- Email: [email protected]
A neuropsychological profile of attention deficits in young males with fragile X syndrome
Neuropsychologia 38 (2000) 1261±1270
www.elsevier.com/locate/neuropsychologia
A neuropsychological pro®le of attention de®cits in young males with fragile X syndrome F. Munir a, K.M. Cornish a,*, J. Wilding b a
Section of Developmental Psychiatry, Division of Psychiatry, E Floor, South Block, Queens Medical Centre, University of Nottingham, Nottingham NG7 2UH, UK b Department of Psychology, Royal Holloway, University of London, Egham, Surrey, UK Received 20 August 1999; received in revised form 27 January 2000; accepted 2 February 2000
Abstract Dierent processes of attention were examined in a group of 25 fragile X boys with FMR-1 full mutation and compared with three control groups: a learning disabled comparison group comprising 25 boys with Down's syndrome, matched to the fragile X boys on verbal mental age; and 50 mainstream school boys (controls) matched to the fragile X boys on verbal mental age. The controls were further divided into those matched on ``poor attention'' to the fragile X boys and a ``good'' attention group, as rated by the ACTeRS questionnaire. Four categories of attention tasks were employed: selective attention, divided attention, sustained attention and executive functioning. The main ®ndings of the study indicate that fragile X boys display an attention de®cit at higher levels of attention function/executive functioning and that this pro®le is dierent from the pro®le identi®ed in Down's syndrome boys and more extreme than the pro®le identi®ed in the poor attention control group. These ®ndings are discussed in the context of functional neuroimaging and brain-behaviour correlates in fragile X syndrome. 7 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction Fragile X syndrome is the most common inherited genetic condition causing developmental and learning disabilities [34]. Recent evidence suggests that fragile X syndrome aects approximately 1:4000 males and 1:6000 females [24], and is associated with the dysfunction of a gene Ð known as the fragile X mental retardation 1 (FMR-1) gene Ð on the X chromosome related to the fragile site Xq27.3. This site involves the abnormal ampli®cation of a DNA segment containing a trinucleotide sequence of CGG repeats [27,45,48]. Between six and 58 CGG repeats are typically observed in the general population. In individuals who have the fragile X mutation, an expansion or ampli®cation occurs in this segment of
* Corresponding author. Tel.: +44-115-978-1918. E-mail address: [email protected] (K.M. Cornish).
their DNA from 59 to up to several hundred CGG repeats. Those who have between 59 and 200 CGG repeats have a pre-mutation and are known as ``carriers'' but with no overt clinical eects. It appears that the pre-mutation is functionally silent with the non-methylated gene still being transcribed. Individuals with more than 200 CGG repeats however, have a full mutation whereby the gene becomes methylated at the fragile site Xq27.3 leading to a lack of mRNA and FMR-1 protein (FMRP) synthesis [29,45]. Males with the full mutation are invariably clinically aected. Females who carry the full mutation on the other hand, may be clinically unaected. It is the lack of FMRP that results in the clinical manifestation of the fragile X syndrome. While the speci®c function of FMRP is not known, it is strongly expressed in brain tissues [45] and may be responsible for an impairment or lack of structural development in brain functioning in aected fragile X males and females.
0028-3932/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 3 2 ( 0 0 ) 0 0 0 3 6 - 1
1262
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
1.1. Neuropsychological phenotype in fragile X males The neuropsychological-cognitive pro®le of fragile X boys (full mutation) appears to be more complex than originally thought. This directly impacts on the current understanding of the disorder in terms of brain development and function and the genetic factors underlying the disorder. Past studies suggested de®cits in short-term non-verbal memory, visual motor and spatial skills, with relative strengths in verbal labelling, verbal memory and comprehension [12,13,17] thus indicating cognitive de®cits that may be attributed to poor right-hemisphere functioning [9]. More recent comprehensive studies, however, suggest a pattern of de®cits that are task speci®c rather than global in nature. Spatial skills for example, are not globally impaired in fragile X males. Instead, the de®cit appears to be speci®c to those skills that require visuo-spatial orientation and constructional abilities, with visuo-perceptual skills relatively intact [7,8,12]. Similarly, verbal skills are not a global strength in fragile X males, with speci®c de®cits reported on tasks that require short-term memory for complex sequential information, and particular strengths on tasks that require short-term memory for simple, meaningful information [13,37]. Further evidence for a possible discrepancy in cognitive functioning suggests fragile X males display relatively intact skills for learning simple verbal and non-verbal tasks, but are greatly impaired on tasks which require the manipulation of internal representations [16]. Likewise, on tasks of working memory (the ability to temporarily hold information in mind while processing same or other information), fragile X males can temporarily maintain and recall simple meaningful information, but not abstract nonsequential information irrespective of the verbal or non-verbal working memory subsystem [26]. Current studies therefore suggest that fragile X males can perform relatively simple tasks but cannot manipulate more complex information regardless of its modality. It is not known why these speci®c cognitive skills are aected in fragile X males. One possibility may lie in the direction of higher control processes of attention such as the ability to modulate complex or novel information and plan and organise information in the mind. Indeed, neuroanatomical studies of full mutation fragile X males and females report abnormalities in both left and right parietal regions as well as in the prefrontal regions [37,40] and the caudate nucleus [33]. Signi®cantly, these structures appear to be involved in attention [5,6,30] and higher control processes of attention such as executive functioning [2] indicating that these mechanisms may be impaired in those with fragile X syndrome. These neuroanatomical ®ndings are in accord with neuropsychological ®ndings in fragile X females, where a consistent pattern of di-
culties in visual±spatial attention and executive functioning has been documented [21,22,24,39]. In contrast, comparatively few studies have taken a similar direction in terms of identifying a neuropsychological pro®le of the attention de®cit in fragile X males. This is somewhat surprising given that attentional problems, impulsive behaviour and hyperactivity are frequently cited as core features of the behaviour phenotype in fragile X males [4,14,19,42]. As a result, relatively little is known about the precise nature of the attention de®cits, although recent studies have indicated the importance of teasing apart cognitive aspects of attention. For example, selective attention refers to the ability to select relevant stimuli from irrelevant information, while divided attention refers to the ability to attend to more than one source of information simultaneously [36,38]. In contrast, maintaining attention over a period of time is referred to as sustained attention; and problem-solving, planning, inhibiting and manipulating mental representations of tasks and goals are seen as higher control processes of attention or executive functioning [31,46]. Neuroanatomic studies further suggest that these dierent aspects of attention are subserved by dierent regions of the brain [5,18,32]. 1.2. The present study The main objective of the present study was to investigate the nature of cognitive impairment in young fragile X males (full mutation) on a range of attention and executive functioning tasks: selective attention, divided attention, sustained attention, inhibition and response organisation. Young males with Down's syndrome (Trisomy 21) matched with the fragile X group for chronological age and mental age were chosen as the comparison group. This syndrome group was especially included because Down's syndrome also presents a genetic chromosome abnormality, but distinct from fragile X syndrome, and both syndrome groups represent two of the most common causes of mental impairment for which aetiology is known [10]. Two normal functioning control groups matched closely on mental age to the fragile X males were also included. The ®rst group was selected to match the fragile X group on a similar behaviour pro®le of hyperactivity and attention problems but with no known genetic condition. This group was included so that dierences in performance between the two groups would establish a cognitive pro®le of attention de®cit speci®c to the fragile X syndrome or similar to those with hyperactive behaviour without fragile X syndrome. The second control group, also matched on mental age, was selected from boys with above average attentional ability.
