Elsevier

Brain and Language

Volume 120, Issue 3, March 2012, Pages 345-359
Brain and Language

Exceptional lexical skills but executive language deficits in school starters and young adults with Turners syndrome: Implications for X chromosome effects on brain function

https://doi.org/10.1016/j.bandl.2011.12.001Get rights and content

Abstract

TS school starters had enhanced receptive and expressive language on standardised assessment (CELF-P) and enhanced rhyme judgements, spoonerisms, and lexical decision, indicating enhanced phonological skills and word representations. There was marginal but consistent advantage across lexico-semantic tasks. On executive tasks, speeded naming of numbers was impaired but not pictures. Young TS adults had enhanced naming and receptive vocabulary, indicating enhanced semantic skills. There were consistent deficits in executive language: phonemic oral fluency, rhyme fluency, speeded naming of pictures, numbers and colours; sentence completion requiring supression of prepotent responses.

Haploinsufficiency of X-chromosome drives mechanisms that affect the anatomical and neurochemical development of the brain, resulting in enhanced temporal lobe aspects of language. These strengths co-exist with impaired development of frontal lobe executive language systems. This means not only that these elements of language can decouple in development but that their very independence is driven by mechanisms linked to the X-chromosome.

Highlights

► TS school starters had enhanced receptive and expressive language. ► TS school starters had enhanced phonological skills. ► TS adults had enhanced naming and vocabulary but impaired executive language. ► Haploinsufficiency of X leads to enhanced temporal but impaired frontal language. ► Independence of lexico-semantic and executive language is driven by X-chromosome.

Introduction

Turner’s syndrome [TS] is a genetic disorder in females, compatible with normal intelligence, with the majority of girls integrated into mainstream schools and adults assimilated within the normal adult population. Despite normal intelligence, cognitive strengths and weaknesses have been described, particularly weak visuo-spatial skills. However, it is language that is the focus of the current paper. Here, there has been evidence of exceptional strength in middle childhood in vocabulary and reading (Temple, 2002, Temple and Carney, 1996). The current paper explores evidence for any verbal advantage in school starters and young adults. It also explores executive aspects of language in the context of previous reports of weak executive skills in TS.

The incidence of TS is 1/2000–2500 live female births (Jacobs et al., 1990), which is approximately comparable to that of autism, yet it has received much less systematic investigation. Instead of the usual pattern 46XX, the second X chromosome is deleted or abnormal (Ford, Jones, Polini, de Almeida, & Briggs, 1959). In approximately half of cases the karyotype is 45XO in all sampled cells, with the other half having mosaicism, partial deletions, isochromosome of X, or ring X, in which broken ends of an X chromosome join to form a ring. In reality, a 2nd cell line may be present in all cases of TS, but not always detectable by conventional cytogenetic techniques, with nonmosaic 45XO being a prenatal lethal condition (Hook & Warburton, 1983). In 70% of cases the intact X chromosome is of maternal origin and in 30% paternal (Jacobs et al., 1990). The physical phenotype associated with the X chromosome haploinsufficiency includes short stature and ovarian dysgenesis (Turner, 1938).

Candidate genes for TS are thought to escape X-inactivation, and cytogenetic and molecular studies implicate the short arm of the X chromosome. Studying nonmosaic cases of TS with partial deletions, a critical area of Xp11.2–p22.1 was identified (Zinn et al., 1998). In a further study of Xp deletions, half had the phenotype of TS, including six cases where the deletion involved only Xp22.3 (Ross, Roeltgen, Kushner, Wei, & Zinn, 2000). In contrast two cases, with interstitial Xp deletions that spared distal Xp22.3 did not have the phenotype. There was no relationship between either small stature or ovarian function and neuropsychological phenotype, suggesting that it has a genetic basis, independent of hormonal influences.

TS is therefore a genetic disorder that enables insight into the neuropsychological impact of haploinsufficiency within a portion of the short arm of the X chromosome, with implications for knowledge of X chromosome effects on normal brain function. TS also provides evidence concerning the limitations of biological and functional plasticity in the context of explicit genetic drivers.

