Acta Psychologica 124 (2007) 139–158 www.elsevier.com/locate/actpsy

Developmental changes in oculomotor control and working-memory eYciency Rena M. Eenshuistra a,¤, K. Richard Ridderinkhof a,b, Maaike A. Weidema a, Maurits W. van der Molen b a

Cognitive Psychology Unit, University of Leiden, P.O. Box 9555, 2300 RB Leiden, The Netherlands b Department of Psychology, University of Amsterdam, The Netherlands Available online 17 November 2006

Abstract In the present study, we examined the developmental changes in the eYciency of saccadic inhibitory control. More speciWcally, the contribution of age-related changes in working-memory engagement was investigated. We manipulated the eYciency of inhibitory oculomotor control in antisaccade tasks by using Wxation-oVset conditions, which are supposed to aVect inhibitory demands, and by adding increasing working-memory loads to the antisaccade task. In general, in comparison to antisaccade performance of adults, the antisaccade performance of 8year-old and 12-year-old children was characterized by an increase in direction errors, and/or longer saccadic onset latencies on correct antisaccades. However, this pattern was not altered by the Wxation-oVset manipulations. In contrast, increased working-memory demands deteriorated 8-year-olds’ antisaccade performance unequally as compared to older children and young adults. These Wndings suggest that – at least in young children – the available functional working-memory capacity is engaged in oculomotor inhibition. © 2006 Elsevier B.V. All rights reserved. PsycINFO classiWcation: 2323; 2340; 2343; 2346; 2820 Keywords: Development; Antisaccade task; Inhibitory control; Working memory

*

Corresponding author. Tel.: +31 71 527 4815; fax: +31 71 527 3783. E-mail address: [email protected] (R.M. Eenshuistra).

0001-6918/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.actpsy.2006.09.012

140

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

1. Working memory and inhibitory control Developmental studies frequently report on two marked changes in the development of cognitive control: First, the increasing ability to inhibit reXexive or prepotent responses (Dempster, 1992; Diamond, 1990), and second, the increasing quantity of information that can be held in working memory (WM) (e.g., Case, 1985; Dempster, 1981, 1985; Gathercole, 1999; Luciana & Nelson, 1998; Pascual-Leone, 1970). While inhibitory control and WM function (especially updating) are assumed to be independent (at least in part; e.g., Miyake et al., 2000), many studies have interrelated the eYciency of inhibitory control with the capacity of WM. As Pennington (1994) argued, one of the key functions of WM is action selection, which requires that task-relevant information has to be maintained active until the correct action has been selected. However, WM is assumed to be restricted by a limited capacity, which means that the activation of task-relevant information necessarily implies the inhibition of inappropriate actions or task-irrelevant information. EYcient inhibitory control increases functional WM capacity by preventing the restricted WM capacity from being polluted with irrelevant information. Furthermore, the implication of limited capacity has consequences for the eYciency of cognitive processing, that is, correct task performance will only occur if suYcient functional WM capacity is available for maintaining the appropriate task demands. The assumed relationship between inhibitory control and WM capacity has led to several, rather opposite, theories about the mechanisms that control inhibition. On the one hand, inhibitory control is seen as a central mechanism that can be held responsible for the eVectiveness of WM function in a sense that such an inhibitory mechanism selects the information that enters WM (Hasher & Zacks, 1988). On the other hand, some researchers came to believe that the eYciency of inhibitory control depends on the availability of functional WM capacity (e.g., Conway & Engle, 1994, 1996; Pennington, 1994; Roberts, Hager, & Heron, 1994). In this perspective, inhibition can be seen as intrinsic to WM, that is, inhibitory control itself is supposed to consume WM capacity. Thus, the eYciency of inhibitory control depends on the amount of available WM capacity. Roberts et al. (1994) and Roberts and Pennington (1996) suggested that the inhibition of prepotent responses requires that the instruction be refrained from performing the prepotent response and carry out an alternative correct response that is to be maintained in WM throughout task execution. Successful inhibition, then, would be dependent on the prepotency of the to-beinhibited actions, the WM demand of the executed task, and the availability of WM resources needed for task execution. This perspective resembles in many respects the general capacity theory proposed by Conway and Engle (1994, 1996). In this theory, WM capacity reXects the ability to draw domain-free attentional resources on cognitive tasks, regardless of the nature of these tasks. Tasks that require controlled attention, such as tasks that are sensitive to interference from irrelevant information, are assumed to be especially dependent on the availability of attentional resources. Subjects who score high on measures of WM capacity have more attentional resources available for task performance than subjects with low WM capacity. Although the nature of attentional resources remains somewhat unspeciWed, diVerences in the availability of these attentional resources are assumed to originate from individual and developmental diVerences. In a broader perspective, this view Wts in nicely with more general theories about goaldirected biasing and control of elementary cognitive processes (e.g., Desimone & Duncan, 1995; Miller & Cohen, 2001). These theories propose that eYcient task execution depends

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

141

on the top–down regulation by the PFC, mainly by activating and maintaining task-relevant information. In such a perspective, activation and inhibition can be considered as two sides of the same coin. Activation can be seen as the biased enhancement of task-relevant cognitive processes to resolve the competition with task-relevant information. Inhibition, then, is the diminishing or suppression of processing task-irrelevant information. Recently, fMRI work has shown that in certain conXict situations, cognitive control mechanisms increase performance by temporarily intensifying cortical responses to task-relevant information, instead of increasing cortical responses to task-irrelevant information (Egner & Hirsch, 2005). The case of the antisaccade task, to be examined in detail below, is of particular interest for present concerns, because more than in many other “conXict” tasks, this task incurs such a potent pull of the inappropriate action (a reXexive saccade) that goaldirected processes aimed at enhancing the activation of the correct antisaccade require complementary goal-directed processes aimed at suppressing the activation of the incorrect reXexive prosaccade. The strong relation between inhibitory control and WM may not come as a surprise if one considers the brain areas involved in WM and inhibitory control. Using brain-imaging techniques, numerous adult studies have provided signiWcant evidence for the involvement of the prefrontal cortex (PFC), and in particular the lateral prefrontal cortex (LPFC), in both inhibitory control and WM function (e.g., Goldman-Rakic, 1987; Miller & Cohen, 2001; O’Reilly, Braver, & Cohen, 1999; Smith & Jonides, 1998; for a review, see Ridderinkhof, Van den Wildenberg, Segalowitz, & Carter, 2004). Although the inhibition of prepotent responses likely involves other brain areas as well (such as the basal ganglia; e.g., Casey et al., 1997), LPFC appears to be dominantly involved in prompting inhibitory control (cf. Aron, Robbins, & Poldrack, 2004). Furthermore, developmental researchers have suggested that the improvement in cognitive control functions such as WM and inhibitory control could be related to the maturation of the PFC and frontostriatal and frontoparietal networks (Casey, Tottenham, Liston, & Durston, 2005; Durston et al., 2002; Goldman-Rakic, 1987; Olesen, Westerberg, & Klingberg, 2004). These researchers point at the slow maturation of the PFC in comparison to other brain areas, which could explain the late prospering of certain cognitive control functions. Children often activate similar brain areas as adults do during performance of WM tasks (Casey et al., 1997; Nelson et al., 2000; Thomas et al., 1999), but the increase in WM capacity is related to an increase in activity in these areas (Klingberg, Forssberg, & Westerberg, 2002). These Wndings neatly add to the assumption that WM can be seen as a general capacity system, the expansion of which depends on the maturation of PFC. To recap, developmental studies that focus on WM and inhibitory control show us that the developments of inhibitory and WM functions are closely related to each other. However, the nature of the interaction between WM and inhibition is still a point of discussion. Some studies (e.g., Harnishfeger & Bjorklund, 1993) suggest that the age-related improvement in the eYciency of inhibitory control is the primitive that drives the progress of other cognitive functions (such as WM), whereas other studies (e.g., Swanson, 1996) suggest that the increased availability of WM capacity during childhood accounts for the amelioration of other cognitive functions (including inhibitory control). The present study was designed to advance our understanding of developmental aspects in the divergent relation between WM and inhibitory control by using the antisaccade (AS) paradigm. AS tasks, which involve the suppression of prepotent eye-movements, provide a suitable tool to study inhibitory abilities in relation to WM (Roberts et al., 1994).

