Visuo-spatial cognition 1 Running head: VISUO-SPATIAL COGNITION IN WILLIAMS SYNDROME Visuo-Spatial Cognition in Williams Syndrome: Reviewing and Accounting for the Strengths and Weaknesses in Performance

نویسندگان

  • Emily K. Farran
  • Christopher Jarrold
چکیده

Individuals with Williams syndrome typically show relatively poor visuo-spatial abilities in comparison to stronger verbal skills. However, individuals‟ level of performance is not consistent across all visuo-spatial tasks. The studies assessing visuo-spatial functioning in Williams syndrome are critically reviewed, in order to provide a clear pattern of the relative difficulty of these tasks. This prompts a possible explanation of the variability in performance seen which focuses on the processing demands of some of these tasks. Individuals with Williams syndrome show an atypical processing style on tests of construction, which does not affect tests of perception. Visuo-spatial cognition 3 Introduction Ever since the pioneering work of Bellugi and her colleagues (e.g., Bellugi, Sabo, & Vaid, 1988) it has been clear that a fundamental aspect of the psychological profile of individuals with Williams syndrome (WS) is their relatively poor performance on tests of visuo-spatial cognition (see Karmiloff-Smith, Klima, Bellugi, Grant, & Baron-Cohen, 1995; Mervis, 1999). Many researchers describe the cognitive profile of WS by detailing the marked contrast that is seen between these individuals‟ verbal and visuo-spatial abilities. For example, Bellugi, Wang, and Jernigan (1994) describe a "Pattern of linguistic preservation and marked spatial cognitive deficit” (p. 44), whilst Udwin and Yule (1991) suggest that "overall their verbal abilities are markedly superior to their visuo-spatial and motor skills" (p. 233). In general it is the case that the verbal abilities of individuals with WS are superior to their non-verbal abilities (Grant et al., 1997; Howlin, Davies & Udwin, 1998). However, Jarrold, Baddeley, & Hewes (1998) and Jarrold, Baddeley, Hewes, & Phillips (in press) claim from cross-sectional and longitudinal data respectively, that in WS, verbal ability improves at a faster rate than non-verbal ability, so that as individuals develop an increasing discrepancy between these two domains emerges. This is supported by Atkinson et al. (in press) who report steeper slopes in improvement with age in vocabulary and grammar ability, than the comparatively slow rate of improvement in ability with age on three visuo-spatial tasks. While the discrepancy between verbal and non-verbal ability is, broadly speaking, characteristic of WS, the situation is complicated by the considerable variance between the composite measures of any IQ score. These composite scores often hide an interesting pattern of differences in abilities shown on tests that measure particular aspects of cognition within the verbal or visuo-spatial domains. Karmiloff-Smith et al. (1997) have shown, for example, that performance is not uniform on a range of tasks measuring different aspects of verbal ability. The purpose of this article is to critically review the research carried out to date in the area of visuo-spatial cognition in WS, with the aim of explaining the reasons for the varying levels of performance between different visuo-spatial tasks. As this is arguably the weakest of all aspects of cognition in WS it is particularly important to provide potential explanations for the difficulties these individuals encounter on visual and spatial tests. Visuo-spatial cognition 4 Methodological Issues Methodological issues need to be taken into account when evaluating the findings of any study; however, particular methodological problems arise when working with special populations in general, and with individuals with WS specifically. Consequently, before reviewing the studies of visuo-spatial cognition in WS we provide a brief outline of these methodological concerns and their potential effects on these studies. Floor and ceiling effects can be a major problem when testing atypical populations. Clearly when either effect occurs, the test used may be artificially constraining the possible range of performance. Floor effects are particularly prevalent when testing visuo-spatial processing in WS, as this is such a weak area of cognition. This is evident in studies employing the Benton Lines Orientation test to assess individuals with WS (Benton, Varney, & Hamsher, 1978; see Bellugi et al., 1988; Rossen, Klima, Bellugi, Bihrle, & Jones, 1996; Wang, Doherty, Rourke, & Bellugi, 1995). Such results can only tell us that a group‟s abilities are at or below the lowest level that the test purports to measure, and thus that the group‟s scores may not be truly representative of their actual skills. In the present context, ceiling effects are commonly seen in comparison or control groups. When WS participants are matched to a control group for chronological age (CA), it is difficult to find a test which encompasses the range of abilities seen across the two groups. If the test is too easy for the control group they will score at ceiling, and as a consequence the performance of the WS group may erroneously appear to be close to that of controls. This possibility will be discussed in relation to much of the research that addresses face recognition in WS, where CA matched control groups are often employed (e.g. Karmiloff-Smith, 1997). The problem of floor and ceiling effects can be overcome somewhat by matching individuals with WS to typically developing (TD) groups for mental age (MA). This approach has been adopted by Bertrand, Mervis, and Eisenberg (1997) for example who, in addition to typically developing CA matched controls, also employed a typically developing control group matched for MA. This reduces the problem of differing levels of ability, but creates a new concern due to the discrepancy that will necessarily arise between the CAs of each group. The higher CA of the WS group will equate to them also having more Visuo-spatial cognition 5 experience, more practice in using their skills and more strategic coping styles (although this may also be linked to MA) which can introduce confounds into the experiment. This can be overcome by employing groups of individuals with learning difficulties as controls (Crisco, Dobbs, & Mulhern, 1988), or in addition to TD controls (Jarrold, Baddeley, & Hewes, 1999), because these individuals can potentially be matched to WS groups for both CA and MA. However, an additional issue that arises whenever a control group is employed concerns the criteria used to match groups. This follows from the uneven profile of abilities of individuals with WS in contrast to the flat profile seen in typical development. A CA matched control group is likely to differ from a WS group in all areas of intelligence, but more so in the visuo-spatial domain and less so in the verbal domain. When matching groups by general MA, a control group will have higher visuo-spatial skills, and lower verbal skills than the WS group. Any discrepancies in results that then emerges between groups could be primarily due to these differences in levels of ability, rather than in performance on the task in question. Visuo-spatial cognition in WS is particularly susceptible to the problems of matching by general MA, as level of ability in this area is so low. A related problem occurs when test batteries