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
2. Method 2.1. Participants The present study involved four groups of participants: 1. 25 boys with fragile X syndrome (age range 8.06± 15.09 years; mean age 10.88 years) recruited from the UK parent support group. Diagnosis of FXS was established by DNA testing which con®rmed the presence of FMR-1 full mutation; 2. 25 boys with Down's syndrome (age range 7.04± 15.09 years; mean age 11.17 years) recruited from a UK parent support group. Diagnosis of Down's syndrome had previously been established by cytogenetic testing which con®rmed a karyotype with a free trisomy 21; 3. 25 control boys matched on mental age and therefore biologically younger than the fragile X group (age range 5.02±10.09; mean age 7.58 years). These boys were matched to the fragile X group on degree of hyperactivity and poor attention abilities as rated by their teachers on the Comprehensive Teacher Rating Scale (ACTeRS) [43]; and therefore were referred to as the ``poor attention'' group; and 4. 25 control boys also matched on mental age and therefore biologically younger than the fragile X group (age range 5.02±10.09 years; mean age 7.97 years) but rated by teachers as demonstrating average-good attention ability and average-low hyperactivity and were referred to as the ``good attention'' group. All boys with fragile X syndrome and Down's syndrome were receiving special education and none were living in institutional settings. None of the children in the control groups had a history of speci®c learning disability. Furthermore, none of the children in any of the groups had sensory impairments including hearing de®cits and decreased visual acuity. Finally, none of the children were on stimulant drugs such as methylphenidate (Ritalin) for hyperactivity which might in¯uence cognitive performance. Matching on verbal mental age (VMA) was done using the British Picture Vocabulary Scale (BPVS) Short Form [11]. 2.2. Neuropsychological measures of attention A comprehensive battery of attention tasks speci®cally designed for use with children was selected to assess the dierent cognitive aspects of attention: selective, divided and sustained attention, and executive functioning. The battery of tasks employed consists of recently developed standardised neuropsychological tasks of attentional functioning; the Test of Everyday
1263
Attention for children (TEAch) [20], adapted from the adult Test of Everyday Attention [35] and the computerised Wilding Attention Test for Children (WATT) [47]. The computer-based tasks were presented on a Twinhead, Slimnote±610TX laptop computer with a coloured visual display screen 23 cm wide 18.5 cm high. These tasks were performed using an externally attached mouse. 2.2.1. Selective attention measures The Visearch task (WATT) was a computerised visual search task where each child was presented with a picture of a forest consisting of trees and other shapes on a computer screen. Each child was required to search for a certain type of shape (target) to reveal a hidden monster, by clicking onto the target with a mouse. There were other similar and dissimilar shapes on the screen that served as distractors, and had to be ignored. There were two trials with a maximum of 20 targets to be located in each. In the ®rst trial, each child was told; ``You need to ®nd the king monster, by ®rst ®nding all the other hidden monsters, who are hiding behind the black eggs standing upright'' (a vertical, black ellipse). In the second trial, the hidden monsters had to be found ``behind the pink eggs lying on their sides'' (a horizontal, pinkish-brown ellipse). An example of each shape and a practice run was given on the screen before each trial began. The game terminated when either all the monsters had been found or when a maximum of 50 clicks had been reached. The total number of correct targets found (maximum 20), the time taken to complete the task, the number of false alarms and the mean distance moved between detected targets was recorded by the computer program for each trial. 2.2.2. Divided attention measures The Visearch dual task (WATT) was also a computerised task based on the visearch task (selective attention). Using the same format as Visearch (see above), this divided attention condition required each child to search for two dierent types of targets (shapes) alternately, to reveal a hidden monster by clicking on the appropriate target with the mouse. There were other shapes on the screen that served as distractors and had to be ignored. Each child was told; ``You need to ®nd the king monster, by ®rst ®nding all the other hidden monsters. However, they are now hiding behind two shapes; the black eggs standing upright'' (a vertical, black ellipse from the ®rst trial in Visearch). ``and the pink eggs lying on their sides'' (a horizontal, pinkishbrown ellipse from the second trial in Visearch); ``However, you must ®rst click on a black egg and then on a pink egg and keep clicking on one and then the other until you ®nd the king monster. If you click on two black eggs or two pink eggs then you won't
1264
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
®nd a monster. It must be one black then one pink egg''. An example of each shape, and a practice run was given on the screen before the trial began. There was one trial only, with a maximum of 20 targets to be found (10 for the ®rst shape and 10 for the second). The game terminated when either all the monsters had been found or when a maximum of 50 clicks had been reached. The total number of correct targets found (maximum 20), the time taken to complete the task, the number of false alarms and the mean distance moved between detected targets was recorded by the computer program. 2.2.3. Sustained attention measures The Vigilan task (WATT) was a computerised vigilance task. Using the same screen layout as Visearch (see above), each child was required to watch for a monster appearing randomly on the screen. They were each told, ``In a minute you will see a monster pop out of one of these black eggs. Every time you see a monster appear, you have to quickly move your mouse and click onto it before it disappears. You have to keep watching for the monsters, until the game says you have ®nished''. A total of 20 targets were presented one at a time, at irregular intervals. There was one trial only, with a pre-demonstration. The number of targets detected (maximum 20), the mean time (s) taken to click onto a target, the number of false alarms and the total distance wandered while awaiting an appearance of a target (random wandering) were recorded by the computer program. 2.2.4. Executive function measures The Walk task (TEA-Ch), similar to the Stop task, primarily assesses the ability to withhold a response and to delay a response. In this task, each child was required to remain vigilant as they marked (placed a dot) along an A4 sheet of printed footprints (presented in columns of 15) with each ``bleep'' heard from an audio tape recording. If the same ``bleep'' was heard, followed by a sound similar to a man screaming, each child had to stop, and move on to the next column. This task was performed with a felt tip pen on two A4 sized sheets with ten columns of 15 footprints in each (20 columns in total). The ®rst two practice trials were demonstrated by the experimenter, and each child was asked to do the same on the next two practice trials, and for the actual trials that followed. The task speeded up incrementally as it progressed. There were four practice trials followed by 20 actual trials (20 columns) with a maximum possible score of 20 correct responses. Same±Opposite task (TEA-Ch) is divided into two parts where each child is ®rst required to read out a series of number ``ones'' and ``twos'' as they appeared on a sheet of A4 paper e.g., ``1, 2, 2, 1 . . .'' (non-com-
peting baseline condition). In the second part of this task, they were asked to read the opposite of each number that appeared, so that each one would be read as ``two'', and each two would be read as ``one'' (competing condition). Each child was required to stay on each item until the correct response was given. Both parts of this task were ®rst demonstrated by the experimenter. After the ®rst part had been demonstrated, they were asked to do the same on a practice trial. The ®rst part of the task was then administered. The experimenter then demonstrated the second part, and each child was asked to do the same on the practice trial, and for the trial that followed. For each part of this task, there were two practice trials and two actual trials (a total of four actual trials). The time taken to complete each series of number reading was recorded. All children were tested individually, in a quiet private room. The battery took approximately 1 h to complete and was administered in one session. Before each testing session it was established through parents and teachers whether each child had previous ``mouse'' experience for the computer tasks. All children in each group had some previous mouse experience either at school or at home. All tests were carried out in the same order for each participant. Reaction times for performance on paper and pencil tasks were recorded by a stopwatch to the nearest second. For the computer-based tasks all reaction times were recorded by the program. 3. Results 3.1. Intellectual functioning Verbal mental age diered signi®cantly between the groups [F(3,96)=5.14; P < 0.002], with the Down's syndrome group demonstrating a signi®cantly lower mental age than the good attention control group (P < 0.003; post-hoc Student Newman±Keuls test). There were no signi®cant dierences between the fragile X group and the two control groups on mental age (Table 1). 3.2. Neuropsychological measures 3.2.1. Selective attention Scores from the two trials of the visearch task were ®rst averaged to obtain a single score for each measure. The time taken (s) to complete the task was divided by the total number of targets found (maximum 20) in order to obtain a time per target for analysis (s/n targets). The correct number of targets found, false alarms (which measured the number of errors made and impulsive clicking) and the mean distance moved between detected targets (which measured search eciency independent of mouse skill) were also
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
1265
Table 1 Comparison of participants with fragile X syndrome (FXS), Down's syndrome (DS), poor attention control group (PAC) and good attention control group (GAC) matched on mental age Neuropsychological measures Chronological age (CA) (SD) Mental age (MA) (SD) Selective attention Correct targets Time per target search Distance moved False alarms Divided attention (proportional dierence) Correct targets Time per target search Distance moved False alarms Sustained attention Targets found Mean time taken False alarms Executive functioning Inhibition Response±organisation
Fragile X syndrome (FXS)
Down's syndrome (DS)
Poor attention control group (PAC)
Good attention control group (GAC)
Signi®cance test
Post-hoc analyses
10.88 (2.12) 6.77 (1.60)
11.17 (2.53) 6.09 (1.51)
7.58 (1.58) 6.96 (1.38)
7.97 (1.66) 7.77 (1.61)
11.18 (6.16) 10.35 (6.52) 42.03 (17.36) 36.72 (10.70)
16.62 (3.01) 8.41 (2.28) 34.89 (3.42) 26.88 (9.14)
19.48 (0.92) 5.29 (1.84) 29.04 (5.53) 11.84 (7.36)
19.52 (0.81) 3.86 (1.39) 25.08 (4.79) 4.40 (2.11)
P < 0.0001
FXS < DS < PAC, GAC
P < 0.0001
FXS, DS > PAC, GAC
P < 0.0001
FXS, DS > GAC; FXS > PAC
P < 0.0001
FXS > DS > PAC, GAC; PAC > GAC
1.49 (0.72) 1.42 (0.63) 1.93 (0.61) 1.26 (3.44)
1.41 (0.24) 1.37 (0.24) 1.52 (0.37) 1.58 (0.54)
1.19 (0.15) 1.25 (0.20) 1.30 (0.25) 1.33 (0.63)
1.10 (0.12) 1.27 (0.21) 1.23 (0.22) 1.32 (0.81)
P < 0.0001
FXS,DS < PAC, GAC
18.28 (1.51) 4.15 (1.91) 29.88 (24.95)
18.08 (1.26) 3.90 (0.78) 10.32 (7.36)
18.76 (1.13) 3.60 (0.90) 6.16 (4.62)
19.12 (0.78) 2.94 (0.82) 2.84 (1.62)
NS
8.60 (5.25) 12.88 (3.99)
13.48 (4.91) 24.48 (4.45)
12.36 (4.06) 10.48 (4.92)
16.04 (2.15) 6.20 (2.45)
analysed. A multivariate analysis of variance (MANOVA) was used with group as the independent variable on these four scores. There was a signi®cant main eect of group (F = 46.15; df=3; P < 0.0001), and a signi®cant group by score interaction (F = 66.26; df=3; P < 0.0001) indicating that the pattern of performance across the groups was dierent for dierent scores. Separate ANOVAs across the dierent selective attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the main and interaction eects in detail. However, to reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an
NS P < 0.0001
FXS > DS, PAC, GAC
NS
P < 0.004
FXS, DS, PAC > GAC
P < 0.0001
FXS > all groups; DS > GAC
P < 0.0001
FXS < all groups; DS, PAC < GAC
P < 0.0001
DS > FXS > PAC, GAC; PAC > GAC
alpha level of 0.05/4=0.01 were considered statistically signi®cant. Performance diered signi®cantly between the groups on all selective attention scores: number of correct targets found [F(3,96)=34.13; df=3; P < 0.0001], time per target [F(3,96)=16.33; df=3; P < 0.0001], False alarms [F(3,96)=82.69; df=3; P < 0.0001] and mean distance moved [F(3,96)=14.83; df=3; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group found signi®cantly fewer correct targets and made signi®cantly more false alarms than the Down's syndrome group and both syndrome groups found signi®cantly fewer correct targets and
1266
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
made more false alarms than the two control groups. For time per target and mean distance moved between detected targets, there was no signi®cant dierence between the fragile X syndrome and the Down's syndrome group. Both groups, however, took signi®cantly longer to search for each target in comparison to the control groups. They also moved signi®cantly more distance between detected targets in comparison with the good attention control group and in comparison with the poor attention control group for the fragile X group only. Across the two control groups, the poor attention control group made signi®cantly more false alarms than the good attention control group. There were no other signi®cant dierences between the two control groups. 3.2.2. Divided attention In order to examine a possible divided attention deficit, scores from the visearch dual (divided) attention condition Ð correct targets found, time per target, mean distance moved between detected targets and false alarms Ð were divided by their counterpart scores from the single (selective) attention condition.1 The divided scores measure the relative dierence in performance between the selective and divided attention condition. A multivariate analysis of variance (MANOVA) was used on these scores with group as the independent variable. There was a signi®cant main eect of group (F = 6.56; df=3; P < 0.0001), but no signi®cant group by task score interaction (F = 2.28; df=3; P > 0.05; NS). Separate ANOVA's across the dierent divided attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the main eects in detail. To reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an alpha level of 0.05/4=0.01 were considered statistically signi®cant. There were no signi®cant group dierences for time per target [F(3,96)=1.19; P > 0.314] and false alarms [F(3,96)=1.41; P > 0.244] suggesting that there was no signi®cant dierence in the proportional increase in search ratio or in false alarms from the selective to the divided attention condition between the groups. However, there were signi®cant group dierences for: mean distance moved [F(3,96)=15.76; P < 0.0001] and correct targets found [F(3,96)=12.29; P < 0.0001]. A post hoc Newman±Keuls test revealed that both syndrome groups demonstrated a signi®cantly greater decrease in 1 Except for correct targets found, where the selective attention score was divided by the divided attention score since this score re¯ects a positive measure (i.e. the greater the score, the better the performance) and therefore expected to decrease from the selective to the divided attention condition.
