Williams Syndrome Research Paper

1.    Williams Syndrome: Genetic Profile

Williams syndrome  is a rare  developmental  disorder and occurs in approximately 1 in 20,000 live births. It is caused  by a submicroscopic deletion  on  chromosome 7q11.23. The deleted  region  contains  some 20 genes, about 17 of which have been identified (Ewart et al. 1993, Frangiskakis et al. 1996, Meng  et al. 1998, Tassabehji  et al. 1996). Only a few of these genes are expressed  in  the  brain  and  are  therefore  of  special interest  to the cognitive neuroscientist. Others  affect the physical development of patients, particularly with respect to impairments  to the cardio-vascular system.

Initial  excitement  came from the discovery that  in the majority  of cases of WS one copy of the elastin gene (ELN)  (Ewart  et al. 1993) and  one copy of the Limkinase1  gene (LIMK1)  (Frangiskakis et al. 1996, Tassabehji et al. 1996) were consistently deleted. ELN is important for  elasticity  of the  heart,  skin,  blood vessels, and  lungs. Its deletion  was therefore  rapidly linked to the facial dysmorphology and supra-valvular aortic  stenosis  (SVAS) found  consistently  in individuals with WS (Ewart et al. 1993). LIMK1 is expressed in the developing brain and its deletion was claimed to explain  the  typical  pattern  of  spatial  impairments found  in  the  cognitive  profile  of  such  individuals (Frangiskakis et al. 1996, Mervis et al. 1999). Despite the  claims  about   the  role  in  WS  of  the  ELN  and LIMK1 genes in  facial  dysmorphology and  spatial cognition  respectively,  direct  one-to-one  genotype/phenotype  mappings  are highly unlikely in cognitive neuroscience, as a subsequent study by Tassebehji and collaborators showed (Tassabehji  et al. 1996). Three patients  with SVAS were examined  who had  partial deletions on chromosome 7 in the same region as the clinical  groups   with  WS.  The  study   showed  that despite the ELN and LIMK1 deletions, none of these patients  had the facial dysmorphology typical of WS, nor did they display the uneven WS cognitive profile of impaired  visuo-spatial   cognition  and  enhanced  linguistic  capacities.   They  all  had  an  even  cognitive profile within the normal  range.  The results indicate that the ELN deletion does not alone explain the facial dysmorphology found  in WS. They also suggest that the  LIMK1 deletion  is either  irrelevant  to  the  development  of spatial  cognition  or that  its expression interacts with a number of other genes to contribute to the spatial impairment. But it is clear that one-to-one mapping  between specific genes and higher cognitive outcomes  does not hold.

2.    Williams Syndrome: Phenotypic Profile

The pioneering work of Bellugi and her collaborators initially pointed  to some clear-cut dissociations  in the cognitive architecture of WS. Language  and face processing appeared to be surprisingly preserved in the face  of  both   general  retardation  and   particularly serious  problems  with visuo-spatial  cognition,  number,  planning,   and  problem   solving  (Bellugi  et  al. 1994). Researchers  in the field of WS have been fairly cautious   about   their   claims,   referring   to   relative strengths  and  weaknesses rather  than  absolute  ones (Bellugi et al. 1999, Karmiloff-Smith 1998, Klein  & Mervis 1999, Mervis 1999, Tager-Flusberg et al. 1998, Vicari et al. 1996, Volterra  et al. 1996). By contrast, secondary  sources  cited in writings  by linguists,  developmental  psychologists,  neuropsychologists of adult brain damage, and philosophers have often used Williams syndrome to bolster claims about innate and independently functioning   modules,  some  of  which are  purportedly  intact   and   others   impaired   (e.g., Pinker 1994, 1999). This stems from the view that the pattern of behavioral  performance found in the phenotypic  outcome  is a direct  window  on the purported innately specified, modular structure of the cognitive architecture of the brain (Baron-Cohen 1998, Leslie 1992, Temple 1997). Such reasoning  treats  the genetically impaired brain as if it were a normal brain with parts intact and parts impaired, ignoring the dynamic  role of genetic mutation in interaction with environmental input in fostering overall brain growth. This  has  been  particularly the  case with  studies  of autism and Williams syndrome, in which cognitive impairments  in older  children  and  adults  have been used  to  make  claims  about  gene expression,  in the absence of studies of the starting  state in infants.