such as the Wechsler scales (Wechsler, 1974, 1981) or the Differential Ability Scales (DAS; Elliot, 1990) are employed. These batteries use a large number of tests to determine an individual‟s full scale IQ (FSIQ) or level of development (MA). The uneven profile of abilities in WS means that performance will not be equivalent across all of the individual subtests of a test battery. It is therefore important to consider the number and type of subtests used in order to determine which specific abilities are contributing to the composite measure of FSIQ. The Wechsler scales contain 5 non-verbal and 5 verbal subtests, whilst the DAS contains 6 subtests in total. Shortened versions of the WISC are also used (e.g., Grant et al. 1997). Speculatively, the more subtests employed, the more likely the average score will encompass the full range of abilities, thus producing a more reliable measure of general ability. However, although FSIQ can reliably assess general ability, due to the imbalance in skill in individuals with WS, an IQ score is not necessarily a particularly valid measure. It merely represents an averaged score of many differing levels of ability, and is therefore unlikely to be a powerful predictor of functioning on other tests of Visuo-spatial cognition 6 interest (as in typical development). In the case of WS, the individual subtests arguably provide more informative details of cognition than a composite IQ score. These problems of matching are exacerbated when individuals with learning difficulties are used as controls, because the cognitive profile of these controls may also be less uniform than that seen in typical development. For example, individuals with Down syndrome (DS) are often used as controls for individuals with WS (e.g., Bellugi, Bihrle, Neville, Doherty, & Jernigan, 1992; Bellugi et al., 1988; Rossen et al., 1996; Wang et al., 1995). The choice of DS controls is based on the assumption that these individuals exhibit a flat profile of abilities. However, Klein and Mervis (1999) present evidence against this assumption. They suggest that DS individuals have a relative strength in the area of visuo-spatial construction, and a relative weakness in verbal ability (see also Chapman, 1995; Fowler, 1990; Jarrold & Baddeley, 1997; Miller, 1987). In light of this, where studies have used DS individuals as controls, any observed differences in scores could be due to strengths or weaknesses in the DS control group as much as in the WS group. To overcome these difficulties control groups can be matched by their performance on a single measure. In order for matching to be appropriate the measure used must be drawn from the same area of cognition as the area under investigation. This ensures that the scores of the control group are predictive of the expected level of performance of the WS group in the test condition, although none of the studies reviewed here adopt this approach. A further problem can arise if the predictive MA measure and the testing measure are too closely related. In this case the experimenter might just be testing the same abilities twice in both groups, and any interesting results that might indicate a deviation from typical development will be wiped out (Bishop, 1997). Bishop claims that even when a difference is noted in these cases, one cannot be sure that this is not simply due to differences in the relative reliability of the two tasks. In summary, every method has some weaknesses, and research with individuals with WS is particularly susceptible to the problems of floor and ceiling effects and of matching controls appropriately. This emphasises the importance of employing a number of methodological techniques in a single study, with the intention of counteracting the weaknesses of one methodology with the strengths of another. For example, matching by Visuo-spatial cognition 7 both MA and by CA as in Bertrand et al. (1997), or by employing both TD controls and controls with moderate learning difficulties (see Jarrold et al., 1999). In the following two sections of the paper we review the main body of research in the area of visuo-spatial cognition in WS, discussing first studies which have used test batteries such as the Wechsler scales, and secondly those employing tests of specific aspects of visuo-spatial ability. Clearly all of these studies need to be interpreted in the light of the methodological concerns raised here. Studies Employing Standardised Test Batteries Non-verbal Subtests of the WAIS / WISC The WS cognitive profile has been primarily documented using standardised test batteries such as the Wechsler Intelligence Scale for Children and the Wechsler Adult Intelligence Scale (WISC-R, WAIS-R; Wechsler, 1974, 1981). Five studies provide information on the individual scores of each subtest of the battery (Arnold et al., 1985; Dall'oglio & Milani, 1995; Howlin et al., 1998; Udwin & Yule, 1991; Udwin, Yule & Martin, 1987), although the sample employed by Udwin and Yule (1991) is a subset of that employed by Udwin et al. (1987). Mean subtest scores where provided (2 studies: Howlin et al., 1998; Udwin et al., 1987) are given in Table 1. The five non-verbal subtests of the WISC-R and the WAIS-R provide a measure of Performance IQ (PIQ). These are: Picture Completion; where a picture is presented and the participant has to indicate what is missing; Picture Arrangement, which involves placing a series of pictures into a sequential order of events; Block Design, where the participant is instructed to use coloured blocks to model an example pattern; Object Assembly, which is a jigsaw type task; and Coding which is a timed task where the participant uses a key to draw specified symbols below a set of numbers. Dall'oglio and Milani (1995) assessed 16 individuals with WS aged 4;10 to 15;4 years (no mean age given) using the WISC-R. Their participants showed poor performance on the Block Design, Coding, and Picture Arrangement subtests, in contrast to better performance on the Picture Completion and Object Assembly subtests. Arnold et al. (1985) measured the performance of 23 participants (mean age: 10;4 years, range: 7;2 to 13;1). They report that Coding was significantly poorer than Picture Completion, Block Design, and Object Visuo-spatial cognition 8 Assembly. In addition, performance on Picture Completion was higher than Picture Arrangement performance, a result which is consistent with the data presented by Dall'oglio and Milani (1995). Howlin et al. (1998) studied 62 individuals with WS who were of mean age: 26.5 years (range: 19 to 39 years). In contrast to the studies described above, they found that scores on the Picture Arrangement task were the highest amongst the PIQ subtests and was significantly higher than scores on the Coding subtest (referred to as Digit Symbol) where the lowest scores were achieved (see Table 1). Udwin et al. (1987) originally tested 44 participants with a mean age of 11;1 years (range 6;0 to 15;9 years). They did not find highest scores on the Picture Arrangement task, but a subsequent study of 20 of these individuals, who had a mean age of 10;4 years (range: 6;5 to 14;5 years) did achieve their highest mean score on the Picture Arrangement subtest (Udwin & Yule, 1991). In both studies individuals showed significantly poorer ability in Coding than a combined mean of the scores achieved on the other four Performance subtests (see Table