the number of correct targets found than the two control groups. The fragile X group further displayed a signi®cantly greater increase in the mean distance moved between detected targets than the Down's syndrome group and the two control groups. There were no other signi®cant dierences between the groups. 3.2.3. Sustained attention The three Vigilan scores Ð targets detected, mean time taken to respond and false alarms were analysed with multivariate analysis of variance (MANOVA) with group as the independent variable. Two of the scores were ®rst transformed before any analyses to reduce their skewness. A natural log transformation was made for the mean time taken and a square root transformation for the number of false alarms. There was a signi®cant main eect of group (F = 14.78; df=3; P < 0.0001), and a signi®cant group by task score interaction (F = 16.37; df=3; P < 0.0001) indicating that the pattern of performance across the groups was dierent for dierent scores. Separate ANOVA's on the dierent sustained attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the eects in detail. However, to reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an alpha level of 0.05/3=0.02 were considered statistically signi®cant. There were no signi®cant group dierences for targets detected [F(3,96)=3.83; P > 0.03; NS] although there was a trend with the fragile X group detecting fewer targets than the other three groups. However, performance between the groups diered signi®cantly for: mean time taken to respond to targets [F(3,96)=4.80; P < 0.004] and for false alarms [F(3, 96)=22.66; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group displayed signi®cantly more false alarms in comparison with all three groups. This was followed by the Down's syndrome group who displayed signi®cantly more false alarms in comparison with the good attention control group only. For mean time taken to respond, the fragile X group, Down's syndrome group and poor attention control group did not dier signi®cantly but all displayed greater mean times to respond to a target in comparison with the good attention control group. 3.2.4. Executive functioning The total number of correct inhibitions made on the Walk task (maximum 20) were selected for analysis. For the Same±Opposite task, the time taken to complete the two ``same'' conditions (non-competing baseline conditions Ð reading the numbers one and two as presented) of the task were added together; and the time taken to complete the ``opposite'' conditions (competing conditions Ð reading the opposite of the
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
numbers presented) were added together. The total sum time for the ``same'' condition was then subtracted from the total sum time of the ``opposite'' condition to obtain a measure of response±organisation time for analysis ([opptime1+oppti2 1 2 me ]ÿ[sametime +sametime ]). Unfortunately, only 17 out of 25 (68%) fragile X boys in comparison to 22 out of 25 (88%) Down's syndrome boys were able to complete the opposite condition of the Same±Opposite task. As a result, in order to retain equal n, 17 participants were randomly selected from the Down's syndrome group and each of the two control groups on this task for purpose of analysis.2 Two univariate analyses of variance (ANOVAs) were performed, one for each variable, with group as the independent variable. The ANOVA revealed a signi®cant group eect for the Walk task [F(3,96)=13.12; P < 0.0001] and a signi®cant group eect for the Same±Opposite task [F(3,96)=92.77; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group demonstrated signi®cantly poorer inhibition on the Walk task than all the other three groups. This was followed by the Down's syndrome group and the poor attention control group who both displayed signi®cantly poorer inhibition than the good attention control group. In contrast, the Down's syndrome group demonstrated signi®cantly more diculty in organising responses on the Same±Opposite task than all the other three groups. This was followed by the fragile X group who displayed signi®cantly greater dif®culty in organising their responses in comparison to the two control groups. Finally, the poor attention control group displayed signi®cantly greater diculty in organising their responses than the good attention control group.
4. Discussion This study examined dierent components of attention skills in a sample of young fragile X males, a learning disabled comparison group and two normal control groups. The ®ndings of the present study suggest that fragile X males display a pro®le of attention that is signi®cantly dierent from that seen in Down's syndrome males and the two normal control groups. In comparison to Down's syndrome males, fragile X males performed signi®cantly worse on aspects of selective attention, divided attention, sustained attention and inhibition. In contrast, they per2 The pattern of results did not change when these cases were not excluded from the Down's syndrome group and each of the four control groups.
1267
formed better than Down's syndrome males on the response±organisation task (executive functioning). In comparison to the poor attention and good attention normal control groups, fragile X males performed signi®cantly worse than both control groups on all attention measures on which dierences between groups occurred, except response time in the sustained attention task where they were not signi®cantly slower than the PAC group. In terms of a selective attention de®cit, poor performance by fragile X males compared with Down's syndrome males appeared to be speci®c to selecting relevant information rather than speed of search and distance moved in search. This impairment is re¯ected in the signi®cantly smaller number of targets detected and the high number of false alarms produced by this group which indicates that they had greater diculty either in discriminating between target and distractor or in shifting attention and inhibiting impulsive responding to incorrect targets. The impulsive behaviour of fragile X males is also evident on tasks of sustained attention (shown by the high false alarm rate) and inhibition in the Walk task and is in accordance with their behavioural phenotype of reported hyperactivity and impulsivity [4,14]. In terms of a divided attention de®cit, measured as the proportional dierence in performance between the selective and divided attention condition, fragile X males displayed impaired performance, compared with Down's syndrome, in the mean distance moved across the computer screen while locating correct targets. An impairment on this measure suggests fragile X males had greater diculty in organising search at the same time as shifting attention from one type of target search to another, therefore displaying a de®cit in their ability to divide or switch attention. In contrast to a de®cit in selective and divided attention, fragile X males performed at a comparable level to Down's syndrome males and the two control groups in the speed of detecting targets in the sustained attention task, although the two learning disabled groups and the poor attention control group were slower than the good attention control group in responding to targets. This ®nding is consistent with studies that suggest those with learning disabilities or hyperactive behaviour do not demonstrate a sustained attention de®cit but indicate slower information-processing skills in comparison to controls [28,44]. This is also consistent with recent ®ndings of adult fragile X males displaying relatively intact performance on a task of sustained attention, but an impairment on a task that required attentional shift from one stimulus to another in comparison with learning disabled controls [8]. On a task of inhibition, fragile X males demonstrated a signi®cant impairment in the ability to inhibit or delay responding in comparison to the other three
1268
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
groups. This ®nding is in accordance with the overall performance of fragile X males who consistently show impulsive behaviour on all attention tasks in this study (except the response organisation task). This present ®nding further adds to the growing body of literature that indicates executive dysfunctions to be the most persistent de®cit in fragile X syndrome both males and females [22,25,39]. Interestingly, the poor attention controls also show a consistent de®cit in the ability to inhibit impulsive responding on nearly all tasks in comparison to the good attention controls. They produced more false alarms in both selective and sustained attention and worse performance in the Walk task and response organisation task, suggesting a similar but less profound pattern of impulsive behaviour to the fragile X males. On a task of response±organisation, however, the Down's syndrome males were more impaired than fragile X males. This is somewhat surprising given that a consistent de®cit in organising and planning is frequently cited as a core weakness in individuals with fragile X syndrome [22,39]. One explanation might be that the present task primarily measured speed at which a novel stimulus±response linkage could be processed and errors corrected, rather than ability to cope with sudden unexpected changes in response demands, as in the walk task. Thus, greater impairment on this task might re¯ect a de®cit in speed of information processing rather than a de®cit in response organisation. Moreover, fewer fragile X males (68%) than Down's syndrome males (88%) were able to complete this task, providing further support for the hypothesis that at least some fragile X males have greater diculty in planning and organising information than Down's syndrome males. Across the two normal control groups, the poor attention controls displayed a signi®cant de®cit on this task in comparison with the good attention controls, again suggesting a similar but less profound pattern of de®cit to fragile X males. The results from the poor attention controls are in line with reports that ascribe those with hyperactive behaviour or AD/ HD to a de®cit in higher-processes such as executive function and behavioural inhibition [1,2]. Overall, both the learning-disabled groups were impaired in many respects compared with the good attention control group and poor attention on measures of selective attention, divided attention and response organisation. Hence there are aspects of attention that seem to be vulnerable in the learningdisabled groups, particularly selective attention, indicating a possible limited general attentional capacity that is allocated ineciently across complex task demands [15]. This ®nding is in accordance with studies on those with learning disabilities displaying a smaller pool of attentional capacity on tasks of divided attention [23,28].
