Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/more-to-adhd-than-meets-the-eye-observable-abnormalities-in-search-mLK2sTdd8B
- Publisher
- Springer Journals
- Copyright
- Copyright © 2005 by Sonuga-Barke et al; licensee BioMed Central Ltd.
- Subject
- Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
- eISSN
- 1744-9081
- DOI
- 10.1186/1744-9081-1-10
- pmid
- 16033644
- Publisher site
- See Article on Publisher Site
Abstract
Children with Attention Deficit/Hyperactivity Disorder (ADHD) often perform poorly on tasks requiring sustained and systematic attention to stimuli for extended periods of time. The current paper tested the hypothesis that such deficits are the result of observable abnormalities in search behaviour (e.g., attention-onset, -duration and -sequencing), and therefore can be explained without reference to deficits in non-observable (i.e., cognitive) processes. Forty boys (20 ADHD and 20 controls) performed a computer-based complex discrimination task adapted from the Matching Familiar Figures Task with four different fixed search interval lengths (5-, 10-, 15- and 20- s). Children with ADHD identified fewer targets than controls (p < 0.001), initiated searches later, spent less time attending to stimuli, and searched in a less intensive and less systematic way (p's < 0.05). There were significant univariate associations between ADHD, task performance and search behaviour. However, there was no support for the hypothesis that abnormalities in search carried the effect of ADHD on performance. The pattern of results in fact suggested that abnormal attending during testing is a statistical marker, rather than a mediator, of ADHD performance deficits. The results confirm the importance of examining covert processes, as well as behavioural abnormalities when trying to understand the psychopathophyiology of ADHD. Attention Deficit/Hyperactivity Disorder (ADHD; [1]) is a secutively presented distractors (e.g., Continuous Per- disorder of childhood and adolescence characterised by a formance Task; CPT – [3]). It is also true of more complex pattern of extreme, pervasive, persistent and debilitating tasks that require self-directed and controlled search for inattention, overactivity and impulsiveness. Children targets amongst multiple concurrently presented distrac- with ADHD are more likely than their peers to experience tors (e.g., Matching Familiar Figures Task; [4]). educational under-achievement, social isolation and anti- social behaviour during the school years and to go on to In trying to explain the causes of this commonly observed have significant difficulties in the post-school years. Chil- pattern of performance deficit a range of different mecha- dren with ADHD often perform poorly on tasks requiring nisms operating at different levels of analysis have been the sustained and systematic allocation of attention over invoked. For instance, cognitive accounts link deficits in periods of extended time [2]. This appears to be true of performance to impairments in covert processes such as tasks that require vigilance for rare targets amongst con- information encoding and retrieval as well as the 'holding Page 1 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 in mind' of targets and their systematic comparison to dis- has been almost no enquiry into the structure of ADHD tractors [5-7]. Such analyses fall firmly within the domain children's search or its level of organisation. Furthermore occupied by contemporary models of ADHD which there has been no formal analysis of whether any abnor- emphasise its cognitive character [[8-10] for a discussion] malities in the quantitative aspects of search described and appear to receive considerable support from experi- above in fact account for, or in statistical terms mediate, mental studies of cognitive performance: ADHD children performance deficits in ADHD. perform poorly on tasks thought to tap a range of execu- tive and non-executive cognitive skills such as working The study presented here compares the quality and timing and spatial memory, planning, attentional flexibility and of aspects of ADHD and control children's search behav- inhibition [11-17]. iour while they are performing a computer-based fixed- trial version of the MFFT. The key questions were; (i) Despite this strong body of evidence for the existence of whether ADHD children differ from controls in these cognitive deficits and the compelling nature of the cogni- aspects of search behaviour and (ii) whether these differ- tive deficit account, performance on complex discrimina- ences in search behaviour, if identified, would mediate tion tasks such as those described above can, in fact, be the association between ADHD and poor task perform- explained much more straightforwardly without invoking ance. According to the statistical concept of mediation deficits in non-observable cognitive processes. This