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Spatial location 1 Running head: SPATIAL LOCATION AND WILLIAMS SYNDROME Evidence for Unusual Spatial Location Coding in Williams Syndrome: An Explanation for the Local Bias in Visuo-Spatial Construction Tasks?

Individuals with Williams syndrome (WS) display poor visuo-spatial cognition relative to verbal abilities. Furthermore, whilst perceptual abilities are delayed, visuo-spatial construction abilities are comparatively even weaker, and are characterised by a local bias. We investigated whether this differentiation in visuospatial abilities can be explained by a deficit in coding spatial location i...

متن کامل

Texture segmentation and Williams syndrome 1 RUNNING HEAD: Texture segmentation and Williams syndrome Texture Segmentation in Williams Syndrome

Texture segmentation and Williams syndrome 2 Abstract Williams syndrome (WS) is a developmental disorder in which visuo-spatial cognition is poor relative to verbal ability. At the level of visuo-spatial perception, individuals with WS can perceive both the local and global aspects of an image. However, the manner in which local elements are integrated into a global whole is atypical, with rela...

متن کامل

What makes counting count? Verbal and visuo-spatial contributions to typical and atypical number development.

Williams Syndrome (WS) is marked by a relative strength in verbal cognition coupled with a serious impairment in non-verbal cognition. A strong deficit in numerical cognition has been anecdotally reported in this disorder; however, its nature has not been systematically investigated. Here, we tested 14 children with WS (mean age=7 years 2 months), 14 typically developing controls individually m...

متن کامل

Strategies and biases in location memory in Williams syndrome.

Individuals with Williams syndrome (WS) demonstrate impaired visuo-spatial abilities in comparison to their level of verbal ability. In particular, visuo-spatial construction is an area of relative weakness. It has been hypothesised that poor or atypical location coding abilities contribute strongly to the impaired abilities observed on construction and drawing tasks [Farran, E. K., & Jarrold, ...

متن کامل

Memory abilities in Williams syndrome: dissociation or developmental delay hypothesis?

Williams syndrome (WS) is a neurodevelopmental genetic disorder often described as being characterized by a dissociative cognitive architecture, in which profound impairments of visuo-spatial cognition contrast with relative preservation of linguistic, face recognition and auditory short-memory abilities. This asymmetric and dissociative cognition has been also proposed to characterize WS memor...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009