Recent studies of infants, children, and adults with WS strongly  suggest that  the starting  state cannot  be simply inferred from the phenotypic  outcome  (Paterson et al. 1999). Infants  with WS were compared  with chronological (CA)  and  mental-age  (MA)  matched infants with Down syndrome (DS), as well as MA and CA-matched typically  developing  controls.   Despite the  fact  that  WS adults  perform  significantly  better than DS adults on vocabulary tasks, WS infants are as seriously  impaired  on  vocabulary tasks  as  are  DS infants.  Moreover,  despite WS adults  having  significantly  worse problems  with judging  numerosities  in adulthood than  DS  adults  and  control  groups,  WS infants  perform  normally,  like the CA-controls, and significantly  better  than  DS infants.  In other  words, the patterns  obtaining in infancy  turn  out  to be the opposite  of the patterns  in adulthood, pointing to the importance of the dynamics  of developmental  trajectories over time, rather  than a static view of the infant starting  state and the phenotypic  outcome.

Even in cases where the phenotypic  outcome  seems to display spared performance, in-depth analyses suggest   that   people   with   WS  process   inputs   via different cognitive processes (Karmiloff-Smith 1998). In the domain  of vocabulary, individuals with WS do not   obey  the  same  lexical  constraints  as  normal controls  when  learning  new  words.  In  the  domain of syntax, WS adults tend to display patterns typical of much younger children rather than intact performance (Klein  and  Mervis  1999). For  example,  it has  been claimed that  individuals  with WS have intact  regular past tense formation of verbs, alongside impaired associative  lexical processes  (Clahsen  and  Almazan 1998). Recent  research  challenges  this  claim in that once  verbal  mental  age  is taken  into  account,   WS patients  display no selective deficit for irregular  verbs (Thomas et al. 2001). Specifically, the WS data can be placed on the normal  developmental  pathway  found in much  younger  subjects.  These various  results  are consistent  with the hypothesis  that  the WS language system is seriously delayed because it has developed under different constraints.

WS  language   is  not   simply   delayed,   however. Several studies now suggest that there is an imbalance in WS at different times in development between phonology  and semantics (Karmiloff-Smith 1998, Thomas  et al. 2001). For  example, when WS participants monitor sentences for a target word, they do not show sensitivity to certain  sentential  violations,  suggesting that in WS semantic information may become available  too slowly to be integrated  with the online processing of syntax. A study of reading in WS came to similar conclusions about the weak role of semantics in  learning   to   read   new  words.   The   WS  group displayed equal levels of reading for both concrete and abstract words. By contrast, the controls  found  concrete, imageable words much easier to read. In general, imageability  effects have been shown to be weaker in people  with  WS  (Karmiloff-Smith  1998).  A  more recent study by Vicari and his colleagues demonstrated that,  compared  to normal  controls,  word learning  is superior in WS if the auditory  presentation of a word is accompanied by the simultaneous presentation of a photograph depicting  the object (Vicari et al. 2000). This  seems  to  be  because,  unlike  normal  controls, people with WS are defective in spontaneously forming a visual (semantic)  image of auditorily  presented words. Finally, when learning new spoken words and despite  a vocabulary test  age of 8, people  with  WS behave like 4–5 year olds, and do not show the pattern typical  from  6 years  onwards  in the  normal  population. Like very young children, WS patients  are less influenced by the semantics of the real words that the nonce terms resemble. Rather, they rely more on phonology.  Taken   together,   a  variety   of  studies suggest that, unlike typical development, phonological representations are  at  times  stronger  than  semantic representations in their influence on the way in which WS language develops.