is support for this hypothesis requires four predictions be because effective performance on these tasks requires the confirmed. provision, protection and, systematic and skilled use of available processing time. This means one could account I – that there is an association between ADHD and per- for the poor performance of ADHD children on such tasks formance with children with ADHD performing more purely in terms of their tendency to (i) start to attend later, poorly than controls. and to terminate searches earlier – so producing a shorter duration of attention than controls (i.e., quantitative II – that there is an association between ADHD and search aspects of attending), and/or (ii) employ less systematic style with children with ADHD engaging in less systematic sequencing of attention to individual stimuli and to look or efficient search behaviour than controls at a smaller proportion of stimuli before trying to identify a target (i.e., qualitative aspects of search). Importantly if III – that search-related style and behaviour are related to these quantitative and qualitative abnormalities in performance in the sample as a whole. observed search behaviour exist they could affect the per- formance of ADHD children whether, or not, underlying, IV – that the association between ADHD and performance unobservable cognitive abilities are intact. Furthermore is lost when the association between ADHD and search the theoretical significance of such a finding would be style and behaviour is controlled. considerable. This is because it would in principle offer an explanation of poor performance on many laboratory In the current study we set out to test these predictions tests; even those that have been used to index the non- using 4 different fixed trial conditions in which children observable cognitive processes discussed above. This in had 5-, 10-, 15- or 20 s to identify the target from amongst turn would caste doubt on the essentially cognitive nature foils. These different time intervals were used to examine of ADHD. the extent to which ADHD and control children adapt their search style to different temporal constraints and to Given the potential significance of the search-based see whether the impact of restricted search behaviours account of task performance it is surprising that there has becomes more important at particular search intervals. not been more study of the quantitative and qualitative characteristics of search behaviour in ADHD. There is In this study we used a novel approach to measuring some evidence that children with ADHD fail to provide search behaviour that did not employ eye gaze measure- and exploit sufficient processing time during search. This ments. It was based on an analysis of different aspects of appears to be partly because of difficulty either sustaining the timing and order in which initially 'covered' individ- attention or modulating attentional fluctuations that ual targets on the MFFT were accessed for viewing. The occur over time or in protecting such time from interfer- assumption in this study, therefore, is that targets accessed ence by extraneous stimuli [3,18-20]. Furthermore Kara- for inspection were actually inspected. This assumption tekin and Asarnow, [21] found that ADHD children could be tested in future studies using more traditional initiated searches later than controls and fixated for eye-tracking techniques. shorter periods of time on more demanding tasks. Frank et al., [22] suggested that ADHD children's searches are self-terminating rather than exhaustive. However, there Page 2 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 Table 1: Performance and search characteristics of the ADHD and control groups. Mean Performance & Search Style (standard deviation) F Statistics 5-s 10-s 15-s 20-s Group Interval GxI correct Control 4.15 (2.21) 6.10 (1.68) 6.30 (2.20) 5.94 (1.84) 13.71* 13.61** 0.6 responses 10.25** AD/HD 2.56 (1.13) 4.31 (1.80) 4.60 (2.26) 5.15 (2.36) number Control 1.54 (0.37) 3.31 (0.50) 4.55 (1.04) 4.22 (1.14) 7.09* 138.42** 5.12** boxes open 36.82** AD/HD 1.30 (0.37) 2.51 (0.79) 3.42 (1.10) 4.14 (1.27) look time per Control 3.03 (0.24) 6.58 (0.68) 10.10 (1.50) 14.38 (1.47) 6.10* 548.9** 0.71 1.54 trial (s) AD/HD 2.67 (0.47) 5.91 (1.11) 9.96 (1.45) 12.99 (2.71) search Control 1.49 (0.29) 1.58 (0.43) 1.61 (0.45) 1.65 (0.42) 4.85* 3.64* 1.77 1.07 Initiation (s) AD/HD 1.59 (0.51) 2.02 (0.91) 1.99 (0.75) 2.00 (0.77) systematic Control 5.00 (2.61) 7.95 (1.43) 7.46 (1.23) 7.62 (1.72) 25.76** 39.25** 1.89 searches 36.07** AD/HD 2.35 (1.98) 5.20 (1.57) 6.15 (2.10) 5.52 (2.11) Note: AD/HD = Attention Deficit/Hyperactivity Disorder. Figures in parentheses are standard deviations. * p < 0.05; ** p < 0.01. Italicised figures in F column relate to quadratic contrasts. 