[Correction added after publication 5 July 2010: Author name Umbertina Conti Reed was added to the list of authors and is affiliated to the Department of Neurology, Medical School, University of São Paulo, São Paulo, Brazil.]
KATIA OSTERNACK PINTO
Department of Psychology, Clinical Hospital of University of São Paulo, São Paulo, Brazil.
Maria Clara Drummond Soares de Moura at Av. Professor Lineu Prestes, 1524, Room 238, Prédio Biomédicas I, Cidade Universitária, Butantã, São Paulo/SP, CEP:05508-900 Brazil. E-mail: email@example.com
Aim The cognitive deficits present in the Duchenne muscular dystrophy (DMD) are not yet well characterized. Attention, considered to be the brain mechanism responsible for the selection of sensory stimuli, could be disturbed in DMD, contributing, at least partially, to the observed global cognitive deficit. The aim of this study was to investigate attentional function in individuals with DMD.
Method Twenty-five males (mean age 12y; SD 2y 2mo) with DMD and 25 healthy males (mean age 12y; SD 2y) were tested in a visuospatial task (Posner computerized test). They were instructed to respond as quickly as possible to a lateralized visual target stimulus with the ipsilateral hand. Their attention was automatically orientated by a peripheral prime stimulus or, alternatively, voluntarily orientated by a central spatially informative cue.
Results The main result obtained was that the attentional effect (sum of the benefit and the cost of attention) did not differ between the two groups in the case of automatic attention (p=0.846) but was much larger for individuals with DMD than for comparison individuals in the case of voluntary attention (p<0.001).
Interpretation The large voluntary attentional effect exhibited by the participants with DMD seems similar to that of younger children, suggesting that the disease is associated with delayed maturation of voluntary attention mechanisms.
• This paper extends our knowledge about the cognitive deficits present in Duchenne muscular dystrophy.
• It demonstrates an important alteration of voluntary attention in patients with IQ in the normal range.
• Its results suggest that this alteration is associated with a developmental lag of voluntary attention function.
Duchenne muscular dystrophy (DMD) is a recessively inherited genetic disease with the second highest incidence of all genetic diseases. It is primarily characterized clinically by progressive and irreversible weakening of muscle strength due to skeletal muscle loss. The muscle loss is a result of the deficiency of a protein called dystrophin, secondary to a mutation of the Xp21 (or DMD) gene located on the short arm of the X chromosome.1
In addition to skeletal muscle, dystrophin is also expressed in the central nervous system (CNS). In the normal brain, the complete protein (Dp427) and its isoforms are concentrated in the postsynaptic pyramidal cells of the cerebral cortex (mainly those of the deep layers of the frontal cortex), in neurons in the hippocampus, and in the soma and dendrites of Purkinje cells in the cerebellum.2
It has been shown that the protein is not expressed in the brain of mdx (muscular dystrophy X-linked) rats (an animal model of DMD).1,3 It was also found to be not expressed in the cerebral cortex of an individual with DMD whose brain was submitted to an autopsy.4 A deficiency of dystrophin in the cerebral cortex, possibly leading to disturbances in synaptic transmission, might explain the commonly observed significant reduction (about 1SD) in the IQ of individuals with DMD and the fact that almost 35% of such individuals present mild to severe learning disability.*5
There have been attempts to characterize more precisely the cognitive deficits of individuals with DMD. A consistent finding has been their impaired performance in the Digit Span Test, which requires immediate recall of verbal information. According to Hinton et al.,6,7 this would indicate a deficit in the verbal component of working memory. This problem could also be caused by an attentional or language disturbance.8
Attention can be defined as neural activity that selects information for further processing. Evidence presumably indicating the existence of an important attentional deficit in individuals with DMD was obtained by Cotton et al.9 using the Symbol Digit Modality Test, and, more recently, by Cyrulnick et al.10 using the visual attention subtest (visual search) of the Developmental Neuropsychological Assessment (NEPSY).11 Marini et al.12 linked the story-telling disturbance found in individuals of normal intelligence with DMD to a visuospatial attention deficit, which would cause a reduction in the accuracy of visual inspection. In agreement with the idea is the high incidence of signs of attention-deficit–hyperactivity disorder in individuals with DMD. Steele et al.13 presented evidence that these signs could be present in as many as 50% of such individuals. In other studies, however, individuals with DMD have been observed to be able to concentrate and sustain their attention as well as typically developed individuals, suggesting that this function is mostly preserved.6
According to Knudsen,14 attention controls the access of information to working memory. This mechanism, in turn, helps to keep the selected information under the focus of attention. As might be expected, some brain areas mediate both functions. LaBar et al.15 considered that to be the case for the frontal eye fields, the ventral premotor cortex, the supplementary motor cortex, the intraparietal sulcus cortex, the thalamus, and the cerebellum.
