Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Moment-to-moment dynamics of ADHD behaviour in South African children

Moment-to-moment dynamics of ADHD behaviour in South African children Background: The behaviour of children with Attention-Deficit/Hyperactivity Disorder is characterized by low predictability of responding. Low behavioural predictability is one way of operationalizing intra-individual ADHD- related variability. ADHD-related variability may be caused by inefficient behavioural selection mechanisms linked to reinforcement and extinction, as suggested by the recently published dynamic developmental theory (DDT) of ADHD. DDT argues that ADHD is a basic neurobehavioural disorder, caused by dysfunctioning dopamine systems. For establishing ADHD as a neurobehavioural disorder, findings from studies conducted in Western countries should be replicated in other cultural populations. The present study replicated the study conducted in Norway, with children from the Limpopo province in the Republic of South Africa. Methods: Boys and girls, aged 6–9 yr, from seven ethnic groups participated. Scores by teachers on the Disruptive Behavior Disorders rating scale defined participation in either ADHD-hyperactive/impulsive (-HI), th ADHD-predominantly inattentive (-PI), or ADHD-combined (-C) groups. Children below the 86 percentile were matched on gender and age and comprised the non-ADHD group. The children completed a computerized game-like task where mouse clicks on one of two squares on the screen resulted in delivery of a reinforcer according to a variable interval schedule of reinforcement. Reinforcers were cartoon pictures presented on the screen together with a sound. Predictability of response location and timing were measured in terms of explained variance. Results: Overall, the results replicated findings from Norway. Specifically, the ADHD-C group showed significantly lower predictability of responding than the non-ADHD group, while the ADHD-HI and the ADHD- PI groups were in-between. In accordance with the previous study, response location, but not response timing, was a sensitive behavioural measure. There were no significant gender differences. Cartoon pictures were effective reinforcers as the non-ADHD group showed learning of the task. There was no relation between behavioural predictability and motor functions. Conclusion: The present study makes a strong case for ADHD as a basic, neurobehavioural disorder, not a cultural phenomenon, by replicating findings from a wealthy Western country in a poor province of a developing country. The results were, generally, in line with predictions from the dynamic developmental theory of ADHD by indicating that reinforcers were less efficient in the ADHD group than in the non-ADHD group. Finally, the results substantiated ADHD-related variability as an etiologically important characteristic of ADHD behaviour. Page 1 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 on a microlevel in order to identify possible mechanisms Background Attention-deficit/hyperactivity disorder (ADHD) [1] is a underlying the observed intra-individual variability. behavioural disorder affecting about 2–5% of grade school children [2], making it one of the most common ADHD-related variability is mainly observed at the behav- child psychiatric diagnoses in the United States of Amer- ioural level [16], so a thorough behavioural account is ica and in Europe. In childhood, the diagnosis is more fre- warranted in order to guide the investigation of, for quent among boys. Depending on the referral practice, instance, underlying neurobiological processes. Castel- boys to girls ratios vary between 10:1 (clinical samples) lanos and colleagues [16] raised a number of key ques- and 3:1 (community samples) [3]. Research indicates that tions about ADHD-related variability that need to be the major aetiological factor is genetic [4], probably addressed in a basic research programme, including its mainly expressed as alterations in catecholaminergic reg- robustness in its association with ADHD; whether it is a ulation of brain activity [5,6]. The disorder places the random or dynamic-periodic phenomenon, whether it child at risk for school failure and dropout, juvenile delin- varies dynamically as a function of context, whether it is quency, criminality, substance abuse, and sexual promis- unique to ADHD, and finally, whether it reflects processes cuity with HIV/AIDS and teenage pregnancies as possible causal to ADHD. In the recently published dynamic devel- consequences. In this way, the disorder is extremely opmental theory of ADHD (DDT) [19] it is argued that costly, both to the afflicted individuals and their families, the main behavioural selection mechanisms, reinforce- and to the society [7,8]. Although it has been discussed ment and extinction, are altered in ADHD. These altera- whether ADHD is a phenomenon of the Western culture, tions bring about a different learning style resulting in e.g., [9], its worldwide existence is well documented [10- increased behavioural variability, in addition to hyperac- 13]. However, it seems that inconsistent assessment crite- tivity, impulsiveness, and deficient sustained attention. ria and procedures affect prevalence rates of ADHD across Some of the variability may be the result of deficient countries and cultural groups [10]. The reliance on behav- acquisition of reinforced behaviour combined with defi- ioural observations and clinical descriptions of the behav- cient extinction of non-reinforced behaviour causing iour makes the diagnosis vulnerable to subjective and shorter and less predictable behavioural sequences cultural perceptions, and the need for more objective cri- [15,19]. Thus, the DDT places ADHD-related variability teria for diagnosing is long needed. No biological marker within a causal model [19]. The DDT thereby suggests that has yet been found, and no existing neuropsychological ADHD-related variability is not a random phenomenon, test can reliably define a case of ADHD [14]. However, but can be predicted in the combined and the hyperactive/ theory-driven research aimed at identifying dysfunctions impulsive subtypes of ADHD. Further, the DDT argues in basic behavioural mechanisms may provide an empiri- that ADHD-related variability will vary dynamically as a cal basis for understanding processes and functions on function of context, task, and motivational preferences. other levels of analysis (e.g., developmental, neuropsy- Recently, we showed that boys with ADHD combined and chological, psychosocial), and for generating more hyperactive/impulsive subtypes had significantly less pre- sophisticated tests for early and reliable identification of dictable response sequences than normal boys [15] and affected individuals. The main purpose of the present that the overall variability was observed during infrequent study is to replicate an earlier study [15] showing that reinforcement as opposed to frequent [20]. However, only moment-to-moment dynamics of ADHD behaviour may boys from a culturally homogeneous population partici- represent a new way of understanding underlying behav- pated, so it is of vital importance that the phenomenon ioural mechanisms basic to ADHD. can be replicated in other samples. ADHD is characterized by age-inappropriate hyperactiv- The present study is a replication and extension of that ity, impulsiveness, and deficient sustained attention [1]. study, with a larger sample including both genders. Fur- Most clinical and experimental reports show increased ther, the study is conducted in a developing country with variability, both between and within subjects in the a culturally less homogeneous population. However, clin- ADHD group. While the group heterogeneity suggests ical resources are sparse, and assessment methods devel- multiple independent pathways to the disorder [16], the oped and validated in Western countries may not be intra-individual variability might be a key characteristic of relevant or feasible. A recent prevalence study conducted an endophenotype of ADHD and has recently become a in various language groups in South Africa showed that topic of particular interest [15-17]. A purportedly impor- when using a teacher rating scale, similar figures for prev- tant step in the direction of describing and explaining the alence as in Western countries were obtained. Also, rather role and function of this variability is more fine-grained similar distributions across the three ADHD subtypes analyses of behaviour as opposed to the traditional statis- (inattentive, hyperactive/impulsive, and combined) were tical summary measures of means and standard devia- found, in addition to a similar gender distribution [12]. tions [18]. In the present study, response data is analyzed Page 2 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 The primary aim of the present study was to investigate if Both teachers and parents were given the rating scale to the results from the Norwegian study could be replicated complete. Only the teacher's ratings were used for the in a different sample derived from various language screening, since the return of the parent's rating scale was groups in the Limpopo province of South Africa. If repli- below 50%, probably because many children either did cated, the combined results from the Norwegian and the not live with their parents or the parents were illiterate. South African studies would make a strong case for a Teacher ratings are usually regarded as an accurate meas- basic, neurobiological mechanism differentiating ADHD ure of assessment [24,25]. The teachers returned ratings of from non-ADHD children, refuting the argument that close to 100% of their pupils. The children meeting the ADHD is a cultural phenomenon resulting from Western criteria for inclusion into the clinical group (~7%) were way of life, or from Western conceptualization of psychi- selected for further testing. They were matched for gender, atric problems. In addition, the findings might lend sup- age, and language with children without ADHD as meas- port to the recently published dynamic developmental ured by the screening process. theory of ADHD [19] arguing that the main behavioural selection mechanisms, reinforcement and extinction, are Children were divided into a group classified as ADHD, altered in ADHD and will result in increased behavioural the "ADHD group" and a comparison group without variability. ADHD symptoms (Table 1), based on teacher ratings on the DBD rating scale. The cut off point for the ADHD th The present sample made it possible to investigate poten- group (95 percentile or above) was based on the results tial differential effects of reinforcers on response patterns from the prevalence study [12] in which more than 6000 in the three subtypes of ADHD [2]: the predominantly children in the Limpopo Province were rated on the DBD inattentive type (ADHD-PI), the hyperactive/impulsive scale. Norms were developed for a high cut-off based on type (ADHD-HI), and the combined type (ADHD-C). In the qualified assumption that ADHD exists in about 5% addition, potential gender differences could be studied. of the population [12]. According to these norms, scores higher than 18 on the hyperactive/impulsive (H/I) items ADHD-related variability was defined as reduced predict- were classified as ADHD-HI and higher than 21 on the ability of consecutive responses. The task used was a com- inattentive (Inatt) items were classified as the ADHD-PI puterized game where mouse clicks on either of two group. If the criteria were met on both types of items, the squares on the screen resulted in the presentation of a child was classified as ADHD-C. The cut off point for the th reinforcer. Reinforcers were cartoon pictures displayed on comparison group was set at the 85 percentile or below in the screen for a short period accompanied by a sound. order to decrease the risk for false negatives in this group, These were delivered according to variable interval (VI) as the DBD rating was the only measure. Thus, children schedules of reinforcement. VI schedules specify that with scores on H/I-related items less than 15 and Inatt responses result in a reinforcer after varying time intervals. items less than 17 were regarded as comparisons. Thus, reinforcers are presented at unpredictable times [21], avoiding any confounding with time discrimination The final sample consisted of children from seven ethnic problems. All mouse clicks were recorded both in terms of groups inhabiting the Limpopo Province of South Africa. where on the screen responses were placed (response loca- For the Afrikaans and English speaking groups, the IQ was tion; the spatial dimension) and response timing (the established with the Senior South African Individual Scale temporal dimension). Thus, the present task allowed for (SSAIS-R) [26]. As there are no standardized IQ tests for analysis of both spatial and temporal aspects of behav- the indigenous African populations, Raven's progressive iour. matrices were used to estimate IQ. This test is also consid- ered to be "culture-fair" [27,28]. Children with IQ lower Methods than 80 and/or with a history of neurological problems Participants (e.g. epilepsy, head injuries, cerebral palsy, or cerebral Children from seven ethnic groups of the Limpopo Prov- malaria) were excluded. None of the children were on psy- ince of South Africa (Northern Sotho, Venda, Tsonga, chostimulant medication at the time of testing. Tswana, North Ndebele, Afrikaans, and English) were recruited from a school-based population. The 179 chil- Procedure dren (128 boys and 51 girls) were recruited following Written permission was obtained from the Department of screening of the general population of primary school Education, Limpopo Province, as well of the principals of children representative of all socio-economic levels for the selected schools. Participation was voluntary. ADHD-like behaviour. The Disruptive Behavior Disorders Informed consent was obtained from the child's parents (DBD) rating scale [22,23] was standardized for the pop- or guardians. ulations of the Limpopo province of South Africa in an earlier study [12] and used as the screening instrument. Page 3 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 Table 1: Means and Standard Deviations, for Age and DBD Scores, by Subtype ADHD-C ADHD-HI ADHD-PI Non-ADHD Boys Girls Boys Girls Boys Girls Boys Girls Total Age (mo)* 102.4 ± 10.6 104.4 ± 11.8 109.5 ± 20.7 103.6 ± 9.0 104.5 ± 15.7 102.9 ± 19.2 108.0 ± 14.9 101.5 ± 11.8 106. 