However, the fragile X group demonstrated additional problems speci®c to this disorder on measures of selective attention, divided attention and inhibition, with a particular impairment in the ability to plan and organise search, shift attention from one concept to another, delay responding and inhibit task-irrelevant responses. These de®cits are typically associated with executive dysfunction or higher control processes of attention; and are considered to be core characteristics of frontal lobe functioning [41]. For example, the prefrontal cortex and its connections are known to underlie response inhibition. The dorsolateral prefrontal cortex appears to be involved with adjusting attention and plays a crucial role in the organisation of visually guided behaviour during visual search [3]. The orbito-frontal cortex also plays a part in inhibiting behavioural response to external contingencies. The present ®ndings suggest that the pattern of attention de®cits in fragile X males may indicate a frontal lobe dysfunction consistent with the structural abnormalities reported in the prefrontal regions in those with fragile X syndrome [33]. These ®ndings further suggest that this pattern of attention de®cits in fragile X males, more severely aected in this group than in Down's syndrome males or the poor attention normal controls, is speci®c to their syndrome and adds to the growing literature on distinct neurodevelopmental phenotypes. Additional bene®ts also accrue from such data by providing information to professionals that might guide and assist the timing of early intervention and treatment programmes that could maximise developmental potential and improve prognosis.
Acknowledgements This work was supported by Mental Health Foundation research grant R593/3 awarded to Dr Kim Cornish. The authors wish to thank all participating families for their support in this research.
References [1] Barkley RA. Behavioural inhibition, sustained attention and executive functions: toward a more comprehensive theory. Developmental and Behavioural Pediatrics 1997;18:271±9. [2] Barkley RA. Attention de®cit/hyperactivity disorder, self-regulation and time: toward a more comprehensive theory. Developmental and Behavioural Pediatrics 1997;18:271±9. [3] Barone P, Joseph JP. Role of the dorsolateral prefrontal cortex in organising visually guided behaviour. Brain, Behaviour and Evolution 1989;33:132±5. [4] Baumgardner T, Reiss AL, Freund LS, Abrams MT. Speci®cations of the neurobehavioural associations in males with fragile X syndrome. Pediatrics 1995;95:744±52. [5] Corbetta M, Miezin FM, Dobmeyer S, Shulman GL, Peterson
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
[6] [7] [8]
[9] [10] [11] [12] [13] [14] [15] [16] [17]
[18]
[19]
[20] [21]
[22]
[23] [24] [25]
E. Selective and divided attention during visual discrimination of colour, shape and speed: functional anatomy by positron emission tomography. Journal of Neuroscience 1991;11:2383± 402. Corbetta M, Miezin FM, Shulman GL, Peterson SE. A PET study of visuospatial attention. Journal of Neuroscience 1993;13:1202±26. Cornish KM, Munir F, Cross G. Spatial cognition in males with fragile X syndrome: evidence for a neuropsychological phenotype. Cortex, in press. Cornish KM, Munir F, Cross G. Dierential impact of the FMR-1 full mutation on memory and attention functioning: evidence for a molecular-neuropsychological phenotype? Brain, submitted. Crowe SF, Hay A. Neuropsychological dimensions of the fragile X syndrome: support for a non-dominant hemisphere dysfunction hypothesis. Neuropsychologia 1990;28:9±16. Davids JR, Hagerman RJ, Eilert RE. Orthopaedic aspects of fragile X syndrome. Journal of Bone Joint Surgery 1990;72:889± 96. Dunn L, Dunn P, Whetton C, Pintilie D. British picture vocabulary scale. Windsor: NFER, 1982. Dykens EM, Hoddap RM, Leckman JF. Strengths and weaknesses in the intellectual functioning of males with fragile X syndrome. American Journal of Medical Genetics 1987;28:13±5. Freund LS, Reiss AL. Cognitive pro®les associated with the fragile X syndrome in males and females. American Journal of Medical Genetics 1991;38:542±7. Hagerman RA. Fragile X syndrome. Current Problems in Pediatrics 1987;17:627±74. Kahneman D. Attention and eort. Englewood Clis, NJ: Prentice-Hall, 1973. Kaufman PM, Leckman JF, Ort SI. Delayed response performance in males with fragile X syndrome. Journal of Clinical Experimental Neuropsychology 1990;12:69. Kemper MB, Hagerman RJ, Ahmad RS, Mariner R. Cognitive pro®les and the spectrum of clinical manifestations in heterozygous fragile X females. American Journal of Medical Genetics 1986;23:139±56. Konishi S, Nakajima K, Uchida I, Kameyama M, Nakahara K, Sekihara K, Miyashita Y. Transient activation of inferior prefrontal cortex during cognitive set shifting. Nature Neuroscience 1998;1:80±4. Lachiewicz AM, Spiridigliozzi A, Guillion CM, Ransford SN, Rao K. Aberrant behaviours of young boys with fragile X syndrome. American Journal on Mental Retardation 1994;98:567± 79. Manly T, Robertson IH, Anderson V, Nimmo-Smith I. The test of everyday attention for children: TEA-Ch. Bury St Edmunds: Tames Valley Test Company, 1999. Mazzocco MMM, Hagerman RJ, Cronister-Silverman A, Pennington B. Speci®c frontal lobe de®cits among women with the fragile X gene. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31:1141±8. Mazzocco MMM, Pennington BF, Hagerman RJ. The neurocognitive phenotype of female carriers of fragile X: additional evidence for speci®city. Developmental and Behavioural Pediatrics 1993;14:328±34. Merrill EC, Peacock M. Allocation of attention and task diculty. Journal of Mental Retardation 1994;98:588±93. Morton JE, Bundey S, Webb T, MacDonald F, Rindl PM, Bullock S. Fragile X syndrome is less common than previously estimated. Journal of Medical Genetics 1997;34:1±5. Munir F, Cornish KM, Wilding J. Patterns of attention de®cit in fragile X syndrome: a neuropsychological perspective. In: Proceedings from the Sixth International Fragile X Syndrome Conference, Ashville, USA, 1998.
1269
[26] Munir F, Cornish KM, Wilding J. Nature of the working memory de®cit in fragile X syndrome. Brain and Cognition, submitted. [27] Oberle L, Rousseau F, Heitz D, Kretz C, Devys D, Hanaver A, Boue J, Bertheas MF, Mandel JL. Instability of a 550-base pair DNA segment and abnormal methylation in fragile X syndrome. Science 1991;252:1097±102. [28] Pearson DA, Yaee LS, Loveland KA, Lewis KR. Comparison of sustained and selective attention in children who have mental retardation with and without Attention De®cit Disorder. American Journal of Mental Retardation 1996;100:592±607. [29] Pieretti M, Zhang F, Fu Y-H, Warren ST, Oostra BA, Caskey CT, Nelson DL. Absence of expression of the FMR1 gene in the fragile X syndrome. Cell 1991;66:817±22. [30] Posner MI. Interaction of arousal and selection in the posterior attention network. In: Baddeley A, Weiskrantz L, editors. Attention: selection, awareness and control Ð a tribute to Donald Broadbent. Oxford Science Publications, 1993. p. 390± 405. [31] Posner MI, Dehaene S. Attentional networks. Trends in Neurosciences 1994;17:75±9. [32] Posner MI, Peterson SE. The attention system of the human brain. Annual Review of Neuroscience 1990;13:25±42. [33] Reiss AL, Abrams MT, Greenlaw R, Freund L, Denckla MB. Neurodevelopmental eects of the FMR1 full mutation in humans. Nature medicine 1995;1:159±67. [34] Riddle JE, Cheema A, Sobesky WE, Gardner SC, Taylor AC, Pennington BF, Hagerman RJ. Phenotypic involvement in females with the FMR-1 gene mutation. American Journal of Mental Retardation 1998;102:590±601. [35] Robertson IH, Ward T, Ridgeway V, Nimmo-Smith I. The test of everyday attention. University of Cambridge: MRC Applied Psychology Unit, 1994. [36] Schneider W, Shirin RM. Controlled and automatic human processing: I. Detection, search and attention. Psychological Review 1977;92:424±8. [37] Shapiro MB, Murphy DGM, Hagerman RJ, Azari NP, Alexander GE, Miezejeski CM, Hinton VJ, Horwitz B, Haxby JV, Kumar A, White B, Grady CL. Adult fragile X syndrome: neuropsychology, brain anatomy and metabolism. American Journal of Medical Genetics 1995;60:480±93. [38] Shirin RM, Schneider W. Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychological Review 1977;84:127±90. [39] Sobesky WE, Pennington BF, Porter D, Hull CE, Hagerman RJ. Emotional and neurocognitive de®cits in fragile X. American Journal of Medical Genetics 1994;51:378±84. [40] Sobesky WE, Taylor AK, Pennington BF, Riddle J, Hagerman RJ. Molecular/clinical correlations in females with fragile X. In: Presented at the 7th International Workshop on fragile X and X-linked Mental retardation, Tromso, Norway, 1995. [41] Stuss DT, Benson DF. The frontal lobes. New York: Raven Press, 1986. [42] Turk J. Fragile X syndrome and attentional de®cits. Journal of Applied Research in Intellectual Disabilities 1998;11:175±91. [43] Ullmann RK, Sleator EK, Sprague RL. A new rating scale for diagnosis and monitoring of ADD children. Psychopharmacology Bulletin 1984;21:339±44. [44] Van der Meere J, Wekking E, Sergeant J. Sustained attention in pervasive hyperactivity. Journal of Child Psychology and Psychiatry 1991;32:275±84. [45] Verkerk AJMH, Pieretti M, Sutclie JS, Fu Y-H, Kuhl DPA, Pizzuti A, Reiner O, Richards S, Victoria MF, Zhang F, Eussen BE, Van Ommen G-JB, Blonden LAJ, Riggins GJ, Chestain JL, Kurst CB, Galjaard H, Caskey CT, Nelson DL, Oostra BA, Warren ST. Identi®cation of a gene (FMR-1)
1270
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell 1991;65:905±14. [46] Welsh MC, Pennington BF. Assessing frontal lobe functioning in children: views from developmental psychology. Developmental Psychology 1988;4:199±230.