0.05. The effect of group and interval interacted signifi- Results Do AD/HD children perform less well than controls? cantly only in the case of the number of boxes opened. The number of correct responses at each inspection inter- This interaction was associated with a levelling off of the val is reported in Table 1 along with the F statistics and number searches made by children in the control group significance from a two-way ANOVA with group (ADHD between the 15- and 20-s intervals. However, this interac- v controls) as the between subjects factor and inspection tion was much reduced when conduct and emotional interval (5-, 10-, 15-, 20-s) as the within subjects factor. problems were controlled, F (3,34) = 3.41; p < 0.05. There were effects of both inspection interval and group but no interaction between these factors. Controls outper- Was AD/HD search style associated with poorer formed children with ADHD at each level although these performance in the whole sample? differences only reached significance at 5-, t(38) = -2.85; p In order to address this question in an easily comprehen- < 0.01, 10-, t(38) = -3.22; p < 0.005 and 15-s, t(38) = - sible way search style and performance variables were col- 2.44; p < 0.05, but not at 20-s, t(39) = -1.18; p > 0.20. The lapsed across search intervals (5-, 10-, 15- and 20-s). This significance of the quadratic term suggested that as one was justified on the grounds that in general there might expect children's performance demonstrated a appeared to be no interaction between inspection interval diminishing level of gain for each additional 5 s of and group. This suggested that differences in search style processing time. Introducing conduct or emotional prob- and performance demonstrated between the groups were lems as covariates into the analysis had no effect on this not systematically influenced by search interval (except pattern of significance, F(1,36) = 13.84; p < 0.001. for number of boxes opened). Furthermore interval spe- cific measures of each variable showed good internal con- Was ADHD search behaviour less systematic, intensive and sistency. Cronbach's α for number of systematic searches sustained than that of controls? (.81), number of boxes opened (.75), search initiation There was an effect of interval on all four search measures (.77), total look time (.55) and number of correct (.63) with the number of boxes opened, the length of time the were all in the acceptable range. These aggregate scores for boxes were looked at, the number of systematic searches performance was significantly correlated with systematic and time at which searches were initiated all increasing as searches (.30), number of boxes opened (.32), search ini- function of interval length. There was also an effect of tiation (-.42) and average look time (.34). group on all measures with ADHD children tending to begin searches later, open fewer boxes, look for less time Did search behaviour mediate the association between overall and search in a less systematic way than controls. ADHD and poor performance? The effect of group persisted when conduct and emotional For this analysis the number of search related-variables problems were entered as covariates; F(1,36) > 5.79; p < was reduced in order to simplify the test of the Page 3 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 mediational hypothesis. The four measures were submit- evidence about the abnormalities in their style of search. ted to Principal Components Analysis with Oblimin rota- However, they give no support to the assertion that their tion. The use of this technique for reducing the number of performance on such tasks is the result of abnormalities in variables is justified on the basis of the current ratio of 1 observable characteristics of their search behaviour; either item or search-related variable for every 10 subjects (i.e., 4 in terms of the reduced amount of time allocated to search variables and 40 subjects) although it needs to rec- search, because of a late onset and a premature off-set of ognise that the solution may be unstable given the abso- searching, or the more chaotic and sluggish search style. lute size of the sample. Two factors with eigen values Even taking into account these factors ADHD children still greater than 1 were extracted. These had the following performed more poorly than controls. While it is possible loadings: Factor I (intensive and systematic search; 59.0 per- that aspects of search-related behaviour other than those cent of the variance) had loadings of .83 for number of observed and coded might have been the constraining ele- systematic searches, .98 number of boxes opened, -.43 for ment it is difficult to identify what these additional char- search initiation and -.14 for average look time. The load- acteristics might have been. It is also possible the ings for Factor II (late start – short look; 27.6 percent of the approach taken to measuring search behaviour was insen- variance) were -.17 for number of