142

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

1.1. Working-memory and inhibitory control in the antisaccade task Substantial age-related eVects have been observed in AS task performance of children as compared to young adults, that is, inhibitory control over prepotent eye-movements is still ineYcient in young children, but steadily improves until adulthood (e.g., Fischer, Biscaldi, & Gezeck, 1997; Fukushima, Hatta, & Fukushima, 2000; Klein & Foerster, 2001; Munoz, Broughton, Goldring, & Armstrong, 1998). The purpose of the present study was to examine which of the two developmental processes – the growth of WM capacity or the perfection of the inhibitory control mechanism – is the most likely candidate to account for the developmental changes in AS performance. Studies that emphasize available WM capacity as a primitive that drives the eYciency of inhibitory control have frequently involved the use of an AS task. Roberts et al. (1994) demonstrated that AS performance of young adults could easily be disrupted when a concurrent task was to be performed concurrent to the AS task. According to Roberts et al., the load posed on WM by the concurrent task reduces the amount of attentional resources or functional WM capacity that can be used for the AS task. Recently, two further AS studies conWrmed this idea. First, Kane, Bleckley, Conway, and Engle (2001) demonstrated that adult subjects with high WM capacity outperformed adult subjects with low WM capacity in terms of the frequency of non-suppressed saccadic errors. Secondly, Eenshuistra, Ridderinkhof, and van der Molen (2004) replicated the Roberts et al. Wnding that AS performance of young adults was deteriorated when a concurrent task was added to the AS task, and observed that AS performance of elderly subjects was disproportionately hampered by the concurrent task load. In elderly subjects, the decline in inhibitory oculomotor control thus is best understood in terms of age-related changes in functional WM capacity. The AS task has been used mainly to investigate inhibitory control in adults. In comparison, little research has been done with children. Since the use of the AS task has supplied us with an interesting insight in to the role of WM capacity in successful suppression of prepotent eye movements, this task may provide a suitable tool to examine the nature of the relationship between WM and inhibitory control in children. 1.2. Neural circuits involved in AS performance In an AS task, while Wxating on a central point, the abrupt onset of a peripherally presented cue elicits a strong reXex-like pull to make an eye-movement towards the cue. This strong prepotent response, or the so-called reXexive saccade, has to be inhibited voluntarily in order to generate a correct intentional AS in the opposite direction. AS task performance is usually characterized by an increase in saccade onset latencies (i.e., it takes longer to initiate antisaccades compared to prosaccades toward the cue), and by the occurrence of incorrect prosaccades in the direction of the cue or the so-called direction errors (Hallett, 1978; Roberts et al., 1994). A full description of the neural oculomotor system involved in AS performance is beyond the scope of the present article, but a brief introduction is useful for framing the possible developmental trends. In essence, when a saccade is required in the direction opposite to the visual hemiWeld in which a stimulus onset occurs, several distinct but interrelated oculomotor processes may come into play to modulate elementary exogenous saccadic eye-movement initiation circuits (cf. Everling & Fischer, 1998): (1) active Wxation of

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

143

the oculomotor system, (2) saccade activation, or the programming and initiation of saccade execution, and (3) selective suppression of saccades until the program for the appropriate eye movement has been fully developed. First, oculomotor control requires an elementary saccade Wxation system (for reviews see Moschovakis, Scudder, & Highstein, 1996; Schall, Morel, King, & Bullier, 1995; Wurtz & Goldberg, 1989). The system must be able to Wxate rather than respond to any trivial change of information in the visual Weld. During Wxation, movement neurons tend to be silent, but just before saccade initiation the movement neurons (in areas of the SC that correspond retinotopically to the intended saccade endpoint) start Wring intensely while Wxation neuron activity is almost completely suppressed (Anderson, Keller, Gandhi, & Das, 1998; Dorris & Munoz, 1998; Munoz & Wurtz, 1993, 1995). Second, oculomotor control requires an intentional saccade initiation system. The frontal eye Welds (FEF) contribute to the endogenous preparation of a saccadic eye movement by supporting the maintenance of Wxation (to prevent inappropriate reXexive saccades) while programming and initiating the intended saccade by imposing the desired pattern of activation onto the intermediate layers of the SC (Burman & Bruce, 1997; Pierrot-Deseilligny et al., 2003; Schall, Stuphorn, & Brown, 2002; Schlag-Rey, Schlag, & Dassonville, 1992). Third, oculomotor control requires selective saccade suppression systems. Several diVerent pathways may be involved in bringing about the selective suppression of saccadic eye movements. One pathway involves direct inhibitory projections to movement neurons in the SC from the DLPFC. The DLPFC may selectively suppress the activation of reXexive saccades by inhibiting activity in movement neurons in the SC (Everling, Dorris, Klein, & Munoz, 1999; Matsuda et al., 2004). Another pathway involves the basal ganglia. Both the FEF and the DLPFC project to the caudate nucleus which in turn biases the oculomotor system by indirect propagation to the superior colliculus (from the caudate to the substantia nigra pars reticulata (SNr) and then on to the SC; Hikosaka & Watanabe, 2000; Sato & Hikosaka, 2002). This caudate-driven bias may selectively strengthen or remove the tonic inhibition from SNr on SC movement neurons with a particular (leftward or rightward) movement Weld. The net result of caudate activity would be to selectively de-activate the region of the SC involved in generating the inappropriate eye movement, thus keeping it from initiating a saccade (cf. Takikawa, Kawagoe, & Hikosaka, 2002). To date, the developmental maturation of these neural components of the oculomotor control system has not been extensively charted. While elementary saccade generation systems appears to be in place early in infancy (Johnson, 1994), less is known about the maturation of more sophisticated functional mechanisms such as the DLPFC-driven selective suppression systems. 1.3. The present study Two diVerent lines of thought have been invoked in explaining the nature of the relationship between WM capacity and inhibitory control. First, it was hypothesized that WM can be seen as the primitive that drives the eYciency of inhibition. In the present study, this hypothesis will be examined by adding a concurrent WM task to the regular prosaccade (PS) and AS tasks. It was expected that posing a load on WM will leave less functional WM capacity available for eYcient inhibition of reXexive saccades in the AS task, and therefore will aVect AS task performance in children more than in younger adults. Several studies have shown that AS task performance is subject to a dramatic age-related decrease in the