4 ± 5.1 DBD** Inatt 24.0 ± 3.8 23.8 ± 2.7 13.8 ± 4.2 14.8 ± 2.5 22.2 ± 4.4 25.8 ± 2.6 6.6 ± 5.7 5.2 ± 5.9 13.5 ± 9.2 DBD** H/I 22.0 ± 3.2 20.9 ± 3.2 20.2 ± 1.8 20.6 ± 2.1 10.0 ± 6.0 9.9 ± 4.1 4.1 ± 3.6 4.4 ± 5.1 10.6 ± 8.4 Afrikaans 2 3 1 1 3 1 10 5 26 English 000 0 102 1 4 N Sotho 1 3 4 1 0 1129 31 Tsonga 313 1 823 1 22 Venda 4 2 2 4 112178 50 N Ndebele 343 0 428 3 27 Tswana 303 1 109 2 19 Total 16 13 16 8 28 8 61 29 179 * There were no statistically significant differences in age between the subtypes ** Differences between groups were not tested, as the groups were defined by these scores. DBD: Disruptive Behavior Disorder rating scale [22]; Inatt: Scores on inattentive items; H/I: Scores on hyperactive/impulsive items The children were always tested by a tester fluent in the the two squares switched sides at random, but were never child's own language. Most assessments were done at their displayed on the same side more than twice in a row. The schools during school hours, where one of the classrooms number of presentations on each side was the same. was made available. The exceptions were children whose school was within a radius of 2 km from the University The task was designed according to a multiple variable and children referred for clinical assessment. These were interval (VI) schedule of reinforcement. A schedule is tested at the University Clinic. called multiple when two or more schedule components alternate and are signalled by discriminative stimuli. In an The SSAIS-R IQ test [26] and the Raven Progressive Matri- interval schedule some time must elapse before a response ces Test were administered by Masters students in Clinical will result in delivery of a reinforcer. In VI schedules, the Psychology who were doing their hospital internship. intervals will vary around a specified arithmetic mean [21]. The reinforcer-dense schedule was a VI 2 s and the As many of the children were not acquainted with com- reinforcer-lean schedule was a VI 20 s schedule of rein- puters, a 'mouse-training' session was part of the testing forcement. The screen's background colour changed procedure. according to the contingency in operation and functioned as the conditioned discriminative stimulus for the two Reinforcement task contingencies. A navy blue background signalled the VI 2 The task was designed as a computer "game" and was run s, while a yellow background signalled the VI 20 s. The on one of three similar laptops. The screen resolution was dark grey square was always associated with reinforce- set to 640 by 480 pixels. The response device was a stand- ment (the correct square), thus this was the discriminative ard computer mouse and clicks on either left or right but- stimulus. Reinforcers were cartoon pictures displayed on ton were recorded as responses. Moving the mouse made the screen for 1.5 s simultaneously with a sound (different the cursor move on the screen, the cursor being in the computer-generated sounds). shape of an arrow. Two squares (140 × 140 pixels) 120 pixels apart, one in a light and one in a dark shade of grey, The child was introduced to the test with the following were displayed on the screen 120 pixels from the vertical instruction (told in the child's own native language): sides (left-right) and 170 pixels from the top and from the bottom. A click within one of the squares induced a brief "This is a game you may play now. It is a little strange, because change in colour as feedback; the dark grey square turned I will not tell you how to play the game. Your task is to find out into a lighter shade and the light grey square turned into how the game works. You may use this mouse and move the a darker shade. Clicks outside the squares were recorded, arrow across the screen like this (experimenter demonstrates but gave no feedback. Following reinforcer presentations, how to move the mouse and cursor). If you want to point, you Page 4 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 can click with one of these buttons (experimenter points to the ders of the squares, disregarding whether the square was mouse buttons). You may talk while you are playing, but I will correct or not. 3) The distance-to-correct-centre measure was not answer any questions about the game. I will sit back here based on the distance from where the response was placed and write a little while you play. Do you understand your task? to the centre of the correct square. Thus, this measure was You may start now." a variant of the distance-to-nearest-centre measure, anchored to the centre of the correct square. Both distance The task started with a shaping sequence where every cor- scores (measure 2 and 3) were in terms of vertical and hor- rect response was reinforced. The screen background was izontal pixels, with 0,0 defining the centre of the square. blue and the correct square was always on the same side 4) Timing response patterns were based on consecutive (right). After six correct responses, the VI 2 s schedule interresponse times (IRTs). came into effect without any signal. The child received four reinforcers upon responding during the VI 2 s contin- The predictability of responses spatially and temporally gency before the schedule changed to the VI 20 s contin- was assessed by explained variance, i.e., autocorrelations gency (and the background changed from blue to yellow) squared. Explained variance is a better measure for overall where four reinforcers were to be obtained. The time it predictability than autocorrelations by itself since the lat- took to obtain these eight reinforcers constitute the first ter will have both positive and negative values cancelling segment of the test. The schedules alternated so that each each other when added. Autocorrelations (serial correla- was displayed five times; i.e., there were five segments, tions) of each measure were correlations of consecutive resulting in a total of 40 reinforcers per child (plus six response measures across five lags. Thus, correlations from the shaping sequence). The entire task, including between the value of response n and of response n+1 is instruction and shaping, took about 10–13 min to com- the first lag, between n and n+2 is the second lag, and so plete. on up to correlations between response n and n+5 being the fifth lag. For programming reasons, autocorrelations Data recording and statistics were computed on consecutive responses through a full Data were recorded by the laptops. Percentage of all segment and not reset at reinforcer delivery because the responses within the correct square, response side (left or number of responses was huge compared to the number right), response coordinates (i.e. the horizontal and verti- of reinforcers. Changes in predictability of responding cal pixel that the tip of the arrow-shaped cursor touched throughout the experiment could be observed in the when the child clicked a mouse button), and interre- explained variance for each child from segment to seg- sponse times (IRTs) were the recorded dependent varia- ment. An increase in autocorrelations over segments bles. The individual IRT distributions were highly skewed would indicate that performance became more and more with a long tail towards long IRTs. IRTs were therefore predictable throughout the experiment, and thus be an normalized by log transformations prior to analysis indirect measure of learning. We hypothesised that com- [logIRT = log10 ((IRT/1000) + 0.001)]. pared to the non-ADHD group less of the behaviour of children with ADHD would be predictable, indicated by Behavioural measures generally less explained variance. In addition, predictabil- The same behaviours were analyzed in the previous [15] ity of responses in the ADHD group should, to a larger and the present study. Response sequences were analyzed extent than in the non-ADHD group, be restricted to the in the VI 20 s condition only; as the short schedule usually first lag, indicating shorter behavioural sequences. would not allow for enough responses in a segment (i.e., four reinforcer deliveries within the VI 2 s condition) to In addition to the response sequences, learning was meas- calculate autocorrelations and explained variance (see ured as mean percent of all responses that were placed below). The following behaviours were analyzed: 1) A within the correct square. A high score on percent correct general side response pattern, i.e., whether consecutive would indicate that the dark grey square exerted good responses were on the left or right side of the screen. stimulus control over the responding of the children, and Highly predictable responding would typically be on the is thus a measure of attention. side where the correct target was positioned, and indicates good stimulus control. Likewise, low predictability Statistics implies that responses are unsystematically distributed on Data were analyzed by means of SPSS 11.0 for Windows the two sides and means poor stimulus control. 2) The dis- (SPSS) and Statistica 6.1 [29] program packages. The dis- tance-to-nearest-centre measure was based on the distance tance scores were computed as the square root of the sum from the pixel where the response was placed, to the cen- of squared horizontal and vertical distances. Explained tre of the selected square, whether it was the correct square variance (autocorrelations squared) was analyzed using or not. This measure indicates whether the children devel- repeated measures ANOVA across segments and lags. The oped strategies of responding that was related to the bor- ANOVA was supplemented with MANOVA. A multivari- Page 5 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 responding was found for the three spatial behavioural Non-ADHD measures, but not for the timing measure. The ADHD 0,30 ADHD Combined group had the lowest explained variance of all groups on the spatial measures. 0,25 0,20 Side response pattern This measure assessed whether the child's choice of side of 0,15 the screen (left or right) was predictable irrespective of on which side the correct response target was displayed. 0,10 Highly predictable responding across lags would imply that behaviour was orderly related to side of screen. 0,05 Generally, predictability from one response to the imme- 0,00 123 45 123 45 1234 5 1 2345 12 345 diate next (first lag) was in the low range (0.31 > mean R Seg. 1 2 3 4 5 > 0.17; median R = 0.22) compared to previously pub- lished results [15]. The side response pattern of the non- Response predictability accord Figure 1 ing to side of the screen ADHD group was more predictable compared to the Response predictability according to side of the ADHD group (Fig. 1). There were significant main effects screen. Predictability of consecutive responses according to of Group and of Segment (Table 2). The ANOVA showed side of the screen (left or right), depicted as mean explained significant interaction effects between Group and Seg- variance (autocorrelations squared), across segments (Seg. ment, between Group and Lag, and between Group, Seg- 1–5) and lags (1–5 per segment), for ADHD and non-ADHD ment, and Lag. All interactions were confirmed with the groups. Abbreviations: Seg.: segment of session. Lag: number of responses that has to be correlated to the present one, multivariate analysis (Table 2). The significant main effect i.e., correlations between response n and n+1 is the first lag, of Segment implies that there was a general upward trend between n and n+2 is the second lag, and so on up to corre- in predictability from segment 1 to segment 4. lations between response n and n+5 being the fifth lag. The follow-up ANOVA of the three ADHD subtypes and the non-ADHD group showed no significant main effect ate approach to repeated measures is recommended when of Subtype, but the main effects of Segment and of Lag variables have more than two levels because MANOVAs were significant. There was a significant interaction correct for the assumption of compound symmetry and between Subtype and Lag (F(12, 720) = 2.99; p < .001), sphericity in ANOVAs [29]. and between Subtype, Segment, and Lag (F(48, 2880) = 1.62; p < .005). Only the three-way interaction was con- Analyses relevant for the primary aim were performed firmed by the MANOVA. The interaction effects implied first, with Group (2: ADHD and non-ADHD) as the that predictability in responding for the three ADHD sub- between-group variable; and segment (5) and lag (5) as types and the non-ADHD group changed differently within-group variables. The ADHD group consisted of across segments and lags. The ADHD-C group showed the children with ADHD-C and ADHD-HI in order to repli- least predictable responding across segments. No other cate the Norwegian study. Then, as follow-up analyses, interaction effects were statistically significant. subtypes of ADHD (ADHD-HI, ADHD-PI, and ADHD-C) versus non-ADHD were analysed across segments and Distance to nearest centre lags, in order to investigate potential differences between This measure assessed to what degree the children subtypes of ADHD. Results were followed up with post responded in predictable patterns in terms of the distance hoc Scheffé tests where relevant. Non-published results from where the response was placed to the centre of the may be obtained from the first author upon request. nearest square irrespective of whether it was the correct response target or not. Demographic data Demographic and diagnostic measures of the ADHD- The generally low values of explained variance in the first 2 2 related subtypes and the non-ADHD comparison group lag (0.21 > mean R > 0.12; median R = 0.17) show a are displayed in Table 1. rather low predictability from one response to the imme- diate next. As can be seen in Figure 2, the ADHD group was generally lower than the non-ADHD group. There was Results Generally, there was no effect of Gender, neither main no significant main effect of Group, however, but a signif- effects nor interaction effects. Consequently, the reported icant main effect of Lag. The interaction between Group findings combine boys and girls. Further, predictable Page 6 of 13 (page number not for citation purposes) Ex plained Variance Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 Table 2: Results from repeated measures ANOVA and multivariate tests for repeated measures, of explained variance (squared autocorrelations) Measure Variable ANOVA Multivariate Df F Df F Side Response Pattern Group (G) 1, 146 13.857* Segment (Seg) 4, 584 3.769** 4, 143 2.650* Lag 4, 584 167.626*** 4, 143 49.352*** G * Seg 4, 584 3.941** 4, 143 2.792* G * Lag 4, 584 4.206** 4, 143 2.697* G * Seg * Lag 16, 2336 1.802* 16, 131 2.090** Distance to centre of nearest square G 1, 146 2.588 Seg 4, 584 0.150 4, 143 0.146 Lag 4, 584 135.577*** 4, 143 41.892*** G * Seg 4, 584 0.425 4, 143 0.428 G * Lag 4, 584 1.242 4, 143 2.863* G * Seg * Lag 16, 2336 1.410 16, 131 1.487 Distance to centre of correct square G 1, 146 3.974* Seg 4, 584 0.875 4, 143 0.723 Lag 4, 584 187.974*** 4, 143 57.819*** G * Seg 4, 584 4.475*** 4, 143 3.663** G * Lag 4, 584 4.996*** 4, 143 1.943 G * Seg * Lag 16, 2336 1.505 16, 131 1.203 Timing Response Pattern G 1, 146 0.843 Seg 4, 584 1.270 4, 143 0.929 Lag 4, 584 60.096*** 4, 143 31.501*** G * Seg 4, 584 0.994 4, 143 0.944 G * Lag 4, 584 0.660 4, 143 0.922 G * Seg * Lag 16, 2336 1.107 16, 131 1.374 *: p < .05; **: p < .01; ***: p < .001 and Lag was significant in the multivariate analysis (Table behaviour compared to the two other spatial measures. 