systematic searches, .15 sitive to more subtle differences between the groups that number of boxes opened, .70 for search initiation and - might become obvious if more fine grained approaches to .99 for average look time. measuring search such as eye-tracking were used. How- ever, one possibility is that deficits in performance dis- The mediational hypothesis was tested using a standard played by the children with ADHD were the result of three stage procedure using single and multiple regres- deficits in non-observable processes that either were unre- sions for each search factor separately and with both lated to search behaviour or at least did not have their search factors together. In stage one ADHD status was effect on performance via an impact on search behaviour. regressed onto the number of correct responses and the In this sense the current data add support to the idea that standardised β coefficient derived indicating the strength ADHD is a disorder with distinct cognitive and behav- of the association. In stage 2 the completeness of the chain ioural elements which will probably affect different of associations linking ADHD to performance via search domains of functioning in different ways. From a practical factors was tested by regressing ADHD status onto search point of view this suggests that interventions targeted at behaviour and search behaviour onto the number of cor- improving attending or search behaviour rather than rect responses. In stage 3 the search factor was introduced improving underlying processes are unlikely to be success- alongside ADHD into a multiple regression analysis to ful in improving task performance. test whether the pathway between ADHD and perform- ance remained significant once search behaviour was con- The performance data is interesting in itself for a number trolled. ADHD status was entered in step 1 on the model of reasons. First, it does not provide support for accounts and search behaviours in step 2. The results are presented of ADHD performance on tasks of extended duration that diagrammatically in figure 1 along with the regression sta- stress the role of premature task disengagement and the tistics. As indicated by the preceding analyses ADHD is existence of a deficit in sustained attention [23]. Such related to performance and search behaviour. Also search accounts would have predicted an interaction between behaviour is related to performance for both search fac- group and inspection interval with errors increasing tors and their combination. However, the association across interval duration at a greater rate in the ADHD, between ADHD and performance, although slightly but than the control group. In contrast to this prediction the not significantly reduced, remained clearly significant size of the deficit displayed by children with ADHD, rela- when search behaviour and ADHD were introduced tive to controls, was largely independent of interval alongside each other into the analysis. In fact in each case length. In fact deficits were most marked on the 5-, 10- it was the link between search behaviour and performance and 15-s intervals, the opposite of what would be pre- that was reduced to non-significant levels. This suggests dicted by a task disengagement account. At first glance, that this association was partly the result of a common this seems to suggest that, relative to that of controls, effect of ADHD status on both search behaviour and per- ADHD performance improved on the 20-s trial. A closer formance. These effects were not affected by entering con- inspection of the results reveals that the narrowing gap duct and emotional problems into the analysis alongside between the two groups on this interval is due as much to ADHD status into the multiple regression analyses. a slight decline in the control performance as it is to improvements in the ADHD group. Second, the results provide no support for the state regulation account of Discussion The results of the current are consistent with the previous ADHD [24]. According to this account ADHD children's evidence that ADHD children perform poorly on complex difficulties arise because of problems modulating their tasks requiring systematic search. They also provide new energetic state to meet the changing demands of different Page 4 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 STEP ONE STEP TWO STEP THREE MODEL I: INTENSE AND SYSTEMATIC SEARCH ISS ISS ISS -.51 .34 .11 -.51 ADHD CORRECT ADHD CORRECT ADHD CORRECT -.46 -.52 R = .23; F(2,37)=6.97; p<0.005 MODEL II: LATE START SHORT LOOK LSSL LSSL LSSL .37 .3 -.22 -.38 ADHD CORRECT ADHD CORRECT ADHD CORRECT -.52 -.43 R =.31; F(2,37)=8.16; p<0.005 MODEL III: BOTH SEARCH FACTORS COMBINED ISS & ISS&LSSL LSSL ISS&LSSL -.56 .45 -.56 -.23 ADHD CORRECT ADHD CORRECT ADHD CORRECT -.52 -.39 R =.30;F(2,37)=7.95; p<0.005 Figure 1 Diagrammatic representation of the three steps of the mediational