (ditty chanted while stimulus serve their accuracy on more difficult trials. Thus they showed was visible) but performed no better than in the basic condition sizeable KI' differences and small accuracy differences. An ele- when the extra time could be used to remind themselves of the gant analysis of this tendency of adults to alter their response rules but not to instantiate the correct rule for the current trial times to preserve a constant level of accuracy in the face of vari- (ditty chanted between trials, before stimulus was visible). ations in task difficulty is provided by Usher and McClelland On the other hand, we have evidence here that if a task is (2001) and Usher, Olami, and McClelland (2002). In our study, sufficiently easy that 4-5-years-olds can compute the answer younger children often showed less of a change in speed and in roughly a second, they will modulate their speed to preserve hence showed very large differences in accuracy across trials of their accuracy. On the Pictures test, for example, children of differing difficulty. For example, older participants were better 4-5 years slowed way down on Incongruent trials relative to able to modulate their speed and slowed down in the difficult Congruent ones, thereby preserving their accuracy so that the Dots-Mixed condition relative to the easy Dots-Congruent con- difference in their accuracy on Incongruent versus Congruent dition to minimize a reduction in accuracy. Younger participants trials was smaller than that seen by older children of 6-8 years (even those given a very long time to respond) kept their response given less time to compute their responses (see Fig. 3). Simi- speed more constant across conditions, perhaps because they larly, on the Arrows test, children of 4-5 years used the ample were too impulsive to take more time when they needed it, time allowed them to maintain an accuracy level of over 80%, at the cost of accuracy in the difficult conditions. Hence, for a level of accuracy not seen when given less time to respond example, the accuracy difference between the Dots-Mixed and until children were 10-11 years old. Children of 4-6 years also Dots-Congruent conditions decreased with age but RT differ- showed smaller local switch costs on the Arrows test than did the ences between those two conditions increased over age (see older children; they achieved this by using their allotted time to Fig. 6). Similarly, the mixing cost of Congruent and Incongru- slow down on the switch trials; their KI' switch costs were over ent trials being mixed together declined over age for accuracy twice those of participants at any other age. but increased over age for RT (see Fig. 7). Likewise over age, Certainly there is considerable evidence that 6-year-olds ben- differences in performance on the six- and two-Abstract-Shapes efited from having a longer time to respond (3000 ms versus conditions declined in accuracy but increased in RT. 1250ms). The results clearly show that by 6 years, if allowed The very youngest children (4-5 years of age) were given a more time to respond, children will take advantage of that to long time to respond (3000 ms) so it is unlikely that they lacked reduce their errors. sufficient time to modulate their response speed. It is more likely The relative lack of response speed modulation in children that they had difficulty inhibiting impulsive responding, i.e., of 6-8 years tested in the adult condition probably had a dif- difficulty withholding their response long enough to take the ferent cause than that for children of 4-5. In the case of the time they really needed. For instance, the RTs for children of 6—8-year-olds, the response window (1250 ms ISI; 750 ms stim- 4-5 years on nonswitch Incongruent trials in the Dots-Mixed ulus presentation) was likely too brief to allow them the time they condition differed little from their RTs in the easier single-task needed to slow down sufficiently in the more difficult conditions Dots-Incongruent block. This was true even though their KI's to preserve their accuracy. Thus, even on the easy Pictures task, in both conditions were on average less than half of the time they could not slow down sufficiently on incompatible trials to allotted, so they had time to compute their responses but did preserve their accuracy, and so although their KI's were longer not make use of that extra time. Their inhibitory problems can on incompatible trials, their accuracy suffered on those trials also be seen in their greater likelihood to respond impulsively more than was found at any other age. before a stimulus appeared and to fail to promptly stop pressing Given our hypothesis that working memory and inhibition a response button after responding. are independent, we had predicted that performance on tasks EFTA01098911 M.C. Davidson a al. /Neuropsychologia 44 (2006)2037-2078 2071 that tax primarily memory or primarily inhibition would not remained strong throughout the age range for children, though be highly correlated. Instead, when the tasks were matched for the accuracy difference disappeared among young adults. This difficulty, speed on working memory and inhibition tasks was may suggest that much of the age-related reduction in the cost highly correlated. Individuals who were fast at exercising inhi- of exercising cognitive flexibility comes after 13 years of age. bition also tended to be fast on working memory measures, even Consistent with the "all or none" principle (Diamond, in after accounting for age effects. Accuracy across working mem- preparation), it should be easier to inhibit a dominant response all ory and inhibition measures was also correlated, though not as the time than only some of the time. We thus predicted that per- strongly. formance at all ages would be better in Dots-Incongruent (where Finally, we had predicted that the most difficult condition inhibition was consistently required on all trials) than in Mixed at all ages would be the one that taxed inhibition and memory blocks of Dots or Arrows (where inhibition is only required on in a switching context (Dots-Mixed) and that that would even the 50% of trials that are Incongruent), and that this difference be more difficult than having to hold much more information would be greater the younger the children. Indeed, we found that in mind but with no inhibition or switching component (six- inhibiting the spatially-compatible response some of the time Abstract-Shapes). Indeed, as predicted, we found that at every in a switching context despite minimal memory requirements age, including for young adults, holding two superordinate (the Arrows task) took a greater toll on speed and accuracy at rules in mind and switching randomly between the rule for every age than did inhibiting the spatially-compatible response making a prepotent response and the rule for inhibiting that to consistently on all trials (Dots-Incongruent). Not surprisingly, make the opposite response (Dots-Mixed) was the most difficult the differences were even larger between Dots-Mixed and Dots- condition, harder even than holding six arbitrary rules in mind Incongruent. Our prediction that the difference in performance for stimuli that did not easily lend themselves to verbal labels between Dots-Mixed and Dots-Incongruent would decrease over (six-Abstract-Shapes). age as cognitive flexibility improved fared less well as these differences remained large at all ages, though the difference 212. How our