Although it is becoming increasingly clear that vocabulary development  does  not  follow  a  normal developmental  trajectory in WS and  that  semantics places  a  weaker  constraint on  WS  language  development  than  in typical  controls,  it remains  possible that  WS syntax is intact,  as many have claimed (e.g., Pinker 1994, 1999). There are, however, a number  of lines of evidence to doubt this. First, vocabulary levels are  usually  better   than   syntactic  levels  in  WS  on various standardized tasks, although both are significantly   below   chronological  age   (Karmiloff-Smith 1998). Second,  even in very simple  imitation  tasks, participants with WS show impairment with complex syntactic constructions such as embedded relative clauses. These and various  other  findings in different laboratories are hardly  consistent  with the view that WS  syntax  is intact.  Even  in  an  area  of  relatively simple  syntax—grammatical concord  over  sentence elements—which young, normal French-speaking children  acquire  easily  and  early,  people  with  WS show impairment even in adulthood (Karmiloff-Smith 1998). Although the WS children learn the local gender marker (correct article) for a nonce term easily (in fact, more easily than  control  children),  their capacity  for gender agreement across sentence elements such as agreement on adjectives or pronouns is seriously impaired.  Studies  of  Italian-speaking children  have also revealed that  grammatical gender is a particular problem,  with  WS children  displaying  errors  never encountered in normal  development  (Volterra  et al. 1996). Several studies  (e.g., Klein  and  Mervis  1999) now suggest that  the problems  that  people with WS have with semantics and syntax are often camouflaged by their good verbal memory. This again demonstrates that overt behavior  is not necessarily an index of underlying  cognitive competence.

A similar conclusion  holds for the domain  of face processing. Despite reports that WS face processing is intact (Bellugi et al. 1994, Rossen et al. 1996), in-depth studies of face processing suggest that individuals with WS use different strategies from normal controls (Karmiloff-Smith 1998). Several studies  (Deruelle  et al. 1999, Karmiloff-Smith 1998, Udwin & Yule 1991) have replicated earlier work, revealing normal or near normal  behavioral  scores on standardized tasks  like the  Benton  Facial  Recognition Test  (Benton  et  al. 1983) and  the Rivermead  Behavioural  Memory  Test (Wilson et al. 1985). However, these studies have seriously challenged the notion that the behavioral success displayed  in WS face processing  capacities  is subserved by normal  cognitive processes. Where normal controls  use predominantly configural  strategies for processing upright faces and featural strategies for processing inverted faces, the WS patients tend to process   the   featural   details   of  both   upright   and inverted  faces.  These  different  strategies  have  been further  explored in brain imaging studies (Mills et al. 2000). When older children and adults with WS have to match faces in an event related potential study, they display temporal  processes found at no age in normal controls. They also tend to process faces bilaterally or predominantly  with   the   left  hemisphere,   whereas normal  controls  show  a  right  hemisphere  bias  for faces.  All  of  these  findings  point   to  atypical  face processing   in  WS,  despite  the  normal   behavioral scores.

3.    Concluding Comments

Developmental cognitive neuroscience  must take development  seriously.  The  WS brain  is 20%  smaller than normal brains and qualitatively  different in terms of brain anatomy (Bellugi et al. 1999), brain chemistry (Rae et al. 1998), and computational processing (Mills et al. 2000). This holds throughout embriogenesis and postnatal brain  development  and  means  that  interaction with environmental stimuli will be subtly different. Given a very different brain, it is unsurprising that  even when  overt  behavior  seems normal,  as in some  aspects  of  WS language  and  face processing, these  skills actually  turn  out  to  be underpinned  by cognitive processes that are different from the normal case (Karmiloff-Smith 1998).