[47] Wilding J. Wilding attention tests. University of London: Department of Psychology, Royal Holloway, 1997. [48] Yu S, Pritchard M, Kremer E, Lynch M, Nancarrow J, Baker E, Holmen K, Mulley JC, Warren ST, Schlessinger D. Fragile X genotype characterised by an unstable region of DNA, Science 1991;252:1179±81.
www.elsevier.com/locate/neuropsychologia
A neuropsychological pro®le of attention de®cits in young males with fragile X syndrome F. Munir a, K.M. Cornish a,*, J. Wilding b a
Section of Developmental Psychiatry, Division of Psychiatry, E Floor, South Block, Queens Medical Centre, University of Nottingham, Nottingham NG7 2UH, UK b Department of Psychology, Royal Holloway, University of London, Egham, Surrey, UK Received 20 August 1999; received in revised form 27 January 2000; accepted 2 February 2000
Abstract Dierent processes of attention were examined in a group of 25 fragile X boys with FMR-1 full mutation and compared with three control groups: a learning disabled comparison group comprising 25 boys with Down's syndrome, matched to the fragile X boys on verbal mental age; and 50 mainstream school boys (controls) matched to the fragile X boys on verbal mental age. The controls were further divided into those matched on ``poor attention'' to the fragile X boys and a ``good'' attention group, as rated by the ACTeRS questionnaire. Four categories of attention tasks were employed: selective attention, divided attention, sustained attention and executive functioning. The main ®ndings of the study indicate that fragile X boys display an attention de®cit at higher levels of attention function/executive functioning and that this pro®le is dierent from the pro®le identi®ed in Down's syndrome boys and more extreme than the pro®le identi®ed in the poor attention control group. These ®ndings are discussed in the context of functional neuroimaging and brain-behaviour correlates in fragile X syndrome. 7 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction Fragile X syndrome is the most common inherited genetic condition causing developmental and learning disabilities [34]. Recent evidence suggests that fragile X syndrome aects approximately 1:4000 males and 1:6000 females [24], and is associated with the dysfunction of a gene Ð known as the fragile X mental retardation 1 (FMR-1) gene Ð on the X chromosome related to the fragile site Xq27.3. This site involves the abnormal ampli®cation of a DNA segment containing a trinucleotide sequence of CGG repeats [27,45,48]. Between six and 58 CGG repeats are typically observed in the general population. In individuals who have the fragile X mutation, an expansion or ampli®cation occurs in this segment of
* Corresponding author. Tel.: +44-115-978-1918. E-mail address: [email protected] (K.M. Cornish).
their DNA from 59 to up to several hundred CGG repeats. Those who have between 59 and 200 CGG repeats have a pre-mutation and are known as ``carriers'' but with no overt clinical eects. It appears that the pre-mutation is functionally silent with the non-methylated gene still being transcribed. Individuals with more than 200 CGG repeats however, have a full mutation whereby the gene becomes methylated at the fragile site Xq27.3 leading to a lack of mRNA and FMR-1 protein (FMRP) synthesis [29,45]. Males with the full mutation are invariably clinically aected. Females who carry the full mutation on the other hand, may be clinically unaected. It is the lack of FMRP that results in the clinical manifestation of the fragile X syndrome. While the speci®c function of FMRP is not known, it is strongly expressed in brain tissues [45] and may be responsible for an impairment or lack of structural development in brain functioning in aected fragile X males and females.
0028-3932/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 3 2 ( 0 0 ) 0 0 0 3 6 - 1
1262
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
1.1. Neuropsychological phenotype in fragile X males The neuropsychological-cognitive pro®le of fragile X boys (full mutation) appears to be more complex than originally thought. This directly impacts on the current understanding of the disorder in terms of brain development and function and the genetic factors underlying the disorder. Past studies suggested de®cits in short-term non-verbal memory, visual motor and spatial skills, with relative strengths in verbal labelling, verbal memory and comprehension [12,13,17] thus indicating cognitive de®cits that may be attributed to poor right-hemisphere functioning [9]. More recent comprehensive studies, however, suggest a pattern of de®cits that are task speci®c rather than global in nature. Spatial skills for example, are not globally impaired in fragile X males. Instead, the de®cit appears to be speci®c to those skills that require visuo-spatial orientation and constructional abilities, with visuo-perceptual skills relatively intact [7,8,12]. Similarly, verbal skills are not a global strength in fragile X males, with speci®c de®cits reported on tasks that require short-term memory for complex sequential information, and particular strengths on tasks that require short-term memory for simple, meaningful information [13,37]. Further evidence for a possible discrepancy in cognitive functioning suggests fragile X males display relatively intact skills for learning simple verbal and non-verbal tasks, but are greatly impaired on tasks which require the manipulation of internal representations [16]. Likewise, on tasks of working memory (the ability to temporarily hold information in mind while processing same or other information), fragile X males can temporarily maintain and recall simple meaningful information, but not abstract nonsequential information irrespective of the verbal or non-verbal working memory subsystem [26]. Current studies therefore suggest that fragile X males can perform relatively simple tasks but cannot manipulate more complex information regardless of its modality. It is not known why these speci®c cognitive skills are aected in fragile X males. One possibility may lie in the direction of higher control processes of attention such as the ability to modulate complex or novel information and plan and organise information in the mind. Indeed, neuroanatomical studies of full mutation fragile X males and females report abnormalities in both left and right parietal regions as well as in the prefrontal regions [37,40] and the caudate nucleus [33]. Signi®cantly, these structures appear to be involved in attention [5,6,30] and higher control processes of attention such as executive functioning [2] indicating that these mechanisms may be impaired in those with fragile X syndrome. These neuroanatomical ®ndings are in accord with neuropsychological ®ndings in fragile X females, where a consistent pattern of di-
culties in visual±spatial attention and executive functioning has been documented [21,22,24,39]. In contrast, comparatively few studies have taken a similar direction in terms of identifying a neuropsychological pro®le of the attention de®cit in fragile X males. This is somewhat surprising given that attentional problems, impulsive behaviour and hyperactivity are frequently cited as core features of the behaviour phenotype in fragile X males [4,14,19,42]. As a result, relatively little is known about the precise nature of the attention de®cits, although recent studies have indicated the importance of teasing apart cognitive aspects of attention. For example, selective attention refers to the ability to select relevant stimuli from irrelevant information, while divided attention refers to the ability to attend to more than one source of information simultaneously [36,38]. In contrast, maintaining attention over a period of time is referred to as sustained attention; and problem-solving, planning, inhibiting and manipulating mental representations of tasks and goals are seen as higher control processes of attention or executive functioning [31,46]. Neuroanatomic studies further suggest that these dierent aspects of attention are subserved by dierent regions of the brain [5,18,32]. 1.2. The present study The main objective of the present study was to investigate the nature of cognitive impairment in young fragile X males (full mutation) on a range of attention and executive functioning tasks: selective attention, divided attention, sustained attention, inhibition and response organisation. Young males with Down's syndrome (Trisomy 21) matched with the fragile X group for chronological age and mental age were chosen as the comparison group. This syndrome group was especially included because Down's syndrome also presents a genetic chromosome abnormality, but distinct from fragile X syndrome, and both syndrome groups represent two of the most common causes of mental impairment for which aetiology is known [10]. Two normal functioning control groups matched closely on mental age to the fragile X males were also included. The ®rst group was selected to match the fragile X group on a similar behaviour pro®le of hyperactivity and attention problems but with no known genetic condition. This group was included so that dierences in performance between the two groups would establish a cognitive pro®le of attention de®cit speci®c to the fragile X syndrome or similar to those with hyperactive behaviour without fragile X syndrome. The second control group, also matched on mental age, was selected from boys with above average attentional ability.