predictions concerning cognitive flexibility in accuracy was larger the younger the children. Similarly, the and task switching fared accuracy difference between Dots-Incongruent and Arrows was smaller in adults than in children, but otherwise the markedly Cognitive flexibility (switching, overcoming inertial tenden- better performance on Dots-Incongruent than on Arrows was cies) was far harder than consistent inhibition in steady state or equally true across all ages. than holding and manipulating a couple of items in mind, and Also consistent with the "all or none" principle is that perfor- showed a much longer developmental progression. The cost, and mance should be better on not-switching anything (repeat-rule, longer developmental progression, of cognitive flexibility can be repeat-response trials) and on switching everything (switch tri- seen most clearly on the Arrows test, where little or no memory als where the response-site also switches) than on trials where was required as the arrow pointed to the correct response on either the rule or response-site changes but not the other. We every trial. Since we hypothesized that switching is so difficult, had predicted that these effects, heretofore documented only in we had predicted that having to switch between tasks even when adults and older children (Kleinsorge, 1999; Meiran, 2000a,b; memory demands were minimized (as in the Arrows test) would Rogers & Monsell, 1995; Schuch & Koch, 2004), would also show a long developmental progression. This was confirmed. be found in young children. We predicted that throughout our Even by age 10, the percentage of correct responses did not age span, participants would do better at switching tasks if the exceed 80% on the Arrows test, and even by the age of 13, chil- response-site also changed and would be slower and less accu- dren were not yet performing at adult levels on the Arrows task. rate on switch trials when the response-site remained the same At all ages, the RT costs for having to exercise inhibition as on the previous trial. In both Dots-Mixed and Arrows, older in a switching context on the Arrows task versus not having to children and adults were indeed better at switching tasks if the exercise inhibition or switch (Dots-Congruent condition) took a response-site also changed than if the response-site remained much greater toll onresponse speed than did increasing the mem- the same as on the preceding trial. However, contrary to our ory load from two to six arbitrary rules However, consistent with prediction, the youngest children (children of 4-8 years) per- inhibition in a switching context being disproportionately dif- formed better on switch trials where the response-site remained ficult for young children and memory being disproportionately the same. They were faster on switch than nonswitch trials and difficult for adults, even young adults, the larger accuracy dif- on response-switch rather than response-stay trials and those ference between Arrows versus Dots-Congruent than between effects tended to be additive. These results raise the possibility two- versus six-Abstract-Shapes was found only for the younger that perhaps the hypothesized "all or none" default of cogni- two-thirds of the subjects (children ≤9 years). tive systems is an efficient characteristic of the mature cognitive Consistent with cognitive flexibility improving with age, system. It appears, in this particular context anyway, that piece- performance differences on Dots-Incongruent and Dots-Mixed meal, additive effects are more characteristic of young children's decreased over age. If cognitive flexibility is improving, performance. however, one would also expect the difference in perfor- Research in adults has shown that performance on nonswitch mance on Dots-Incongruent and Arrows to decrease over age. trials (where the rule remains the same as on the previous trial) The markedly faster speed and better accuracy in the Dots- is worse when these trials are presented in the context of peri- Incongruent condition compared with the Arrows test, however, odically having to switch than in the context of a block of all EFTA01098912 2072 MC. Davidson a /Neumps)rhologia 44 (2006)2037-2078 nonswitch trials (e.g., Fagot, 1994; Mayr, 2000a). We found that Cepeda et al. (2001) calculated global switch costs differ- indeed performance was worse-slower, less accurate, and char- ently from the studies above. They compared performance on acterized by more anticipatory errors-on nonswitch trials within all trials in Mixed blocks (not just the nonswitch trials) to perfor- the Mixed block of the Dots task than within either single-task mance in single-task blocks. Just as we found that the difference block of the task. Such global switch costs were among the in performance on Dots-Mixed versus Dots-Congruent or Dots- strongest effects found in this study. It is not that participants Incongruent decreased over age from 6 years to young adult- forgot the rules when they had to hold both the Congruent and hood, so too Cepeda and colleagues found that the difference Incongruent rules in mind for the Mixed block. Indeed, chil- in performance in their Mixed blocks versus their single-task dren often called out the correct higher-order rule on trials in the blocks decreased from their youngest age (7 years) to young Mixed condition (e.g., "same," "opposite," "opposite," "same") adulthood. Cohen et al. (2001) similarly report a linear decline in even as they were making errors. The problem seemed to be in the difference in performance in Mixed blocks versus single-task quickly translating that rule into the correct response. The pres- blocks for both speed and accuracy, with their oldest children ence of global switch costs at all ages in our study is consistent (age I 1 years) still showing a larger difference in both dependent with task-switching studies in children, young adults, and older measures than young adults. adults; all studies consistently find global switch costs through- This illustrates an important point. Inhibiting a dominant out the age spectrum. response requires effort, but it is not nearly as difficult if that We had predicted that global switch costs would decrease inhibition needs to be consistently maintained (as in Dots- over age. That prediction was only partially confirmed. The Incongruent). What is far more demanding is switching back global switch cost in accuracy declined from 9 to 13 years of and forth between sometimes inhibiting a dominant response age, as predicted, but the global switch cost in RT increased from and sometimes making it. What is truly difficult is overcom- 6 years through early adulthood (see Fig. 7). Adults adjusted ing one's inertial tendency to continue in the same mindset, their speed to preserve their accuracy; younger children did that switching between one mental set and another. Even now many much less hence the difference in the speed-accuracy trade-off investigators still administer the Stroop task in single-task blocks with age. This mirrors exactly what was found by Cohen et (blocks of always reading the word and blocks of always nam- al. (2001) using a very different task-switching paradigm. They ing the ink color). While it requires effort to focus on the ink used Meiran's (1996) task-switching paradigm presented as a color (and one can see that toll in slowed responding) one can computer game. A smiley face appeared at one of four quadrants get in the mode of always focusing on the ink color and the of a square, preceded by a cue indicating the relevant dimension task is quite manageable. It is far harder not to be able to rely (horizontal ["is the cue in the left or right half?"