Williams  syndrome  is an  excellent  model  for  the neurocognitive  study of genetic disorders,  because of its strikingly unusual  cognitive profile. The syndrome is especially important because of the way in which in-depth research highlights the need to go beyond both observable  behavior  and  static  descriptions  of snapshots of developmental  outcomes,  to the charting  of neurocognitive  trajectories  from infancy onwards.

Bibliography:

1. Baron-Cohen S 1998 Modularity in developmental   cognitive neuropsychology: Evidence from autism and gilles de la Tourette syndrome.  In: Burack  J A, Hodapp R M, Zigler E (eds.)  Handbook  of  Mental   Retardation  and  Cambridge  University  Press, Cambridge,  UK,  p. 335
2. Bellugi U, Wang P P, Jernigan T 1994 Williams Syndrome:  An unusual  neuropsychological profile.  In: Broman  S H,  Grafman  J  (eds.)  Atypical   Cognitive  Deficit  in  Developmental Disorders: Implications for Brain Function. Lawrence Erlbaum Associates Inc., Hillsdale, NJ, pp. 23–56
3. Bellugi U, Lichtenberger L, Mills D, Galburda A, Korenberg J R 1999 Bridging cognition the brain and molecular genetics: evidence from  Williams     Trends  in  the  Neurosciences 22(5): 197–207
4. Benton A L,  Hamsher   K,  de S Varney  N R,  Spreen  O  1983 Contributions to Neuropsychological Assessment. Oxford University Press, New York
5. Clahsen H,  Almazan   M  1998  Syntax   and   morphology  in Williams syndrome.  Cognition 68(3): 167–98
6. Deruelle C, Mancini J, Livet M O, Casse-Perrot C, de Schonen S 1999 Configural  and  local processing  of faces in children with Williams syndrome.  Brain and Cognition 41: 276–98
7. Ewart A K,  Morris  C A, Atkinson  D,  Weishan  J, Sternes  K, Spallone   P,  Stock  D A,  Leppert   M,  Keating   M T  1993 Hemizygosity at the elastin locus in a developmental  disorder Williams syndrome.  Nature Genetics 5: 11–6
8. Frangiskakis J M, Ewart A, Morris C A, Mervis C B, Bertrand J, Robinson B F,  Klein  B P, Ensing  G, Everett  L A, Green E D,  Proschel  C,  Gutowski  N J,  Noble  M,  Atkinson  D L, Odelberg  S J, Keating  M T 1996 LIM-kinase1 hemizygosity implicated  in  impaired  visuospatial   constructive   Cell 86: 59–69
9. Karmiloff-Smith A 1998 Development itself is the key to understanding developmental   Trends in Cognitive Sciences 2: 389–98
10. Klein B P, Mervis C B 1999 Contrasting patterns of cognitive abilities  of 9 and  10-years-olds  with  Williams  syndrome  or Down       Developmental   Neuropsychology   16(2): 177–96
11. Leslie A M  1992  Pretence   autism   and   the  theory-of-mind-module. Current Directions in Psychological Science 1: 18–21
12. Meng X, Lu X, Zhizhong L, Green E D, Massa  H, Trask  B J, Morris C A, Keating M T 1998 Complete physical map of the common deletion region in Williams syndrome and identification  and  characterization of three  novel genes. Human Genetics 103: 590–9
13. Mervis C B  1999  The  Williams  syndrome   cognitive  profile: Strengths weakness and interrelations among auditory  short-term  memory  language  and  visuospatial   constructive     In: Winograd  E, Fivush  R, Hirst  W (eds.) Ecological Approaches to Cognition: Essays in Honor of Ulric Neisser. Lawrence Erlbaum Associates Inc. Mahwah,  NJ, pp. 193–227
