Williams Syndrome: A Brief Description of the Genotype and Phenotype

Williams syndrome is a neurodevelopmental genetic disorder with a specific physical, behavioral, and cognitive profile. It is caused by a microdeletion on the long arm of chromosome 7, affecting some 28 genes (Donnai & Karmiloff-Smith, Reference Donnai and Karmiloff-Smith2000; Semel & Rosner, Reference Semel and Rosner2005; Tassabehji, Reference Tassabehji2003). In most cases, the deletion is sporadic, although a few cases of parent-to-child transmission and concordant monozygotic twins have been reported (Morris, Thomas, & Greenberg, Reference Morris, Thomas and Greenberg1993; Pankau, Gosch, Simeoni, & Wessel, Reference Pankau, Gosch, Simeoni and Wessel1993). For a long time, the prevalence of WS was estimated to be between 1 in 20,000 and 1 in 50,000 (Morris, Demsey, Leonard, Dilts, & Blackburn, Reference Morris, Demsey, Leonard, Dilts and Blackburn1988), but more recently a Norwegian study has situated the prevalence much higher, at closer to 1 in 7,500 (Strømme, Bjørnstad, & Ramstad, Reference Strømme, Bjørnstad and Ramstad2002). WS was first described by two cardiology groups (Beuren, Apitz, & Harmjanz, Reference Beuren, Apitz and Harmjanz1962; Williams, Barratt-Boyes, & Lowe, Reference Williams, Barratt-Boyes and Lowe1961), both identifying the association of several clinical features in affected individuals: narrowing of the aorta (supravalvular aortic stenosis [SVAS]), distinctive facial dysmorphology, slow physical growth, and learning difficulties. It took another couple of decades before the syndrome started to be extensively investigated by cognitive psychologists and neuroscientists, hoping thereby to gather data that might contribute directly to the question of whether or not the human mind/brain starts out with innately specified, cognitive-level modules (like grammar, face processing, spatial cognition, number, etc.) that operate independently of one another and of general intelligence. The widely held view was, and still is to some extent, that WS involves a juxtaposition of intact and impaired modules, including intact grammar, as the quotation from Pinker at the beginning of this article suggests. But, as we shall see, the real picture of the WS phenotype is far more complex.

At the level of the brain, WS is characterized by structural abnormalities, with total brain volume being about 80% of normal brains (Neville et al., Reference Neville, Mills, Bellugi, Broman and Grafman1994). Once overall brain volume is controlled for, structural imaging studies point to specific reductions in parietal cortex (Eckert et al., Reference Eckert, Hu, Eliez, Bellugi, Galaburda and Korenberg2005) as well as in the brainstem (Reiss et al., Reference Reiss, Eliez, Schmitt, Straus, Lai and Jones2000). The corpus callosum is particularly small (Schmitt, Eliez, Warsofsky, Bellugi, & Reiss, Reference Schmitt, Eliez, Warsofsky, Bellugi and Reiss2001) but primary auditory cortex is proportionally large, with atypical cytoarchitecture (Holinger et al., Reference Holinger, Bellugi, Mills, Korenberg, Reiss and Sherman2005). Although the cerebellum is also large in WS brains (Jones et al., Reference Jones, Hesselink, Courchesne, Duncan, Matsuda and Bellugi2002), this is an area in which we identified abnormal brain biochemistry (Rae et al., Reference Rae, Karmiloff-Smith, Lee, Dixon, Grant and Blamire1998). Atypical neuronal size, orientation, and density have also been identified (Galaburda, Holinger, Bellugi, & Sherman, Reference Galaburda, Holinger, Bellugi and Sherman2002), as has increased cortical gyrification in the right parietal and occipital regions and the left frontal areas (Schmitt et al., Reference Schmitt, Watts, Eliez, Bellugi, Galaburda and Reiss2002). Gray-matter volume is close to that found in normal controls, but white-matter volume is reduced (Reiss et al., Reference Reiss, Eliez, Schmitt, Straus, Lai and Jones2000). Finally, the hippocampus in WS brains appears to have an atypical shape (Meyer-Lindenberg, Mervis, et al., Reference Meyer-Lindenberg, Hariri, Munoz, Mervis, Mattay and Morris2005). Apart from these structural abnormalities, other studies using functional magnetic resonance imaging (fMRI; Meyer-Lindenberg, Hariri, et al., Reference Meyer-Lindenberg, Mervis, Sarpal, Koch, Steele and Kohn2005) and high-density event-related potentials (ERPs; Grice et al., Reference Grice, Michael, Karmiloff-Smith, Hanife, Csibra and de Haan2001, Reference Grice, de Haan, Hanife, Johnson, Csibra and Grant2003; Mills et al., Reference Mills, Alvarez, George, Appelbaum, Bellugi and Neville2000) point to atypical functioning in the WS brain. It is worth noting, however, that most of the studies of the WS brain structure and function have been based on adult brains. We have as yet little understanding of the developmental mechanisms that yield these brain abnormalities. This is obviously vital when attempting to account for language acquisition in a developmental disorder like WS.