144

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

amount of direction errors and substantial age-related drop in AS onset latencies (e.g., Fischer et al., 1997; Fukushima et al., 2000; Klein & Foerster, 2001; Munoz et al., 1998). As Roberts et al. (1994) argued, adding a secondary task will only be eVective in changing AS task performance when the load of that task is suYciently high. In a previous study, we used the running memory paradigm (Pollack, Johnson, & Knaft, 1959) to create an appropriate secondary task to study the inXuence of WM capacity on AS task performance in older adults (Eenshuistra et al., 2004). In a running memory task (or a WM updating task), a subject watches or listens to a sequence of items of unknown and variable length. The subject is asked to recall the most recent items of the just presented list serially and in the correct order. Continuously updating WM provides a continuous dynamic load on functional WM capacity, since both the WM storage component and the central executive are taxed (for an overview, see Baddeley, 1986, 2003). In the present study, we will apply our previous design to the developmental questions outlined above. An alternative hypothesis was also evaluated: it could be argued that an increase in the eYciency of inhibitory control can be held responsible for the expansion of available functional WM capacity. If developmental changes in the AS task were a consequence of a suboptimal eYciency of an inhibitory mechanism, then adding Wxation-oVset manipulations could modulate performance on the saccade tasks (e.g., Dorris & Munoz, 1995; Fischer et al., 1997; Forbes & Klein, 1996; Reuter-Lorenz, Hughes, & Fendrich, 1991). Since gap manipulations are thought to facilitate elementary saccade initiation processes and thus to eVectively disfacilitate suppression of saccades, such manipulations should be expected to have a relatively large impact on children’s AS performance as compared to the performance of young adults. In other words, in a gap manipulation children are expected to experience a disproportional increase in direction errors and a disproportional decrease in saccadic onset latencies as compared to young adults. Similarly, since overlap manipulations are thought to facilitate the active saccade Wxation system, children should be expected to beneWt more from the overlap manipulation than young adults, that is, children are expected to make relatively fewer direction errors and have disproportionably longer saccadic onset latencies. A full factorial combination of gap/overlap conditions and PS/AS tasks in combination with a control condition, in which a Wxation point is directly followed by the cue presentation (sometimes referred to as a step condition), will allow us to determine the nature of developmental eVects on the saccade Wxation and suppression system. This is one of the purposes of the present study. 2. Method 2.1. Participants The subjects were 19 8–9-year-old children (mean age D 8.8, SD D .41, 11 girls and 8 boys), 19 11–13-year-old children (mean age D 12.1, SD D .58, 11 girls and 8 boys), and 21 young adults (mean age D 22.0, SD D 1.7; 14 women and 7 men). Children were enlisted from a local primary school. Normal functioning in the classroom as determined by the child’s teacher was held as a criterion for participating in the experiment. Children were tested during school hours at the university laboratory. All children received a present for participating and the school received book tokens for every participant. Young adults were undergraduate students from the University of Amsterdam and they gained course credit for their participation.

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

145

A standard health questionnaire revealed that all of the participants were relatively healthy. For children the health questionnaire was Wlled in by their parents. None of the children had shown intellectual or learning problems. All subjects had normal or corrected-to-normal vision. Informed consent was obtained from all subjects prior to the experiment. For children informed consent was obtained from their parents. Data of four 8–9-year-olds, three 11–13-year-olds, and two young adults were discarded from the analyses, because their data were incomplete due to recording failures. In total, a sample size of 50 subjects was used for further analyses. 2.2. Tasks and apparatus The experiment was set up in conformity with the experimental design in the study of Eenshuistra et al. (2004). Four diVerent conditions were to be carried out: a WM update condition, two Wxation-oVset conditions, that is, a gap and an overlap condition, and a control condition. Each condition consisted of two diVerent eye-movement tasks: a PS task and an AS task. A subject thus had to perform a total of eight tasks. An individual measure of WM capacity was determined with two digit span tasks from the WAIS-R (Wechsler, 1981): the forward digit span task and the backward digit span task. 2.2.1. Stimuli Stimuli in the eye-movement tasks were displayed on a 15-in. VGA color monitor, controlled by an IBM compatible Pentium PC. The same stimulus set was used for both eyemovement tasks. This stimulus set was composed of 16 pictures from the standardized Snodgrass and Vanderwart (1980) stimulus set. For the memory-updating condition, recognition of the stimuli had to be unambiguous. Therefore, stimuli were chosen on the basis of an approximately equal score on image agreement, name agreement, and familiarity of the concept (see Eenshuistra et al., 2004; Table 1). The familiarity of the selected concepts was also established for the Dutch equivalents (De Vries, 1986; see also Eenshuistra et al., 2004; Table 1). Furthermore, it was determined whether the Dutch concept was already known by 8-year-olds. This was ascertained on the basis of a target vocabulary for 6-yearolds (Kohnstamm, 1981; see also Eenshuistra et al., 2004; Table 1). The identiWcation of the stimuli was facilitated further by coloring the images in prototypical colors (e.g., a red apple, a yellow star, and a brown chair). All stimuli were presented on a computer monitor against a white background. 2.2.2. Timing parameters All trials had the same basic structure. Each trial started with a black Wxation cross that remained at the center of the screen for 1667 ms. During or after the presentation of the Wxation cross, depending on the condition, a cue was displayed 13° to the left Table 1 Mean scores and standard deviations for children and young adults on the forward and backward digit span task

Forward digit span Backward digit span

8-Year-olds

12-Year-olds

Young adults

5.00 (1.81) 3.80(1.57)

5.63 (1.96) 5.25 (1.24)

7.21 (2.28) 6.89 (1.15)

146

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

or right of the Wxation cross. The cue, consisting of a black dot (with a diameter of 1.3 cm), was presented for 417 ms, and was followed by a stimulus. The stimulus remained on screen for 417 ms. Immediately after the oVset of the stimulus, a new trial started. 2.2.3. Eye-movement tasks The subjects were instructed to look at the Wxation cross until the cue was presented. In the PS task, the stimulus was presented at the same location as the cue location, and the subjects were asked to make as fast as possible an eye movement towards the position of the cue. In the AS task the stimulus was presented at the opposite site of the cue location, and the instruction was to make as fast as possible an eye movement to the opposite side of the cue location. PS trials and AS trials were presented in separate blocks. Eye movements were recorded with an infrared-based iView eyetracker (SMI) with 50 Hz temporal resolution and a <0.1° spatial resolution. The head was stabilized by means of a chin rest, which was located 40 cm in front of the monitor. 2.2.4. Conditions In each of the four conditions, a PS task and an AS task were to be performed. In both the control condition and the WM updating condition, the Wxation oVset was immediately followed by the onset of the cue. In the WM updating condition, stimuli were presented in seven lists with lengths of 2, 4 (twice), 6, 10, 12 or 16 stimuli, respectively. The subjects were not informed about the length of each list. The occurrence of these list lengths was randomized, as was the order of stimuli within the lists (with the restriction that each stimulus appeared once-only within a given list). The instruction was to perform a PS or AS task, and simultaneously remember the most recently presented stimulus in each trial. At the end of each list, a randomly selected image from the stimulus set was presented. The subject was asked if this image was the same as the one presented most recently. A response could be given by means of pressing one of two response buttons (labeled YES and NO). The subjects were to carry out this task as accurate as possible. In the gap condition, the cue was displayed 217 ms after the oVset of the Wxation cross. In the overlap condition, the cue temporally overlapped with the Wxation cross for 217 ms. In all other respects, the gap and overlap conditions were identical to the control condition (including the PS and AS instructions). 2.3. Procedure In each condition a PS task and an AS task had to be performed, and both tasks were to be carried out twice: once before and once after a break. The sequence of the tasks was balanced. Each task started with a verbal instruction, which was accompanied by 3 trial examples on screen. Before the break, the eye-movement tasks started with 13 practice trials. After the break, 7 practice trials were presented before each task. Prior to task presentation, the subjects were presented with 15 calibration trials, which served to calibrate the eyetracker’s measurement units to visual angle for that subject for that task. Each task consisted of 54 randomized trials (plus seven questions, one at the end of each list, in the memory-update conditions). The experimental session lasted for approximately three and a half hours; each individual task took about 10–15 min to complete. Short breaks were also allowed after each experimental task. The timing and pace was such that the children and adults could