Given this intimate relationship between working memory and attention, one can easily see the difficulty in distinguishing deficits related to one or the other function. It is very likely that most neuropsychological tests used so far to evaluate attention also measure working memory to a certain extent. Considering the importance of attention for cognitive activity, it would be interesting to attempt to isolate attentional abnormalities in individuals with DMD.
In the current study, we evaluated automatic and voluntary visuospatial attention using Posner tests.16 Automatic attention is normally mobilized by sudden salient sensory stimuli in the environment, whereas voluntary attention is commonly mobilized by some endogenous signal or some information provided to the individual by someone else. Posner tests have been widely used in the literature and might be considered a criterion standard for the evaluation of visuospatial attention. They are, presumably, able to indicate attentional competence with minimum dependence on working memory or language.
Twenty-five individuals with DMD as indicated by progressive loss of muscle force, high levels of creatinine phosphokinase, characteristic muscle histology, and/or DMD gene alterations (14 participants with deletion of one to six of the following exons: 41, 43, 45, 47, 48, 50, 51, 52, 60, and Pm346), who were attending the Clinical Hospital of São Paulo University Medical School, and 25 healthy individuals, attending lessons in the Socio-Educational Nucleus of Jaguaré in São Paulo City, were invited to participate in the study. The patients with DMD were selected according to their age and still relatively good upper-limb motor performance. The comparison participants were selected to match the participants with DMD in terms of age, socio-economic status, and educational level. All participants were aged 10 to 16 years; the mean age in both groups was 12 years (SD: DMD group, 2y 1mo, and comparison group, 2y; t-value<0.001, p>0.05). Only individuals who were right handed, had normal or corrected to normal vision, and were not taking any drug with central nervous system action were included in the study. The mean (SD) full-scale IQ, verbal IQ, and performance IQ in the DMD group were, respectively, 83.6 (15.0), 89.3 (14.20, and 80.4 (14.5), and in the comparison group were, respectively, 88.9 (10.4), 87.0 (8.9), and 93.3 (SD). All the participants were attending elementary school or middle school at the time of testing. None of the participants was taking drugs with CNS action. The total IQ of the participants with DMD did not differ from that of the comparison individuals (t-value=−0.866, p=0.39). In addition, the verbal IQ of the participants with DMD did not differ from that of the comparison individuals (t-value=0.372, p=0.71), but the performance IQ of the participants with DMD was somewhat lower than that of the comparison individuals (t-value=−2.109, p=0.043).
We did not evaluate motor function of the participants with DMD. We assumed that all of them had some degree of motor impairment. However, as the motor response we required from them was very simple, and a difference score between conditions was our major dependent variable (see below), we felt this motor impairment would interfere only minimally with the critical results.
The legal representatives of all participants authorized their testing by signing a free and informed consent form. The Ethics Committee for Human Research of the Institute of Psychology of the University of São Paulo and the ethics committee of the São Paulo University Medical School approved this study.
The patients with DMD were tested in the hospital (where they were being attended) and the comparison participants in the school where they were attending lessons. Participants were tested individually in a dimly lit room. They remained seated in front of a video monitor (with a dark-grey background screen and luminance equal to 3cd/m2) on which visual stimuli were presented. An adapted mouse coupled to a computer game port allowed them to respond to some of these stimuli. The stimuli were presented, and the responses were recorded, by means of a computer and by programs developed using MEL2 language software (Psychology Software Tools, Pittsburgh, PA, USA).
The testing session comprised two parts: one evaluating automatic attention and another evaluating voluntary attention, with the order of the two parts being varied among participants. Before each part, the participant was given the appropriate instructions (see below) and asked to perform around 15 practice trials.
The automatic attention test consisted of two blocks of 48 trials each; there was a short resting interval between the two blocks. Each trial began with the appearance of a small central cross (fixation point) and two empty squares (with 1.50 degrees of visual angle of side, luminance of 28cd/m2) centred 6.8 degrees of visual angle from the fixation point on each side (Fig. 1a). After 1750 to 2250ms, a prime stimulus appeared, represented by the brightening (40cd/m2) of the border of one of the squares (one-third of the trials on the left side and one-third of the trials on the right side) or of both squares (one-third of the trials, the bilateral condition) for 100ms. The target stimulus, represented by a vertical line (0.8 degrees of visual angle in length and 0.4 degrees of visual angle in width, luminance of 45cd/m2), then appeared inside the left (half of the trials) or right (half of the trials) square for 100ms. In the unilateral prime stimulus trials, the target stimulus could appear with equal probability on the same side (same condition) or on the opposite side (opposite condition). The participant was instructed to keep his eyes on the fixation point, to hold the mouse with both hands, and to press as quickly as possible, with his left or right thumb, respectively, the left or right button of the mouse, depending on whether the target stimulus appeared on the left or right side.