In twins discordant for TS, magnetic resonance imaging (MRI) indicated reduced grey matter volume in right parieto-occipital and left parieto-perisylvian areas, in the twin with TS compared to her unaffected twin. There was also reduced grey matter volume in the twin with TS in the right frontal area (Reiss et al., 1993). Further reports detail reduced volume of parieto-occipital cortex (Brown et al., 2002, Cutter et al., 2006 [though see Raznahan et al., 2010], Marzelli et al., 2011, Murphy et al., 1993), prefrontal cortex (left-sided prefrontal Cutter et al., 2006; bilateral, Molko et al., 2004), hippocampus (right-sided, Cutter et al., 2006, Kesler et al., 2004; bilaterally, Murphy et al., 1993) and temporal lobe (left superior, Molko et al., 2004; inferior volume bilaterally, Cutter et al., 2006). Contrasting reports are of increased volume in amygdala (bilaterally, Good et al., 2003, left-sided, Kesler et al., 2006, Kesler et al., 2004), temporal lobe (left-sided, Marzelli et al., 2011; pole, Molko et al., 2004; bilaterally, Rae et al., 2004) and orbitofrontal cortex (Good et al., 2003, Molko et al., 2004). Reduced volume of the splenium and genu of the corpus callosum are reported (Cutter et al., 2006, Fryer et al., 2003), though also increased volume of the genu (Cutter et al., 2006). There is reduced volume of occipito-frontal fasciculus on the left (connecting temporal and occipital lobes) (Cutter et al., 2006).

Most MRI studies depend upon manual measurement or voxel-based morphometry but surface-based morphometry [SBM] enables greater sensitivity by looking at both surface area and cortical thickness (Raznahan et al., 2010). Using SBM, decreased surface area but increased cortical thickness in parietal and occipital lobes was reported in TS, so that overall lobe volume was normal (Raznahan et al., 2010). There was reduced gyrification overall yet increased cortical thickness in three areas of prefrontal cortex (including orbitofrontal), inferior temporal and lateral occipital cortex and intraparietal sulcus. Reduced cortical thickness was seen bilaterally in parahippocampal cortex. Raznahan et al. suggested that the X-chromosome is differentially impacting upon mechanisms that affect surface area (cortical folding) and cortical thickness (dendritic pruning / arborisation) with X-linked genes or sex steroids directly impacting upon cortical folding. Overall, there was significantly increased temporal lobe volume bilaterally, confirming the previous results of Rae et al. (2004).

Interpretation of anatomical studies in neuropsychology often assumes that reduced volumes reflect underdevelopment and increased volumes reflect similarly detrimental processes such as failure of pruning and that both have negative consequences. In view of the enhancement of some neuropsychological skills in TS (to be discussed below), the possibility that enlarged anatomical volume, particularly in left temporal areas actually reflects enhanced systems must also to be retained.

Positron Emission Tomography and functional MRI studies in TS support dysfunction in occipitoparietal regions (Clark et al., 2002, Elliott et al., 1996, Haberecht et al., 2001, Hart et al., 2006, Kesler et al., 2004), though Hart et al. (2006) show effects only on a spatial and not a verbal task. Abnormal prefrontal activation has also been reported (Beaton et al., 2010, Haberecht et al., 2001, Hart et al., 2006, Murphy et al., 1997, Tamm et al., 2003).

TS leads to ovarian dysgenesis, with gonial cells undergoing a high rate of atresion after the 3rd embryonic month so that by infancy the ovaries are absent or present merely as vestigial streaks (Polani, 1981). Consequently, during fetal development, the brain develops with reduced exposure to the endogenous sex hormones. This means that TS offers the additional capacity to explore hormonal modulation of genetic influences upon brain development and neuropsychological function. Sex hormones are thought to play a significant role in normal cerebral organisation and cognitive development (Arnold and Gorski, 1984, Clark and Goldman-Rakic, 1989, Denenberg et al., 1988, Goy and McEwen, 1980, Peper et al., 2009, Witte et al., 2010). Of potential recent significance for TS, study of grey matter volume and circulating levels of sex hormones found that left superior parietal volume was positively associated with estrogen levels in normal adults (Witte et al., 2010). Nevertheless, the visuospatial weaknesses that might have been attributed to this regional substrate in TS are thought to be genetically influenced as visuospatial performance is not affected by exogenous hormonal therapy in TS (Ross et al., 2000).