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

147

complete the session without signs of substantial fatigue or vexation. Halfway through the experimental session, a health questionnaire was Wlled out during a 15 min break. 2.4. Statistical analyses Two dependent variables of primary interest were determined from the eye-movement data: saccade onset latencies and the proportion of direction errors. Saccade onset latencies were deWned as the time, relative to the onset of the cue, at which the velocity signal exceeded 18°/s, and the fovea displaced 8° from Wxation. The deWnition of direction errors was limited to those trials in which a saccade was made in the opposite direction of the stimulus (direction errors were typically followed by a corrective saccade in the correct direction). Two sets of ANOVA analyses were conducted on saccade onset latencies and proportion of direction errors with Age (8-year-olds, 12-year-olds and young adults) as a betweensubjects variable and Task (PS versus AS task) and Manipulation (the two Wxation-oVsets versus control in one set of analyses; WM updating versus control in the other set) as withinsubject variables. In addition, the digit span scores and the proportion correct answers on the WM-update-task questions were entered into an ANOVA with Age as a between-subjects variable. Greenhouse–Geisser adjustments were applied when appropriate. 3. Results 3.1. Digit span tasks Table 1 shows the scores on the forward and backward digit span tasks. As age increases, scores on both digit span tasks increase as well. This overall eVect of age was signiWcant for both digit span tasks, F(2, 47) D 5.37, p D .008, and F(2, 47) D 23.48, p < .001 for the forward and backward digit span task, respectively. However, post hoc analyses show that age diVerences in the forward digit span task only arise between the 8-year-old children and the young adults, p D .010, whereas all age groups diVer from each other in the backward digit span task, all p’s < .025. 3.2. Analysis of the WM updating manipulation 3.2.1. Manipulation check of WM updating task Table 2 shows the percentage correct answers on the questions that were posed in the WM update task. As can be seen all groups performed accurately on this task, indicating that all subjects properly adhered to the instructions. No signiWcant diVerences between the age groups were found, F(2, 49) D .51, p > .6. Table 2 Accuracy rates (%) and standard deviations for children and young adults on the working-memory update task questions

Prosaccade task Antisaccade task

8-Year-olds

12-Year-olds

Young adults

95.4 (9.55) 96.0 (4.62)

98.7 (2.88) 93.0 (11.07)

97.4 (4.89) 97.0 (4.95)

148

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

3.2.2. Saccade onset latencies Fig. 1a shows the mean onset latencies of the correct saccades in the control condition and the WM updating condition. Overall, PS onset latencies were faster than AS latencies, F(1, 46) D 485.18, p < .001. As age increases saccade onset latencies signiWcantly decrease, F(2, 46) D 10.58, p < .001. Age was also found to aVect AS task performance more than PS performance, Age £ Task: F(2, 46) D 9.92, p < .001. The WM updating manipulation caused a signiWcant increase in response latencies, F(1, 46) D 44.85, p < .001, and this eVect interacted signiWcantly with age (Manipulation £ Age), F(2, 46) D 9.00, p D .001. However, the WM manipulation only caused a marginally signiWcant eVect on task (Manipulation £ Task), F(1, 46) D 3.60, p D .064. The interaction of primary interest, the 3-way interaction between Age, Task and WM Manipulation did not reach signiWcance, F(2, 46) D 1.62, p D .208. 3.2.3. Direction errors Fig. 1b shows the accuracy in the WM updating condition as compared to the control condition. First of all, it is shown that greater proportions of errors were found in the AS

Fig. 1. (a) Mean saccade onset latencies as a function of age, saccade task and working-memory update manipulation. (b) Proportion of errors. (c) Saccadic onset latencies as a function of accuracy (saccadic latency⬘ D saccadic latency/(1 ¡ proportion of errors)).

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

149

task as compared to the PS task, F(1, 46) D 183.82, p < .001, and that the proportion of errors declines with age, F(2, 46) D 17.05, p < .001. Taxing WM led to a signiWcant increase in direction errors, F(1, 46) D 4.49, p D .024. This eVect diVered between PS and AS tasks (Manipulation £ Task), F(1, 46) D 8.61, p D .005, and age groups (Manipulation £ Age), F(2, 46) D 8.61, p D .016. More interesting, the 3-way interaction between Manipulation, Task and Age also reached signiWcance, F(2, 46) D 5.04, p D .010. It was found that the diVerence in AS performance between 8-year-old children and young adults primarily accounted for the signiWcant 3-way interaction eVect, p D .004. No age eVect was found between the 12-year-old children and young adults, p D .273, whereas the diVerence between 8-year-old children and 12-year-old children was marginally signiWcant, p D .07. The relationship between WM capacity and AS performance was furthermore expressed in the signiWcant correlation between AS accuracy and the scores on the backward digit span task (which is a particularly sensitive WM capacity measure): r2 D ¡ .45, p D .001. This means that subjects with lower scores on the digit span scores made more errors in the AS task. Closer inspection of Fig. 1a suggests that the WM manipulation aVected AS latencies of young adults relatively more than AS latencies of 8-year-old children. In other words, AS latencies increased substantially in young adults when WM was taxed, whereas AS latencies of 8-year-old children remained nearly unaVected by the WM manipulation. (An intermediate eVect was found in 12-year-old children.) It could be argued that the signiWcant diVerence in AS accuracy between young adults and 8-year-old children was obscured by a diVerence in speed-accuracy strategy. To control for such a possible speed-accuracy diVerence, saccadic onset latencies were adjusted according to a speed-accuracy correction proposed by Townsend and Ashby (1983), that is, saccadic latency⬘ D saccadic latency/ (1 ¡ proportion of errors). When saccade onset latencies were adapted, they do show an age-related pattern (see Fig. 1c). Saccade onset latencies of young adults signiWcantly diVered from those of 8-year-old children, F(1, 31) D 4,71, p D .004, when WM was taxed. The diVerence in accuracy performance between 12-year-old children and young adults remained non-signiWcant, F(1, 28) D 1.57, p D .22. 3.3. Analysis of the Wxation-oVset manipulations 3.3.1. Saccade onset latencies In Fig. 2a the mean onset latencies of the correct saccades in the Wxation-oVset conditions and the control condition are depicted. Increasing age is associated with faster saccade onset latencies in both saccade tasks. Both the main eVects of Task, F(1, 47) D 512.98, p < .001, and Age, F(2, 47) D 21.22, p < .001. The overall eVect of the Wxation oVset manipulations was also signiWcant, F(2, 94) D 71.00, p < .001. Planned comparisons revealed faster saccade onset latencies in the gap condition, (F(1, 47) D 33.50, p < .001, and slower saccade onset latencies in the overlap condition, F(1, 47) D 59.73, p < .001, as compared to the control condition. In general, the eVect of Task diVered signiWcantly across the Wxation-oVset manipulations, F(2, 94) D 10.65, p < .001, with more pronounced gap and overlap eVects in the PS tasks as compared to the AS tasks. Age failed to interact with the Wxation-oVset manipulations (Age £ Manipulation), F(4, 94) D 1.16, p D .334, and, more importantly, the 3-way interaction between Task, Age and Manipulation did not reach signiWcance either, F(4, 94) D .76, p D .552. As we already mentioned, the gap and overlap eVects were more pronounced in the PS tasks. When we deWne the gap eVect as the diVerence between overlap and gap performance

150

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

Fig. 2. (a) Mean saccade onset latencies as a function of age, saccade task and Wxation-oVset manipulation. (b) Proportion of errors. (c) The gap eVect (saccade onset latencies in the overlap condition – saccade onset latencies in the gap condition).