2). Predictability from one response to the immediate next (explained variance in the first lag) was generally in the 2 2 The follow-up ANOVA of the three subtypes and the non- lower middle range (0.35 > mean R > 0.19; median R = ADHD group showed no significant effects except a signif- 0.25) compared to previous results [15]. The ADHD icant main effect of Lag (F(4, 720) = 149.27; p < .001). In group had less predictable behaviour than the non-ADHD order to check if there were statistically significant differ- group. ences between any two subtypes, a post hoc Scheffé test of Subtype and Lag was performed. The main results from The ANOVA showed a significant main effect of Group, in this test indicate a larger decrease in explained variance addition to the main effect of Lag (Table 2). There were from the first to the next lags in the non-ADHD group and also significant interaction effects between Group and in the ADHD-PI subtype compared to the other subtypes, Segment, and between Group and Lag. The interaction and that there was a larger difference between the ADHD- between Group, Segment, and Lag did not meet conven- C subtype and the non-ADHD group compared to any tional levels of significance (p > .08). Only the interaction other combination of subtypes. involving Group and Segment was confirmed by the mul- tivariate analysis (Table 2). This interaction showed that Distance to correct centre while the non-ADHD group's behaviour increased in pre- This measure assessed patterns in response placements in dictability over segments, the ADHD group's behaviour terms of distance from the centre of the correct square. did not improve over segments. Again, low explained variance indicated high variability in spatial responding. As can be seen in Figure 3, responding The follow-up ANOVA of the three subtypes and the non- was somewhat more predictable with this measure of ADHD group only showed a significant interaction effect Page 7 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 showed a significant interaction effect between Group, Non-ADHD ADHD Segment, and Lag (F(48, 2880) = 1.39; p < .05), which was 0,20 not confirmed by the multivariate analysis. The interac- tion implied that patterns of explained variance changed 0,15 between the groups both across segments and lags in an uninterpretable way. No further follow-up analyses were run for this measure. 0,10 Learning Percentage of all responses that were placed within the 0,05 square associated with reinforcement (correct square) was used as a measure of learning. Overall, the non-ADHD group had higher percent correct scores than the ADHD 0,00 12345 12345 1234 5 12345 12345 group (58.6% vs 46.2%). The non-ADHD group increased Seg. 1 2 3 4 5 their scores with about 10% (from 52.8% to 62.5%) across segments, while the ADHD group only increased Respo tr Figure 2 e of a square nse predictability according to distance from the cen- with less than 2%. The ANOVA of percent correct choice Response predictability according to distance from of square showed significant main effects of Group (F(1, the centre of a square. Predictability of consecutive 147) = 16.142; p < .001) and of Segment (F(4, 588) = responses according to distance from the centre of a 7.635; p < .001). The interaction effect between Group response square, whether correct or not, to where on the and Segment was also significant (F(4, 588) = 4.297; p < screen the responses were placed. Curves show mean .002); and was confirmed by the MANOVA (F(4, 144) = explained variance (autocorrelations squared) across seg- ments (Seg. 1–5) and lags (1–5 per segment), for ADHD and 2.964; p < .03). non-ADHD groups. Abbreviations: Seg.: segment of session. Lag: number of responses that has to be correlated to the The follow-up ANOVA of the three subtypes and the non- present one, i.e., correlations between response n and n+1 is ADHD group showed a main effect of Subtype (F(3, 180) the first lag, between n and n+2 is the second lag, and so on = 6.14; p < .001) and of Segment (F(4, 720) = 7.18; p < up to correlations between response n and n+5 being the .001) (Fig. 4). The interaction effect between Subtype and fifth lag. Segment did not meet conventional levels of significance (F(12, 720) = 1.72; p = .058). The main effect of Segment was confirmed by the MANOVA. between Subtype and Lag (F(12, 720) = 2.64; p < .002) in addition to a main effect of Lag (F(4, 720) = 198.39; p < A post-hoc Scheffé test of the four subtypes showed that .001). The interaction effect indicated that the ADHD-C the non-ADHD group had significantly higher mean per- group showed less behavioural predictability than any of cent correct than the ADHD-C and the ADHD-HI sub- the other groups, while the ADHD-PI did not differ from types (p < .008 and p < .04, respectively). The ADHD-PI the non-ADHD group. The interaction was not confirmed group was not significantly different from any of the other by the MANOVA. subtypes, and the -C and -HI subtypes were not signifi- cantly different from each other. In order to check what contributed to the interaction effect, a post hoc Scheffé test was performed involving Response predictability and motor functions Subtype and Lag. The main results from this test con- A recent study of motor functions of the present South- firmed some overlap between the non-ADHD and African sample showed significant differences between ADHD-PI subtypes, and that explained variance for these children with ADHD-related symptoms and those with- groups was significantly higher than that of the ADHD-C out (Meyer & Sagvolden, subm. to BBF). Increased behav- subtype. ioural variability may be a result of motor coordination problems. This relation was tested by correlating first lag Timing response pattern explained variance of all spatial measures with normal- The development of patterns in response timing was ized scores on Grooved Pegboard test and Mazes (see investigated in terms of consecutive interresponse times Meyer & Sagvolden for details) for both dominant and (IRTs). Explained variance of the first lag was generally non-dominant hand, for the ADHD-C group and the non- very low and not significantly different between the ADHD group separately. For the ADHD group, scores on groups (0.08 > mean R > 0.05). Besides a significant main response predictability was mainly negatively related to effect of Lag, there were no other main or interaction scores on motor tests (on the motor tests, low scores were effects (Table 2). The follow-up analysis of the subtypes preferred), but the relation was weak (Pearson correla- Page 8 of 13 (page number not for citation purposes) Explained Variance Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 0,35 Non-ADHD ADHD-C ADHD ADHD-HI ADHD-PI 0,30 Non-ADHD 0,25 0,20 0,15 50 0,10 0,05 0,00 12345 12345 12345 12345 12 345 Segment Segm. 1 2 3 4 5 Mean percent correct Figure 4 Respo tre Figure 3 of the correct target nse predictability according to distance from the cen- Mean percent correct. Mean percent correct choice of Response predictability according to distance from response target across consecutive segments (Seg) for the centre of the correct target. Predictability of consec- ADHD and non-ADHD groups. Abbreviations: ADHD-C: utive responses according to distance from the centre of the ADHD combined type; ADHD-HI: ADHD hyperactive/ correct target square to where on the screen the responses impulsive type; ADHD-PI: ADHD predominantly inattentive were placed. Curves show mean explained variance (auto- type. correlations squared) across segments (Seg. 1–5) and lags (1– 5 per segment), for ADHD and non-ADHD groups. Abbrevi- ations: Seg.: segment of session. Lag: number of responses that has to be correlated to the present one, i.e., correla- ince. Actually, when comparing the results from the two tions between response n and n+1 is the first lag, between n studies, the ADHD groups from the two populations seem and n+2 is the second lag, and so on up to correlations more similar to each other than the non-ADHD groups between response n and n+5 being the fifth lag. are to each other. Combined, the results suggest that the phenomenon might pertain to the ADHD-C subtype in particular, although some results indicate that children with less severe ADHD, as those with hyperactive/impul- tions .234 > r > -.352). For the non-ADHD group, the rela- sive subtype, may show weaker forms of the phenome- tion was even weaker (Pearson correlations .237 > r > - non. This conclusion is in line with predictions made .175). from the dynamic developmental theory, DDT [19], applying specifically to ADHD-C and -HI subtypes. Discussion The present study was designed to follow up the results of Another striking similarity between the two studies was a study in a wealthy European country, Norway, which the finding that predictability of consecutive responding showed that response sequences in children with ADHD were found for the spatial behavioural measures only and symptoms is less predictable than those of children with- not for the temporal measure. We speculated that this out ADHD [15]. The present study was conducted in the finding might be related to the visuo-spatial nature of the poor Limpopo province in South Africa, a developing task, alternatively that striatal dysfunction might explain country with a more heterogeneous population, and with deficient habit learning in ADHD [15]. The fact that simi- fewer resources and assessment options compared to lar patterns of results were found in two extremely diverse Western countries. The children's behaviour problems samples, recruited by different methods and in different were rated with the DBD [22] as psychiatric services are cultures, indicates a biologically-founded mechanism generally not available in developing countries. A strong rather than a culturally-imposed response style, and that case for ADHD as a basic, neurobehavioural disorder, not the mechanism might be specific to ADHD with symp- a cultural phenomenon, could be made if the results from toms of hyperactivity/impulsiveness. Imaging and neu- Norway were replicated in a developing country. ropsychological research suggest a right hemisphere frontal-striatal circuitry involvement in ADHD, e.g., The most striking finding in the present study was that the [30,31], indicating a specific dysfunction in perception lower predictability of consecutive responses of boys with and treatment of visuo-spatial stimuli. An alternative ADHD compared to controls in the Norwegian study [15] hypothesis, highly compatible with the DDT, has sug- was replicated in boys and girls from the Limpopo prov- gested that cortical basal-ganglionic neuronal modules Page 9 of 13 (page number not for citation purposes) Explained Variance Mean Percent Correct Responses Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 learn to recognize and register complex contextual pat- from those often described in the -HI and -C subtypes terns that are relevant to behaviour, and that this learning [19,34]. The present findings thus support the assump- is mediated by dopaminergic reinforcement signals [32]. tion in the DDT that altered learning mechanisms related The contextual information includes the state of the to a shorter delay gradient mainly apply to the -HI and -C organism, the location of targets of action, the desirability subtypes and that the present attention measure mainly of an action, motor intentions, and sensory inputs apt for relates to the learning style of these subtypes in particular. either selecting or triggering motor programs. The pattern recognition going on in the striatum is guided by In general, the behaviour of all groups was less predictable dopaminergic reward prediction signals [33]. The recur- in the present study than in the previous study [15], par- sive process in the cortico-striatal module enables the ticularly when comparing the non-ADHD groups. Learn- basal ganglia to encode even more complex contexts ing was also poorer, as mean percent correct never based on those initially recognized [32]. With dysfunc- exceeded 63% even in the non-ADHD group, while the tioning dopamine systems, learning the association non-ADHD group in the Norwegian study performed up between different contextual cues and between environ- to 90% correct [20]. Several factors may explain this mental signals and relevant motor programs will be ham- result. For instance, the groups were probably more heter- pered [19]. ogeneous in the present study. The DBD rating scale is most likely less sensitive than the standard comprehen- It is important to notice that the cartoons acted as rein- sive diagnosis performed in the Norwegian study. Also, forcers in the non-ADHD children as shown by their the non-ADHD group was less well described and allowed th learning curves (Fig. 4). Learning could also be observed for DBD scores up to the 85 percentile. In addition, chil- on the distance-to-correct-centre measure (Fig. 3), show- dren in a developing country may not be as used to com- ing increasingly better prediction from one response to puters as Norwegian children. While most Norwegian the next (first lag) in the non-ADHD group over segments, children are acquainted with computer games, including but not in the ADHD group. The higher and increasing clicking on items in order to bring up other items or hap- percent correct scores of the non-ADHD group, and the penings on the screen, the South African children may not lower, flatter learning curves of the ADHD groups, repli- be that familiar with computer games. This will result in cate other findings from Norway [20]. In the present study more explorative and somewhat less systematic behaviour the ADHD-C and -HI groups had more incorrect in the South African children. Further, an unintended pro- responses than correct throughout the session (Fig. 4). cedural difference between the two studies might have This indicates no stimulus control over behaviour by the influenced the results. In Norway, testers were instructed dark grey square in these groups. Stimulus control is dem- to add verbal feedback (like "Wow", "Look at that", etc.) onstrated when performance is predictably related to the when the cartoons were displayed on the screen. This was stimulus signalling reinforcement, and the stimulus has not done in South Africa. Thus the reinforcing effect might gained conditioned reinforcing properties. For a stimulus have been less, particularly for the non-ADHD children as to become a conditional reinforcer, it must be coupled this group deviated more from their Norwegian counter- with a primary reinforcer within a certain time, delimited parts than the ADHD group. by the length of a delay-of-reinforcement gradient (Cata- nia's precommentary in [19]). The present findings may The findings demonstrate, nonetheless, that the task be explained by a short delay gradient in the -C and -HI could be run in just one session and with no extra tangible groups, resulting in no association between the stimulus reinforcers. Hence, a quick (less than 15 min altogether), and the response-produced reinforcer, as predicted by the and easy task as the present actually showed basic behav- dynamic developmental theory [19]. For the ADHD-PI ioural differences between children with ADHD and chil- group, performance improved from about 45% to about dren without symptoms. 