analysis of the effects of search style on performance Diagrammatic representation of the three steps of the mediational analysis of the effects of search style on performance. In each case step one involves the regression on ADHD group status on to the number of correct responses: Step two involved regressing ADHD into the search factor and then the search factor score on the number of correct responses: Step three involved simultaneously regressing ADHD status and search factor scores onto the number of correct responses. ISS = intense and systematic search; LSSL = Late start short look. settings [25]. A key prediction from this account is that the differences between the studies in terms of task difficulty. performance of children with ADHD will be disrupted in For instance, in the task version used in the Sonuga-Barke settings and tasks with either high (over-arousing) or low [26] paper all test stimuli were visible throughout every (under-activating) rates of stimulus presentation. Support trial as in the original MFFT. In the version used in the cur- for the state regulation account recently came from a rent only one stimulus was visible at any one time. The paper reporting the results of two experiments using a current task was therefore more likely to tap working similar version of the MFFT to that employed in the cur- memory capacities more directly as participants had to rent study [26]. In these studies a quadratic interaction hold one stimulus in mind if they wanted to compare it between ADHD group status and inspection interval was directly with another. The task used in the current paper reported (5-, 10-, and 15-s) with ADHD children perform- was therefore likely to have a higher cognitive load and be ing less well than controls on 5- (short) and 15-s (long) considerably more demanding than the task used previ- intervals but as well as them on 10-s intervals. The failure ously. This suggestion is confirmed by a direct compari- of the current study to replicate this pattern may be due to son of performance on the two tasks. The average error Page 5 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 rate in the current study was 54 percent while in the pre- Method vious studies it was 43 and 44 percent respectively Participants Forty boys between the ages of eight and twelve (20 (F(1,78) = 8.85; p < 0.001). It is possible that in the cur- rent study the increased levels of task difficulty masked ADHD and 20 typically developing controls) took part in cognitive-energetic effects associated with interval dura- the study. The ADHD children were recruited through tion. A direct comparison of the two versions of the MFFT National Health Service child psychiatric and paediatric is required to test this hypothesis. There was no evidence outpatient clinics. All had a diagnosis of Hyperkinetic Dis- to support the view that children with AD/HD produce order [31] the equivalent of severe combined type ADHD better than expected performance on shorter trials (i.e., 5- which affects approximately only 30 percent of the most s intervals) because they employ compensatory strategies severe ADHD cases [32]. Diagnosis was made against developed in response to the impact of their own impul- research criteria on the basis of a thorough clinical inves- sive cognitive style on processing opportunities [26]. Def- tigation that involved parental and child interviews, par- icits were as great on the 5-s, as they were on the 10-, 15- ent and teacher rating scales and direct observation of the and 20-s, intervals. child. In addition, children were only included in the ADHD group if they scored above the clinical cut-off The data on search behaviour are also interesting in its (seven or more out of ten) on the hyperactivity scale of own right. ADHD children took longer to initiate their both the parent and teacher versions of the Strengths and search and spent less time attending to stimuli during the Difficulties Questionnaire [SDQ; [33]]. task. Second, they tended to be less systematic and slower during searches when they were actively on task. These The SDQ is a development of the Rutter questionnaire findings are consistent with the impression given by pre- [34]. It has an extended hyperactivity scale and scales vious studies of ADHD children search-related perform- measuring more general conduct and emotional prob- ance [21,22,27]. However, they do not fit well with lems [35]. The psychometric properties and norms of the accounts of ADHD that emphasize the impulsive and dis- SDQ have been extensively examined in large epidemio- inhibited nature of ADHD children's cognitive perform- logical studies within the pre-school age group [36]. ance [28]. On the current task ADHD children's Together these criteria would place the ADHD children performance was marked by a slow start to searching, a included in the current sample within the top two or