(see also Klein & Foerster, 2001; Munoz et al., 1998), this Wnding is expressed more clearly (see Fig. 2c). This diVerence in gap eVect in the PS tasks and AS tasks is highly signiWcant, F(1, 47) D 19.93, p < .001, but is once more not inXuenced by age, F(2, 47) D 1.04, p D .360. 3.3.2. Direction errors Fig. 2b represents the mean proportion of direction errors in the control condition as compared to both Wxation-oVset conditions. On the whole, more errors were made in the AS task as compared to the PS task, F(1, 47) D 151.90, p < .001. With increasing age the proportion of errors declined, F(2, 47) D 11.59, p < .001. Age also interacted with Task (Age £ Task), F(2, 47) D 13.10, p < .001. Further inspection of Fig. 2b shows that the agerelated decrease in proportion of direction errors mainly stems from diVerences in AS task performance, F(2, 47) D 12.49, p < .001, whereas performance on the PS task remains more constant across age, F(2, 47) D 2.77, p D .073.

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

151

The Wxation oVset manipulations yielded a signiWcant main eVect, F(2, 94) D 13.37, p < .001. As was expected less errors were made in the overlap condition as compared to the control condition, F(1, 47) D 15.67, p < .001, whereas the amount of direction errors in the gap condition and the control condition did not diVer, F(1, 47) D .58, p D .450. A signiWcant interaction between Wxation-oVset manipulation and task was found (Manipulation £ Task), F(2, 94) D 11.06, p < .001. Planned contrasts reveal that this interaction is obtained in the overlap condition as compared to the control condition, F(1, 47) D 16.33, p < .001, but not in the gap versus control condition, F(1, 47) D .001, p D .972. Importantly, Age did not interact with the Wxation-oVset manipulations (Age £ Manipulation), F(4, 94) D 1.96, p D .107, nor with the Wxation-oVset manipulations and task together (Age £ Manipulation £ Task), F(4, 94) D 1.40, p D .241. 4. Discussion This study was undertaken to establish the contributions of two diVerent constructs, inhibitory control and WM capacity, to developmental changes observed in AS performance. As was argued by Roberts et al. (1994), successful inhibition of prepotent reXexive PSs consumes a considerable proportion of WM capacity, and under the assumption that WM capacity is limited, inhibitory control becomes easily disturbed when WM capacity is insuYcient. We hypothesized that smaller WM capacity in children could explain age diVerences in inhibition in the AS task. Adding an external WM load, then, was expected to pronounce the age-related eVects. An alternative hypothesis was also examined. In contrast to the former hypothesis, it could be argued that ineYcient performance in the AS task is a result of ineYcient functioning of an inhibitory control mechanism (Bjorklund & Harnishfeger, 1990). In this case, Wxation-oVset manipulations could adjust developmental eVects, since Wxation-oVset manipulations have been argued to aVect the mechanisms of inhibitory oculomotor control in a direct manner. The Wndings of the present study reveal that the Wrst hypothesis had greater descriptive power with respect to the age-related performance in the AS task. Analysis of the direction errors showed a signiWcant age-related eVect only in the condition with the concurrent WM-updating task. By contrast, no speciWc age-related diVerences in the proportion of direction errors were found in gap or overlap conditions. These latter Wndings contradict the hypothesis that age-related changes in inhibitory control would be exaggerated in the gap condition and alleviated in the overlap condition relative to the control/step condition. Analyses of saccade latencies yielded less decisive results with respect to the two alternative hypotheses. The data displayed the anticipated eVects of age (AS latencies were slower than PS latencies, and this eVect was more pronounced for children than for young adults) and task manipulations (compared to the control condition, the decrease in AS latencies was marginally larger in the WM condition, larger in the gap condition, and smaller in the overlap condition). However, the age-related decrease in the AS eVect on eye movement latencies was similar across WM, gap, overlap, and control versions of the task. 4.1. Inhibition of eye-movements In general, the Wxation-oVset manipulations yielded a pattern of results comparable to several other studies in which developmental eVects in gap and overlap conditions were

152

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

examined (Fischer et al., 1997; Klein & Foerster, 2001; Munoz et al., 1998). The results are in accordance with the model of saccade generation outlined in the Introduction. That is, gap conditions resulted in AS performance declines presumably because they facilitate the reXexive saccade generation system and impair the elementary Wxation system, while overlap conditions resulted in AS performance improvements presumably because they disfacilitate the exogenous saccade initiation system and facilitate the Wxation system. The elementary saccade initiation and Wxation systems are implemented predominantly at subcortical levels. While frontal cortical regions (most prominently DLPFC) are especially characterized by considerable developmental changes (e.g., Fuster, 1997; Van der Molen & Ridderinkhof, 1998), the subcortical mechanisms (capitalizing on the SC) are already (functionally) matured in the infant (Stampalija & Kostovic, 1981; Yakovlev & Lecours, 1967), suggesting a relatively intact mechanism for active Wxation at an early age. Indeed, gap and overlap conditions were not observed to interact with age changes in AS performance. It remains possible however that more pronounced developmental changes will be observed when experimental manipulations are used that challenge the endogenous saccade generation system and, especially, the selective saccade suppression system. If such manipulations reveal developmental changes in oculomotor inhibition, this might even suggest that oculomotor inhibition is diVerentiated and that some oculomotor inhibitory processes, such as inhibitory requirements tapped by Wxation oVset manipulations, are relatively unaVected (cf. Nigg, 2000). Some reservations regarding the interpretation of Wxation-oVset manipulations have been reported. Our Wndings show that the Wxation oVset manipulations mainly aVected PS latencies (see Fig. 2c). This Wnding was reported previously by Reuter-Lorenz et al. (1991), Reuter-Lorenz, Oonk, Barnes, & Hughes (1995), and also in developmental (Klein & Foerster, 2001) and aging studies (Forbes & Klein, 1996). If the WxationoVset conditions manipulate the eYciency of an active Wxation mechanism, and if this mechanism serves to inhibit the initiation of saccades in general, then it is not clear why the Wxation-oVset manipulations should inXuence the latencies of PSs more than those of ASs. Reuter-Lorenz et al. (1991), Reuter-Lorenz et al. (1995) suggested that the release of inhibition in Wxation-oVset conditions is collicular-based, and since the superior colliculus presumably plays a greater role in the generation of reXexive PSs compared to voluntary ASs, it makes sense that the eVect of Wxation-oVset manipulations should be greater in PSs. And since ASs have longer latencies, it can be assumed that processes underlying the generation of ASs have a more protracted time course as compared to the relatively fast process of Wxation release. Fixation release eVects as a result of Wxation-oVset manipulations, then, could well occur with ASs, but the eVect is more or less absorbed by the other slower AS processes. It remains possible that the Wxation oVset manipulations were not powerful enough to detect age-related changes in oculomotor inhibition. However, the present manipulations have been advocated and used widely in the literature as an eVective manipulation of oculomotor inhibition. Making the gap condition more diYcult, for example by prolonging the gap period, will not likely result in more speciWc age-related eVects. The few studies that investigated the gap paradigm with longer gap periods found that the most pronounced gap eVects are observed when gap periods between 150 and 250 ms are used. Longer gap intervals (between 250 and 600) weaken the gap eVect, instead of increasing it (e.g., Fischer & Weber, 1997; Krauzlis & Miles, 1996).