[ or vertical ["is on always ignoring the word; to have to switch back and forth the cue in the top or bottom bairn). This was administered to between sometimes reading the word and sometimes naming the 150 children (ages 5-11 years) and 16 young adults. They found ink color. that global switch costs in accuracy decreased from 5 to II years, Because of floor effects (people are already slower and more and even 11-year-olds were not as accurate in mixed blocks as error-prone in the Incongruent-only block), we predicted that young adults, but global switch costs in speed of responding the effect of context (the Mixed block versus single-task block) increased over age (just as we found here). would be greater on Congruent than Incongruent trials. We fur- Contrary to our findings, however, though with only a few ther predicted that this should be more evident the younger the overlapping ages, Reimers and Maylor (2005) found that global child. That is, we predicted that the younger the child, the closer RT switch costs decreased linearly from 10 to 18 years. Like performance on "easy" (Congruent, nonswitch) trials would fall us, Crone et al. (in press) found significant global switch costs to the level of "harder" trials in the context of sometimes hav- in both speed and accuracy. However, unlike us, Cohen et al. ing to switch back and forth. Consistent with this prediction, we (2001), or Reimers and Maylor (2005), Crone and colleagues found that the cost of mixing nonswitch trials in with switch found no change in the size of global switch costs with age trials and mixing Congruent trials in with Incongruent ones, in either speed or accuracy from 7 to 8 years to 23 years. versus having single-task blocks, was greater for the Congru- Most studies in older adults report greater global RT switch ent (easier) nonswitch trials than the Incongruent nonswitch costs in elders than in young adults (Kray et al., 2004; Kray & trials. When these trials were administered in separate single- Lindenberger, 2000; Mayr% 2000; van Asselen & Ridderinlchof, task blocks, fewer errors occurred on Congruent trials (except 2000), though Kray et al. (2002) report that no difference in for adults where accuracy did not differ in the Congruent and global RI' switch costs is found between younger and older Incongruent single-task blocks) but when they were intermixed adults when switches between tasks are unpredictable. The dif- within the same block comparable numbers of errors occurred ference between predictable and unpredictable switches would on Congruent and Incongruent trials. Participants were able to also explain the differences across studies in whether global RT respond much faster on nonswitch Congruent trials when all switch costs differ between young children and adults. The only trials in the block were Congruent than when some were Incon- study to find smaller global RT switch costs with increasing gruent, though only Congruent trials following a Congruent trial age from young children to adults was the one study that also were included in these analyses. The same was true for Incon- predictably switched between tasks (Reimers & Maylor, 2005), gruent trials but to a lesser extent. The effect of context (Mixed where a predictable double-alternation switching pattern was block versus single-task block) appears to have been larger for used in the Mixed block. the faster, more automatic response (responding on the same side EFTA01098913 M.C. Davidson a al. /Neuromrhologia 44 (2006)2037-2078 2073 as the stimulus) than for the slower, more demanding response looking straight down or diagonally to the other side. It is likely (inhibiting the dominant response and responding opposite to it). that adults reduced the separate rules for vertical and diagonal However, contrary to the portion of our prediction concerning Arrows to one rule ("Press where the arrow is pointing") but development, the size of the greater effect of context on Congru- that children did not spontaneously do that. We will see what ent versus Incongruent trials did not change significantly over effect explicitly instructing children to code this as one rule has age. on their performance. Some of our adults had a little difficulty The difficulty of the harder condition is underestimated in distinguishing the gray and striped Dots. We will therefore use single-task blocks (always having to respond opposite to the stimuli of different shapes (hearts and flowers) in future testing. side of the stimulus tends to reduce its difficulty because you Among the most important additional work to be done is to get in the mode of doing that) and the ease of the easier condi- complement the work of O'Craven, Davidson, Bergida, Savoy tion is underestimated in Mixed blocks (because people tend to and Diamond (in preparation) on the neural systems recruited slow down across the board on such blocks). Comparing non- for performance of the various conditions tested here in adults switch Incongruent trials in the Dots-Mixed block to nonswitch with neuroimaging studies of the neural systems recruited by Congruent trials in the Dots-Congruent block may come closest children in performance of these conditions and how, and why, to approximating the full difficulty of inhibiting the tendency to that changes over age. Certainly there are important functional make the spatially compatible response. and structural changes in the neural network recruited for cog- A striking difference in our findings for children and adults nitive control and executive functions throughout the age range was that while RT was unquestionably a more sensitive mea- investigated here (Diamond, 2002). Functional changes in the sure than percentage of correct responses for adults, the latter neural basis for cognitive control appear to be characterized by was often the more sensitive measure for children, especially increasingly focal activation during early childhood and then younger children. For example, for our youngest children (4-6 decreasingly intense activation of the focal regions during ado- years of age). age-related improvements in each of the three lescence (e.g., Brown et al., 2005; Casey, Galvan, & Hare, 2005; conditions of the Dots task and in each of the two conditions Durston et al., 2005). of our Abstract Shapes task were far more evident in accuracy Besides characterizing effects (such as local switch costs, than in speed. Age-related improvements in the ability to inhibit global switch costs, and asymmetric switch costs), previously spatially compatible responses were far more evident in reduced studied only in older children or adults, in young children and accuracy differences between Congruent and Incongruent trials throughout a wide age span, some of the most important findings over age than in reduced RT differences in each of the tasks that to come out of this study include the following: tested this (Pictures, Arrows, and Dots; see Fig. 1I). Similarly, Inhibiting the tendency to make a spatially compatible age-related improvements in the ability to hold multiple items in response exacted a greater toll on young children's performance mind were far more evident in the reduced change in accuracy than did memory demands and than