The Importance of Tracing Full Developmental Trajectories

Although research into normal language acquisition has focused on infant precursors, it is still relatively rare to find such a focus in work on developmental disorders. Yet, in previous research we have shown that the profile of proficiencies and deficits in infancy can be very different from the subsequent profiles in adolescence and adulthood (Karmiloff-Smith, Reference Karmiloff-Smith1998; Paterson, Brown, Gsödl, Johnson, & Karmiloff-Smith, Reference Paterson, Brown, Gsödl, Johnson and Karmiloff-Smith1999). Pinker's (1999) aforementioned claim of “spared” language in WS is based on adult data where some individuals with WS do display surprisingly proficient (albeit not intact) language abilities by adolescence or adulthood. But if language were spared in this clinical group, as Pinker and others claim, why is its onset so frequently delayed? Is it just a case of the delayed maturation of a “language module,” or is there a developmental explanation?

In fact, many interacting delays and deviances are contributing factors. For example, infants with WS are extremely delayed in the onset of babbling (Masataka, Reference Masataka2001) and in the segmentation of the speech stream (Nazzi, Paterson, & Karmiloff-Smith, Reference Nazzi, Paterson and Karmiloff-Smith2003). Curiously, unlike typically developing children, production precedes referential pointing in WS (Mervis & Bertrand, Reference Mervis, Bertrand, Adamson and Romski1997), suggesting that early productive language may not be referential in nature. Comprehension does not show the usual advantage over production in this clinical group (Paterson et al., Reference Paterson, Brown, Gsödl, Johnson and Karmiloff-Smith1999). Moreover, despite their fascination with faces, toddlers with WS do not follow the eye gaze of their interlocutors, nor do they pay attention to the referential targets of pointing (Laing, Hulme, Grant, & Karmiloff-Smith, Reference Laing, Butterworth, Ansari, Gsödl, Longhi and Panagiotaki2002). And, even when they have greater vocabularies than their healthy controls, they show the opposite pattern from the controls and rely more on perceptual cues than on linguistic labels (Nazzi, Gopnik, & Karmiloff-Smith, Reference Nazzi, Gopnik and Karmiloff-Smith2005), pointing to an synchrony in the cognitive and lexical development of young children with WS (see also Nazzi & Karmiloff-Smith, Reference Nazzi and Karmiloff-Smith2002). In other words, the early stages of language acquisition in WS are not spared; multiple causes contribute to its late onset and atypical trajectory. This highlights how important it is to trace developmental trajectories across time, ideally longitudinally (Annaz, Reference Annaz2006; Annaz, Karmiloff-Smith, & Thomas, Reference Annaz, Karmiloff-Smith, Thomas, Reed and Warner Rogers2008; Ansari & Karmiloff-Smith, Reference Ansari and Karmiloff-Smith2002; Karmiloff-Smith, Reference Karmiloff-Smith1992, Reference Karmiloff-Smith1998, Reference Karmiloff-Smith2007; Thomas et al., Reference Thomas, Grant, Barham, Gsödl, Laing and Lakusta2001, in press).