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

153

Some studies have used analogues of the antisaccade task in four-month-old infants (Johnson, 1995) and 12–30 month-old toddlers (Scerif et al., 2005). When a target stimulus was always paired with the contralateral presentation of a cue at the opposite location, both groups of young children learnt to avoid looking at the cue location. Fixation oV-set manipulations have not been tested systematically in combination with the antisaccade paradigm for this younger age group, leaving open the possibility that the eVects of Wxation oVsets on oculomotor control may be subject to development in this age range (e.g., Csibra, Tucker, & Johnson, 1998; Csibra, Tucker, Volein, & Johnson, 2000; Hood & Atkinson, 1993). 4.2. Working-memory capacity Adding a WM load to the AS task led to a signiWcant disturbance in accuracy of the youngest children, whereas performance of the other two age groups was less aVected. If the amount of available WM capacity is the primitive that drives cognitive performance (Conway & Engle, 1994, 1996), including inhibitory control, and if children are aZicted with less functional WM capacity as compared to young adults, then this diVerence in availability of WM capacity could account for developmental diVerences in cognitive performance, and more speciWcally in developmental diVerences in inhibitory control. In view of Kane et al. (2001) demonstration that the eYciency of AS performance depends on the amount of available WM capacity in young adults, the present study shows that diVerences in WM capacity oVer an elegant explanation for developmental diVerences in AS performance. Twelve-year-old children diVered signiWcantly from young adults in their scores on the digit span tasks, indicating that 12-year-olds have less WM capacity at their disposal as compared to young adults. Nevertheless, the present study yielded an absence of any speciWc age-related eVects in 12-year-old children as compared to young adults. It is possible that the WM manipulation was not diYcult enough to aVect accuracy performance in subjects aged 12 or older. In other words, 12-year-old children had enough functional WM capacity available for correct AS performance despite the load that was posed on WM capacity. Support for such an explanation comes from the study of Roberts et al. (1994), in which it was demonstrated that a less diYcult WM manipulation barely aVected the AS performance of young adults. Alternatively, the assumption of a diVerentiated development of AS performance oVers another explanation (cf., Klein & Foerster, 2001). They showed that when the performance of 11-year-old children was compared to that of young adults, some aspects of AS performance were already matured at age 11, whereas other aspects show global aging eVects, and again others show speciWc age-related eVects. A limitation of the present study concerns the use of the WM-update task. In this study, the subjects only had to remember the last item of each series of items of the WM-update task. Strictly speaking, simply adding a task with a Wxed load to the AS task creates a dual task situation. It could be argued then that the decrease in AS performance can be ascribed to developmental diVerences in switching attention between tasks or dual-task interference. However, such a conclusion is not necessarily in contradiction with our view, since one might assume that both divided attention and resistance to interference depend on available WM capacity as well. The pattern of results found in the present study corresponds with the results of an aging study of Eenshuistra et al. (2004) in which the same tasks were to be performed by

154

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

older subjects. Older subjects showed longer saccadic onset latencies and a disproportional increase in direction errors as compared to young adults in the WM updating condition, suggesting to the authors that an age-related diminution in WM capacity underlies these age-related changes in AS task performance. The Wndings of Roberts et al. (1994) & Kane et al. (2001), together with the aging study of Eenshuistra et al. (2004), strengthen the view of the present study that development of eYcient inhibition over reXexive saccades might be mediated by developmental diVerences in available functional WM capacity, since developmental deWciencies in the control over reXexive responses only emerge when WM requirements are such that insuYcient capacity is left for inhibitory control. In a recent study, Kramer, Gonzalez de Sather, & Cassavaugh (2005) also concentrated on the developmental changes in oculomotor inhibition. Kramer et al. compared children’s performance on an oculomotor capture task with antisaccade task performance. The former task does require the inhibition of eye movements to an abrupt onset of a stimulus; however, this stimulus never serves as target, nor does it predict the location of the target as in the AS task. Contrary to antisaccade performance, performance on the oculomotor capture task was relatively age-invariant. These results were interpreted in terms of WM functioning. Put diVerently, whereas the AS task requires that various instructions (i.e., to detect the cue, inhibit an eye movement towards the cue, and make an eye movement in the opposite direction of the cue towards the target) have to be maintained in WM, the WM demands of the oculomotor capture task are minimal, since this task only requires an eye movement towards a readily distinguishable target. Interestingly, the absence of age eVects in the oculomotor capture task, together with the presently observed absence of any age-speciWc eVects on the Wxation-oVset manipulations, seems to support the assumption that the eYciency of inhibitory oculomotor control does not alter as children are growing older. 4.3. Conclusion Consistent with several recent Wndings, the present data foster support for the hypothesis that age diVerences in inhibition in the AS task are explained by smaller WM capacity in children rather than by ineYcient functioning of inhibitory control mechanisms. Thus, despite the observations of pronounced developmental maturation in some domains of inhibitory control (e.g., in response inhibition as measured with stop tasks; Ridderinkhof, Band, & Logan, 1999), such Wndings do not necessarily generalize to inhibitory oculomotor control. As suggested by Nigg (2000) inhibitory control should not be seen as one construct, instead diVerent kinds of inhibitions should be distinguished each being sensitive to diVerent kinds of tasks and factors such as development and pathology. Future research eVorts in the booming Weld of developmental cognitive neuroscience (e.g., Munakata, Casey, & Diamond, 2004) may aim at further examining the extent to which age changes in inhibition are fundamental in and of themselves (cf. Kirkham & Diamond, 2003), or secondary to developmental trends in other mechanisms such as functional WM or perhaps even higher-order levels of reXection (cf. Zelazo, 2004). Acknowledgement The assistance of Monique Geers in setting up the experiments and collecting the data is gratefully acknowledged.