The voluntary attention test consisted of two blocks of 60 trials each; a short resting interval separated these two blocks. Each trial began with the appearance of the fixation point and the two empty squares (Fig. 1b). After 750 to 1250ms, a central cue appeared, represented by an arrowhead (0.4 degrees of visual angle in height and 0.2 degrees of visual angle in width, luminance of 45cd/m2) pointing to the left (one-third of the trials) or right (one-third of the trials) side, or two arrowheads each one pointing to one side (one-third of the trials, the neutral condition) for 1000ms. The target stimulus, represented by a vertical line, then appeared inside the left (half of the trials) or right (half of the trials) square for 100ms. In the trials in which the arrowhead pointed to one side, the target stimulus could appear either on the pointed side (three-fifths of the trials, valid condition) or on the opposite side (one-fifth of the trials, invalid condition). The participant was instructed to keep his eyes on the fixation point and to orientate his attention to the side indicated by the cue – or to let it diffuse in case of the neutral cue – and press as quickly as possible, with his left or right thumb, the left or right button of the mouse, respectively, according to the side of appearance of the target stimulus.
After each response of the participant, a message appeared at the centre of the screen indicating his reaction time in milliseconds or, in case of an error, whether an anticipated response (before 150ms after target appearance), a wrong hand response, or a slow or absent response (reaction time longer than 1000ms) had occurred.
The block median reaction time was submitted to an analysis of variance (ANOVA) with the groups (comparison and DMD) as a between-participant factor and kind of attention (automatic or voluntary) and kind of prime stimulus (same/valid, opposite/invalid, or bilateral/neutral) as within-participant factors. A second ANOVA considered the benefit (difference between reaction time in the bilateral/neutral condition and reaction time in the same/valid condition) and cost (difference between reaction time in the opposite/invalid condition and reaction time in the bilateral/neutral condition) of attention. The factors in this analysis were the group, the kind of attention, and the kind of effect (benefit or cost). When appropriate, data were further analysed with the Tukey test. For all these analyses, a significance level of 0.05 was used.
The automatic/voluntary attentional effect (difference between the reaction times in the opposite/invalid and the same/valid conditions) was correlated with age for both the participants with DMD and the comparison group. Pearson’s correlation analysis was used with a significance level of 0.05. This analysis would allow a comparison between the development of attention in the participants and that in the comparison group.
The number of each kind of error (anticipation, inversion, or omission) for each kind of prime stimulus and for each kind of attention was compared between the two groups using the Kolmogorov–Smirnov test. The significance level in this case was reduced to 0.008 to account for multiple comparisons.
The reaction time analysis revealed a main effect of group (F1,48=8.99, p=0.004), kind of attention (F1,48=13.41, p<0.001), and kind of prime stimulus (F2,96=151.11, p<0.001). Reaction time was longer among participants with DMD than in the comparison group and shorter for voluntary attention than for automatic attention.
There was a triple interaction (F2,96=7.456, p<0.001). The post-hoc analysis showed that reaction time was longer in individuals with DMD than in comparison individuals for the same, opposite, and bilateral conditions, and also for the neutral and invalid conditions (p<0.001 in all cases). For the valid condition, reaction time did not differ (p=0.991) between the two groups (Fig. 2).
The benefit–cost analysis showed a main effect of group (F1,48=21.786, p<0.001) and kind of effect (F1,48=7.186, p=0.010). The participants with DMD presented a larger attentional effect (benefit plus cost) than the comparison individuals. The cost was larger than the benefit.
There was an interaction of group and kind of attention (F1,48=10.818, p=0.002). The post-hoc analysis showed that the effect of voluntary attention was larger for the participants with DMD than for the comparison individuals (p<0.001; Fig. 3). There was also an interaction between kind of attention and kind of effect (F1,48=4.401, p=0.041). The post-hoc analysis showed that the cost was larger than the benefit in the case of automatic attention (p=0.014).
No significant correlation between the automatic attentional effect and the age of the participants with DMD was observed (r=0.27, p>0.05). The voluntary attentional effect diminished with increase in age of participants (r=−0.40, p<0.05). Among the comparison individuals, no significant correlation between the two variables was observed (r=0.2, p>0.05, for the automatic attentional effect; r=−0.06, p>0.05, for the voluntary attentional effect). These results are shown in Figure 4.
Only one significant difference was found in the accuracy analysis: the number of inversion errors in the invalid condition was higher for the individuals with DMD than for the comparison group (p<0.001).