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
2. Method 2.1. Participants The present study involved four groups of participants: 1. 25 boys with fragile X syndrome (age range 8.06± 15.09 years; mean age 10.88 years) recruited from the UK parent support group. Diagnosis of FXS was established by DNA testing which con®rmed the presence of FMR-1 full mutation; 2. 25 boys with Down's syndrome (age range 7.04± 15.09 years; mean age 11.17 years) recruited from a UK parent support group. Diagnosis of Down's syndrome had previously been established by cytogenetic testing which con®rmed a karyotype with a free trisomy 21; 3. 25 control boys matched on mental age and therefore biologically younger than the fragile X group (age range 5.02±10.09; mean age 7.58 years). These boys were matched to the fragile X group on degree of hyperactivity and poor attention abilities as rated by their teachers on the Comprehensive Teacher Rating Scale (ACTeRS) [43]; and therefore were referred to as the ``poor attention'' group; and 4. 25 control boys also matched on mental age and therefore biologically younger than the fragile X group (age range 5.02±10.09 years; mean age 7.97 years) but rated by teachers as demonstrating average-good attention ability and average-low hyperactivity and were referred to as the ``good attention'' group. All boys with fragile X syndrome and Down's syndrome were receiving special education and none were living in institutional settings. None of the children in the control groups had a history of speci®c learning disability. Furthermore, none of the children in any of the groups had sensory impairments including hearing de®cits and decreased visual acuity. Finally, none of the children were on stimulant drugs such as methylphenidate (Ritalin) for hyperactivity which might in¯uence cognitive performance. Matching on verbal mental age (VMA) was done using the British Picture Vocabulary Scale (BPVS) Short Form [11]. 2.2. Neuropsychological measures of attention A comprehensive battery of attention tasks speci®cally designed for use with children was selected to assess the dierent cognitive aspects of attention: selective, divided and sustained attention, and executive functioning. The battery of tasks employed consists of recently developed standardised neuropsychological tasks of attentional functioning; the Test of Everyday
1263
Attention for children (TEAch) [20], adapted from the adult Test of Everyday Attention [35] and the computerised Wilding Attention Test for Children (WATT) [47]. The computer-based tasks were presented on a Twinhead, Slimnote±610TX laptop computer with a coloured visual display screen 23 cm wide 18.5 cm high. These tasks were performed using an externally attached mouse. 2.2.1. Selective attention measures The Visearch task (WATT) was a computerised visual search task where each child was presented with a picture of a forest consisting of trees and other shapes on a computer screen. Each child was required to search for a certain type of shape (target) to reveal a hidden monster, by clicking onto the target with a mouse. There were other similar and dissimilar shapes on the screen that served as distractors, and had to be ignored. There were two trials with a maximum of 20 targets to be located in each. In the ®rst trial, each child was told; ``You need to ®nd the king monster, by ®rst ®nding all the other hidden monsters, who are hiding behind the black eggs standing upright'' (a vertical, black ellipse). In the second trial, the hidden monsters had to be found ``behind the pink eggs lying on their sides'' (a horizontal, pinkish-brown ellipse). An example of each shape and a practice run was given on the screen before each trial began. The game terminated when either all the monsters had been found or when a maximum of 50 clicks had been reached. The total number of correct targets found (maximum 20), the time taken to complete the task, the number of false alarms and the mean distance moved between detected targets was recorded by the computer program for each trial. 2.2.2. Divided attention measures The Visearch dual task (WATT) was also a computerised task based on the visearch task (selective attention). Using the same format as Visearch (see above), this divided attention condition required each child to search for two dierent types of targets (shapes) alternately, to reveal a hidden monster by clicking on the appropriate target with the mouse. There were other shapes on the screen that served as distractors and had to be ignored. Each child was told; ``You need to ®nd the king monster, by ®rst ®nding all the other hidden monsters. However, they are now hiding behind two shapes; the black eggs standing upright'' (a vertical, black ellipse from the ®rst trial in Visearch). ``and the pink eggs lying on their sides'' (a horizontal, pinkishbrown ellipse from the second trial in Visearch); ``However, you must ®rst click on a black egg and then on a pink egg and keep clicking on one and then the other until you ®nd the king monster. If you click on two black eggs or two pink eggs then you won't
1264
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
®nd a monster. It must be one black then one pink egg''. An example of each shape, and a practice run was given on the screen before the trial began. There was one trial only, with a maximum of 20 targets to be found (10 for the ®rst shape and 10 for the second). The game terminated when either all the monsters had been found or when a maximum of 50 clicks had been reached. The total number of correct targets found (maximum 20), the time taken to complete the task, the number of false alarms and the mean distance moved between detected targets was recorded by the computer program. 2.2.3. Sustained attention measures The Vigilan task (WATT) was a computerised vigilance task. Using the same screen layout as Visearch (see above), each child was required to watch for a monster appearing randomly on the screen. They were each told, ``In a minute you will see a monster pop out of one of these black eggs. Every time you see a monster appear, you have to quickly move your mouse and click onto it before it disappears. You have to keep watching for the monsters, until the game says you have ®nished''. A total of 20 targets were presented one at a time, at irregular intervals. There was one trial only, with a pre-demonstration. The number of targets detected (maximum 20), the mean time (s) taken to click onto a target, the number of false alarms and the total distance wandered while awaiting an appearance of a target (random wandering) were recorded by the computer program. 2.2.4. Executive function measures The Walk task (TEA-Ch), similar to the Stop task, primarily assesses the ability to withhold a response and to delay a response. In this task, each child was required to remain vigilant as they marked (placed a dot) along an A4 sheet of printed footprints (presented in columns of 15) with each ``bleep'' heard from an audio tape recording. If the same ``bleep'' was heard, followed by a sound similar to a man screaming, each child had to stop, and move on to the next column. This task was performed with a felt tip pen on two A4 sized sheets with ten columns of 15 footprints in each (20 columns in total). The ®rst two practice trials were demonstrated by the experimenter, and each child was asked to do the same on the next two practice trials, and for the actual trials that followed. The task speeded up incrementally as it progressed. There were four practice trials followed by 20 actual trials (20 columns) with a maximum possible score of 20 correct responses. Same±Opposite task (TEA-Ch) is divided into two parts where each child is ®rst required to read out a series of number ``ones'' and ``twos'' as they appeared on a sheet of A4 paper e.g., ``1, 2, 2, 1 . . .'' (non-com-
peting baseline condition). In the second part of this task, they were asked to read the opposite of each number that appeared, so that each one would be read as ``two'', and each two would be read as ``one'' (competing condition). Each child was required to stay on each item until the correct response was given. Both parts of this task were ®rst demonstrated by the experimenter. After the ®rst part had been demonstrated, they were asked to do the same on a practice trial. The ®rst part of the task was then administered. The experimenter then demonstrated the second part, and each child was asked to do the same on the practice trial, and for the trial that followed. For each part of this task, there were two practice trials and two actual trials (a total of four actual trials). The time taken to complete each series of number reading was recorded. All children were tested individually, in a quiet private room. The battery took approximately 1 h to complete and was administered in one session. Before each testing session it was established through parents and teachers whether each child had previous ``mouse'' experience for the computer tasks. All children in each group had some previous mouse experience either at school or at home. All tests were carried out in the same order for each participant. Reaction times for performance on paper and pencil tasks were recorded by a stopwatch to the nearest second. For the computer-based tasks all reaction times were recorded by the program. 3. Results 3.1. Intellectual functioning Verbal mental age diered signi®cantly between the groups [F(3,96)=5.14; P < 0.002], with the Down's syndrome group demonstrating a signi®cantly lower mental age than the good attention control group (P < 0.003; post-hoc Student Newman±Keuls test). There were no signi®cant dierences between the fragile X group and the two control groups on mental age (Table 1). 3.2. Neuropsychological measures 3.2.1. Selective attention Scores from the two trials of the visearch task were ®rst averaged to obtain a single score for each measure. The time taken (s) to complete the task was divided by the total number of targets found (maximum 20) in order to obtain a time per target for analysis (s/n targets). The correct number of targets found, false alarms (which measured the number of errors made and impulsive clicking) and the mean distance moved between detected targets (which measured search eciency independent of mouse skill) were also
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
1265
Table 1 Comparison of participants with fragile X syndrome (FXS), Down's syndrome (DS), poor attention control group (PAC) and good attention control group (GAC) matched on mental age Neuropsychological measures Chronological age (CA) (SD) Mental age (MA) (SD) Selective attention Correct targets Time per target search Distance moved False alarms Divided attention (proportional dierence) Correct targets Time per target search Distance moved False alarms Sustained attention Targets found Mean time taken False alarms Executive functioning Inhibition Response±organisation
Fragile X syndrome (FXS)
Down's syndrome (DS)
Poor attention control group (PAC)
Good attention control group (GAC)
Signi®cance test
Post-hoc analyses
10.88 (2.12) 6.77 (1.60)
11.17 (2.53) 6.09 (1.51)
7.58 (1.58) 6.96 (1.38)
7.97 (1.66) 7.77 (1.61)
11.18 (6.16) 10.35 (6.52) 42.03 (17.36) 36.72 (10.70)
16.62 (3.01) 8.41 (2.28) 34.89 (3.42) 26.88 (9.14)
19.48 (0.92) 5.29 (1.84) 29.04 (5.53) 11.84 (7.36)
19.52 (0.81) 3.86 (1.39) 25.08 (4.79) 4.40 (2.11)
P < 0.0001
FXS < DS < PAC, GAC
P < 0.0001
FXS, DS > PAC, GAC
P < 0.0001
FXS, DS > GAC; FXS > PAC
P < 0.0001
FXS > DS > PAC, GAC; PAC > GAC
1.49 (0.72) 1.42 (0.63) 1.93 (0.61) 1.26 (3.44)
1.41 (0.24) 1.37 (0.24) 1.52 (0.37) 1.58 (0.54)
1.19 (0.15) 1.25 (0.20) 1.30 (0.25) 1.33 (0.63)
1.10 (0.12) 1.27 (0.21) 1.23 (0.22) 1.32 (0.81)
P < 0.0001
FXS,DS < PAC, GAC
18.28 (1.51) 4.15 (1.91) 29.88 (24.95)
18.08 (1.26) 3.90 (0.78) 10.32 (7.36)
18.76 (1.13) 3.60 (0.90) 6.16 (4.62)
19.12 (0.78) 2.94 (0.82) 2.84 (1.62)
NS
8.60 (5.25) 12.88 (3.99)
13.48 (4.91) 24.48 (4.45)
12.36 (4.06) 10.48 (4.92)
16.04 (2.15) 6.20 (2.45)
analysed. A multivariate analysis of variance (MANOVA) was used with group as the independent variable on these four scores. There was a signi®cant main eect of group (F = 46.15; df=3; P < 0.0001), and a signi®cant group by score interaction (F = 66.26; df=3; P < 0.0001) indicating that the pattern of performance across the groups was dierent for dierent scores. Separate ANOVAs across the dierent selective attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the main and interaction eects in detail. However, to reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an
NS P < 0.0001
FXS > DS, PAC, GAC
NS
P < 0.004
FXS, DS, PAC > GAC
P < 0.0001
FXS > all groups; DS > GAC
P < 0.0001
FXS < all groups; DS, PAC < GAC
P < 0.0001
DS > FXS > PAC, GAC; PAC > GAC
alpha level of 0.05/4=0.01 were considered statistically signi®cant. Performance diered signi®cantly between the groups on all selective attention scores: number of correct targets found [F(3,96)=34.13; df=3; P < 0.0001], time per target [F(3,96)=16.33; df=3; P < 0.0001], False alarms [F(3,96)=82.69; df=3; P < 0.0001] and mean distance moved [F(3,96)=14.83; df=3; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group found signi®cantly fewer correct targets and made signi®cantly more false alarms than the Down's syndrome group and both syndrome groups found signi®cantly fewer correct targets and
1266
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
made more false alarms than the two control groups. For time per target and mean distance moved between detected targets, there was no signi®cant dierence between the fragile X syndrome and the Down's syndrome group. Both groups, however, took signi®cantly longer to search for each target in comparison to the control groups. They also moved signi®cantly more distance between detected targets in comparison with the good attention control group and in comparison with the poor attention control group for the fragile X group only. Across the two control groups, the poor attention control group made signi®cantly more false alarms than the good attention control group. There were no other signi®cant dierences between the two control groups. 3.2.2. Divided attention In order to examine a possible divided attention deficit, scores from the visearch dual (divided) attention condition Ð correct targets found, time per target, mean distance moved between detected targets and false alarms Ð were divided by their counterpart scores from the single (selective) attention condition.1 The divided scores measure the relative dierence in performance between the selective and divided attention condition. A multivariate analysis of variance (MANOVA) was used on these scores with group as the independent variable. There was a signi®cant main eect of group (F = 6.56; df=3; P < 0.0001), but no signi®cant group by task score interaction (F = 2.28; df=3; P > 0.05; NS). Separate ANOVA's across the dierent divided attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the main eects in detail. To reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an alpha level of 0.05/4=0.01 were considered statistically signi®cant. There were no signi®cant group dierences for time per target [F(3,96)=1.19; P > 0.314] and false alarms [F(3,96)=1.41; P > 0.244] suggesting that there was no signi®cant dierence in the proportional increase in search ratio or in false alarms from the selective to the divided attention condition between the groups. However, there were signi®cant group dierences for: mean distance moved [F(3,96)=15.76; P < 0.0001] and correct targets found [F(3,96)=12.29; P < 0.0001]. A post hoc Newman±Keuls test revealed that both syndrome groups demonstrated a signi®cantly greater decrease in 1 Except for correct targets found, where the selective attention score was divided by the divided attention score since this score re¯ects a positive measure (i.e. the greater the score, the better the performance) and therefore expected to decrease from the selective to the divided attention condition.