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

155

References Anderson, R. W., Keller, E. L., Gandhi, N. J., & Das, S. (1998). Two-dimensional saccade-related population activity in superior colliculus in monkey. Journal of Neurophysiology, 80, 798–817. Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition and the right inferior frontal cortex. Trends in Cognitive Sciences, 8, 170–177. Baddeley, A. (1986). Working memory. London: Oxford University Press. Baddeley, A. (2003). Working memory: looking back and looking forward. Nature Reviews Neuroscience, 4(10), 829–839. Bjorklund, D. F., & Harnishfeger, K. K. (1990). The resources construct in cognitive development: diverse sources of evidence and a theory of ineYcient inhibition. Developmental Review, 10, 48–71. Burman, D. D., & Bruce, C. J. (1997). Suppression of task-related saccades by electrical stimulation in the primate’s frontal eye Weld. Journal of Neurophysiology, 77(5), 2252–2267. Case, R. (1985). Intellectual development. Birth to adulthood New York: Academic Press. Casey, B. J., Castellanos, F. X., Giedd, J. N., Marsh, W. L., Hamburger, S. D., Anne, B., et al. (1997). Implication of right frontostriatal circuitry in response inhibition and attention-deWcit/hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 374–383. Casey, B. J., Tottenham, N., Liston, C., & Durston, S. (2005). Imaging the developing brain: what have we learned about cognitive development? Trends in Cognitive Sciences, 9(3), 104–110. Conway, A. R. A., & Engle, R. W. (1994). Working memory and retrieval: A resource-dependent inhibition model. Journal of Experimental Psychology: General, 123, 354–373. Conway, A. R. A., & Engle, R. W. (1996). Individual diVerences in working memory capacity: more evidence for a general capacity theory. Memory, 4(6), 577–590. Csibra, G., Tucker, L. A., & Johnson, M. H. (1998). Neural correlates of saccade planning in infants: a high-density ERP study. International Journal of Psychophysiology, 29, 201–215. Csibra, G., Tucker, L. A., Volein, A., & Johnson, M. H. (2000). Cortical development and saccade planning: the ontogeny of the spike potential. Neuroreport, 11, 1069–1073. Dempster, F. N. (1981). Memory span: sources of individual and developmental diVerences. Psychological Bulletin, 89, 63–100. Dempster, F. N. (1985). Short-term memory development in childhood and adolescence. In C. J. Brainerd & M. Pressley (Eds.), Basic processes in memory development (pp. 208–248). New York: Springer. Dempster, F. N. (1992). The rise and fall of the inhibitory mechanism: toward a uniWed theory of cognitive development and aging. Developmental Review, 12, 45–75. Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual attention. Annual Review of Neuroscience, 18, 193–222. De Vries, K. L. M. (1986). Vertrouwdheid van woorden en woordherkenning [Familiarity of words and word recognition]. Nederlands Tijdschrift voor de Psychologie, 41, 295–304. Diamond, A. (1990). Developmental time course in human infants and infant monkeys, and the neural bases of inhibitory control in reaching. In A. Diamond (Ed.), The development and neural bases of higher cognitive functions. Annals of the New York Academy of Sciences, vol. 608 (pp. 267–317). Dorris, M. C., & Munoz, D. P. (1995). A neural correlate for the gap eVect on saccadic reaction times in the monkey. Journal of Neurophysiology, 73, 2558–2562. Dorris, M. C., & Munoz, D. P. (1998). Saccadic probability inXuences motor preparation signals and time to saccadic initiation. Journal of Neuroscience, 18(17), 7015–7026. Durston, S., Thomas, K. M., Yang, Y. H., Ulug, A. M., Zimmerman, R. D., & Casey, B. J. (2002). A neural basis for the development of inhibitory control. Developmental Science, 5(4), F9–F16. Eenshuistra, R. M., Ridderinkhof, K. R., & van der Molen, M. W. (2004). Age-related changes in antisaccade task performance: inhibitory control or working-memory engagement? Brain & Cognition, 56, 177–188. Egner, T., & Hirsch, J. (2005). Cognitive control mechanisms resolve conXict through cortical ampliWcation of task-relevant information. Nature Neuroscience, 8, 1784–1790. Everling, S., Dorris, M. C., Klein, R. M., & Munoz, D. P. (1999). Role of primate superior colliculus in preparation and execution of anti-saccades and pro-saccades. Journal of Neuroscience, 19(7), 2740–2754. Everling, S., & Fischer, B. (1998). The antisaccade: a review of basic research and clinical studies. Neuropsychologia, 36(9), 885–899. Fischer, B., Biscaldi, M., & Gezeck, S. (1997). On the development of voluntary and reXexive components in human saccade generation. Brain Research, 754, 285–297.

156

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

Fischer, B., & Weber, H. (1997). EVects of stimulus conditions on the performance of antisaccades in man. Experimental Brain Research, 116, 191–200. Forbes, K., & Klein, R. M. (1996). The magnitude of the Wxation oVset eVect with endogenously and exogenously controlled saccades. Journal of Cognitive Neuroscience, 8(4), 344–352. Fukushima, J., Hatta, T., & Fukushima, K. (2000). Development of voluntary control of saccadic eye movements I. Age-related changes in normal children. Brain & Development, 22, 173–180. Fuster, J. M. (1997). The prefrontal cortex. Anatomy, physiology, and neuropsychology of the frontal lobe Philadelphia: Lippincott-Raven Publishers. Gathercole, S. E. (1999). Cognitive approaches to the development of short-term memory. Trends in Cognitive Sciences, 3, 410–419. Goldman-Rakic, P. S. (1987). Circuitry of primate prefrontal cortex and regulation of behavior by representational memory. In F. Plum (Ed.), Higher functions of the brain, Part 1 (Vol. 5). Handbook of physiology: The nervous system: Section 1 (pp. 373–417). Bethesda: American Physiological Society. Hallett, P. E. (1978). Primary and secondary saccades to goals deWned by instructions. Vision Research, 18, 1279– 1296. Harnishfeger, K. K., & Bjorklund, D. F. (1993). The ontogeny of inhibition mechansisms: a renewed approach to cognitive development. In M. L. Howe & R. Pasnak (Eds.), Emerging themes in cognitive development: Foundations (Vol. 1, pp. 28–49). New York: Springer. Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging: a review and a new view. In G. H. Bower (Ed.), The psychology of learning and motivation (vol. 22, pp. 193–225). New York: Academic Press. Hikosaka, K., & Watanabe, M. (2000). Delay activity of orbital and lateral prefrontal neurons of the monkey varying with diVerent rewards. Cerebral Cortex, 10(2), 263–271. Hood, B. M., & Atkinson, J. (1993). Disengaging visual attention in the infant and adult. Infant Behaviour and Development, 16, 405–422. Johnson, M. H. (1994). Brain and cognitive development in infancy. Current Opinion in Neurobiology, 4, 218–225. Johnson, M. H. (1995). The inhibition of automatic saccades in early infancy. Developmental Psychobiology, 28, 281–291. Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001). A controlled-attention view of workingmemory capacity. Journal of Experimental Psychology: General, 30, 169–183. Kirkham, N. Z., & Diamond, A. (2003). Sorting between theories of perseveration: performance in conXict tasks requires memory, attention and inhibition. Developmental Science, 6, 474–476. Klein, C., & Foerster, F. (2001). Development of prosaccade and antisaccade task performance in participants aged 6 to 26 years. Psychophysiology, 38, 179–189. Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Increased brain activity in frontal and parietal cortex underlies the development of visuospatial working memory capacity during childhood. Journal of Cognitive Neuroscience, 14(1), 1–10. Kohnstamm, G. A., Schaerlaekens, A. M., de Vries, A. K., Akkerhuis, G. W., & Froonincksx, M. (1981). Nieuwe StreeXijst Woordenschat voor 6-jarigen (New word vocabulary list for six year olds). Lisse: Swets & Zeitlinger. Kramer, A. F., Gonzalez de Sather, J. C. M., & Cassavaugh, N. D. (2005). Development of attentional and oculomotor control. Developmental Psychology, 41(5), 760–772. Krauzlis, R. J., & Miles, F. A. (1996). Decreases in the latency of smooth pursuit and saccadic eye movements produced by the “gap paradigm” in the monkey. Vision Research, 36, 1973–1985. Luciana, M., & Nelson, C. A. (1998). The functional emergence of prefrontally guided working memory systems in four- to eight-year-old children. Neuropsychologia, 36, 273–293. Matsuda, T., Matsuura, M., Ohkubo, T., Ohkubo, H., Matsushima, E., Inoue, K., et al. (2004). Functional MRI mapping of brain activation during visually guided saccades and antisaccades: cortical and subcortical networks. Psychiatry Research: Neuroimaging, 131, 147–155. Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24, 167–202. Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: a latent variable analysis. Cognitive Psychology, 41, 49–100. Moschovakis, A. K., Scudder, C. A., & Highstein, S. M. (1996). The microscopic anatomy and physiology of the mammalian saccadic system. Progress in Neurobiology, 50(2–3), 133–254. Munakata, Y., Casey, B. J., & Diamond, A. (2004). Developmental cognitive neuroscience: progress and potential. Trends in Cognitive Sciences, 8, 122–128.