Average IQs in TS are between 95 and 102 (Mazzocco, 2006). Within this, there are consistent reports of lower performance IQ than verbal IQ (e.g. Cornoldi et al., 2001, Downey et al., 1991, Mazzocco, 2006, Shaffer, 1962, Shucard et al., 1992, Temple and Carney, 1993), though Temple and Carney (1993) demonstrated that the discrepancy was greater for the 45XO karyotype, than other karyotypes. There is no consistent difference between girls with TS and their mothers in verbal IQ but girls with TS consistently score at or below their mothers in performance IQ (Mazzocco, 2006).

The dominant focus in exploration of neuropsychological function in TS has been of non-verbal deficits in space form perception (McGlone, 1985, Money, 1963); directional sense (Alexander, Walker & Money, 1964) drawing (Alexander et al., 1966, Temple and Carney, 1995); right–left discrimination (Rovet and Netley, 1981, Shucard et al., 1992); visuo-spatial memory (Bishop et al., 2000, Cornoldi et al., 2001, Pennington et al., 1985, Shucard et al., 1992); locating objects in space (Shucard et al., 1992); memory for matrix positions (Rovet and Netley, 1981, Rovet and Netley, 1982) and visuo-motor integration (McCauley et al., 1987, Temple and Carney, 1995), though there is intact performance on measures of tactile spatial skill and on some traditional measures of visuo-perceptual skill (Mazzocco et al., 2006, Temple and Carney, 1995).

The cognitive phenotype also includes impaired executive functions, though different studies focus upon different tasks (Haberecht et al., 2001, Kirk et al., 2005, Lasker et al., 2007, Romans et al., 1997, Tamm et al., 2003, Temple, 2002, Temple et al., 1996). It has been argued that there is a motor deficit (e.g. Nijhuis-van der Sanden, Eling, & Otten, 2003), though motor response times are intact (Simon et al., 2008). Specific difficulties with mathematics are also common (Alexander and Money, 1966, Borsel et al., 1999, Butterworth et al., 1999, Mazzocco, 2006, Rovet, 1993, Rovet et al., 1994).

In contrast to studies of spatial ability and executive functions in TS, language has been the focus of few investigations (Murphy, 2009). Early studies reported verbal skills that were above average (e.g. Money and Alexander, 1966, Shaffer, 1962). Lahood and Bacon (1985) also found that verbal IQ and subtest scores on vocabulary, verbal comprehension and verbal similarities were significantly above normative expectation. Temple (2002) found elevated levels of vocabulary in middle childhood, and greater understanding of low frequency words e.g. protractor. Bishop et al. (2000) reported elevated verbal memory in some adults with TS. The Bishop et al. (2000) study is frequently misquoted, so it is worth emphasising that at no point in this study are the verbal memory skills of those with TS found to be weaker than those of controls. For some cases, initial verbal recall is elevated but the advantage is not sustained with time. Thus, an initial superiority returns to a normal level. To quote this as reflecting enhanced forgetting and therefore impaired performance is misleading as it implies a deficiency in final performance where there is absolutely none, simply an unsustained superiority over time. In literacy, early studies noted no difficulties (Alexander & Money, 1965). Temple and Carney (1996) found that 8–12 year old children with TS, had elevated reading, with both lexical word recognition and phonological reading significantly better than peers, though Rae et al. (2004) report strength in word recognition but weak phonological reading.

Thus, haploinsufficiency of the X chromosome is linked to verbal strength in word knowledge, initial verbal memory and word recognition in reading. These areas lie within lexico-semantic aspects of language. Strength in reading including long regular and rare words (Temple & Carney, 1996) argues for strength in phonological skills but results are inconsistent (Rae et al., 2004). In relation to general language strength, by the time of adulthood, Borsel et al. (1999) report that of those progressing to advanced education, 43% (16/37) chose training with a strong emphasis on language (e.g. translator/interpreter, speech language pathologist, secretary, communication studies).

Strength in lexico-semantic aspects of language contrasts with consistent reports of weak performance on verbal fluency tasks in TS, often considered to be measures of executive verbal processes (McGlone, 1985, Rae et al., 2004, Reiss et al., 1993, Romans et al., 1998, Romans et al., 1997, Temple, 2002, Temple et al., 1996, Waber, 1979). There is also weak performance, in 8–12 year olds, on the Stroop, another measures of executive verbal processes (Temple et al., 1996) as well as in the production of written narrative, at the age of 10–11 years (Reiss et al., 1993) and on some tasks in 9–12 year olds, involving oral production of narrative, though not all (Temple, 2002). Rapid automatised naming is also significantly slower than normal (Mazzocco, 2001). Thus haploinsufficiency of X also seems linked to areas of weak executive language processes.