three longer time to look at each stimuli and a slower passage percent of the childhood population in terms of severity from one stimulus to another. This is consistent with the of ADHD symptoms and impairment. No child in the view that ADHD children process information more ADHD group received a diagnosis of Hyperkinetic Con- slowly, rather than more quickly, than controls [29]. duct Disorder although many showed some signs of con- However, it must be born in mind that the task used in duct problems (see table 2). Children were identified this study was externally paced, rather than, self-paced. from a region of predominantly Caucasian ethnic compo- Trial length could not be shortened by 'impulsive sition that had a mixed socio-economic background. Sev- responding', a factor that appears to be crucial in deter- enty-five boys with a diagnosis were originally identified, mining the extent to which children with ADHD are will- 35 of these agreed to take part in the study. Fifteen chil- ing to trade accuracy for speed [30]. dren from the original group were excluded because they did not meet entry criteria for the study (i.e., they had an The difference between the groups in terms of the ISS fac- IQ below 80 and/or were outside the required age range). tor was considerable for the 5-, 10-, and 15-s intervals. Children refrained from taking any medication prescribed However, there was no difference between the groups at for ADHD in the 24 hours prior to testing. Controls were the 20-s interval value. This was related to a drop in the selected at random from local schools that reflected the levels of ISS displayed by controls rather than an increase ethnic and socio-economic composition of the region in by children with ADHD. This decrease in ISS by the con- general. They were children who did not meet either par- trols was largely due to a slowing down of the average rate ent or teacher borderline cut-offs for hyperactivity on the of search rather than a drop in other elements of the ISS SDQ (i.e., less than six on both parent and teacher scales; construct (e.g., numbers of systematic searches). Such a [33]). Of the 50 control children who were initially con- change is perhaps not surprising given the fact that rapid tacted 26 agreed to participate. Four of these children were search becomes less important on longer trials. In fact excluded because they failed to meet the inclusion criteria slowing search down could be regarded as a sign that con- for controls. Two children were excluded because they trols were able to adapt their search behaviour to the exist- failed to co-operate during testing. ing time constraints. IQ was measured using four sub-scales of the Wechsler Intelligence Scale for Children (WISC-IIIR; [37]); similar- ities, vocabulary, block design, object assembly. The four Page 6 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 Table 2: Age, IQ and behaviour ratings for the ADHD and control children. CONTROLS AD/HD Mean age 9.9 (0.7) 10.5 (1.5) t = 1.65 Mean IQ 110.3(17.8) 106.7 (13.1) t = 0.71 % conduct problems 0 35 χ = 8.48** % emotional problems 0 20 χ = 4.44* Note: AD/HD = Attention Deficit/ Hyperactivity Disorder; * p < 0.05; ** p < 0.01; % behaviour problems based on children meeting SDQ criteria for presence of difficulties at home and school. Figures in parentheses are standard deviations. sub-tests were pro-rated and an estimate of the child's full achieved by moving the cursor over the chosen stimulus scale IQ was derived. Table 2 shows the mean IQ scores and pressing the button on the top of the mouse. Children and ages for members of the two groups as well as the pro- were allowed one attempt to select the target on each trial. portion of children who met SDQ cut-offs for borderline The experiment was presented on, and data collected conduct or emotional problems. While groups did not using, a Pentium 75 computer connected to a mouse. Var- differ in terms of either age or IQ the ADHD group had a iables collected included; time to initiation of search; higher proportion of children with conduct and emo- number of boxes 'opened' during each trial; order in tional problems. Analyses were performed both with and which boxes were opened and length of time which boxes without behavioural and emotional problems as were opened. Two higher order variables were derived: covariates. The number of exhaustive searches (i.e., the number of times all boxes were opened during a trial) and the Procedure number of systematic searches. A search was deemed to be Children took part in one session consisting of four systematic when boxes were opened in a systematic order blocks of 10 trials. At the start of each trial the target stim- (i.e., going across columns or up or down rows) with no ulus was presented in a box measuring 5.5 × 6.5 cm in the box being opened twice until all other boxes had been centre at the top of the