We found that, in the automatic attention test, neither the benefit nor the cost of spatial attention orienting differed between individuals with DMD and the comparison group. In the voluntary attention test, although there was no significant benefit or cost of spatial attention orienting for the comparison individuals, both these effects were found to be very large for the participants. In addition, the participants made a larger number of inversion errors in the invalid condition than the comparison individuals. These results indicate that voluntary attention but not automatic attention is disturbed in the disease.
The participants with DMD, who had a mean age of 12 years, behaved in some ways like much younger children. Perche and Garcia-Larrea17 demonstrated that 6- to 9-year-old children exhibit a much larger voluntary attention effect than young adults. Wetzel and Schröger18 showed that 6- to 8-year-old children make more distraction errors than 10- to 12-year-olds and adolescents. Perche and Garcia-Larrea17 suggested that young children, in contrast to young adults, have more trust in the spatial information provided by the cue and fully orient their attention to the indicated hemifield, which would tend to leave their attention more spread across the visual field. Our participants with DMD might well have adopted that attentional strategy.
Results very similar to ours were reported by Wilson et al.19 when testing children with developmental coordination disorder, whose most obvious characteristic was delayed development of motor abilities. The automatic attention effect exhibited by these children was not different from that exhibited by typically developing children. On the other hand, the voluntary attention effect was larger than that exhibited by typically developing children. Wilmut et al.20 pointed out the similarity of this attentional pattern to that of younger children and concluded that developmental coordination disorder could be associated with developmental delay of attentional abilities.
It is possible that a developmental lag of voluntary attention function also occurs in DMD. The adult pattern of attention orienting in healthy children is achieved by the age of 10 to 12 years (Wetzel and Schröger18). The fact that the large voluntary attentional effects exhibited by the participants with DMD decreased with age, tending to become similar to those of the comparison individuals by the age of 15 or 16 years, is in accordance with this idea. It should be noted that other studies have shown evidence of a characteristic delay in the development of cognitive function,10 including social abilities21 in participants with DMD. A longitudinal study must be conducted in the future to confirm (or not) our hypothesis.
Dystrophin deficiency could explain the occurrence of a slower maturation of the neural systems that control voluntary attention, as well as other cognitive mechanisms, in children and adolescents with DMD. Experimental evidence strongly suggests that dystrophin is very important for normal development of the CNS and might contribute to synaptogenesis, synapse maturation, and synapse stabilization, as suggested by the dramatic increase in astrocytes and neurons that occurs before the fifth gestational week, and the associated increase in postsynaptic density.22,23 A particularly affected area might be the frontal cortex, which, as mentioned before, normally expresses dystrophin in its deep layers.2 As this area is very important for voluntary attention control,24 immaturity of this area could cause serious disturbance of this function.
The fact that reaction time was longer in individuals with DMD than in comparison individuals was expected, considering the characteristic motor deficits of the disease.
The somewhat lower performance IQ of the participants with DMD might raise the question of whether their larger voluntary attention effects are in some way related to a general disturbance of visuospatial ability. This possibility is unlikely given that automatic attention effects in this group did not differ from those of the comparison individuals. Alternatively, the possibility that the lower performance IQ of the participants with DMD is, at least partly, related to their altered ability to control voluntary attention should also be considered. Voluntary attention is necessary for proper execution of all the tests used to evaluate performance IQ, and its disturbance should thus affect this parameter. In a future study, it would be interesting to investigate specifically the influence of the voluntary attentional change we demonstrated on the performance IQ of participants with DMD.
The literature indicates that the verbal components of intelligence are disturbed in participants with DMD. Our participants showed no disturbance of this kind. When identified, a measurement of verbal attention should be performed. As language could be primarily affected by the disease, it would be desirable first to evaluate language separately from attention and then, later, to evaluate verbal attention. In this way, it would be easier to interpret eventual changes found in the verbal attention tests.
A limitation of our study is the fact that we did not use the siblings of the participants as comparison individuals. Although this is highly recommended, there are studies in which this advice was not followed.7,10 In our case, only few participants with DMD had male siblings of an appropriate age. Thus, our comparison group comprised non-sibling males matched for age and socio-economic status.
What are the consequences of this attentional disturbance to normal life for individuals with DMD? Presumably, they would have difficulty in dealing with situations that involve a main task with one or more secondary tasks, as many common daily situations do. An appropriate strategy to help these individuals would seem to require them to perform only one task at a time whenever possible.
Overall, our results suggest that 10- to 16-year-old individuals with DMD are somewhat immature in the way they control their voluntary attention, owing to a delayed maturation of voluntary attention mechanisms with the disease.
North American usage: mental retardation.
We thank Elaine Cristina Zachi for her technical help in the study.