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

157

Munoz, D. P., Broughton, J. R., Goldring, J. E., & Armstrong, I. T. (1998). Age-related performance of human subjects on saccadic eye movement tasks. Experimental Brain Research, 121, 391–400. Munoz, D. P., & Wurtz, R. H. (1993). Fixation cells in monky superior colliculus. I. Characteristics of cell discharge. Journal of Neurophysiology, 70, 559–575. Munoz, D. P., & Wurtz, R. H. (1995). Saccade-related activity in monkey superior colliculus. I. Characteristics of burst and buildup cells. Journal of Neurophysiology, 73, 2313–2333. Nelson, C. A., Monk, C. S., Lin, J., Carver, L. J., Thomas, K. M., & Truwit, C. L. (2000). Functional neuroanatomy of spatial working memory in children. Developmental Psychology, 36, 109–116. Nigg, J. T. (2000). On inhibition/disinhibition in developmental psychopathology: views from cognitive and personality psychology and a working inhibition taxonomy. Psychological Bulletin, 126, 220–246. Olesen, J. P., Westerberg, H., & Klingberg, T. (2004). Increased prefrontal and parietal activity after training of working memory. Nature Neuroscience, 7(1), 75–79. O’Reilly, R. C., Braver, T. S., & Cohen, J. D. (1999). In A. Miyake & P. Shah (Eds.), A biologically based computational model of working memory. Models of working memory: Mechanisms of active maintenance and executive control. New York: Cambridge University Press. Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget’s developmental stages. Acta Psychologica, 32, 301–345. Pennington, B. F. (1994). The working memory function of the prefrontal cortices: implications for developmental and individual diVerences in cognition: Implications for developmental and individual diVerences in cognition. In M. M. Haith, J. B. Benson, R. J. Roberts Jr., & B. F. Pennington (Eds.), Future-oriented processes in development. Chicago: University of Chicago Press. Pierrot-Deseilligny, C., Müri, R. M., Ploner, C. J., Gaymard, B., Demeret, S., & Rivaud-Pechoux, S. (2003). Decisional role of the dorsolateral prefrontal cortex in ocular motor behaviour. Brain, 126, 1460–1473. Pollack, I., Johnson, L., & Knaft, P. (1959). Running memory span. Journal of Experimental Psychology, 57, 137– 146. Reuter-Lorenz, P. A., Hughes, H. C., & Fendrich, R. (1991). The reduction of saccadic latency by prior oVset of the Wxation point: an analysis of the “gap eVect”. Perception & Pscyhophysics, 49, 167–175. Reuter-Lorenz, P. A., Oonk, H. M., Barnes, L. L., & Hughes, H. C. (1995). EVects of warning signals and Wxation point oVsets on the latencies of pro-versus antisaccades: implications for an interpretation of the gap eVect. Experimental Brain Research, 103, 287–293. Ridderinkhof, K. R., Band, G. P. H., & Logan, G. D. (1999). A study of adaptive behavior: eVects of age and irrelevant information on the ability to inhibit one’s actions. Acta Psychologica, 101, 315–337. Ridderinkhof, K. R., Van den Wildenberg, W. P. M., Segalowitz, S. J., & Carter, C. S. (2004). Neurocognitive mechanisms of cognitive control: the role of prefrontal cortex in action selection, response inhibition, performance monitoring, and reward-based learning. Brain and Cognition, 56, 129–140. Roberts, R. J., Jr., Hager, L. D., & Heron, C. (1994). Prefrontal cognitive processes: working memory and inhibition in the antisaccade task. Journal of Experimental Psychology: General, 123(4), 374–393. Roberts, R. J., Jr, & Pennington, B. F. (1996). An interactive framework for examining prefrontal cognitive processes. Developmental Neuropsychology, 12, 105–126. Sato, M., & Hikosaka, O. (2002). Role of primate substantia nigra pars reticulata in reward-oriented saccadic eye movement. Journal of Neuroscience, 22(6), 2363–2373. Scerif, G., KarmiloV-Smith, A., Campos, R., Elsabbagh, M., Driver, J., & Cornish, K. (2005). To look or not to look? Typical and atypical development of oculomotor control. Journal of Cognitive Neuroscience, 4, 591– 604. Schall, J. D., Morel, A., King, D. J., & Bullier, J. (1995). Topography of visual–cortex connections with frontal eye Weld in macaque – convergence and segregation of processing streams. Journal of Neuroscience, 15(6), 4464– 4487. Schall, J. D., Stuphorn, V., & Brown, J. W. (2002). Monitoring and control of action by frontal lobes. Neuron, 36, 309–322. Schlag-Rey, M., Schlag, J., & Dassonville, P. (1992). How the frontal eye Weld can impose a saccade goal on superior colliculus neurons. Journal of Neurophysiology, 67, 1003–1005. Smith, E. E., & Jonides, J. (1998). Neuroimaging analyses of human working memory. Proceedings of the National Acadamy of Science, 95, 12061–12068. Snodgrass, J. G., & Vanderwart, M. (1980). A standardized set of 260 pictures: norms for name agreement, image agreement, familiarity, and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6(2), 174–215.

158

R.M. Eenshuistra et al. / Acta Psychologica 124 (2007) 139–158

Stampalija, A., & Kostovic, I. (1981). The laminar organization of the superior colliculus (SC) in the human fetus. In A. Huber & D. Klein (Eds.), Neurogenetics and neuro-opthalmology. North-Holland: Elsevier. Swanson, L. H. (1996). Individual and age-related diVerences in children’s working memory. Memory & Cognition, 24(1), 70–82. Takikawa, Y., Kawagoe, R., & Hikosaka, O. (2002). Reward-dependent spatial selectivity of anticipatory activity in monkey caudate neurons. Journal of Neurophysiology, 87, 508–515. Thomas, K. M., King, S. W., Franzen, P. L., Welsh, T. F., Berkowitz, A. L., Noll, D. C., et al. (1999). A developmental functional MRI study of spatial working memory. Neuroimage, 10, 327–338. Townsend, J. T., & Ashby, F. G. (1983). The stochastic modeling of elementary psychological processes. Cambridge: Cambridge University Press. Van der Molen, M. W., & Ridderinkhof, K. R. (1998). The growing and aging brain: life-span changes in brain and cognitive functioning. In A. Demetrious, W. Doise, & C. F. M. van Lieshout (Eds.), Life-span developmental psychology: A European perspective (pp. 35–99). London: Wiley. Wechsler, D. (1981). Wechsler Adult Intelligence Scale – Revised. San Antonio, TX: The Psychological Corporation. Wurtz, R. H., & Goldberg, M. E. (Eds.). (1989). The neurobiology of saccadic eye-movements. New York: Elsevier. Yakovlev, P. I., & Lecours, A. (1967). The myelogenetic cycles of regional maturation of the brain. In A. Minokowski (Ed.), Regional development of the brain in early life. Philadelphia: Davis. Zelazo, P. D. (2004). The development of conscious control in childhood. Trends in Cognitive Sciences, 8, 12–17.