computer monitor screen. Six other opened. Ratings of these variables were extremely reliable boxes of the same dimensions were presented in two rows with 100 percent agreement between two raters. All ses- of three under the target stimuli. These contained the six sions were video taped and periods of on- and off-task test stimuli (five foils and a second copy of the target stim- behaviour was recorded. Computer generated, synchro- ulus). The position of the target copy was varied randomly nised, real-time information about search behaviour (i.e., amongst the foils. At the start of each trial all six boxes if boxes were open or not at a particular point in time) was containing test stimuli were 'covered' by a white square printed on to the video tape of each session so that the equal in size to the surrounding box. The target stimulus time open boxes were being attended to could be esti- was visible throughout the inspection period of each trial. mated. Inter-rater reliability for these estimates was high To 'open' a box to view one of the six test stimuli partici- with 98 percent agreement between raters. pants were required to click on the area within the box with a mouse controlled cursor. When participants had References 1. American Psychiatric Association: Diagnostic and statistical man- finished inspecting the stimuli in the open box a second ual of mental disorders, 4th ed. Washington: American Psychiat- click closed the box and re-covered the stimuli. Partici- ric Association; 1994. pants could not open a second or subsequent box until 2. Manly T, Anderson V, Nimmo-Smith I, Turner A, Watson P, Robert- son IH: The differential assessment of children's attention: they had closed the box they were inspecting. In block The Test of Everyday Attention for Children (TEA-Ch), nor- one, children had 5-s in which to inspect the stimuli on mative sample and ADHD performance. J Child Psychol each trial, in block two the viewing time was 10-s, in block Psychiatry 2001, 42:1065-1081. 3. Epstein JN, Erkanli A, Conners CK, Klaric J, Costello JE, Angold A: three it was 15-s and in block four it was 20-s. The order Relations between continuous performance test perform- of presentation of these blocks was completely ran- ance measures and ADHD behaviours. J Abnorm Child Psychol 2003, 31:543-554. domised across subjects. At the end of the inspection 4. Sonuga-Barke EJS, Williams E, Hall M, Saxton T: Hyperactivity and period (5-, 10-, 15- or 20-s) the squares 'covering' the Delay Aversion III; The effect on cognitive style of imposing stimuli were removed so that all stimuli were visible. At delay after errors. J Child Psychol Psychiatry 1996, 37:189-194. 5. Karatekin C: A test of the integrity of the components of Bad- the same time the target was covered in order to remove deley's model of working memory in attention-deficit/hyper- the opportunity for children to continue to search after activity disorder (ADHD). J Child Psychol Psychiatry 2004, 45:912-926. the end of the viewing period. The participants were 6. Kourakis IE, Katachanakis CN, Vlahonikolis IG, Paritsis NK: Exami- prompted to identify the copy of the target by the words nation of verbal memory and recall time in children with "choose now" written on the computer screen. This was attention deficit hyperactivity disorder. Dev Neuropsychol 2004, 26:565-570. Page 7 of 8 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:10 http://www.behavioralandbrainfunctions.com/content/1/1/10 7. Brocki KC, Bohlin G: Executive functions in children aged 6 to 29. Landau YE, Auerbach JG, Gross-Tsur V, Shalev RS: Speed of per- 13: A dimensional and developmental study. Dev Neuropsychol formance of children with developmental right hemisphere 2004, 26:571-593. syndrome and with attention-deficit hyperactivity disorder. 8. Mirsky AF, Pascualvaca DM, Duncan CC, French LM: A model of J Child Neurol 2003, 18:264-268. attention and its relation to ADHD. Mental Retardation Dev Dis- 30. Sonuga-Barke EJS, Houlberg K, Hall M: When is impulsiveness not abil Res Rev 1999, 5:169-176. impulsiveness?; The case of children's cognitive style. J Child 9. Barkley RA: Behavioral inhibition, sustained attention, and PsycholPsychiatry 1994, 35:1247-1253. executive functions: Constructing a unifying theory of 31. World Health Organisation: International classification of Dis- ADHD. Psycholl Bull 1997, 121:65-94. eases: Tenth Edition 1993. Chapter 5. Mental and behavioural dis- 10. Sonuga-Barke EJS (2002b): Psychological Heterogeneity in AD/ orders. Diagnostic criteria for research. Geneva: Author . HD; A dual pathway model of behaviour and cognition. Behav 32. Swanson J, Sergeant J, Taylor E, Sonuga-Barke EJS, Jensen P, Cantwell Brain Res 2002, 130:29-36. D: Attention Deficit Hyperactivity Disorder and Hyperki- 11. Willcutt EG: The neuropsychology of ADHD: Validity of the netic Disorder. The Lancet 1998, 351:429-433. executive