14. Mervis C B,  Morris   C A,  Bertrand  J,  Robinson  B F  1999 Williams syndrome: Findings  from an integrated  program of research.   In:  Tager-Flusberg  H  (ed.)  Neuro-Developmental Disorders. MIT Press, Cambridge,  MA, pp. 65–110
15. Mills D L, Alvarez T D, St. George M, Appelbaum L G, Bellugi U, Neville H 2000 Electrophysiological studies of face processing in Williams syndrome. Journal of Cognitive Neuroscience 12(11): 47–64
16. Pani J R,  Mervis  C B,  Robinson  B F  1999  Global   spatial organization by individuals with Williams syndrome. Psychological Science 10(5): 453–8
17. Paterson S J, Brown J H, Gsodl M K, Johnson M H, KarmiloffSmith  A 1999 Cognitive  modularity and  genetic  Science 286(5448): 2355–8
18. Pinker S 1994 The Language Instinct. Penguin, Harmondsworth, UK
19. Pinker S 1999 Words and Rules. Weidenfeld & Nicolson, London
20. Pober B R, Dykens E M 1996 Williams syndrome: An overview of medical cognitive  and  behavioural features.  Mental  Retardation 5: 929–43
21. Rae C,  Karmiloff-Smith A,  Lee M A,  Dixon  R M,  Grant   J, Blamire  A M,  Thompson C H,  Styles P,  Radda G K  1998 Brain biochemistry in Williams syndrome: Evidence for a role of the cerebellum in cognition? Neurology 51: 33–40
22. Rossen M,  Klima  E S,  Bellugi  U,  Bihrle  A,  Jones  W  1996 Interaction between language  and  cognition:  Evidence from Williams        In:    Beitchman    J H,    Cohen    J H, Konstantareas M M,  Tannock R  (eds.) Language Learning and Behavior Disorders: Developmental Biological and Clinical Perspectives.  Cambridge   University  Press,  New  York,  pp. 367–92
23. Tager-Flusberg H, Boshart J, Baron-Cohen S 1998 Reading the windows to the soul: Evidence of domain-specific sparing  in Williams     Journal  of  Cognitive Neuroscience  10: 631–9
24. Tassabehji M,  Metcalfe  K,  Fergusson   W D,  Carette   M J A, Dore  J K, Donnai  D, Read  A P, Proschel C, Gutowski  N J, Mao X, Sheer D 1996 LIM-kinase deleted in Williams syndrome.  Nature Genetics 13: 272–3
25. Tassabehji M,  Metcalfe  K,  Karmiloff-Smith A,  Carette  M J, Grant  J, Dennis N, Reardon W, Splitt M, Read A P, Donnai D 1999 Williams syndrome:  Use of chromosomal microdeletions as a tool to dissect cognitive and  physical phenotypes. American Journal of Human Genetics 64: 118–25
26. Temple C M    1997   Cognitive    neuropsychology   and    its applications to    Journal  of  Child  Psychology  and Psychiatry 38: 27–52
27. Thomas M S C, Grant J, Gsodl M, Laing E, Barham Z, Lakusta L, Tyler L K, Grice S, Paterson  S, Karmiloff-Smith A 2001 Can atypical phenotypes  be used to fractionate the language system?  The  case  of  Williams      Language  and Cognitive Processes 16(2  3): 143–76
28. Udwin O, Yule W 1991 A cognitive and behavioural phenotype in Williams syndrome. Journal of Clinical and Experimental Neuropsychology 13: 232–44
29. Vicari S, Carlesimo G, Brizzolara D, Pezzini G 1996 Short-term memory in  children  with  Williams  syndrome:   A  reduced contribution of  lexical-semantic  knowledge  to  word  Neuropsychologia 34: 919–25
30. Vicari S, Bellucci S, Carlesimo G A 2001 Procedural learning deficit in children with Williams syndrome.  Neuropsychologia 39(7): 665–77
31. Volterra V, Capirci  O, Pezzini G, Sabbadini  L, Vicari S 1996 Linguistic abilities in Italian children with Williams syndrome. Cortex 32: 663–77
32. Wilson B, Cockburn J, Baddeley A 1985 Rivermead Behavioural Memory Thames Valley Test Co., Reading,  UK