55% across segments and showed a parallel improvement to the non-ADHD group though at a 10% lower level The present study found few statistically significant differ- overall. It might be that the non-ADHD group benefited ences between the three subtypes. When differences were more than the -PI group from the reappearing short VI indicated, they generally showed that the ADHD-C group segments, where reinforcers were presented more fre- performed with lower predictability than the other quently, at approximately every 2 s (not shown), and thus groups, and often the results for the ADHD-HI group fell learned the association with the dark grey square more between the -C and the -PI groups. The most likely expla- quickly. Interestingly, the ADHD-PI group showed better nation of this result is that it reflects that the behavioural attention than the other ADHD subtypes when using disturbances in the ADHD-PI and ADHD-HI groups are stimulus control as a measure of sustained attention (cf., less severe than those of the ADHD-C group. [20]). This suggests that the clinically described attention problems characterizing the -PI subtype may be different Page 10 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 There were no gender differences. This supports other present ADHD group and the non-ADHD group did not findings with population-based samples, showing that show significant and systematic relations between motor non-referred subjects of both genders present with similar functions and response predictability for any of the clinical and cognitive profiles [35]. These authors con- groups. Thus, deficient motor functions do not seem to be cluded that gender differences in comorbidities (includ- a significant predictor for predictability of response ing learning deficits) frequently found in clinical samples sequences in the task at hand. more likely are caused by referral biases and not by real differences between girls and boys with ADHD. The present task provides an objective measure of basic behavioural processes that is not confounded with timing ADHD-related variability has been demonstrated in a skills, motor functions, performance requirements plethora of tasks, particularly response time tasks [16]. (including emotional reactions to failure experiences), or These authors suggested that ADHD-related variability the correct understanding of more or less complicated should be demonstrated simultaneously at different levels instructions. As such, the task may prove easy to carry out; of analysis, like neurophysiological levels in addition to both for testers and subjects; and results are interpretable the behavioural; and, equally important, that such studies within a theoretical framework. Future studies will show should be developed within a sound, theoretical frame- its predictability and diagnostic utility. work. The present and earlier findings [15] suggest that predictability of response sequences is another potential Some obvious limitations to the present study should be operationalization of ADHD-related variability that might mentioned. Group membership was decided solely on the represent an etiologically important characteristic of basis of teacher scores on the Disruptive Behavior Disor- ADHD. Further, these studies were designed within the ders (DBD) rating scale [22,23] and not on a comprehen- framework of the dynamic developmental theory, a com- sive diagnostic assessment. This might mean, for instance, prehensive theory of ADHD [19], arguing that the main that only the ADHD-C group, as defined by DBD-scores, behavioural selection mechanisms, reinforcement and captured "real" ADHD to a degree found through clinical extinction, are less efficient in ADHD. Specifically, the evaluations. Inadequately defined groups most likely theory holds that altered reinforcement mechanisms; affected the results by bringing about increased within- depicted by a shorter delay-of-reinforcement gradient, group variability. Despite this, the present findings repli- result in the accumulation of responses that are selected cated to a large extent those from a well-described group by both scheduled and unscheduled reinforcers. This from a different cultural background, and suggest that the occurs because consequences (which might be unsched- task is sensitive to ADHD-related symptoms. However, uled or "accidental") in close proximity to a response will the task has not been applied with other diagnostic groups have a larger impact on future behaviour than a delayed than ADHD, so its specificity needs to be investigated. consequence (which may be the planned or scheduled Finally, the findings only pertain to children within a nar- reinforcer). The finding that reinforcers affect the latest row age range, which imply that future studies need to be response more than an overall response pattern in ADHD conducted with more age groups. (cf., [36]) supports this. Thus, immediate reinforcers may increase the future probability of any response that hap- Conclusion pened to be emitted before its delivery, resulting in aug- The present study makes a strong case for ADHD as a basic, neurobehavioural disorder, not a cultural phenom- mented behavioural variability. In addition, dysfunctioning extinction mechanisms will curb pruning enon, by the overall replication of the results from Nor- of inefficient (i.e., non-reinforced) responses so that way, a wealthy Western European country, in the very behaviour that is no longer functional is retained in the different, poor Limpopo province of South Africa and person's behavioural repertoire for an extended period, with a large majority of native African children. thereby adding to the variability. The dynamic develop- mental theory relates the altered selection mechanisms of Overall, the results were in line with the predictions from behaviour to dysfunctioning dopamine systems with cor- the dynamic developmental theory of ADHD by indicat- responding predictions about other behavioural and neu- ing that reinforcers were less efficient in the ADHD group robiological outcomes [19], which may prove valuable as than in the non-ADHD group in establishing stimulus correlational measures in future studies. control and predictable responding, due to a shorter delay-of-reinforcement gradient. The results also substan- Increased variance may be a result of augmented motor tiated ADHD-related variability as an etiologically impor- difficulties in children with ADHD-related problems. In tant characteristic of ADHD. the Norwegian sample, scores on motor tests did not dif- ferentiate ADHD from non-ADHD groups [20]. Correlat- The present study did not find statistically significant dif- ing first lag scores and scores on motor tests for the ferences between the non-ADHD group and the ADHD- Page 11 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 tion deficit hyperactivity disorder in children referred to a HI or -PI groups. It is likely that this reflects less severe psychiatric clinic. Am J Psychiatry 2002, 159:36-42. behavioural disturbances in these subtypes compared to 4. Faraone SV, Perlis RH, Doyle AE, Smoller JW, Goralnick JJ, Holmgren the ADHD-C subtype. MA, Sklar P: Molecular genetics of Attention-Deficit/Hyperac- tivity Disorder. Biol Psychiatry 2005, 57:1313-1324. 5. Arnsten AFT, Li BM: Neurobiology of executive functions: cat- In spite of basically similar response patterns, behaviour echolamine influences on prefrontal cortical functions. Biol Psychiatry 2005, 57:1377-1385. could not be predicted to the same degree in the South 6. Madras BK, Miller GM, Fischman AJ: The dopamine transporter African children as in the Norwegian children, particularly and Attention-Deficit/Hyperactivity Disorder. Biol Psychiatry when comparing non-ADHD groups. Likely reasons are 2005, 57:1397-1410. 7. Birnbaum HG, Kessler RC, Lowe SW, Secnik K, Greenberg PE, Leong less computer experience in South African children, proce- SA, Swensen AR: Costs of attention deficit-hyperactivity disor- dural differences related to reinforcement, and increased der (ADHD) in the US: excess costs of persons with ADHD group heterogeneity in the present study since the Disrup- and their family members in 2000. Curr Med Res Opin 2005, 21:195-206. tive Behavior Disorders (DBD) rating scale is less sensitive 8. Matza LS, Paramore C, Prasad M: A review of the economic bur- than the standard comprehensive diagnosis performed in den of ADHD. Cost Eff Resour Alloc 2005, 3:5. 9. Timimi S, Taylor E: ADHD is best understood as a cultural con- the Norwegian study. Still, the DBD proves to be a power- struct. Br J Psychiatry 2004, 184:8-9. ful instrument. 10. Dwivedi KN, Banhatti RG: Attention deficit/hyperactivity disor- der and ethnicity. Arch Dis Child 2005, 90 Suppl 1:i10-i12. 11. Leung PW, Luk SL, Ho TP, Taylor E, Mak FL, Bacon-Shone J: The Finally, there were no statistically significant gender differ- diagnosis and prevalence of hyperactivity in Chinese school- ences. This supports other findings with population-based boys. Br J Psychiatry 1996, 168:486-496. 12. Meyer A, Eilertsen DE, Sundet JM, Tshifularo JG, Sagvolden T: Cross- samples. cultural similarities in ADHD-like behaviour amongst South African primary school children. South African Journal of Psychol- Competing interests ogy 2004, 34:123-139. 13. Rohde LA, Szobot C, Polanczyk G, Schmitz M, Martins S, Tramontina The author(s) declare that they have no competing inter- S: Attention-Deficit/Hyperactivity Disorder in a diverse cul- ests. ture: do research and clinical findings support the notion of a cultural construct for the disorder? Biol Psychiatry 2005, 57:1436-1442. Authors' contributions 14. Nigg JT: Neuropsychologic theory and findings in Attention- HA participated in the development of study design, Deficit/Hyperactivity Disorder: the state of the field and sali- development of the reinforcement task used, participated ent challenges for the coming decade. Biol Psychiatry 2005, 57:1424-1436. in preparing the data, performed the statistical analyses, 15. Aase H, Sagvolden T: Moment-to-moment dynamics of ADHD and wrote the manuscript. AM leaded the data collection, behaviour. Behav Brain Funct 2005, 1:12. 16. Castellanos FX, Sonuga-Barke EJ, Scheres A, Di Martino A, Hyde C, participated in preparing the data, participated in writing Walters JR: Varieties of attention-deficit/hyperactivity disor- the manuscript, read and approved the final draft. TS par- der-related intra-individual variability. Biol Psychiatry 2005, ticipated in the development of study design, develop- 57:1416-1423. 17. Castellanos FX, Tannock R: Neuroscience of attention-deficit/ ment of the reinforcement task, wrote the programs for hyperactivity disorder: the search for endophenotypes. Nat statistical analyses, participated in data analyses, partici- Rev Neurosci 2002, 3:617-628. 18. Leth-Steensen C, Elbaz ZK, Douglas VI: Mean response times, var- pated in writing the manuscript, read and approved the iability, and skew in the responding of ADHD children: a final draft. response time distributional approach. Acta Psychol (Amst) 2000, 104:167-190. 19. Sagvolden T, Johansen EB, Aase H, Russell VA: A dynamic develop- Acknowledgements mental theory of attention-deficit/hyperactivity disorder The present study was supported by grants from The National Council for (ADHD) predominantly hyperactive/impulsive and com- Mental Health – Norway (Heidi Aase) and from the Norwegian Universi- bined subtypes. Behav Brain Sci 2005, 28:397-419. ties' Committee for Development Research and Education (NUFU) Psy- 20. Aase H, Sagvolden T: Infrequent, but not frequent, reinforcers produce more variable responding and deficient sustained chology Cooperation Programme between the University of Oslo and the attention in young children with attention-deficit/hyperac- University of Limpopo (Anneke Meyer). tivity disorder (ADHD). J Child Psychol Psychiatry 2006, 47:457-471. We thank Professor Edmund Sonuga-Barke, University of Southampton, for 21. Catania AC: Learning 4th edition edition. N.J. Englewoods Cliffs, Pren- tice Hall,Upper Saddle River; 1998. valuable discussions during the development of the reinforcement task, and 22. Pelham WEJ, Gnagy EM, Greenslade KE, Milich R: Teacher ratings Mr. Martin Hall, University of Southampton, for programming it. The of DSM-III-R symptoms for the disruptive behavior disorders authors also gratefully acknowledge Eunice Mashego, Mudzunga Mathivha, [published erratum appears in J Am Acad Child Adolesc Psy- Ruth Chuene, Tshikani Nkanyani, Gloria Pila, Estelle McAlpine, Dirk Baden- chiatry 1992 Nov;31(6):1177]. J Am Acad Child Adolesc Psychiatry horst and Jan Lekalakala for help in data collection and administration. 1992, 31:210-218. 23. Pillow DR, Pelham WEJ, Hoza B, Molina BS, Stultz CH: Confirma- tory factor analyses examining attention deficit hyperactiv- References ity disorder symptoms and other childhood disruptive 1. American Psychiatric Association: Diagnostic and statistical manual of behaviors. J Abnorm Child Psychol 1998, 26:293-309. mental disorders: DSM-IV-TR Washington DC, Author; 2000. 24. American Psychiatric Association: Clinical practice guideline: 2. American Psychiatric Association: Diagnostic and statistical manual of Diagnosis and evaluation of the child with Attention-Deficit/ mental disorders: DSM-IV 4th edition. Washington, D.C., Author; Hyperactivity Disorder. Pediatrics 2004, 105:1158-1170. 1994:78-85. 25. Wolraich ML, Lambert EW, Baumgaertel A, Garcia-Tornel S, Feurer 3. Biederman J, Mick E, Faraone SV, Braaten E, Doyle A, Spencer T, ID, Bickman L, Doffing MA: Teachers' screening for attention Wilens TE, Frazier E, Johnson MA: Influence of gender on atten- Page 12 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 deficit/hyperactivity disorder: comparing multinational sam- ples on teacher ratings of ADHD. J Abnorm Child Psychol 2003, 31:445-455. 26. Van Eeden R: Manual for the South African Individual Scale - Revised (SSAIS-R). Pretoria, Human Sciences Research Council; 1997. 27. Raven J: The Raven's progressive matrices: change and stabil- ity over culture and time. Cognit Psychol 2000, 41:1-48. 28. Wilkes J, Weigel A: [Comparison of WISC-R and Raven's Pro- gressive Matrices tests in a clinical consultation population]. Z Kinder Jugendpsychiatr Psychother 1998, 26:261-265. 29. Inc. SS: STATISTICA for Windows [Computer software]. Tulsa, OK: StatSoft Inc.; 2003. 30. Giedd JN, Blumenthal J, Molloy E, Castellanos FX: Brain imaging of attention deficit/hyperactivity disorder. Ann N Y Acad Sci 2001, 931:33-49. 31. Stefanatos GA, Wasserstein J: Attention deficit/hyperactivity disorder as a right hemisphere syndrome. Selective litera- ture review and detailed neuropsychological case studies. Ann N Y Acad Sci 2001, 931:172-195. 32. Houk JC, Wise SP: Distributed modular architectures linking basal ganglia, cerebellum, and cerebral cortex: their role in planning and controlling action. Cereb Cortex 1995, 5:95-110. 33. Schultz W: Getting formal with dopamine and reward. Neuron 2002, 36:241-263. 34. Barkley RA: ADHD and the nature of self control New York, Guilford; 35. Biederman J, Kwon A, Aleardi M, Chouinard VA, Marino T, Cole H, Mick E, Faraone SV: Absence of gender effects on attention def- icit hyperactivity disorder: findings in nonreferred subjects. Am J Psychiatry 2005, 162:1083-1089. 36. Tripp G, Alsop B: Sensitivity to reward frequency in boys with attention deficit hyperactivity disorder. J Clin Child Psychol 1999, 28:366-375. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 13 of 13 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral and Brain Functions Springer Journals

Moment-to-moment dynamics of ADHD behaviour in South African children

Loading next page...