The Development of Language in Williams Syndrome

In spite of claims to the contrary (Bellugi, Wang, & Jernigan, Reference Bellugi, Wang, Jernigan, Broman and Grafman1994; Piatelli-Palmarini, Reference Piatelli-Palmarini2001; Pinker, Reference Pinker1994; Rossen, Klima, Bellugi, Bihrle, & Jones, Reference Rossen, Klima, Bellugi, Bihrle, Jones, Beitchman, Cohen, Konstantareas and Tannock1996; Smith & Tsimpli, Reference Smith and Tsimpli1995), it is questionable whether any domain of language, for example, syntax, semantics, phonology, or pragmatics, is intact in WS. Indeed, many empirical studies from different laboratories across the world have challenged intactness claims in respect of all aspects of WS language (e.g., the lexicon: Jarrold, Hartley, Phillips, & Baddeley, Reference Jarrold, Hartley, Phillips and Baddeley2000; Stevens & Karmiloff-Smith, Reference Karmiloff-Smith1997; Temple, Almazan, & Sherwood, Reference Temple, Almazan and Sherwood2002; morphosyntax: Grant, Valian, & Karmiloff-Smith, Reference Grant, Valian and Karmiloff-Smith2002; Karmiloff-Smith et al., Reference Karmiloff-Smith, Grant, Berthoud, Davies, Howlin and Udwin1997; Stojanovik, Perkins, & Howard, Reference Stojanovik, Perkins and Howard2004; Thomas et al., Reference Thomas, Grant, Barham, Gsödl, Laing and Lakusta2001; Volterra, Capirci, & Caselli, Reference Volterra, Capirci and Caselli2001; phonology: Grant et al., Reference Grant, Karmiloff-Smith, Gathercole, Paterson, Howlin and Davies1997; Laing et al., Reference Laing, Hulme, Grant and Karmiloff-Smith2001; pragmatics: Laws & Bishop, Reference Laws and Bishop2004). Clahsen and Almazan (Reference Clahsen and Almazan1998) tried to retain the innateness hypothesis, but for morphosyntax only instead of a whole language module. They argued for a dissociation in WS between impaired vocabulary but intact grammar. However, their claims are questionable because they were based on a very small sample of children with WS (N = 2 for MA = 5 years, and N = 2 for MA = 7 years, with considerable individual variation between the few participants). By contrast, Thomas et al. (Reference Thomas, Grant, Barham, Gsödl, Laing and Lakusta2001) conducted a much broader and in-depth study using the same tasks and compared the performance of 21 individuals with WS with that of four groups of typically developing controls at ages 6, 8, 10 and adult. Thomas and colleagues argued that it is not sufficient to demonstrate that one aspect of language is relatively poorer than another, because this also obtains at younger stages in typical development. Rather, it is necessary to demonstrate that the level of a specific aspect of language is poorer that would be expected in WS, given their overall level of language development. The Thomas et al. study further reported that when verbal mental age was controlled for, the WS group was not only generally impaired, but displayed no selective deficit across vocabulary and morphosyntax. In a subsequent study of rapid naming in WS (Thomas et al., Reference Thomas, Dockrell, Messer, Parmigiani, Ansari and Karmiloff-Smith2006), the results also pointed to atypicality in WS: slower and less accurate naming in the clinical group compared with both chronologically age matched and receptive vocabulary age-matched controls. In fact, Mervis and colleagues (e.g., Klein & Mervis, Reference Klein and Mervis1999) have concluded that the best way to characterize WS language is that it is delayed, reflecting patterns typical of considerably younger children. Indeed, a meta-analysis of a large number of language studies reveals that WS language is neither intact nor at the level expected for chronological age; rather, it is often no better than would be expected for nonverbal mental age (Brock, Reference Brock2007).

A number of other empirical findings suggest that the WS language system is not only delayed but actually develops along a different trajectory compared to typical controls (Karmiloff-Smith et al., Reference Karmiloff-Smith, Grant, Berthoud, Davies, Howlin and Udwin1997; Klein & Mervis, Reference Klein and Mervis1999; Mervis, Morris, Bertrand, & Robinson, Reference Mervis, Morris, Bertrand, Robinson and Tager-Flusberg1999; Thomas & Karmiloff-Smith, Reference Thomas and Karmiloff-Smith2003; Thomas et al., Reference Thomas, Grant, Barham, Gsödl, Laing and Lakusta2001, Reference Thomas, Dockrell, Messer, Parmigiani, Ansari and Karmiloff-Smith2006). For instance, unlike healthy controls, young children with WS use pointing after the appearance of naming. Exhaustive sorting follows the vocabulary spurt in WS, rather than preceding it as in typical development (Mervis et al., Reference Mervis, Morris, Bertrand, Robinson and Tager-Flusberg1999). Furthermore, increased iconic gestures accompanied WS children's performance on a picture-naming task compared to the control group of children (Bello, Capirci, & Volterra, Reference Kuhl2004). It has been suggested that sensitivity to the sound patterns of the language may act as a greater constraint on WS language than sensitivity to meaning (Grant et al., Reference Grant, Karmiloff-Smith, Gathercole, Paterson, Howlin and Davies1997; Klein & Mervis, Reference Klein and Mervis1999; Laing et al., Reference Laing, Hulme, Grant and Karmiloff-Smith2001). Overgeneralization of regular plural -s to irregular plurals was reported to be more prevalent in WS compared with typically developing children (Clahsen & Almazan, Reference Clahsen and Almazan2001). Finally, supporting evidence from several studies across different languages (e.g., Karmiloff-Smith et al., Reference Karmiloff-Smith, Grant, Berthoud, Davies, Howlin and Udwin1997, in French; Klein & Mervis, Reference Klein and Mervis1999, in English; Levy & Bechar, Reference Levy and Bechar2003, in Hebrew; Lukacs, Pleh, & Racsmany, Reference Lukacs, Pleh, Racsmany, Bartke and Siegmüller2004, in Hungarian; Vicari, Carlesimo, Brizzolara, & Pezzini, Reference Vicari, Carlesimo, Brizzolara and Pezzini1996; Volterra, Capirci, Pezzini, Sabbadini, & Vicari, Reference Volterra, Capirci, Pezzini, Sabbadini and Vicari1996; and Volterra, Caselli, Capirci, Tonucci, & Vicari, Reference Volterra, Caselli, Capirci, Tonucci and Vicari2003, in Italian) suggest that the problems that individuals with WS have with semantics, morphosyntax, and pragmatics are often camouflaged by their good verbal memory, challenging the popular belief that their language is intact.