function hypothesis,. In Attention deficit hyperactivity dis- 33. Goodman R: The strengths and difficulties questionnaire: A order: from genes to practice in press Edited by: Gozal D, Molfese D. research note. J Child Psychol Psychiatry 1997, 38:581-586. Humana Press: Totowa, NJ. 34. Rutter M: A childrens' behaviour questionnaire for comple- 12. Cairney S, Maruff P, Vance A, Barnett R, Luk E, Currie J: Contextual tion by teachers: Preliminary findings. J Child Psychol Psychiatry abnormalities of saccadic inhibition in children with atten- 1967, 8:1-11. tion deficit hyperactivity disorder. Exper Brain Res 2001, 35. Goodman R, Scott S: Comparing the Strengths and Difficulties 141:507-518. Questionnaire and the Child Behavior Checklist: Is small 13. Barnett R, Maruff P, Vance A, Luk ESL, Costin J, Wood C, Pantelis C: beautiful? J Abnorm Child Psychol 1999, 27:17-24. Abnormal executive function in AD/HD: the effect of stimu- 36. Goodman R: Psychometric properties of the Strengths and lant medication and age on spatial working memory. Psychol Difficulties Questionnaire (SDQ). J Amer Acad Child Adoles Med 2001, 31:1107-1115. Psychiat 2001, 40:1337. 14. Booth R, Charlton R, Hughes C, Happe F: Disentangling weak 37. Weschler D: Manual for the Weschler Pre-school and Primary coherence and executive dysfunction: planning drawing in Scale Intelligence Test for Children,. New York: The Psychological autism and attention-deficit/hyperactivity disorder. Phil Trans Corporation 2001. Royal Soc London Series B-Biol Sci 2003, 358:387-392. 15. Nigg JT, Blaskey LG, Huang-Pollock CL, Rappley MD: Neuropsycho- logical executive functions and DSM-IV ADHD subtypes. J Amer Acad Child Adoles Psychiat 2002, 41:59-66. 16. Shallice T, Marzocchi GM, Coser S, Del Savio M, Meuter RF, Rumiati RI: Executive function profile of children with attention defi- cit hyperactivity disorder. Dev Neuropsychol 2002, 21:43-71. 17. Williams D, Stott CM, Goodyer IM, Sahakian BJ: Specific language impairment with or without hyperactivity: neuropsychologi- cal evidence for fronto-striatal dysfunction. Dev Med Child Neurol 2000, 42:368-375. 18. Conte R, Kinsbourne M, Swanson J, Zirk H, Samuels M: Presenta- tion rate effects on pair associate learning by attention defi- cit disordered children. Child Dev 1986, 57:681-687. 19. Dalby JT, Kinsbourne M, Swanson JM: Self-paced learning in chil- dren with attention deficit disorder with hyperactivity. J Abnorm Child Psychol 1989, 17:269-275. 20. Lawrence V, Houghton S, Tannock R, Douglas G, Durkin K, Whiting K: ADHD outside the laboratory: Boys' executive function performance on tasks in videogame play and on a visit to the zoo. J Abnorm Child Psychol 2002, 30:447-462. 21. Karatekin C, Asarnow RF: Working memory in childhood onset schizophrenia and attention deficit/hyperactivity disorder. Psychiatry Research 1998, 80:165-176. 22. Frank Y, Seiden J, Napolitano B: Visual event related potentials and reaction time in normal adults, normal children, and children with attention deficit hyperactivity disorder: Differ- ences in short-term memory processing. Int J Neurosci 1996, 88:109-124. 23. Chee P, Logan G, Schachar R, Lindsey P, Wachsmuth R: Effects of event rate and display time on sustained attention in hyper- active, normal and control children. J Abnorm Child Psychol 1989, 17:371-391. 24. Sergeant J: The cognitive-energetic model: an empirical Publish with Bio Med Central and every approach to Attention-Deficit Hyperactivity Disorder. Neu- scientist can read your work free of charge rosci Biobehav Rev 2000, 24:7-12. 25. Van der Meere J, Stemerdink N, Gunning B: Effects of presentation "BioMed Central will be the most significant development for rate of stimuli on response-inhibition in AD/HD children disseminating the results of biomedical researc h in our lifetime." with and without tics. Percep Motor Skills 1995, 81:259-262. Sir Paul Nurse, Cancer Research UK 26. Sonuga-Barke EJS: Interval length and time-use by children with AD/HD: A comparison of four models. J Abnorm Child Your research papers will be: Psychol 2002, 30:257-264. available free of charge to the entire biomedical community 27. Weiler MD, Bernstein JH, Bellinger D, Waber DP: Information processing deficits in children with attention-deficit/hyperac- peer reviewed and published immediately upon acceptance tivity disorder, inattentive type, and children with reading cited in PubMed and archived on PubMed Central disability. J Learn Disabil 2002, 35:448-461. 28. Koschack J, Kunert HJ, Derichs G, Weniger G, Irle E: Impaired and yours — you keep the copyright enhanced attentional function in children with attention def- BioMedcentral icit/hyperactivity disorder. Psychol Med 2003, 33:481-489. Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 8 of 8 (page number not for citation purposes)
Journal
Behavioral and Brain Functions – Springer Journals
Published: Jul 20, 2005