 
/lp/springer-journals/moment-to-moment-dynamics-of-adhd-behaviour-in-south-african-children-JxpZQpv7ef

References (39)

Publisher
Springer Journals
Copyright
Copyright © 2006 by Aase et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
eISSN
1744-9081
DOI
10.1186/1744-9081-2-11
pmid
16569228
Publisher site
See Article on Publisher Site

Abstract

Background: The behaviour of children with Attention-Deficit/Hyperactivity Disorder is characterized by low predictability of responding. Low behavioural predictability is one way of operationalizing intra-individual ADHD- related variability. ADHD-related variability may be caused by inefficient behavioural selection mechanisms linked to reinforcement and extinction, as suggested by the recently published dynamic developmental theory (DDT) of ADHD. DDT argues that ADHD is a basic neurobehavioural disorder, caused by dysfunctioning dopamine systems. For establishing ADHD as a neurobehavioural disorder, findings from studies conducted in Western countries should be replicated in other cultural populations. The present study replicated the study conducted in Norway, with children from the Limpopo province in the Republic of South Africa. Methods: Boys and girls, aged 6–9 yr, from seven ethnic groups participated. Scores by teachers on the Disruptive Behavior Disorders rating scale defined participation in either ADHD-hyperactive/impulsive (-HI), th ADHD-predominantly inattentive (-PI), or ADHD-combined (-C) groups. Children below the 86 percentile were matched on gender and age and comprised the non-ADHD group. The children completed a computerized game-like task where mouse clicks on one of two squares on the screen resulted in delivery of a reinforcer according to a variable interval schedule of reinforcement. Reinforcers were cartoon pictures presented on the screen together with a sound. Predictability of response location and timing were measured in terms of explained variance. Results: Overall, the results replicated findings from Norway. Specifically, the ADHD-C group showed significantly lower predictability of responding than the non-ADHD group, while the ADHD-HI and the ADHD- PI groups were in-between. In accordance with the previous study, response location, but not response timing, was a sensitive behavioural measure. There were no significant gender differences. Cartoon pictures were effective reinforcers as the non-ADHD group showed learning of the task. There was no relation between behavioural predictability and motor functions. Conclusion: The present study makes a strong case for ADHD as a basic, neurobehavioural disorder, not a cultural phenomenon, by replicating findings from a wealthy Western country in a poor province of a developing country. The results were, generally, in line with predictions from the dynamic developmental theory of ADHD by indicating that reinforcers were less efficient in the ADHD group than in the non-ADHD group. Finally, the results substantiated ADHD-related variability as an etiologically important characteristic of ADHD behaviour. Page 1 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 on a microlevel in order to identify possible mechanisms Background Attention-deficit/hyperactivity disorder (ADHD) [1] is a underlying the observed intra-individual variability. behavioural disorder affecting about 2–5% of grade school children [2], making it one of the most common ADHD-related variability is mainly observed at the behav- child psychiatric diagnoses in the United States of Amer- ioural level [16], so a thorough behavioural account is ica and in Europe. In childhood, the diagnosis is more fre- warranted in order to guide the investigation of, for quent among boys. Depending on the referral practice, instance, underlying neurobiological processes. Castel- boys to girls ratios vary between 10:1 (clinical samples) lanos and colleagues [16] raised a number of key ques- and 3:1 (community samples) [3]. Research indicates that tions about ADHD-related variability that need to be the major aetiological factor is genetic [4], probably addressed in a basic research programme, including its mainly expressed as alterations in catecholaminergic reg- robustness in its association with ADHD; whether it is a ulation of brain activity [5,6]. The disorder places the random or dynamic-periodic phenomenon, whether it child at risk for school failure and dropout, juvenile delin- varies dynamically as a function of context, whether it is quency, criminality, substance abuse, and sexual promis- unique to ADHD, and finally, whether it reflects processes cuity with HIV/AIDS and teenage pregnancies as possible causal to ADHD. In the recently published dynamic devel- consequences. In this way, the disorder is extremely opmental theory of ADHD (DDT) [19] it is argued that costly, both to the afflicted individuals and their families, the main behavioural selection mechanisms, reinforce- and to the society [7,8]. Although it has been discussed ment and extinction, are altered in ADHD. These altera- whether ADHD is a phenomenon of the Western culture, tions bring about a different learning style resulting in e.g., [9], its worldwide existence is well documented [10- increased behavioural variability, in addition to hyperac- 13]. However, it seems that inconsistent assessment crite- tivity, impulsiveness, and deficient sustained attention. ria and procedures affect prevalence rates of ADHD across Some of the variability may be the result of deficient countries and cultural groups [10]. The reliance on behav- acquisition of reinforced behaviour combined with defi- ioural observations and clinical descriptions of the behav- cient extinction of non-reinforced behaviour causing iour makes the diagnosis vulnerable to subjective and shorter and less predictable behavioural sequences cultural perceptions, and the need for more objective cri- [15,19]. Thus, the DDT places ADHD-related variability teria for diagnosing is long needed. No biological marker within a causal model [19]. The DDT thereby suggests that has yet been found, and no existing neuropsychological ADHD-related variability is not a random phenomenon, test can reliably define a case of ADHD [14]. However, but can be predicted in the combined and the hyperactive/ theory-driven research aimed at identifying dysfunctions impulsive subtypes of ADHD. Further, the DDT argues in basic behavioural mechanisms may provide an empiri- that ADHD-related variability will vary dynamically as a cal basis for understanding processes and functions on function of context, task, and motivational preferences. other levels of analysis (e.g., developmental, neuropsy- Recently, we showed that boys with ADHD combined and chological, psychosocial), and for generating more hyperactive/impulsive subtypes had significantly less pre- sophisticated tests for early and reliable identification of dictable response sequences than normal boys [15] and affected individuals. The main purpose of the present that the overall variability was observed during infrequent study is to replicate an earlier study [15] showing that reinforcement as opposed to frequent [20]. However, only moment-to-moment dynamics of ADHD behaviour may boys from a culturally homogeneous population partici- represent a new way of understanding underlying behav- pated, so it is of vital importance that the phenomenon ioural mechanisms basic to ADHD. can be replicated in other samples. ADHD is characterized by age-inappropriate hyperactiv- The present study is a replication and extension of that ity, impulsiveness, and deficient sustained attention [1]. study, with a larger sample including both genders. Fur- Most clinical and experimental reports show increased ther, the study is conducted in a developing country with variability, both between and within subjects in the a culturally less homogeneous population. However, clin- ADHD group. While the group heterogeneity suggests ical resources are sparse, and assessment methods devel- multiple independent pathways to the disorder [16], the oped and validated in Western countries may not be intra-individual variability might be a key characteristic of relevant or feasible. A recent prevalence study conducted an endophenotype of ADHD and has recently become a in various language groups in South Africa showed that topic of particular interest [15-17]. A purportedly impor- when using a teacher rating scale, similar figures for prev- tant step in the direction of describing and explaining the alence as in Western countries were obtained. Also, rather role and function of this variability is more fine-grained similar distributions across the three ADHD subtypes analyses of behaviour as opposed to the traditional statis- (inattentive, hyperactive/impulsive, and combined) were tical summary measures of means and standard devia- found, in addition to a similar gender distribution [12]. tions [18]. In the present study, response data is analyzed Page 2 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 The primary aim of the present study was to investigate if Both teachers and parents were given the rating scale to the results from the Norwegian study could be replicated complete. Only the teacher's ratings were used for the in a different sample derived from various language screening, since the return of the parent's rating scale was groups in the Limpopo province of South Africa. If repli- below 50%, probably because many children either did cated, the combined results from the Norwegian and the not live with their parents or the parents were illiterate. South African studies would make a strong case for a Teacher ratings are usually regarded as an accurate meas- basic, neurobiological mechanism differentiating ADHD ure of assessment [24,25]. The teachers returned ratings of from non-ADHD children, refuting the argument that close to 100% of their pupils. The children meeting the ADHD is a cultural phenomenon resulting from Western criteria for inclusion into the clinical group (~7%) were way of life, or from Western conceptualization of psychi- selected for further testing. They were matched for gender, atric problems. In addition, the findings might lend sup- age, and language with children without ADHD as meas- port to the recently published dynamic developmental ured by the screening process. theory of ADHD [19] arguing that the main behavioural selection mechanisms, reinforcement and extinction, are Children were divided into a group classified as ADHD, altered in ADHD and will result in increased behavioural the "ADHD group" and a comparison group without variability. ADHD symptoms (Table 1), based on teacher ratings on the DBD rating scale. The cut off point for the ADHD th The present sample made it possible to investigate poten- group (95 percentile or above) was based on the results tial differential effects of reinforcers on response patterns from the prevalence study [12] in which more than 6000 in the three subtypes of ADHD [2]: the predominantly children in the Limpopo Province were rated on the DBD inattentive type (ADHD-PI), the hyperactive/impulsive scale. Norms were developed for a high cut-off based on type (ADHD-HI), and the combined type (ADHD-C). In the qualified assumption that ADHD exists in about 5% addition, potential gender differences could be studied. of the population [12]. According to these norms, scores higher than 18 on the hyperactive/impulsive (H/I) items ADHD-related variability was defined as reduced predict- were classified as ADHD-HI and higher than 21 on the ability of consecutive responses. The task used was a com- inattentive (Inatt) items were classified as the ADHD-PI puterized game where mouse clicks on either of two group. If the criteria were met on both types of items, the squares on the screen resulted in the presentation of a child was classified as ADHD-C. The cut off point for the th reinforcer. Reinforcers were cartoon pictures displayed on comparison group was set at the 85 percentile or below in the screen for a short period accompanied by a sound. order to decrease the risk for false negatives in this group, These were delivered according to variable interval (VI) as the DBD rating was the only measure. Thus, children schedules of reinforcement. VI schedules specify that with scores on H/I-related items less than 15 and Inatt responses result in a reinforcer after varying time intervals. items less than 17 were regarded as comparisons. Thus, reinforcers are presented at unpredictable times [21], avoiding any confounding with time discrimination The final sample consisted of children from seven ethnic problems. All mouse clicks were recorded both in terms of groups inhabiting the Limpopo Province of South Africa. where on the screen responses were placed (response loca- For the Afrikaans and English speaking groups, the IQ was tion; the spatial dimension) and response timing (the established with the Senior South African Individual Scale temporal dimension). Thus, the present task allowed for (SSAIS-R) [26]. As there are no standardized IQ tests for analysis of both spatial and temporal aspects of behav- the indigenous African populations, Raven's progressive iour. matrices were used to estimate IQ. This test is also consid- ered to be "culture-fair" [27,28]. Children with IQ lower Methods than 80 and/or with a history of neurological problems Participants (e.g. epilepsy, head injuries, cerebral palsy, or cerebral Children from seven ethnic groups of the Limpopo Prov- malaria) were excluded. None of the children were on psy- ince of South Africa (Northern Sotho, Venda, Tsonga, chostimulant medication at the time of testing. Tswana, North Ndebele, Afrikaans, and English) were recruited from a school-based population. The 179 chil- Procedure dren (128 boys and 51 girls) were recruited following Written permission was obtained from the Department of screening of the general population of primary school Education, Limpopo Province, as well of the principals of children representative of all socio-economic levels for the selected schools. Participation was voluntary. ADHD-like behaviour. The Disruptive Behavior Disorders Informed consent was obtained from the child's parents (DBD) rating scale [22,23] was standardized for the pop- or guardians. ulations of the Limpopo province of South Africa in an earlier study [12] and used as the screening instrument. Page 3 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 Table 1: Means and Standard Deviations, for Age and DBD Scores, by Subtype ADHD-C ADHD-HI ADHD-PI Non-ADHD Boys Girls Boys Girls Boys Girls Boys Girls Total Age (mo)* 102.4 ± 10.6 104.4 ± 11.8 109.5 ± 20.7 103.6 ± 9.0 104.5 ± 15.7 102.9 ± 19.2 108.0 ± 14.9 101.5 ± 11.8 106. 