the number of correct targets found than the two control groups. The fragile X group further displayed a signi®cantly greater increase in the mean distance moved between detected targets than the Down's syndrome group and the two control groups. There were no other signi®cant dierences between the groups. 3.2.3. Sustained attention The three Vigilan scores Ð targets detected, mean time taken to respond and false alarms were analysed with multivariate analysis of variance (MANOVA) with group as the independent variable. Two of the scores were ®rst transformed before any analyses to reduce their skewness. A natural log transformation was made for the mean time taken and a square root transformation for the number of false alarms. There was a signi®cant main eect of group (F = 14.78; df=3; P < 0.0001), and a signi®cant group by task score interaction (F = 16.37; df=3; P < 0.0001) indicating that the pattern of performance across the groups was dierent for dierent scores. Separate ANOVA's on the dierent sustained attention scores followed by post-hoc Newman±Keuls planned comparisons were used to analyse the eects in detail. However, to reduce the likelihood of Type I errors, the Bonferroni correction test was used where only those results meeting an alpha level of 0.05/3=0.02 were considered statistically signi®cant. There were no signi®cant group dierences for targets detected [F(3,96)=3.83; P > 0.03; NS] although there was a trend with the fragile X group detecting fewer targets than the other three groups. However, performance between the groups diered signi®cantly for: mean time taken to respond to targets [F(3,96)=4.80; P < 0.004] and for false alarms [F(3, 96)=22.66; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group displayed signi®cantly more false alarms in comparison with all three groups. This was followed by the Down's syndrome group who displayed signi®cantly more false alarms in comparison with the good attention control group only. For mean time taken to respond, the fragile X group, Down's syndrome group and poor attention control group did not dier signi®cantly but all displayed greater mean times to respond to a target in comparison with the good attention control group. 3.2.4. Executive functioning The total number of correct inhibitions made on the Walk task (maximum 20) were selected for analysis. For the Same±Opposite task, the time taken to complete the two ``same'' conditions (non-competing baseline conditions Ð reading the numbers one and two as presented) of the task were added together; and the time taken to complete the ``opposite'' conditions (competing conditions Ð reading the opposite of the
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
numbers presented) were added together. The total sum time for the ``same'' condition was then subtracted from the total sum time of the ``opposite'' condition to obtain a measure of response±organisation time for analysis ([opptime1+oppti2 1 2 me ]ÿ[sametime +sametime ]). Unfortunately, only 17 out of 25 (68%) fragile X boys in comparison to 22 out of 25 (88%) Down's syndrome boys were able to complete the opposite condition of the Same±Opposite task. As a result, in order to retain equal n, 17 participants were randomly selected from the Down's syndrome group and each of the two control groups on this task for purpose of analysis.2 Two univariate analyses of variance (ANOVAs) were performed, one for each variable, with group as the independent variable. The ANOVA revealed a signi®cant group eect for the Walk task [F(3,96)=13.12; P < 0.0001] and a signi®cant group eect for the Same±Opposite task [F(3,96)=92.77; P < 0.0001]. A post hoc Newman±Keuls test revealed that the fragile X group demonstrated signi®cantly poorer inhibition on the Walk task than all the other three groups. This was followed by the Down's syndrome group and the poor attention control group who both displayed signi®cantly poorer inhibition than the good attention control group. In contrast, the Down's syndrome group demonstrated signi®cantly more diculty in organising responses on the Same±Opposite task than all the other three groups. This was followed by the fragile X group who displayed signi®cantly greater dif®culty in organising their responses in comparison to the two control groups. Finally, the poor attention control group displayed signi®cantly greater diculty in organising their responses than the good attention control group.
4. Discussion This study examined dierent components of attention skills in a sample of young fragile X males, a learning disabled comparison group and two normal control groups. The ®ndings of the present study suggest that fragile X males display a pro®le of attention that is signi®cantly dierent from that seen in Down's syndrome males and the two normal control groups. In comparison to Down's syndrome males, fragile X males performed signi®cantly worse on aspects of selective attention, divided attention, sustained attention and inhibition. In contrast, they per2 The pattern of results did not change when these cases were not excluded from the Down's syndrome group and each of the four control groups.
1267
formed better than Down's syndrome males on the response±organisation task (executive functioning). In comparison to the poor attention and good attention normal control groups, fragile X males performed signi®cantly worse than both control groups on all attention measures on which dierences between groups occurred, except response time in the sustained attention task where they were not signi®cantly slower than the PAC group. In terms of a selective attention de®cit, poor performance by fragile X males compared with Down's syndrome males appeared to be speci®c to selecting relevant information rather than speed of search and distance moved in search. This impairment is re¯ected in the signi®cantly smaller number of targets detected and the high number of false alarms produced by this group which indicates that they had greater diculty either in discriminating between target and distractor or in shifting attention and inhibiting impulsive responding to incorrect targets. The impulsive behaviour of fragile X males is also evident on tasks of sustained attention (shown by the high false alarm rate) and inhibition in the Walk task and is in accordance with their behavioural phenotype of reported hyperactivity and impulsivity [4,14]. In terms of a divided attention de®cit, measured as the proportional dierence in performance between the selective and divided attention condition, fragile X males displayed impaired performance, compared with Down's syndrome, in the mean distance moved across the computer screen while locating correct targets. An impairment on this measure suggests fragile X males had greater diculty in organising search at the same time as shifting attention from one type of target search to another, therefore displaying a de®cit in their ability to divide or switch attention. In contrast to a de®cit in selective and divided attention, fragile X males performed at a comparable level to Down's syndrome males and the two control groups in the speed of detecting targets in the sustained attention task, although the two learning disabled groups and the poor attention control group were slower than the good attention control group in responding to targets. This ®nding is consistent with studies that suggest those with learning disabilities or hyperactive behaviour do not demonstrate a sustained attention de®cit but indicate slower information-processing skills in comparison to controls [28,44]. This is also consistent with recent ®ndings of adult fragile X males displaying relatively intact performance on a task of sustained attention, but an impairment on a task that required attentional shift from one stimulus to another in comparison with learning disabled controls [8]. On a task of inhibition, fragile X males demonstrated a signi®cant impairment in the ability to inhibit or delay responding in comparison to the other three
1268
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
groups. This ®nding is in accordance with the overall performance of fragile X males who consistently show impulsive behaviour on all attention tasks in this study (except the response organisation task). This present ®nding further adds to the growing body of literature that indicates executive dysfunctions to be the most persistent de®cit in fragile X syndrome both males and females [22,25,39]. Interestingly, the poor attention controls also show a consistent de®cit in the ability to inhibit impulsive responding on nearly all tasks in comparison to the good attention controls. They produced more false alarms in both selective and sustained attention and worse performance in the Walk task and response organisation task, suggesting a similar but less profound pattern of impulsive behaviour to the fragile X males. On a task of response±organisation, however, the Down's syndrome males were more impaired than fragile X males. This is somewhat surprising given that a consistent de®cit in organising and planning is frequently cited as a core weakness in individuals with fragile X syndrome [22,39]. One explanation might be that the present task primarily measured speed at which a novel stimulus±response linkage could be processed and errors corrected, rather than ability to cope with sudden unexpected changes in response demands, as in the walk task. Thus, greater impairment on this task might re¯ect a de®cit in speed of information processing rather than a de®cit in response organisation. Moreover, fewer fragile X males (68%) than Down's syndrome males (88%) were able to complete this task, providing further support for the hypothesis that at least some fragile X males have greater diculty in planning and organising information than Down's syndrome males. Across the two normal control groups, the poor attention controls displayed a signi®cant de®cit on this task in comparison with the good attention controls, again suggesting a similar but less profound pattern of de®cit to fragile X males. The results from the poor attention controls are in line with reports that ascribe those with hyperactive behaviour or AD/ HD to a de®cit in higher-processes such as executive function and behavioural inhibition [1,2]. Overall, both the learning-disabled groups were impaired in many respects compared with the good attention control group and poor attention on measures of selective attention, divided attention and response organisation. Hence there are aspects of attention that seem to be vulnerable in the learningdisabled groups, particularly selective attention, indicating a possible limited general attentional capacity that is allocated ineciently across complex task demands [15]. This ®nding is in accordance with studies on those with learning disabilities displaying a smaller pool of attentional capacity on tasks of divided attention [23,28].