Taken together, these and many other studies point to the fact that individuals with WS follow a deviant developmental trajectory while acquiring language as children as well as when processing language as adults. Behaviorally WS language may appear to be proficient, but cognitively it seems increasingly likely that it involves different cognitive processes from the language of typically developing individuals. This is also true of the brain. A recent study (Hsu, Karmiloff-Smith, Tzeng, Chin, & Wang, Reference Hsu, Karmiloff-Smith, Tzeng, Chin and Wang2007) of semantics in WS memory reported that while the brains of normal adults differentiated between semantically related words and semantically unrelated words, the brains of individuals with WS did not. Moreover, unlike normal adults, the brains of individuals with WS showed a difference between previously presented words and semantically related words. These findings suggest that the WS proficient performance at the behavioral level is underpinned by a deviant neural pathway, which highlights the need to go beyond behavioral measures and to examine brain processes to understand language in this and other clinical groups.

The Importance of Cross-Domain Studies

Multiple contributing factors to the delay in the onset of language in WS have been listed here in the previous section, on the development of language in WS. But they were all within the general domain of communication. Should scientists merely probe closely related domains when endeavoring to understand atypical development? Again, recalling the early interconnectivity of cortical regions, it is wise not to limit research to the specific areas of language and communication when trying to seek the causes of atypically developing language. It is vital for understanding higher-level linguistic phenotypes to explore how they originated in low-level processing mechanisms early on in development. For example, we have demonstrated a serious deficit in visual saccadic planning in WS (Brown et al., Reference Brown, Johnson, Paterson, Gilmore, Gsödl and Longhi2003), which leads to atypical triadic interaction (Laing et al., Reference Laing, Butterworth, Ansari, Gsödl, Longhi and Panagiotaki2002), thereby altering the way in which toddlers with WS learn vocabulary. So, not only should we explore the communication-relevant contributions to a delay in language, but we should also examine other domains like vision, memory, and attention. Timing plays a critical role in development, particularly if one considers how different domains may interact at different moments across the developmental trajectory. So, something that might at first blush seem irrelevant to the acquisition of grammar, that is, saccadic eye movement planning, may have been critical at the earlier stage of development for subsequent language acquisition.

The myth of intact language in WS is also due to a frequent slippage in which relative differences in abilities are often reported as absolute claims. The fact that a clinical group does better at, say, vocabulary tasks than at spatial tasks, does not confirm the claim that vocabulary is intact, spared, or preserved, and spatial cognition impaired. In many instances, both verbal and nonverbal abilities are below chronological age; it is simply that one is less impaired than the other. And, even when scores fall in the so-called normal range, it is crucial to ascertain whether the behavior may be underpinned by atypical cognitive and brain processes (Karmiloff-Smith, Reference Karmiloff-Smith1998; Karmiloff-Smith et al., Reference Karmiloff-Smith, Thomas, Annaz, Humphreys, Ewing and Brace2004).

The Importance of Cross-Syndrome Comparisons

Interestingly, unlike infants and toddlers with WS, those with Down's syndrome (DS) are not impaired in their planning of saccadic eye movements (Brown et al., Reference Brown, Johnson, Paterson, Gilmore, Gsödl and Longhi2003), yet they are just as delayed in their vocabulary development as are infants and toddlers with WS (Paterson et al., Reference Paterson, Brown, Gsödl, Johnson and Karmiloff-Smith1999). Thus, the reasons for the similar delay and similar overall IQ in these two syndromes must be different, pointing to the importance of tracing full cross-syndrome developmental trajectories. And, in exploring cross-syndrome similarities and differences, it is important to analyze not simply success and failure, but also potential and informative differences in error patterns (Scerif, Cornish, Wilding, Driver, & Karmiloff-Smith, Reference Scerif, Cornish, Wilding, Driver and Karmiloff-Smith2004).