it did on older partici- over age for holding six rather than two arbitrary rules in mind pants' performance. The spatial incompatibility effect (the cost than it was in reduced change in RT (see Fig. 12). of inhibiting the pull to respond on the same side as the stimu- lus) was greater the younger the participant. Even at 13 years of 26. Final comments and conclusions age, children still showed a greater Simon effect than did young adults. Because we wanted to include very young children, we did Inhibitory control was sufficiently problematic for very not test our subjects for nearly as many trials as is typically young children that they took especially long on all trials requir- done in studies of adults. We did not discard the trial following ing inhibition (Incongruent trials and switch trials) and those an error (only error trials), contrary to what is often done in RT effects were additive. Thus, they responded slower when switch- analyses with adult subjects, because with young children error ing to the Incongruent rule than the Congruent one; opposite to rates are sufficiently high that to discard the trials immediately the pattern seen in adults. Intermediate-age children, who were after an error as well as error trials would have resulted in the beginning to exercise better inhibitory control, exerted more loss of too many datapoints. We are impressed, however, with the effort to do that than older children and adults and so showed a consistency of our results with those of other published studies heightened cost to undoing that inhibition (a more exaggerated despite procedural differences. version of the asymmetric switch costs seen in adults and older Certainly there are a number of unanswered questions that children here and reported in the literature for adults). could fruitfully be followed up in later studies. We stopped As inhibitory control improved with age, memory demands testing children at 13 years of age, but on a number of our mea- started to exact a greater cost than inhibitory ones. Beginning sures children of 13 were not yet performing at adult levels. at 10 years of age, increased memory demands (holding of six Older children should be tested to better understand the devel- versus two arbitrary, hard-to-verbalize rules) took a greater toll opmental progression between 13 and 26 years. We will test our on accuracy than did consistently inhibiting the tendency to Simon task (the Pictures task) without the posted memory aids respond on the same side as the stimulus (Dots-Incongruent ver- to see how much our having removed the usual memory demand sus Dots-Congruent)--the opposite of the pattern observed in present in most Simon tasks affected results with our paradigm the youngest children. Also in line with the greater role of mem- in children. Arrows may be a more compelling symbol for adults ory in what is demanding for adults, taxing cognitive flexibility than children, so we will re-administer that condition using eyes and inhibition in a switching context was not that hard for adults EFTA01098914 2074 MC. Davidson et at / Neuropsychologia 44 (2006) 2037-2078 if memory demands were minimal. Unlike the case for children Psychiatry, 2255 Wesbrook Mall, Vancouver, BC, Canada V6T of all ages in our study, the Arrows test was easy for adults. 2A1, or via email: adele.diamond@ubc.ca. Given that we hypothesized that working memory and inhi- bition are independent, we had predicted that performance on References tasks that tax primarily memory or primarily inhibition would not be highly correlated. We were wrong. When matched for dif- Allpon. A.. Styles. E. A.. & Hsieh. S. (1994). Shifting intentional set: Explor- ficulty, these tasks were very highly correlated for RT (0.7-0.8) ing the dynamic control of tasks. In C. Umilta & M. Moscovitch (Eds.). and respectably correlated for accuracy (0.3-0.4). Attention and perfonnance XV (pp. 421-452). Cambridge. MA: MIT Cognitive flexibility (switching between rules), even with Press. Allpon. A.. & Wylie. G. (2000). Task switching. stimulus-response bindings. memory demands minimized, showed a long developmental pro- and negative priming. In S. Monett & J. Driver (Eds.). Comm! of cogni- gression. On the Arrows test, where little or no memory was tive processes: Attention and performance XVI! (pp. 35-70). Cambridge. required since the stimulus pointed to the correct response, even MA: MIT Press. by 13 years, children were not yet performing at adult levels. Anderson. M. C.. & Spellman. B. A. (1995). On the status of inhibitory mechanisms in cognition: Memory retrieval as a model case. Psycholog- Global switch costs were among the strongest effects found in ical Review. 102. 68-100. this study. It is not that participants forgot the rules when they Aron. A. R.. Monsell. S.. Sahakian. B. J.. & Robbins. T. W. (2004). A had to hold both the Congruent and Incongruent rules in mind componential analysis of task-switching deficits associated with lesions for the Mixed block. The problem seemed to be in quickly trans- of left and right frontal cortex. Brain. 127. 1561-1573. lating the rules into the correct response. Aron. A. R.. Sahakian. B. J.. & Robbins. T. W. (2003). Distractibil- The most difficult condition at all ages was the one that taxed ity during selection-for-action: Differential deficits in Huntington's dis- ease and following frontal lobe damage. Neumpsychologia. 41. 1137- inhibition and memory in a switching context (Dots-Mixed). At 1147. every age,holding two superordinate rules in mind and switching Band. G. P.. van der Molen. M. W.. Ovenoom. C. C.. & Verbaten. M. randomly between the rule for making a prepotent response and N. (2000). The ability to activate and inhibit speeded responses: Sepa- the rule for inhibiting that to make the opposite response (Dots- rate developmental trends. Journal of Experimental Child Psychology. 75. Mixed) was harder even than holding six arbitrary rules in mind 263-290. Brass. M.. Derrfuss, J.. Forstmann. B.. & von Cramon. D. Y. (2005). The for stimuli that did not easily lend themselves to verbal labels role of the inferior frontal junction area in cognitive control. Trends in (six-Abstract-Shapes). Cognitive Science. 9.314-316. The speed-accuracy tradeoff changed over age. Across con- Brass. M.. Ruge. H.. Mciran. N.. Rubin. O.. Koch. I.. Zysset. S.. et al. (2003). ditions, older participants slowed down on more difficult trials When the same response has different meanings: Recoding the response to preserve their accuracy. Thus they showed large response meaning in the lateral prefrontal cortex. Neuroimage. 20. 1026-1031. Braver. T. S.. Reynolds. J. R.. & Donaldson. I). I. (2003). Neural mechanisms time differences and small accuracy differences. Younger chil- of transient and sustained cognitive control during task switching. Neuron. dren often showed less of a change in speed and but very large 39. 713-726. differences in accuracy. Young children were often too impul- Brown. A. L. (1975). The development of memory: Knowing, knowing about sive to take the time they needed; their response speed remained knowing. and knowing how to know. In H. W. Reese (Ed.). Advances in child development and behavior.• 10. (10. (pp. 103-152). New York: more constant across conditions, but at the cost of accuracy on Academic Press. the more difficult conditions. Thus, for example, global switch Brown. R. G.. & Marsden. C. D. (1988). An investigation of the phenomenon costs in accuracy declined over age, but global switch costs in of "set' in Parkinson's disease. Movement Disorders. 