4 ± 5.1 DBD** Inatt 24.0 ± 3.8 23.8 ± 2.7 13.8 ± 4.2 14.8 ± 2.5 22.2 ± 4.4 25.8 ± 2.6 6.6 ± 5.7 5.2 ± 5.9 13.5 ± 9.2 DBD** H/I 22.0 ± 3.2 20.9 ± 3.2 20.2 ± 1.8 20.6 ± 2.1 10.0 ± 6.0 9.9 ± 4.1 4.1 ± 3.6 4.4 ± 5.1 10.6 ± 8.4 Afrikaans 2 3 1 1 3 1 10 5 26 English 000 0 102 1 4 N Sotho 1 3 4 1 0 1129 31 Tsonga 313 1 823 1 22 Venda 4 2 2 4 112178 50 N Ndebele 343 0 428 3 27 Tswana 303 1 109 2 19 Total 16 13 16 8 28 8 61 29 179 * There were no statistically significant differences in age between the subtypes ** Differences between groups were not tested, as the groups were defined by these scores. DBD: Disruptive Behavior Disorder rating scale [22]; Inatt: Scores on inattentive items; H/I: Scores on hyperactive/impulsive items The children were always tested by a tester fluent in the the two squares switched sides at random, but were never child's own language. Most assessments were done at their displayed on the same side more than twice in a row. The schools during school hours, where one of the classrooms number of presentations on each side was the same. was made available. The exceptions were children whose school was within a radius of 2 km from the University The task was designed according to a multiple variable and children referred for clinical assessment. These were interval (VI) schedule of reinforcement. A schedule is tested at the University Clinic. called multiple when two or more schedule components alternate and are signalled by discriminative stimuli. In an The SSAIS-R IQ test [26] and the Raven Progressive Matri- interval schedule some time must elapse before a response ces Test were administered by Masters students in Clinical will result in delivery of a reinforcer. In VI schedules, the Psychology who were doing their hospital internship. intervals will vary around a specified arithmetic mean [21]. The reinforcer-dense schedule was a VI 2 s and the As many of the children were not acquainted with com- reinforcer-lean schedule was a VI 20 s schedule of rein- puters, a 'mouse-training' session was part of the testing forcement. The screen's background colour changed procedure. according to the contingency in operation and functioned as the conditioned discriminative stimulus for the two Reinforcement task contingencies. A navy blue background signalled the VI 2 The task was designed as a computer "game" and was run s, while a yellow background signalled the VI 20 s. The on one of three similar laptops. The screen resolution was dark grey square was always associated with reinforce- set to 640 by 480 pixels. The response device was a stand- ment (the correct square), thus this was the discriminative ard computer mouse and clicks on either left or right but- stimulus. Reinforcers were cartoon pictures displayed on ton were recorded as responses. Moving the mouse made the screen for 1.5 s simultaneously with a sound (different the cursor move on the screen, the cursor being in the computer-generated sounds). shape of an arrow. Two squares (140 × 140 pixels) 120 pixels apart, one in a light and one in a dark shade of grey, The child was introduced to the test with the following were displayed on the screen 120 pixels from the vertical instruction (told in the child's own native language): sides (left-right) and 170 pixels from the top and from the bottom. A click within one of the squares induced a brief "This is a game you may play now. It is a little strange, because change in colour as feedback; the dark grey square turned I will not tell you how to play the game. Your task is to find out into a lighter shade and the light grey square turned into how the game works. You may use this mouse and move the a darker shade. Clicks outside the squares were recorded, arrow across the screen like this (experimenter demonstrates but gave no feedback. Following reinforcer presentations, how to move the mouse and cursor). If you want to point, you Page 4 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 can click with one of these buttons (experimenter points to the ders of the squares, disregarding whether the square was mouse buttons). You may talk while you are playing, but I will correct or not. 3) The distance-to-correct-centre measure was not answer any questions about the game. I will sit back here based on the distance from where the response was placed and write a little while you play. Do you understand your task? to the centre of the correct square. Thus, this measure was You may start now." a variant of the distance-to-nearest-centre measure, anchored to the centre of the correct square. Both distance The task started with a shaping sequence where every cor- scores (measure 2 and 3) were in terms of vertical and hor- rect response was reinforced. The screen background was izontal pixels, with 0,0 defining the centre of the square. blue and the correct square was always on the same side 4) Timing response patterns were based on consecutive (right). After six correct responses, the VI 2 s schedule interresponse times (IRTs). came into effect without any signal. The child received four reinforcers upon responding during the VI 2 s contin- The predictability of responses spatially and temporally gency before the schedule changed to the VI 20 s contin- was assessed by explained variance, i.e., autocorrelations gency (and the background changed from blue to yellow) squared. Explained variance is a better measure for overall where four reinforcers were to be obtained. The time it predictability than autocorrelations by itself since the lat- took to obtain these eight reinforcers constitute the first ter will have both positive and negative values cancelling segment of the test. The schedules alternated so that each each other when added. Autocorrelations (serial correla- was displayed five times; i.e., there were five segments, tions) of each measure were correlations of consecutive resulting in a total of 40 reinforcers per child (plus six response measures across five lags. Thus, correlations from the shaping sequence). The entire task, including between the value of response n and of response n+1 is instruction and shaping, took about 10–13 min to com- the first lag, between n and n+2 is the second lag, and so plete. on up to correlations between response n and n+5 being the fifth lag. For programming reasons, autocorrelations Data recording and statistics were computed on consecutive responses through a full Data were recorded by the laptops. Percentage of all segment and not reset at reinforcer delivery because the responses within the correct square, response side (left or number of responses was huge compared to the number right), response coordinates (i.e. the horizontal and verti- of reinforcers. Changes in predictability of responding cal pixel that the tip of the arrow-shaped cursor touched throughout the experiment could be observed in the when the child clicked a mouse button), and interre- explained variance for each child from segment to seg- sponse times (IRTs) were the recorded dependent varia- ment. An increase in autocorrelations over segments bles. The individual IRT distributions were highly skewed would indicate that performance became more and more with a long tail towards long IRTs. IRTs were therefore predictable throughout the experiment, and thus be an normalized by log transformations prior to analysis indirect measure of learning. We hypothesised that com- [logIRT = log10 ((IRT/1000) + 0.001)]. pared to the non-ADHD group less of the behaviour of children with ADHD would be predictable, indicated by Behavioural measures generally less explained variance. In addition, predictabil- The same behaviours were analyzed in the previous [15] ity of responses in the ADHD group should, to a larger and the present study. Response sequences were analyzed extent than in the non-ADHD group, be restricted to the in the VI 20 s condition only; as the short schedule usually first lag, indicating shorter behavioural sequences. would not allow for enough responses in a segment (i.e., four reinforcer deliveries within the VI 2 s condition) to In addition to the response sequences, learning was meas- calculate autocorrelations and explained variance (see ured as mean percent of all responses that were placed below). The following behaviours were analyzed: 1) A within the correct square. A high score on percent correct general side response pattern, i.e., whether consecutive would indicate that the dark grey square exerted good responses were on the left or right side of the screen. stimulus control over the responding of the children, and Highly predictable responding would typically be on the is thus a measure of attention. side where the correct target was positioned, and indicates good stimulus control. Likewise, low predictability Statistics implies that responses are unsystematically distributed on Data were analyzed by means of SPSS 11.0 for Windows the two sides and means poor stimulus control. 2) The dis- (SPSS) and Statistica 6.1 [29] program packages. The dis- tance-to-nearest-centre measure was based on the distance tance scores were computed as the square root of the sum from the pixel where the response was placed, to the cen- of squared horizontal and vertical distances. Explained tre of the selected square, whether it was the correct square variance (autocorrelations squared) was analyzed using or not. This measure indicates whether the children devel- repeated measures ANOVA across segments and lags. The oped strategies of responding that was related to the bor- ANOVA was supplemented with MANOVA. A multivari- Page 5 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 responding was found for the three spatial behavioural Non-ADHD measures, but not for the timing measure. The ADHD 0,30 ADHD Combined group had the lowest explained variance of all groups on the spatial measures. 0,25 0,20 Side response pattern This measure assessed whether the child's choice of side of 0,15 the screen (left or right) was predictable irrespective of on which side the correct response target was displayed. 0,10 Highly predictable responding across lags would imply that behaviour was orderly related to side of screen. 0,05 Generally, predictability from one response to the imme- 0,00 123 45 123 45 1234 5 1 2345 12 345 diate next (first lag) was in the low range (0.31 > mean R Seg. 1 2 3 4 5 > 0.17; median R = 0.22) compared to previously pub- lished results [15]. The side response pattern of the non- Response predictability accord Figure 1 ing to side of the screen ADHD group was more predictable compared to the Response predictability according to side of the ADHD group (Fig. 1). There were significant main effects screen. Predictability of consecutive responses according to of Group and of Segment (Table 2). The ANOVA showed side of the screen (left or right), depicted as mean explained significant interaction effects between Group and Seg- variance (autocorrelations squared), across segments (Seg. ment, between Group and Lag, and between Group, Seg- 1–5) and lags (1–5 per segment), for ADHD and non-ADHD ment, and Lag. All interactions were confirmed with the groups. Abbreviations: Seg.: segment of session. Lag: number of responses that has to be correlated to the present one, multivariate analysis (Table 2). The significant main effect i.e., correlations between response n and n+1 is the first lag, of Segment implies that there was a general upward trend between n and n+2 is the second lag, and so on up to corre- in predictability from segment 1 to segment 4. lations between response n and n+5 being the fifth lag. The follow-up ANOVA of the three ADHD subtypes and the non-ADHD group showed no significant main effect ate approach to repeated measures is recommended when of Subtype, but the main effects of Segment and of Lag variables have more than two levels because MANOVAs were significant. There was a significant interaction correct for the assumption of compound symmetry and between Subtype and Lag (F(12, 720) = 2.99; p < .001), sphericity in ANOVAs [29]. and between Subtype, Segment, and Lag (F(48, 2880) = 1.62; p < .005). Only the three-way interaction was con- Analyses relevant for the primary aim were performed firmed by the MANOVA. The interaction effects implied first, with Group (2: ADHD and non-ADHD) as the that predictability in responding for the three ADHD sub- between-group variable; and segment (5) and lag (5) as types and the non-ADHD group changed differently within-group variables. The ADHD group consisted of across segments and lags. The ADHD-C group showed the children with ADHD-C and ADHD-HI in order to repli- least predictable responding across segments. No other cate the Norwegian study. Then, as follow-up analyses, interaction effects were statistically significant. subtypes of ADHD (ADHD-HI, ADHD-PI, and ADHD-C) versus non-ADHD were analysed across segments and Distance to nearest centre lags, in order to investigate potential differences between This measure assessed to what degree the children subtypes of ADHD. Results were followed up with post responded in predictable patterns in terms of the distance hoc Scheffé tests where relevant. Non-published results from where the response was placed to the centre of the may be obtained from the first author upon request. nearest square irrespective of whether it was the correct response target or not. Demographic data Demographic and diagnostic measures of the ADHD- The generally low values of explained variance in the first 2 2 related subtypes and the non-ADHD comparison group lag (0.21 > mean R > 0.12; median R = 0.17) show a are displayed in Table 1. rather low predictability from one response to the imme- diate next. As can be seen in Figure 2, the ADHD group was generally lower than the non-ADHD group. There was Results Generally, there was no effect of Gender, neither main no significant main effect of Group, however, but a signif- effects nor interaction effects. Consequently, the reported icant main effect of Lag. The interaction between Group findings combine boys and girls. Further, predictable Page 6 of 13 (page number not for citation purposes) Ex plained Variance Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 Table 2: Results from repeated measures ANOVA and multivariate tests for repeated measures, of explained variance (squared autocorrelations) Measure Variable ANOVA Multivariate Df F Df F Side Response Pattern Group (G) 1, 146 13.857* Segment (Seg) 4, 584 3.769** 4, 143 2.650* Lag 4, 584 167.626*** 4, 143 49.352*** G * Seg 4, 584 3.941** 4, 143 2.792* G * Lag 4, 584 4.206** 4, 143 2.697* G * Seg * Lag 16, 2336 1.802* 16, 131 2.090** Distance to centre of nearest square G 1, 146 2.588 Seg 4, 584 0.150 4, 143 0.146 Lag 4, 584 135.577*** 4, 143 41.892*** G * Seg 4, 584 0.425 4, 143 0.428 G * Lag 4, 584 1.242 4, 143 2.863* G * Seg * Lag 16, 2336 1.410 16, 131 1.487 Distance to centre of correct square G 1, 146 3.974* Seg 4, 584 0.875 4, 143 0.723 Lag 4, 584 187.974*** 4, 143 57.819*** G * Seg 4, 584 4.475*** 4, 143 3.663** G * Lag 4, 584 4.996*** 4, 143 1.943 G * Seg * Lag 16, 2336 1.505 16, 131 1.203 Timing Response Pattern G 1, 146 0.843 Seg 4, 584 1.270 4, 143 0.929 Lag 4, 584 60.096*** 4, 143 31.501*** G * Seg 4, 584 0.994 4, 143 0.944 G * Lag 4, 584 0.660 4, 143 0.922 G * Seg * Lag 16, 2336 1.107 16, 131 1.374 *: p < .05; **: p < .01; ***: p < .001 and Lag was significant in the multivariate analysis (Table behaviour compared to the two other spatial measures. 