However, the fragile X group demonstrated additional problems speci®c to this disorder on measures of selective attention, divided attention and inhibition, with a particular impairment in the ability to plan and organise search, shift attention from one concept to another, delay responding and inhibit task-irrelevant responses. These de®cits are typically associated with executive dysfunction or higher control processes of attention; and are considered to be core characteristics of frontal lobe functioning [41]. For example, the prefrontal cortex and its connections are known to underlie response inhibition. The dorsolateral prefrontal cortex appears to be involved with adjusting attention and plays a crucial role in the organisation of visually guided behaviour during visual search [3]. The orbito-frontal cortex also plays a part in inhibiting behavioural response to external contingencies. The present ®ndings suggest that the pattern of attention de®cits in fragile X males may indicate a frontal lobe dysfunction consistent with the structural abnormalities reported in the prefrontal regions in those with fragile X syndrome [33]. These ®ndings further suggest that this pattern of attention de®cits in fragile X males, more severely aected in this group than in Down's syndrome males or the poor attention normal controls, is speci®c to their syndrome and adds to the growing literature on distinct neurodevelopmental phenotypes. Additional bene®ts also accrue from such data by providing information to professionals that might guide and assist the timing of early intervention and treatment programmes that could maximise developmental potential and improve prognosis.
Acknowledgements This work was supported by Mental Health Foundation research grant R593/3 awarded to Dr Kim Cornish. The authors wish to thank all participating families for their support in this research.
References [1] Barkley RA. Behavioural inhibition, sustained attention and executive functions: toward a more comprehensive theory. Developmental and Behavioural Pediatrics 1997;18:271±9. [2] Barkley RA. Attention de®cit/hyperactivity disorder, self-regulation and time: toward a more comprehensive theory. Developmental and Behavioural Pediatrics 1997;18:271±9. [3] Barone P, Joseph JP. Role of the dorsolateral prefrontal cortex in organising visually guided behaviour. Brain, Behaviour and Evolution 1989;33:132±5. [4] Baumgardner T, Reiss AL, Freund LS, Abrams MT. Speci®cations of the neurobehavioural associations in males with fragile X syndrome. Pediatrics 1995;95:744±52. [5] Corbetta M, Miezin FM, Dobmeyer S, Shulman GL, Peterson
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
[6] [7] [8]
[9] [10] [11] [12] [13] [14] [15] [16] [17]
[18]
[19]
[20] [21]
[22]
[23] [24] [25]
E. Selective and divided attention during visual discrimination of colour, shape and speed: functional anatomy by positron emission tomography. Journal of Neuroscience 1991;11:2383± 402. Corbetta M, Miezin FM, Shulman GL, Peterson SE. A PET study of visuospatial attention. Journal of Neuroscience 1993;13:1202±26. Cornish KM, Munir F, Cross G. Spatial cognition in males with fragile X syndrome: evidence for a neuropsychological phenotype. Cortex, in press. Cornish KM, Munir F, Cross G. Dierential impact of the FMR-1 full mutation on memory and attention functioning: evidence for a molecular-neuropsychological phenotype? Brain, submitted. Crowe SF, Hay A. Neuropsychological dimensions of the fragile X syndrome: support for a non-dominant hemisphere dysfunction hypothesis. Neuropsychologia 1990;28:9±16. Davids JR, Hagerman RJ, Eilert RE. Orthopaedic aspects of fragile X syndrome. Journal of Bone Joint Surgery 1990;72:889± 96. Dunn L, Dunn P, Whetton C, Pintilie D. British picture vocabulary scale. Windsor: NFER, 1982. Dykens EM, Hoddap RM, Leckman JF. Strengths and weaknesses in the intellectual functioning of males with fragile X syndrome. American Journal of Medical Genetics 1987;28:13±5. Freund LS, Reiss AL. Cognitive pro®les associated with the fragile X syndrome in males and females. American Journal of Medical Genetics 1991;38:542±7. Hagerman RA. Fragile X syndrome. Current Problems in Pediatrics 1987;17:627±74. Kahneman D. Attention and eort. Englewood Clis, NJ: Prentice-Hall, 1973. Kaufman PM, Leckman JF, Ort SI. Delayed response performance in males with fragile X syndrome. Journal of Clinical Experimental Neuropsychology 1990;12:69. Kemper MB, Hagerman RJ, Ahmad RS, Mariner R. Cognitive pro®les and the spectrum of clinical manifestations in heterozygous fragile X females. American Journal of Medical Genetics 1986;23:139±56. Konishi S, Nakajima K, Uchida I, Kameyama M, Nakahara K, Sekihara K, Miyashita Y. Transient activation of inferior prefrontal cortex during cognitive set shifting. Nature Neuroscience 1998;1:80±4. Lachiewicz AM, Spiridigliozzi A, Guillion CM, Ransford SN, Rao K. Aberrant behaviours of young boys with fragile X syndrome. American Journal on Mental Retardation 1994;98:567± 79. Manly T, Robertson IH, Anderson V, Nimmo-Smith I. The test of everyday attention for children: TEA-Ch. Bury St Edmunds: Tames Valley Test Company, 1999. Mazzocco MMM, Hagerman RJ, Cronister-Silverman A, Pennington B. Speci®c frontal lobe de®cits among women with the fragile X gene. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31:1141±8. Mazzocco MMM, Pennington BF, Hagerman RJ. The neurocognitive phenotype of female carriers of fragile X: additional evidence for speci®city. Developmental and Behavioural Pediatrics 1993;14:328±34. Merrill EC, Peacock M. Allocation of attention and task diculty. Journal of Mental Retardation 1994;98:588±93. Morton JE, Bundey S, Webb T, MacDonald F, Rindl PM, Bullock S. Fragile X syndrome is less common than previously estimated. Journal of Medical Genetics 1997;34:1±5. Munir F, Cornish KM, Wilding J. Patterns of attention de®cit in fragile X syndrome: a neuropsychological perspective. In: Proceedings from the Sixth International Fragile X Syndrome Conference, Ashville, USA, 1998.
1269
[26] Munir F, Cornish KM, Wilding J. Nature of the working memory de®cit in fragile X syndrome. Brain and Cognition, submitted. [27] Oberle L, Rousseau F, Heitz D, Kretz C, Devys D, Hanaver A, Boue J, Bertheas MF, Mandel JL. Instability of a 550-base pair DNA segment and abnormal methylation in fragile X syndrome. Science 1991;252:1097±102. [28] Pearson DA, Yaee LS, Loveland KA, Lewis KR. Comparison of sustained and selective attention in children who have mental retardation with and without Attention De®cit Disorder. American Journal of Mental Retardation 1996;100:592±607. [29] Pieretti M, Zhang F, Fu Y-H, Warren ST, Oostra BA, Caskey CT, Nelson DL. Absence of expression of the FMR1 gene in the fragile X syndrome. Cell 1991;66:817±22. [30] Posner MI. Interaction of arousal and selection in the posterior attention network. In: Baddeley A, Weiskrantz L, editors. Attention: selection, awareness and control Ð a tribute to Donald Broadbent. Oxford Science Publications, 1993. p. 390± 405. [31] Posner MI, Dehaene S. Attentional networks. Trends in Neurosciences 1994;17:75±9. [32] Posner MI, Peterson SE. The attention system of the human brain. Annual Review of Neuroscience 1990;13:25±42. [33] Reiss AL, Abrams MT, Greenlaw R, Freund L, Denckla MB. Neurodevelopmental eects of the FMR1 full mutation in humans. Nature medicine 1995;1:159±67. [34] Riddle JE, Cheema A, Sobesky WE, Gardner SC, Taylor AC, Pennington BF, Hagerman RJ. Phenotypic involvement in females with the FMR-1 gene mutation. American Journal of Mental Retardation 1998;102:590±601. [35] Robertson IH, Ward T, Ridgeway V, Nimmo-Smith I. The test of everyday attention. University of Cambridge: MRC Applied Psychology Unit, 1994. [36] Schneider W, Shirin RM. Controlled and automatic human processing: I. Detection, search and attention. Psychological Review 1977;92:424±8. [37] Shapiro MB, Murphy DGM, Hagerman RJ, Azari NP, Alexander GE, Miezejeski CM, Hinton VJ, Horwitz B, Haxby JV, Kumar A, White B, Grady CL. Adult fragile X syndrome: neuropsychology, brain anatomy and metabolism. American Journal of Medical Genetics 1995;60:480±93. [38] Shirin RM, Schneider W. Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychological Review 1977;84:127±90. [39] Sobesky WE, Pennington BF, Porter D, Hull CE, Hagerman RJ. Emotional and neurocognitive de®cits in fragile X. American Journal of Medical Genetics 1994;51:378±84. [40] Sobesky WE, Taylor AK, Pennington BF, Riddle J, Hagerman RJ. Molecular/clinical correlations in females with fragile X. In: Presented at the 7th International Workshop on fragile X and X-linked Mental retardation, Tromso, Norway, 1995. [41] Stuss DT, Benson DF. The frontal lobes. New York: Raven Press, 1986. [42] Turk J. Fragile X syndrome and attentional de®cits. Journal of Applied Research in Intellectual Disabilities 1998;11:175±91. [43] Ullmann RK, Sleator EK, Sprague RL. A new rating scale for diagnosis and monitoring of ADD children. Psychopharmacology Bulletin 1984;21:339±44. [44] Van der Meere J, Wekking E, Sergeant J. Sustained attention in pervasive hyperactivity. Journal of Child Psychology and Psychiatry 1991;32:275±84. [45] Verkerk AJMH, Pieretti M, Sutclie JS, Fu Y-H, Kuhl DPA, Pizzuti A, Reiner O, Richards S, Victoria MF, Zhang F, Eussen BE, Van Ommen G-JB, Blonden LAJ, Riggins GJ, Chestain JL, Kurst CB, Galjaard H, Caskey CT, Nelson DL, Oostra BA, Warren ST. Identi®cation of a gene (FMR-1)
1270
F. Munir et al. / Neuropsychologia 38 (2000) 1261±1270
containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell 1991;65:905±14. [46] Welsh MC, Pennington BF. Assessing frontal lobe functioning in children: views from developmental psychology. Developmental Psychology 1988;4:199±230.
[47] Wilding J. Wilding attention tests. University of London: Department of Psychology, Royal Holloway, 1997. [48] Yu S, Pritchard M, Kremer E, Lynch M, Nancarrow J, Baker E, Holmen K, Mulley JC, Warren ST, Schlessinger D. Fragile X genotype characterised by an unstable region of DNA, Science 1991;252:1179±81.