3. 152-161. RT increased over age. Brown. T T.. Lugar. H. M.. Coalson. R. S.. Miczin. F. M.. Petersen. S. E.. Effects elicited only in Mixed blocks with adults were found & Schlaggar. B. L (2005). Developmental changes in human cerebral in young children even in single-task blocks. While young chil- functional organization for word generation. Cerebral Cortex. 15. 275- 290. dren could exercise inhibition in steady state, it was sufficiently Bush. G.. Shin. L. M.. Holmes. J.. Rosen. B. R.. & Vogl B. A. (2003). The difficult for them that it exacted a cost not seen in adults. Adults multi-source interference task: Validation study with fMRI in individual (but not young children) seemed to re-set their default response subjects. Molecular Psychiatry. 8. 60-70. when inhibition of the same tendency was required throughout a Campione. J. C.. & Brown. A. L (1974). The effects of contextual changes block. Adults required having to switch between the two condi- and degree of component mastery on transfer of training. Advances in Child Development and Behavior. 9.69-114. tions for a significant effect on their performance to be evident. Casey. B. J.. Galvan. A.. & Hare. T. A.(2005). Changes in cerebral functional organization during cognitive development. Current Opinion in Neurobi- Acknowledgements ology. 15. 239-244. Cepeda. N. J.. Kramer. A. F.. & Gonzalez de Sather. J. C. (2001). Changes in executive control across the life span: Examination of task-switching The authors gratefully acknowledge the support of this work performance. Developmental Psychology. 37. 715-730. by NIH. Grant #R01-HD35453 from NICHD funded the empir- Cohen. J. D.. Dunbar. K.. & McClelland. J. L. (1990). On the control of auto- ical work and grant 0ROI-DA 19685 from NIDA funded prepa- matic processes: A parallel distributed processing account of the Stroop ration of the manuscript. Sarah Munro and Cecil Chau helped effect. Psychological Review. 97.332-361. greatly with the figure preparation. The authors would also like Cohen. S.. Bixenman. M.. Melilla. N.. & Diamond. A. (2001). Task switching in children. In Proceedings of the paper presented at the South Carolina to acknowledge their gratitude to Robert Proctor and two anony- Bicentennial Symposium on Attention. mous reviewers for comments on an earlier draft, and to all the Craft. J. L.. & Simon. J. R. (1970). Processing symbolic information from a parents and children who made this research possible. Reprint visual display: Interference from an irrelevant directional cue. Journal of requests should be addressed to: Adele Diamond, UBC, Dept of Experimental Psychology. 83. 415-420. EFTA01098915 MC. Davidson et al. / Neuropsychologia 44 (2006)2037-2078 2075 Crone. E. A.. Bungc. S. A.. Van der Molen. M. W.. & Ridderinkhof. K. Dreher. J.. & Berman. F. K. (2002). Fractionating the neural substrate of R. (in press). Switching between tasks and responses: A developmental cognitive control processes. Proceedings of the National Academy of Sci- study. Developmental Science. ences. 99. 14595-14600. Crone. E. A.. Ridderinkhof. R. K.. Worm. M.. Somsen. R. J. M.. & van Dreher. J. C.. & Grafman. J. (2003). Dissociating the roles of the rosual der Molen. M. W. (2004). Switching between spatial stimulus-response anterior cingulate and the lateral prefrontal cortices in performing two mappings: A developmental study of cognitive flexibility•. Developmental tasks simultaneously or successively. Cerebral Cortex. 13. 329-339. Science. 7.443-455. Dunton. S.. Fossella. J. A.. Casey. B. J.. Hulshoff Pol. H. E.. Galvan. A.. Crone. E. A.. Wendelken. C.. Donohue. S. E.. & Bunge. S. A. (2005). Neural Schnack. H. G.. et al. (2005). Differential effects of DRD4 and DATI evidence for dissociable components of task-switching. Cerebral Cortex genotype on fronto-striatal gray matter volumes in a sample of subjects (July 6 Epub ahead of print). with attention deficit hyperactivity disorder. their unaffected siblings, and Dassonville. R. Lewis. S. M.. Zhu. X. H.. Ugurbil. K.. Kim. S. G.. & Ashe. controls. Molecular Psychiatry. 10. 678-685. J. (2001). The effect of stimulus—response compatibility on cortical motor Fagot. C. (1994). Chronometric investigations of task switching. Ph.D. Thesis. activation. NeumImage. 131. 1-14. University of California. San Diego. De Jong. R. (1995). Strategical determinants of compatibility effects with Fan. J.. Flombaum. J. 1.. McCandliss, B. D.. Thomas. K. NI.. & Posner. M. task uncertainty. Acta Psychologica. 88. 187-207. 1.(2003). Cognitive and brain consequences of conflict. Neurolmage. 18. DeLuca. C.. Wood. S. J.. Anderson. V.. Buchanan. J. A.. Proffitt. T.. Mahony. 42-57. K.. et al. (2003). Normative data from CANTAB: Development of exec- Fischer. B.. Biscaldi. M.. & Gerack. S. (1997). On the development of vol- utive function over the lifespan. Journal of Clinical and Experimental untary and reflexive components in human saccade generation. Brain Neuropsychology. 25. 242-254. Research. 754. 285-297. Dempster. F. N. (1985). Short-term memory development in childhood and Fitts. P. M.. & Seger. C. M. (1953). S-R compatibility: Spatial characteristics adolescence. In C. J. Brainerd & M. Pressley (Eds.). Basic processes in of stimulus and response codes. Journal of Experimental Psychology. 81. memory development• Progress in cognitive development research. New 174-176. York: Springer—Verlag. Flowers. K. A.. & Robertson. C. (1985). The effect of Paticinson's disease on Diamond. A. (1990). The development and neural bases of higher cognitive the ability to maintain a mental set. Journal of Neurology Neumsurgery functions. NY: New York Academy of Sciences. and Psychiatry. 48. 517-529. Diamond. A. (1991). Neuropsychological insights into the meaning of object Georgopoulos. A. (1994). Population activity in the control of movement. concept development. In S. Carey & R. Gelman (Eds.). The epigenesis International Review of Neurobiology. 37. 103-119. of mind: Essays on biology and knowledge (pp. 67-110). Hillsdale. NJ: Georgopoulos• A. P.. Lurito. J. T.. Petrides, M.. Schwartz. A. B.. & Massey. Lawrence Erlbaum Associates. J. T. (1989). Mental rotation of the neuronal population vector. Science. Diamond. A. (1995). Evidence of robust recognition memory early in life 243. 234-236. even when assessed by reaching behavior. Journal of Experimental Child Gerardi-Coulton. G. (2.(X10). Sensitivity to spatial conflict and the develop- Psychology. 59.419-456. ment of self-regulation in children 24-36 months of age. Developmental Diamond. A. (2002). Normal development of prefrontal cortex from birth to Science. 3. 397-404. young adulthood: Cognitive functions. anatomy, and biochemistry. In D. Gemsbacher. M. A.. & Faust. M. E. (1991). The mechanism of suppression: T. Stuss & R. T. Knight (Eds.). Principles offrontal lobe fr action (pp. A component of general comprehension skill. Journal of Experimental 466-503). London. UK: Oxford University Press. Psychology. 17. 245-262. Diamond. A. (2005). Some generalizations concerning cognition and cog- Gerstadt. C.. Hong. Y.. & Diamond. A. (1994). The relationship between nitive development. Presented to the Psychology Dept.. University• of cognition and action: Performance of 3 112-7 year old children on a Victoria. October 27.2005. Stroop-like day-night tat. Cognition. 53. 