2). Predictability from one response to the immediate next (explained variance in the first lag) was generally in the 2 2 The follow-up ANOVA of the three subtypes and the non- lower middle range (0.35 > mean R > 0.19; median R = ADHD group showed no significant effects except a signif- 0.25) compared to previous results [15]. The ADHD icant main effect of Lag (F(4, 720) = 149.27; p < .001). In group had less predictable behaviour than the non-ADHD order to check if there were statistically significant differ- group. ences between any two subtypes, a post hoc Scheffé test of Subtype and Lag was performed. The main results from The ANOVA showed a significant main effect of Group, in this test indicate a larger decrease in explained variance addition to the main effect of Lag (Table 2). There were from the first to the next lags in the non-ADHD group and also significant interaction effects between Group and in the ADHD-PI subtype compared to the other subtypes, Segment, and between Group and Lag. The interaction and that there was a larger difference between the ADHD- between Group, Segment, and Lag did not meet conven- C subtype and the non-ADHD group compared to any tional levels of significance (p > .08). Only the interaction other combination of subtypes. involving Group and Segment was confirmed by the mul- tivariate analysis (Table 2). This interaction showed that Distance to correct centre while the non-ADHD group's behaviour increased in pre- This measure assessed patterns in response placements in dictability over segments, the ADHD group's behaviour terms of distance from the centre of the correct square. did not improve over segments. Again, low explained variance indicated high variability in spatial responding. As can be seen in Figure 3, responding The follow-up ANOVA of the three subtypes and the non- was somewhat more predictable with this measure of ADHD group only showed a significant interaction effect Page 7 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 showed a significant interaction effect between Group, Non-ADHD ADHD Segment, and Lag (F(48, 2880) = 1.39; p < .05), which was 0,20 not confirmed by the multivariate analysis. The interac- tion implied that patterns of explained variance changed 0,15 between the groups both across segments and lags in an uninterpretable way. No further follow-up analyses were run for this measure. 0,10 Learning Percentage of all responses that were placed within the 0,05 square associated with reinforcement (correct square) was used as a measure of learning. Overall, the non-ADHD group had higher percent correct scores than the ADHD 0,00 12345 12345 1234 5 12345 12345 group (58.6% vs 46.2%). The non-ADHD group increased Seg. 1 2 3 4 5 their scores with about 10% (from 52.8% to 62.5%) across segments, while the ADHD group only increased Respo tr Figure 2 e of a square nse predictability according to distance from the cen- with less than 2%. The ANOVA of percent correct choice Response predictability according to distance from of square showed significant main effects of Group (F(1, the centre of a square. Predictability of consecutive 147) = 16.142; p < .001) and of Segment (F(4, 588) = responses according to distance from the centre of a 7.635; p < .001). The interaction effect between Group response square, whether correct or not, to where on the and Segment was also significant (F(4, 588) = 4.297; p < screen the responses were placed. Curves show mean .002); and was confirmed by the MANOVA (F(4, 144) = explained variance (autocorrelations squared) across seg- ments (Seg. 1–5) and lags (1–5 per segment), for ADHD and 2.964; p < .03). non-ADHD groups. Abbreviations: Seg.: segment of session. Lag: number of responses that has to be correlated to the The follow-up ANOVA of the three subtypes and the non- present one, i.e., correlations between response n and n+1 is ADHD group showed a main effect of Subtype (F(3, 180) the first lag, between n and n+2 is the second lag, and so on = 6.14; p < .001) and of Segment (F(4, 720) = 7.18; p < up to correlations between response n and n+5 being the .001) (Fig. 4). The interaction effect between Subtype and fifth lag. Segment did not meet conventional levels of significance (F(12, 720) = 1.72; p = .058). The main effect of Segment was confirmed by the MANOVA. between Subtype and Lag (F(12, 720) = 2.64; p < .002) in addition to a main effect of Lag (F(4, 720) = 198.39; p < A post-hoc Scheffé test of the four subtypes showed that .001). The interaction effect indicated that the ADHD-C the non-ADHD group had significantly higher mean per- group showed less behavioural predictability than any of cent correct than the ADHD-C and the ADHD-HI sub- the other groups, while the ADHD-PI did not differ from types (p < .008 and p < .04, respectively). The ADHD-PI the non-ADHD group. The interaction was not confirmed group was not significantly different from any of the other by the MANOVA. subtypes, and the -C and -HI subtypes were not signifi- cantly different from each other. In order to check what contributed to the interaction effect, a post hoc Scheffé test was performed involving Response predictability and motor functions Subtype and Lag. The main results from this test con- A recent study of motor functions of the present South- firmed some overlap between the non-ADHD and African sample showed significant differences between ADHD-PI subtypes, and that explained variance for these children with ADHD-related symptoms and those with- groups was significantly higher than that of the ADHD-C out (Meyer & Sagvolden, subm. to BBF). Increased behav- subtype. ioural variability may be a result of motor coordination problems. This relation was tested by correlating first lag Timing response pattern explained variance of all spatial measures with normal- The development of patterns in response timing was ized scores on Grooved Pegboard test and Mazes (see investigated in terms of consecutive interresponse times Meyer & Sagvolden for details) for both dominant and (IRTs). Explained variance of the first lag was generally non-dominant hand, for the ADHD-C group and the non- very low and not significantly different between the ADHD group separately. For the ADHD group, scores on groups (0.08 > mean R > 0.05). Besides a significant main response predictability was mainly negatively related to effect of Lag, there were no other main or interaction scores on motor tests (on the motor tests, low scores were effects (Table 2). The follow-up analysis of the subtypes preferred), but the relation was weak (Pearson correla- Page 8 of 13 (page number not for citation purposes) Explained Variance Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 0,35 Non-ADHD ADHD-C ADHD ADHD-HI ADHD-PI 0,30 Non-ADHD 0,25 0,20 0,15 50 0,10 0,05 0,00 12345 12345 12345 12345 12 345 Segment Segm. 1 2 3 4 5 Mean percent correct Figure 4 Respo tre Figure 3 of the correct target nse predictability according to distance from the cen- Mean percent correct. Mean percent correct choice of Response predictability according to distance from response target across consecutive segments (Seg) for the centre of the correct target. Predictability of consec- ADHD and non-ADHD groups. Abbreviations: ADHD-C: utive responses according to distance from the centre of the ADHD combined type; ADHD-HI: ADHD hyperactive/ correct target square to where on the screen the responses impulsive type; ADHD-PI: ADHD predominantly inattentive were placed. Curves show mean explained variance (auto- type. correlations squared) across segments (Seg. 1–5) and lags (1– 5 per segment), for ADHD and non-ADHD groups. Abbrevi- ations: Seg.: segment of session. Lag: number of responses that has to be correlated to the present one, i.e., correla- ince. Actually, when comparing the results from the two tions between response n and n+1 is the first lag, between n studies, the ADHD groups from the two populations seem and n+2 is the second lag, and so on up to correlations more similar to each other than the non-ADHD groups between response n and n+5 being the fifth lag. are to each other. Combined, the results suggest that the phenomenon might pertain to the ADHD-C subtype in particular, although some results indicate that children with less severe ADHD, as those with hyperactive/impul- tions .234 > r > -.352). For the non-ADHD group, the rela- sive subtype, may show weaker forms of the phenome- tion was even weaker (Pearson correlations .237 > r > - non. This conclusion is in line with predictions made .175). from the dynamic developmental theory, DDT [19], applying specifically to ADHD-C and -HI subtypes. Discussion The present study was designed to follow up the results of Another striking similarity between the two studies was a study in a wealthy European country, Norway, which the finding that predictability of consecutive responding showed that response sequences in children with ADHD were found for the spatial behavioural measures only and symptoms is less predictable than those of children with- not for the temporal measure. We speculated that this out ADHD [15]. The present study was conducted in the finding might be related to the visuo-spatial nature of the poor Limpopo province in South Africa, a developing task, alternatively that striatal dysfunction might explain country with a more heterogeneous population, and with deficient habit learning in ADHD [15]. The fact that simi- fewer resources and assessment options compared to lar patterns of results were found in two extremely diverse Western countries. The children's behaviour problems samples, recruited by different methods and in different were rated with the DBD [22] as psychiatric services are cultures, indicates a biologically-founded mechanism generally not available in developing countries. A strong rather than a culturally-imposed response style, and that case for ADHD as a basic, neurobehavioural disorder, not the mechanism might be specific to ADHD with symp- a cultural phenomenon, could be made if the results from toms of hyperactivity/impulsiveness. Imaging and neu- Norway were replicated in a developing country. ropsychological research suggest a right hemisphere frontal-striatal circuitry involvement in ADHD, e.g., The most striking finding in the present study was that the [30,31], indicating a specific dysfunction in perception lower predictability of consecutive responses of boys with and treatment of visuo-spatial stimuli. An alternative ADHD compared to controls in the Norwegian study [15] hypothesis, highly compatible with the DDT, has sug- was replicated in boys and girls from the Limpopo prov- gested that cortical basal-ganglionic neuronal modules Page 9 of 13 (page number not for citation purposes) Explained Variance Mean Percent Correct Responses Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 learn to recognize and register complex contextual pat- from those often described in the -HI and -C subtypes terns that are relevant to behaviour, and that this learning [19,34]. The present findings thus support the assump- is mediated by dopaminergic reinforcement signals [32]. tion in the DDT that altered learning mechanisms related The contextual information includes the state of the to a shorter delay gradient mainly apply to the -HI and -C organism, the location of targets of action, the desirability subtypes and that the present attention measure mainly of an action, motor intentions, and sensory inputs apt for relates to the learning style of these subtypes in particular. either selecting or triggering motor programs. The pattern recognition going on in the striatum is guided by In general, the behaviour of all groups was less predictable dopaminergic reward prediction signals [33]. The recur- in the present study than in the previous study [15], par- sive process in the cortico-striatal module enables the ticularly when comparing the non-ADHD groups. Learn- basal ganglia to encode even more complex contexts ing was also poorer, as mean percent correct never based on those initially recognized [32]. With dysfunc- exceeded 63% even in the non-ADHD group, while the tioning dopamine systems, learning the association non-ADHD group in the Norwegian study performed up between different contextual cues and between environ- to 90% correct [20]. Several factors may explain this mental signals and relevant motor programs will be ham- result. For instance, the groups were probably more heter- pered [19]. ogeneous in the present study. The DBD rating scale is most likely less sensitive than the standard comprehen- It is important to notice that the cartoons acted as rein- sive diagnosis performed in the Norwegian study. Also, forcers in the non-ADHD children as shown by their the non-ADHD group was less well described and allowed th learning curves (Fig. 4). Learning could also be observed for DBD scores up to the 85 percentile. In addition, chil- on the distance-to-correct-centre measure (Fig. 3), show- dren in a developing country may not be as used to com- ing increasingly better prediction from one response to puters as Norwegian children. While most Norwegian the next (first lag) in the non-ADHD group over segments, children are acquainted with computer games, including but not in the ADHD group. The higher and increasing clicking on items in order to bring up other items or hap- percent correct scores of the non-ADHD group, and the penings on the screen, the South African children may not lower, flatter learning curves of the ADHD groups, repli- be that familiar with computer games. This will result in cate other findings from Norway [20]. In the present study more explorative and somewhat less systematic behaviour the ADHD-C and -HI groups had more incorrect in the South African children. Further, an unintended pro- responses than correct throughout the session (Fig. 4). cedural difference between the two studies might have This indicates no stimulus control over behaviour by the influenced the results. In Norway, testers were instructed dark grey square in these groups. Stimulus control is dem- to add verbal feedback (like "Wow", "Look at that", etc.) onstrated when performance is predictably related to the when the cartoons were displayed on the screen. This was stimulus signalling reinforcement, and the stimulus has not done in South Africa. Thus the reinforcing effect might gained conditioned reinforcing properties. For a stimulus have been less, particularly for the non-ADHD children as to become a conditional reinforcer, it must be coupled this group deviated more from their Norwegian counter- with a primary reinforcer within a certain time, delimited parts than the ADHD group. by the length of a delay-of-reinforcement gradient (Cata- nia's precommentary in [19]). The present findings may The findings demonstrate, nonetheless, that the task be explained by a short delay gradient in the -C and -HI could be run in just one session and with no extra tangible groups, resulting in no association between the stimulus reinforcers. Hence, a quick (less than 15 min altogether), and the response-produced reinforcer, as predicted by the and easy task as the present actually showed basic behav- dynamic developmental theory [19]. For the ADHD-PI ioural differences between children with ADHD and chil- group, performance improved from about 45% to about dren without symptoms. 55% across segments and showed a parallel improvement to the non-ADHD group though at a 10% lower level The present study found few statistically significant differ- overall. It might be that the non-ADHD group benefited ences between the three subtypes. When differences were more than the -PI group from the reappearing short VI indicated, they generally showed that the ADHD-C group segments, where reinforcers were presented more fre- performed with lower predictability than the other quently, at approximately every 2 s (not shown), and thus groups, and often the results for the ADHD-HI group fell learned the association with the dark grey square more between the -C and the -PI groups. The most likely expla- quickly. Interestingly, the ADHD-PI group showed better nation of this result is that it reflects that the behavioural attention than the other ADHD subtypes when using disturbances in the ADHD-PI and ADHD-HI groups are stimulus control as a measure of sustained attention (cf., less severe than those of the ADHD-C group. [20]). This suggests that the clinically described attention problems characterizing the -PI subtype may be different Page 10 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 There were no gender differences. This supports other present ADHD group and the non-ADHD group did not findings with population-based samples, showing that show significant and systematic relations between motor non-referred subjects of both genders present with similar functions and response predictability for any of the clinical and cognitive profiles [35]. These authors con- groups. Thus, deficient motor functions do not seem to be cluded that gender differences in comorbidities (includ- a significant predictor for predictability of response ing learning deficits) frequently found in clinical samples sequences in the task at hand. more likely are caused by referral biases and not by real differences between girls and boys with ADHD. The present task provides an objective measure of basic behavioural processes that is not confounded with timing ADHD-related variability has been demonstrated in a skills, motor functions, performance requirements plethora of tasks, particularly response time tasks [16]. (including emotional reactions to failure experiences), or These authors suggested that ADHD-related variability the correct understanding of more or less complicated should be demonstrated simultaneously at different levels instructions. As such, the task may prove easy to carry out; of analysis, like neurophysiological levels in addition to both for testers and subjects; and results are interpretable the behavioural; and, equally important, that such studies within a theoretical framework. Future studies will show should be developed within a sound, theoretical frame- its predictability and diagnostic utility. work. The present and earlier findings [15] suggest that predictability of response sequences is another potential Some obvious limitations to the present study should be operationalization of ADHD-related variability that might mentioned. Group membership was decided solely on the represent an etiologically important characteristic of basis of teacher scores on the Disruptive Behavior Disor- ADHD. Further, these studies were designed within the ders (DBD) rating scale [22,23] and not on a comprehen- framework of the dynamic developmental theory, a com- sive diagnostic assessment. This might mean, for instance, prehensive theory of ADHD [19], arguing that the main that only the ADHD-C group, as defined by DBD-scores, behavioural selection mechanisms, reinforcement and captured "real" ADHD to a degree found through clinical extinction, are less efficient in ADHD. Specifically, the evaluations. Inadequately defined groups most likely theory holds that altered reinforcement mechanisms; affected the results by bringing about increased within- depicted by a shorter delay-of-reinforcement gradient, group variability. Despite this, the present findings repli- result in the accumulation of responses that are selected cated to a large extent those from a well-described group by both scheduled and unscheduled reinforcers. This from a different cultural background, and suggest that the occurs because consequences (which might be unsched- task is sensitive to ADHD-related symptoms. However, uled or "accidental") in close proximity to a response will the task has not been applied with other diagnostic groups have a larger impact on future behaviour than a delayed than ADHD, so its specificity needs to be investigated. consequence (which may be the planned or scheduled Finally, the findings only pertain to children within a nar- reinforcer). The finding that reinforcers affect the latest row age range, which imply that future studies need to be response more than an overall response pattern in ADHD conducted with more age groups. (cf., [36]) supports this. Thus, immediate reinforcers may increase the future probability of any response that hap- Conclusion pened to be emitted before its delivery, resulting in aug- The present study makes a strong case for ADHD as a basic, neurobehavioural disorder, not a cultural phenom- mented behavioural variability. In addition, dysfunctioning extinction mechanisms will curb pruning enon, by the overall replication of the results from Nor- of inefficient (i.e., non-reinforced) responses so that way, a wealthy Western European country, in the very behaviour that is no longer functional is retained in the different, poor Limpopo province of South Africa and person's behavioural repertoire for an extended period, with a large majority of native African children. thereby adding to the variability. The dynamic develop- mental theory relates the altered selection mechanisms of Overall, the results were in line with the predictions from behaviour to dysfunctioning dopamine systems with cor- the dynamic developmental theory of ADHD by indicat- responding predictions about other behavioural and neu- ing that reinforcers were less efficient in the ADHD group robiological outcomes [19], which may prove valuable as than in the non-ADHD group in establishing stimulus correlational measures in future studies. control and predictable responding, due to a shorter delay-of-reinforcement gradient. The results also substan- Increased variance may be a result of augmented motor tiated ADHD-related variability as an etiologically impor- difficulties in children with ADHD-related problems. In tant characteristic of ADHD. the Norwegian sample, scores on motor tests did not dif- ferentiate ADHD from non-ADHD groups [20]. Correlat- The present study did not find statistically significant dif- ing first lag scores and scores on motor tests for the ferences between the non-ADHD group and the ADHD- Page 11 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 tion deficit hyperactivity disorder in children referred to a HI or -PI groups. It is likely that this reflects less severe psychiatric clinic. Am J Psychiatry 2002, 159:36-42. behavioural disturbances in these subtypes compared to 4. Faraone SV, Perlis RH, Doyle AE, Smoller JW, Goralnick JJ, Holmgren the ADHD-C subtype. MA, Sklar P: Molecular genetics of Attention-Deficit/Hyperac- tivity Disorder. Biol Psychiatry 2005, 57:1313-1324. 5. Arnsten AFT, Li BM: Neurobiology of executive functions: cat- In spite of basically similar response patterns, behaviour echolamine influences on prefrontal cortical functions. Biol Psychiatry 2005, 57:1377-1385. could not be predicted to the same degree in the South 6. Madras BK, Miller GM, Fischman AJ: The dopamine transporter African children as in the Norwegian children, particularly and Attention-Deficit/Hyperactivity Disorder. Biol Psychiatry when comparing non-ADHD groups. Likely reasons are 2005, 57:1397-1410. 7. Birnbaum HG, Kessler RC, Lowe SW, Secnik K, Greenberg PE, Leong less computer experience in South African children, proce- SA, Swensen AR: Costs of attention deficit-hyperactivity disor- dural differences related to reinforcement, and increased der (ADHD) in the US: excess costs of persons with ADHD group heterogeneity in the present study since the Disrup- and their family members in 2000. Curr Med Res Opin 2005, 21:195-206. tive Behavior Disorders (DBD) rating scale is less sensitive 8. Matza LS, Paramore C, Prasad M: A review of the economic bur- than the standard comprehensive diagnosis performed in den of ADHD. Cost Eff Resour Alloc 2005, 3:5. 9. Timimi S, Taylor E: ADHD is best understood as a cultural con- the Norwegian study. Still, the DBD proves to be a power- struct. Br J Psychiatry 2004, 184:8-9. ful instrument. 10. Dwivedi KN, Banhatti RG: Attention deficit/hyperactivity disor- der and ethnicity. Arch Dis Child 2005, 90 Suppl 1:i10-i12. 11. Leung PW, Luk SL, Ho TP, Taylor E, Mak FL, Bacon-Shone J: The Finally, there were no statistically significant gender differ- diagnosis and prevalence of hyperactivity in Chinese school- ences. This supports other findings with population-based boys. Br J Psychiatry 1996, 168:486-496. 12. Meyer A, Eilertsen DE, Sundet JM, Tshifularo JG, Sagvolden T: Cross- samples. cultural similarities in ADHD-like behaviour amongst South African primary school children. South African Journal of Psychol- Competing interests ogy 2004, 34:123-139. 13. Rohde LA, Szobot C, Polanczyk G, Schmitz M, Martins S, Tramontina The author(s) declare that they have no competing inter- S: Attention-Deficit/Hyperactivity Disorder in a diverse cul- ests. ture: do research and clinical findings support the notion of a cultural construct for the disorder? Biol Psychiatry 2005, 57:1436-1442. Authors' contributions 14. Nigg JT: Neuropsychologic theory and findings in Attention- HA participated in the development of study design, Deficit/Hyperactivity Disorder: the state of the field and sali- development of the reinforcement task used, participated ent challenges for the coming decade. Biol Psychiatry 2005, 57:1424-1436. in preparing the data, performed the statistical analyses, 15. Aase H, Sagvolden T: Moment-to-moment dynamics of ADHD and wrote the manuscript. AM leaded the data collection, behaviour. Behav Brain Funct 2005, 1:12. 16. Castellanos FX, Sonuga-Barke EJ, Scheres A, Di Martino A, Hyde C, participated in preparing the data, participated in writing Walters JR: Varieties of attention-deficit/hyperactivity disor- the manuscript, read and approved the final draft. TS par- der-related intra-individual variability. Biol Psychiatry 2005, ticipated in the development of study design, develop- 57:1416-1423. 17. Castellanos FX, Tannock R: Neuroscience of attention-deficit/ ment of the reinforcement task, wrote the programs for hyperactivity disorder: the search for endophenotypes. Nat statistical analyses, participated in data analyses, partici- Rev Neurosci 2002, 3:617-628. 18. Leth-Steensen C, Elbaz ZK, Douglas VI: Mean response times, var- pated in writing the manuscript, read and approved the iability, and skew in the responding of ADHD children: a final draft. response time distributional approach. Acta Psychol (Amst) 2000, 104:167-190. 19. Sagvolden T, Johansen EB, Aase H, Russell VA: A dynamic develop- Acknowledgements mental theory of attention-deficit/hyperactivity disorder The present study was supported by grants from The National Council for (ADHD) predominantly hyperactive/impulsive and com- Mental Health – Norway (Heidi Aase) and from the Norwegian Universi- bined subtypes. Behav Brain Sci 2005, 28:397-419. ties' Committee for Development Research and Education (NUFU) Psy- 20. Aase H, Sagvolden T: Infrequent, but not frequent, reinforcers produce more variable responding and deficient sustained chology Cooperation Programme between the University of Oslo and the attention in young children with attention-deficit/hyperac- University of Limpopo (Anneke Meyer). tivity disorder (ADHD). J Child Psychol Psychiatry 2006, 47:457-471. We thank Professor Edmund Sonuga-Barke, University of Southampton, for 21. Catania AC: Learning 4th edition edition. N.J. Englewoods Cliffs, Pren- tice Hall,Upper Saddle River; 1998. valuable discussions during the development of the reinforcement task, and 22. Pelham WEJ, Gnagy EM, Greenslade KE, Milich R: Teacher ratings Mr. Martin Hall, University of Southampton, for programming it. The of DSM-III-R symptoms for the disruptive behavior disorders authors also gratefully acknowledge Eunice Mashego, Mudzunga Mathivha, [published erratum appears in J Am Acad Child Adolesc Psy- Ruth Chuene, Tshikani Nkanyani, Gloria Pila, Estelle McAlpine, Dirk Baden- chiatry 1992 Nov;31(6):1177]. J Am Acad Child Adolesc Psychiatry horst and Jan Lekalakala for help in data collection and administration. 1992, 31:210-218. 23. Pillow DR, Pelham WEJ, Hoza B, Molina BS, Stultz CH: Confirma- tory factor analyses examining attention deficit hyperactiv- References ity disorder symptoms and other childhood disruptive 1. American Psychiatric Association: Diagnostic and statistical manual of behaviors. J Abnorm Child Psychol 1998, 26:293-309. mental disorders: DSM-IV-TR Washington DC, Author; 2000. 24. American Psychiatric Association: Clinical practice guideline: 2. American Psychiatric Association: Diagnostic and statistical manual of Diagnosis and evaluation of the child with Attention-Deficit/ mental disorders: DSM-IV 4th edition. Washington, D.C., Author; Hyperactivity Disorder. Pediatrics 2004, 105:1158-1170. 1994:78-85. 25. Wolraich ML, Lambert EW, Baumgaertel A, Garcia-Tornel S, Feurer 3. Biederman J, Mick E, Faraone SV, Braaten E, Doyle A, Spencer T, ID, Bickman L, Doffing MA: Teachers' screening for attention Wilens TE, Frazier E, Johnson MA: Influence of gender on atten- Page 12 of 13 (page number not for citation purposes) Behavioral and Brain Functions 2006, 2:11 http://www.behavioralandbrainfunctions.com/content/2/1/11 deficit/hyperactivity disorder: comparing multinational sam- ples on teacher ratings of ADHD. J Abnorm Child Psychol 2003, 31:445-455. 26. Van Eeden R: Manual for the South African Individual Scale - Revised (SSAIS-R). Pretoria, Human Sciences Research Council; 1997. 27. Raven J: The Raven's progressive matrices: change and stabil- ity over culture and time. Cognit Psychol 2000, 41:1-48. 28. Wilkes J, Weigel A: [Comparison of WISC-R and Raven's Pro- gressive Matrices tests in a clinical consultation population]. Z Kinder Jugendpsychiatr Psychother 1998, 26:261-265. 29. Inc. SS: STATISTICA for Windows [Computer software]. Tulsa, OK: StatSoft Inc.; 2003. 30. Giedd JN, Blumenthal J, Molloy E, Castellanos FX: Brain imaging of attention deficit/hyperactivity disorder. Ann N Y Acad Sci 2001, 931:33-49. 31. Stefanatos GA, Wasserstein J: Attention deficit/hyperactivity disorder as a right hemisphere syndrome. Selective litera- ture review and detailed neuropsychological case studies. Ann N Y Acad Sci 2001, 931:172-195. 32. Houk JC, Wise SP: Distributed modular architectures linking basal ganglia, cerebellum, and cerebral cortex: their role in planning and controlling action. Cereb Cortex 1995, 5:95-110. 33. Schultz W: Getting formal with dopamine and reward. Neuron 2002, 36:241-263. 34. Barkley RA: ADHD and the nature of self control New York, Guilford; 35. Biederman J, Kwon A, Aleardi M, Chouinard VA, Marino T, Cole H, Mick E, Faraone SV: Absence of gender effects on attention def- icit hyperactivity disorder: findings in nonreferred subjects. Am J Psychiatry 2005, 162:1083-1089. 36. Tripp G, Alsop B: Sensitivity to reward frequency in boys with attention deficit hyperactivity disorder. J Clin Child Psychol 1999, 28:366-375. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 13 of 13 (page number not for citation purposes)

Journal

Behavioral and Brain FunctionsSpringer Journals

Published: Mar 28, 2006

There are no references for this article.