129-153. Diamond. A. A theory of "all or none:' In preparation. Goldman-Rakic. P. S. (1987). Circuitry of primate prefrontal cortex and Diamond. A.. Carlson. S. M.. & Beck. D. M. (2005). Preschool children's regulation of behavior by representational memory. In F. Plum (Ed.). performance in task switching on the dimensional change card sort task: Handbook ofphysiology: V. (V. (pp. 373-417). Bethesda. MD: American Separating the dimensions aids the ability to switch. Developmental Neu- Physiological Society. mpsychology. 28. 689-729. Gollin. E. S. (1964). Reversal learning and conditional discrimination in Diamond. A.. Kirkham. N. Z.. & Amso. D. (2002). Conditions under which children. Journal of Comparative & Physiological Psychology. 58. 441- young children CAN hold two rules in mind and inhibit a prcpotent 445. response. Developmental Psychology. 38. 352-362. Gollin. E. S. (1965). Factors affecting conditional discrimination in children. I>iamond. A.. Otraven. K. M.. & Savoy. R. L. (1998). Dorsolateral pre- Journal of Comparative and Physiological Pyschology, 60. 422-427. frontal cortex contributions to working memory and inhibition as revealed Gollin. E. S. (1966). Solution of conditional discrimination problems by by fMRI. Society for Neuroscience Abstracts. 24. 1251. young children. Journal of Comparative and Physiological Psychology. Diedrichsen. J.. Mayr. IJ.. Dhaliwal. H.. Keck. S.. & Ivry. R. B. (2000). 62. 454-456. Task-switching deficits in patients with prefrontal lesions or Parkinson's Gollin. E. S.. & Liss. P. (1962). Conditional discrimination in children. Jour- disease. Cognitive Neuroscience Society Annual Meeting Abstracts. I. nal of Comparative and Physiological Psychology. 55.850-855. 99. Hamishfeger. K. K.. & Pope. R. S. (1996). Intending to forget: The devel- DiGirolamo. G. J.. Kramer. A. F.. Barad. B.. Cepeda. N. J.. Weiss- opment of cognitive inhibition in directed forgetting. Journal of Experi- man. D. H.. Milham. M. P.. et al. (2001). General and task- mental Child Psychology. 62. 292-3I5. specific frontal lobe recruitment in older adults during executive pro- Hasher. L.. Stoltzfus. E. R.. Zacks. R. T.. & Rypma. B. (1991). Age and crews: A IMRI investigation of task-switching. Neurorepon. 12. 2065- inhibition. Journal of Experimental Psychology. 17. 163-169. 2071. Hanby. J. V.. Petit. L.. Ungerleider. L G.. & Courtney. S. M. (2000). Dis- Doan. H. M.. & Cooper. D. L. (1971). Conditional discrimination in children: tinguishing the functional roles of multiple regions in distributed neural Two relevant factors. Child Development. 42. 209-220. systems for visual working memory. Neuroimage. It 380-391. Dove. A.. Pollmann. S.. Schubert, T.. Wiggins. C. J.. @ von Cramon. Y. D. Hayes, A. E.. Davidson. M. C.. Keefe. S. W.. & Real. R. D. (1998). Toward (2000). Prefrontal cortex activation in task switching: An event-related a functional analysis of the basal ganglia. Journal of Cognitive Neuro- fNIRI study. Cognitive Brain Research. 9. 103-109. science. 10. 178-198. Downes. J. J.. Roberts. A. C.. Sahakian. B. J.. Evenden. J. L.. Mortis. R. G.. Hedge. A.. & Marsh. N. W. A. (1975). The effect of irrelevant spatial corre- & Robbins. T. W. (1989). Impaired extra-dimensional shift performance in spondence on two-choice response-time. Acta Psychologica. 39.427 39. medicated and unmedicated Paikinson's disease: Evidence for a specific Heidbreder. E. F. (1928). Problem solving in children and adults. Journal of attentional dysfunction. Neuropsyrhologia. 27. 1329-1343. Genetic Psychology. 35. 522-545. EFTA01098916 2076 .41.C. Davidson et al. / Neuropsychologia -14 (2006)2037-2078 Hommel. B. (1993). The relationship between stimulus processing and Los. S. A. (1999). Identifying stimuli of different perceptual categories in response selection in the Simon task: Evidence for a temporal overlap. pure and Mixed blocks of trials: Evidence for stimulus-driven switch Psychological Research/Psychologische Forschung, 55. 280-290. costs. Acta Ps)rhologica, 103. 173-205. Hommel. B. (1994). Effects of irrelevant spatial S compatibility depend on Liu. X.. Banich. M. T. Jacobson. B. L.. & Tanabe. J. L 42004). Common stimulus complexity. Psychological Research/Psychologische Forschung, and distinct neural substrates of attentional control in an integrated Simon 56. 179-184. and spatial Swoop task as assessed by event-related fMR1. Net:mintage. Hommel. B. (1995). Stimulus-response compatibility and the Simon effect: 22. 1097-1106. Toward an empirical clarification. Journal of Experimental Psychology: Lu. C. H.. & Proctor. R. W (1995). The influence of irrelevant location Human Perception and Performance. 21. 764-775. information on performance: A review of the Simon and spatial Stroop Hommel. B., Musseler. J.. Aschersleben. G.. & Prinz. W. (2001). The Theory effects. Psychonomic Bulletin and Review. 2. 174-207. of Event Coding (TEC): a framework for perception and action planning. Luciana. M.. Conklin. H. NI., Hooper. C. J.. & Yager. R. S. (2005). The Behavioral & Brain Sciences. 24(5). 849478. development of nonverbal working memory and executive control pro- Hommel. B.. Proctor. R. W.. & Vu. K. P. (2004). A feature-integration account cesses in adolescents. Child Development. 76.697-712. of sequential effects in the Simon task. Psychological Research. 68.1-17. Luciana. NI., & Nelson. C. A. (2002). Assessment of neuropsychological lacoboni. M.. Woods. R. P.. & Mazziotta. J. C. (1998). Bimodal (auditory function in children using the Cambridge Neuropsychological Testing and visual) le0 frontoparietal circuitry for sensorimotor integration and Automated Battery (CANTAB): Performance in 4 to 12 year-olds. Devel- sensorimotor learning. Brain. 121. 2135-2143. Pt opmental Neumpsychology. 22. 595-623. Jeffrey. W. E. (1961). Variables in early discrimination learning. III. Simul- Luna. B.. Garver. K. E.. Urban. T. A.. Lazar. N. A.. & Sweeney. J. A. (2004). taneous vs. successive stimulus presentation. Child Development. 32. Maturation of cognitive processes from late childhood to adulthood. Child 305-310. Development. 75. 1357-1372. Jersild, A. T. (1927). Mental set and shift. Archives ofPs)rhology. 89. 542. Lyons-Warren. A.. Lillie. R.. & Hershey. T. 42004). Short and long-term Kail. R. (1991a). Development of processing speed in childhood and adoles- spatial delayed response performance across the lifespan. Detrlopmental cence. In H. W. Reese (Ed.). Advances in child development and behavior: Neuropsychology. 26. 661-678. 23. (23. (pp. 151-185). New York: Academic Press. Maclin. E. L.. Gratton. G.. & Fabiani. M. (2001). Visual spatial localization Kail. R. (19916). Developmental change in speed of processing during child- conflict: An fivIRI study. Neuroreport. 12. 3633-3636. hood and adolescence. Psychological Bulletin. 109. 490-501. Mayr. U. (1996). Spatial attention and implicit sequence learning: Evidence Kail. R. (1991c). Processing time declines exponentially during childhood for independent learning of spatial and nonspatial sequences. Journal and adolescence. Developmental Psychology. 27. 259-266. of Experimental Psychology: Learning. Afentory. and Cognition. 22. Kail. R.. & Salthouse. T. A. (1994). Processing speed as a mental capacity. 350-364. Acta Psychologica. 86. 199-225. Map. U. (2001). Age differences in the selection of mental sets: The role of Kecle. S.. & Rafal. R. (2000). Deficits of task set in patients with k0 inhibition, stimulus ambiguity. and response-set overlap. Psychology and prefrontal cortex lesions. In S. Monsell & J. Driver (Eds.). Control Aging. 16. 96-109. of cognitive processes: Attention and performance: XVIII. (XVIII. (pp. Map. U.. & Klieg'. R. (2000a). Complex semantic processing in old age: 627-652). Cambridge. MA: MIT Press. Does it May or does it go? Psychology and Aging• 15. 2943. Kimberg. D. Y.. Aguirre. G. K.. & D'Esposito. M. (2000). Modulation of Map. U.. & Klieg'. R. (200Ob). Task-set switching and long-term memory task-related neural activity in task-switching: An IMRI study. Cognitive retrieval. Journal of Experimental Psychology: Learning. Memory. and Brain Research. 10. 189-196. Cognition. 26. 1124-1140. Kimberg. D. Y.. & Farah. M. J. (1993). A unified account of cognitive Mayr. U.. & Liebscher. T. (2001). Is there an age deficit in the selection of impairments following frontal lobe damage: The role of working memory mental sets? European Journal of Cognitive Psychology. 13. 47-69. in complex. organized behavior. Journal of Experimental Psychology. 122. Mecklinger. A.. & Mueller. N. (1996). Dissociations in the processing of 411-428. "what" and "where" information in working memory: An event-related Kirkham. N. K.. Cruess. L.. & Diamond. A. (2003). Helping children apply potential analysis. Journal of Cognitive Neuroscience. 8. 453-473. their knowledge to their behavior on a dimension-switching task. Devel- Mecklinger. A. D.. von Cramon, D. Y.. Springer. A.. & Matthes-von Cramon. opmental Science. 6(5). 449-476. G. (1999). Executive control functions in task switching: Evidence from Kleinsorge. T. (1999). Response repetition benefits and costs. Acta Psycho- brain injured patients. Journal of Clinical and Experimental Neuropsy- logien. 103. 295-310. chology, 21. 606-6I9. Kleinsorge. T.. & Heuer. H. (1999). Hierarchical switching in a multi- Mciran. N. (1996). Reconfiguration of processing mode prior to task per- dimensional task space. Psychological Research. 62.300-312. formance. Journal of Experimental Psychology: Learning. Memory. and Kramer. A. F.. Hahn. S.. & Gopher. D. (1999). Task coordination and Cognition. 22. 1423-1442. aging: Explorations of executive control processes in the tack switching Mciran. N. (2000a). Modeling cognitive control in task-switching. Psycho- paradigm. Acta Psychologica. 101. 339-378. logical Research. 63.234-249. Kray. J.. Eber. J.. & Lindenberger. U. (2004). Age differences in executive Mciran. N. (20006). Reconfiguration of stimulus task-sets and response task- functioning across the lifespan: The role of verbalization in task prepa- sets during task-switching. In S. Monsell & J. Driver (Eds.). Control ration. Acta Psychologica. 115. 143-165. of cognitive processes: Attention and performance: XVIII. (XVIII. (pp. Kray. J.. Li. K. Z.. & Lindenberger. U. (2002). Age-related changes in task- 377-400). Cambridge. MA: MIT Press. switching components: The role of task uncertainty. Brain and Cognition. Mciran. N. (2005). Task rule congruency and Simon-like effects in switching 49. 363-381. between spatial tasks. Quarterly Journal of Experimental Psychology A. Kray. J.. & Lindenberger. U. (2000). Adult age differences in task switching. 58. 1023-1041. Psychology and Aging. 15. 126-147. Martin. N.. Gotkr. A.. & Perlman. A. (1996). Old age is associated with Levy. R.. & Goldman-Rakic. P. S. (2000). Segregation of working memory pattern of relatively intact and relatively impaired task-set switching abil- functions within the dorsolateral prefrontal cortex. Experimental Brain ities. Journals of Gerontology Series 8-Psychological Sciences and Social Research. /33.23-32. Sciences. 5642). P88-P102. Logan. G. D.. & Zbrodoff. N. J. (1979). When it helps to be misled: Facil- Martin. N.. Chary. L. & Sapir. A. (2000). Component processes in task itative effects of increasing the frequency of conflicting stimuli in a switching. Cognitive Psychokgy. 41. 211-253. Stroop-like task. Memory & Cognition. 7. 166-174. Martin. N.. Gotkr. A.. & Perlman. A. (2001). Old age is associated with Los. S. A. (1996). On the origin of mixing costs: Exploring information a pattern of relatively intact and relatively impaired task-set switching processing in pure and Mixed blocks of trials. Acta Psychologica. 94. abilities. Journals of Gerontology:• Series B: Psychological Sciences & 145-188. Social Sciences B. 56. 88-102. EFTA01098917 MC. Davidson et al. / Neurapsychologia 44 (2006) 2037-2078 2077 Meiran, N.. Levine. J.. Mciran. N.. @ Henik. A. (2000). Task-set switching Schuch. S.. & Koch. I. (2003). The role of response selection for inhibition in schizophrenia. Neumpsychology. 14. 471-482. of task sets in task shifting. Journal of Experimental Psychology: Human Meyer. D. E.. Evans. J. E.. Lauber. E. J.. & Gmeindl. L. (1998). The role of Perception & Performance. 29. 92-105. dorsolateral prefrontal cortex for executive processes in task switching. Schuch. S.. & Koch. I. (2004). The costs of changing the representation of Cognitive Neuroscience Society Annual Meeting Abstracts. action: Response repetition and response-response compatibility in dual Miller. E. K.. & Cohen. J. D. (2001). An integrative theory of prefrontal tasks. Journal of Experimental Psychology: Human Perception and Per- cortex function. Annual Review of Neuroscience. 24. 167-202. formance. 30. 566-582. Miyake. A.. Freidman. N. P.. Emerson. M. J.. Witzki. A. H.. Howener. A.. Shaffer. L. H. (1965). Choice reaction with variable S—R mapping. Journal & Wager. T. D. (2000). The unity and diversity of executive functions of Experimental Psychology. 70. 284-288. and their contributions to complex "frontal lobe" tasks: A latent variable Shallicc. T.. & Burgess. P. W. (1991). Higher-order cognitive impairments analysis. Cognitive Psychology. 41. 49-100. and frontal lobe lesions in man. In H. S. Levin. H. NI. Eisenberg. & A. Monsell. S.. & Driver. J. (Eds.). (2000). Control of cognitive processes: Atten- L Benton (Eds.). Frontal lobe function and dysfunction (pp. 125-138). tion and performance: XVIII. (XVIII.. Cambridge. MA: MIT Press. Oxford: Oxford University Press. Morton. J. B.. & Munakata. Y. (2002). Are you listening? Exploring a devel- Shepard. W. 0. (1957). Learning set in preschool children. Journal of Com- opmental knowledge action dissociation in a speech interpretation task. parative & Physiological Psychology. 50. 15-17. Developmental Science. 5.435 140. Simon. F. R.. & Small. A. M. (1969). Processing auditory irrelevant informa- Munakata. Y. (2000). Challenges to the violation of expectation paradigm: tion: Interference from an irrelevant cue. Journal of Applied Psychology. Throwing the conceptual baby out with thc perceptual processing bath- 53. 433-435. water? Infancy. 1. 471-477. Simon. J. R. (1990). The effects of an irrelevant directional cue on Munoz. D.. Broughton. J.. Goldring. J.. & Armstrong. I. (1998). Age-related human information processing. In R. W. Proctor & T G. Reeve (Eds.). perfomunce of human subjects on saccadic eye movement tasks. Exper- Stimulus-response compatibility,: An integrated perspective (pp. 31-86). imental Brain Research, 217. 1-10. Amsterdam: North-Holland. O'Craven. K. M.. Davidson. M. C.. Bergida. R.. Savoy. R. L. Diamond. A. Simon. J. R.. & !terbium K. (1990). Effect of conflicting cues on information (in preparation). Prefrontal cortex contributions to working memory and processing: the 'Stroop effect' vs. the 'Simon effect. Acta Psychologica. inhibition as revealed by IMRE 73(2). 159-170. O'Craven. K. M.. Savoy. R. L.. & Diamond. A. (1998). Working memory and Sohn. M. H.. Ursa S., Anderson. J. R.. Stenger. V. A.. & Carter. C. S. 42000). inhibition in dorsolateral prefrontal cortex. In Proceedings of the Human Inaugural article: The role of prefrontal cortex and posterior parietal cor- Brain Mapping Annual Meeting. tex in task switching. Proceedings of the National Academy of Sciences. Omori. M.. Yamada. H.. Murata. T.. Sadao. N.. Tanaka. M.. Ishii. 97. 13448-13453. Y.. et al. (1999). Neuronal substrates participating in attentional set- Spector. A.. & Biederman. 1. (1976). Mental set and mental shift revisited. shifting of rules for visual guided motor selection: A functional mag- American Journal of Psychology. 89. 669-679. netic resonance imaging investigation. Neuroscience Research, 33. 317- Stoffels. E. 1. (1996). Uncertainty and processing routes in the selection of 323. a response: an S-R compatibility study. Acta Psychologica. 94(2). 227- Osier. S. F.. & 1Cofsky. E. (1965). Stimulus uncertainty as a variable in 252. the development of conceptual ability. Journal of Experimental Child Stunner. B., Leuthold, H.. Soetens. E.. SchrSter. H.. & Sommer. W. (2002). Psychology. 2. 264-279. Control over location-based response activation in the Simon task: Behav- Owen. A. NI.. Roberts. A. C.. Hodges, J. R.. Summers. B. A.. Polkey. C. E.. ioral and ekctrophysiological evidence. Journal of Experimental Psychol- & Robbins. T. W. (1993). Contrasting mechanisms of impaired attentional ogy: Human Perception & Performance. 28. 1345-1363. set shifting in patients with frontal lobe damage or Parkinson's disease. Sudevan. P.. & Taylor. D. A. (1987). The cueing and priming of cognitive Brain. 116. [159-1175. operations. Journal of Experimental Psychology: Human Perception and Peterson. B. S.. Kane. M. J.. Alexander. G. M.. Lacadic. C.. Skudlarski. Performance. 13. 89-103. R. Leung. H. C.. et al. (2002). An event-related functional MRI study Sylvester. C. C.. Wager. T D.. Lacey. S. C.. Hernandez. L.. Nichols. T. E. comparing interference effects in the Simon and Snoop tasks. Cognitive Smith. E. E.. et al. (2003). Switching attention and resolving interfer- Brain Research. 13. 427-440. ence: IMR1 measures of executive functions. Neumpsyrhologia. 41. 357- Praanistra. P.. Kleine. B. U.. & Schnirzler. A. (1999). Magnetic stimulation 370. of the dorsal premotor cortex modulates the Simon effect. NeumReport. Tagliabue. M.. Zorzi. M.. Unlit/. C.. & Bassignani. F. (2000). The role 10. 3671-3674. of long-term-memory and short-term-memory links in the Simon effect. Proctor. R. W.. & Vu. K.-P. L (2002). Mixing location-irrelevant and location- Journal of Experimental Psychology: Human Perception and Perfor- relevant trials: Influence of stimulus mode on spatial compatibility effects. mance. 26. 648-670. Memory & Cognition. 30. 281-293. Thomas. K. M.. King. S. W.. Frozen. P. I.. Welsh, T. F.• Berkowitz. A. L. Reimers. S.. & Maylor. E. A. (2005). 'Risk switching across the life span: Noll. D. C.. et al. (1999). A developmental functional MRI study of Effects of age on general and specific switch costs. Developmental Psy- spatial working memory. Neumlmage. 10. 327-338. chology. 41. 661-671. Usher. M.. @ McClelland. J. L. (2001). The time course of perceptual choice: Ridderinkhof. K. R. (2002). Activation and suppression in conflict tasks: The leaky, competing accumulator model. Psychological Review. 108. Empirical classification through distributional analyses. In W. Prinz & 550-592. B. Hommel (Eds.). Attention and performance XIX: Common mecha- Usher. NI.. Olami. Z.. & McClelland. J. L. (2002). Hick's law in a stochas- nisms in perception and action (pp. 494-519). Oxford: Oxford University tic race model with speed-accuracy tradeoff. Journal of Mathematical Press. Psychology. 46. 704-715. Rogers. R. D.. & Monsell. S. (1995). Costs of a predictable switch between Valle-Inclfin. F. (1996). The locus of interference in the Simon effect: An simple cognitive tasks. Journal of Experimental Psychology: General. ERP study. Biological Psychology. 43. 147-162. 124. 207-231. Valle-Inclan. E. Hackley. S. A.. & de Labra. C. (2002). Does stimulus- Rogers. R. D.. Sahakian. B. J.. Hodges. J. R.. Polkey. C. E.. Kennard. C.. driven response activation underlie the Simon effect? In W. Prinz & & Robbins. T. W. (1998). Dissociating executive mechanisms of task B. Hommel (Eds.). Attention and performance XIX: Common mecha- control following frontal lobe damage and Parkinson's disease. Brain. nisms in perception and action (pp. 474-493). Oxford: Oxford University 121. 815-842. Press. Salthouse. T. A.. Fristoe. N. M.. Linewcaver. T. T., & Coon. V. E. (1995). van Asselen. M.. & Ridderinkhof. K. R. (2000). Shift costs of predictable and Aging of attention: Does the ability to divide decline? Memory and Cog- unexpected set shifting in young and older adults. Psychologica Belgica. nition. 23. 56-71. 40. 259-273. EFTA01098918 2078 RC Davidson et at / Neuropsychologia 44 (20(16)2037-2078 Verbruggen. F. R.. Licfooghe. B.. Szmalec. A.. & Vandicrendonck. A. (2005). Wascher. E.. & Wolber. M. (2004). Attentional and intentional cueing Inhibiting responses when switching: Does it matter? Experimental Psy- in a Simon task: An EEC-based approach. Psychological Research/ chology. 52. 125-130. Psychologische Forschung. 68. 18-30. Verhaeghen. P., & Dc Mcersman. L. (1998). Aging and the Swoop effect: A Wasiak. E. Hommel. B.. & Allport. A. (2003). Task-switching and long- meta-analysis. Psychology and Aging. 13. 120-126. term priming: Role of episodic stimulus-task bindings in task-shift costs. Verhaeghen. P.. & Salthousc. T. A. (1997). Meta-analyses of age-cognition Cognitive Psychology. 46.361 13. relations in adulthood: Estimates of linear and nonlinear age effects and Wiegand. K.. & Wascher. E. (2005). Dynamic aspects of stimulus—response structural models. Psychological Bulletin. 122. 231-249. correspondence: Evidence for two mechanisms involved in the Simon Vu. K.-P. L.. & Proctor. R. W. (2004). Mixing compatible and incompatible Effect. Journal of Experimental Psychology: Human Perception and Per- mappings: Elimination. reduction. and enhancement of spatial compati- formance. 31. 453-464. bility effects. Quarterly Journal of Experimental Psychology A: Human Wiihr. P. (2004). Sequential modulation of logical-recoding operations in the Experimental Psychology. 57.539-556. Simon task. Experimental Psychology. 58A. 705-731. Wager. T. D.. Reading. S.. & lonides. J. (2004). Neuroimaging studies of Wiihr. P. (2005). Evidence for gating of direct response activation in the shifting attention: A meta-analysis. Nearolmage. 22. 1679-1693. Simon task. Psychonomic Bulletin & Review. 12. 282-288. Wager. T. D.. & Smith. E. E (2003). Neuroimaging studies of working mem- Wylie. G.. & Allpod. A. (2000). Task switching and the measurement of ory: A meta-analysis. Cognitive. Affective and Behavioral Neuroscience. "switch costs". Psychology Research. 63. 2I2-233. 3. 255-274. Yeung. N.. @ Mansell. S. (2003). The effects of recent practice on task Waschcr. E. Schatz. U.. Kudcr. T.. @ Verleger. R. (200O. Validity and bound- switching. Journal of Experimental Psychology: Human Perception and ary conditions of automatic response activation in the Simon task. Journal Performance. 29. 9I9-936. of Experimental Psychology: Human Perception and Mfonnance. 27. Zelazo. P. D.. Craik. F. 1. M.. & Booth. L. (2004). Executive function across 731-751. the life span. Acta Psychologica. 115. 167-183. EFTA01098919