Other language deficits are less reliable. Difficulties in reading comprehension were found in one study but not another (Pennington et al., 1982, Temple and Carney, 1996). Inozemtseva, Matute, Zarabozo, and Ramirez-Duenas (2002) reported difficulty in syntactic processing but the task was confounded by using syntactic structures relating to spatial location and position. No reduction in syntactic complexity, as reflected by mean length of utterance was found in schoolchildren (Temple, 2002). There were raised levels of articulation problems, stuttering and delayed language development in those preselected with head and neck problems (Borsel et al., 1999). Reports of early feeding difficulties and articulatory problems may indicate increased oral dyspraxia. This would be consistent with Bender, Puck, Salbenblatt, and Robinson (1986) who reported increased problems of speech production, though here those with language difficulties also had IQs below 75, so there is a sampling issue. Overall, the evidence for non-executive language difficulties is not compelling though increased rates of oral dyspraxia are possible. The studies of language in TS have interest because they report both exceptional talent and strength in specific areas of language and yet deficits in others. Thus, the X chromosome haploinsufficiency is associated with language skills that fractionate both up and down, in comparison to typical development. This has implications for the normal relationship between the X-chromosome and the development of different potentially modular components of language. To date, the data relevant to this issue is limited. Indeed, several of the language results have emerged from studies that were originally designed to explore other issues.

The current study looked specifically at a wider range of tasks addressing phonological, lexico-semantic and executive elements of language, in both school starters with TS and young adults with TS. By looking at both school starters and young adults, it was possible to assess the way in which TS was expressed, as cognitive functions like executive skills continued to unfold with on-going brain maturation. Anomalies of brain function interact with phases of normal brain development but also serve to illuminate their characteristics (Temple, 1997). The comparison of school starters and young adults also enabled assessment of the degree of functional plasticity or compensation possible within the genetic constraints of X chromosome haploinsufficiency.

Previous studies raised questions about the form of language disorder and/or strength that results from haploinsufficiency of X (Temple, 2002). Studies of children in middle childhood indicated a distinction between strength in verbal skill in some lexico-semantic aspects yet weakness in some tasks with greater executive demands (Temple, 2002, Temple and Carney, 1996). The current study examined whether there was also evidence for this distinction in the emerging lexical skills of school starters and the established lexical skills of young adults. If the dissociation in middle childhood between simple lexical processing and language processing with greater executive demands is sustained in young adulthood, this would have implications for the normal role of the X-chromosome in the development of executive language processes. The study therefore addresses the hypothesis that school starters and adults with TS are enhanced in lexico-semantic aspects of language but impaired in executive aspects of language.

The pattern of skills in both school starters and young adults has theoretical significance, as those who have argued for a neuroconstructivist approach where there is intimate interdependency between emerging skills (Thomas & Karmiloff-Smith, 2002) would predict less sharp dissociations between skills in the early years than those who have argued for a more structured modular system unfolding early in life (Marshall, 1984). Similarly, greater functional plasticity during development would predict less sharp dissociations between skills in the later years, than a more structured modular system appearing relatively early in life (Temple, 1997). Comparisons between areas of strength and weakness in school starters and young adults speak to both the extent and limitation of functional plasticity in development and the form of the cognitive phenotype linked to haploinsufficiency of the X-chromosome, which in turn has implications for the role of the X-chromosome in normal brain maturation. Although it was not the principal objective of the study, the results have additional translational significance in understanding the skills and vulnerabilities of the child with TS as she starts formal education and the adult with TS as she develops independence. More broadly there is potential impact for both normal and abnormal children in increased understanding of language development and its genetic underpinnings.

Section snippets

Child participants with TS [C-TS]

The child cases of TS [C-TS] were in their first year of school and were assessed during their second term at school. The 10 cases were recruited via a parental support organisation after letters of consent were forwarded to the parents. Selection criteria were to be aged 4 or 5 years, to be in the reception year of a mainstream school and to be of normal intelligence with a full scale IQ of at least 80 on the Wechsler Pre-school and Primary Scale of Intelligence [WPPSI-R UK] (Wechsler, 1989).

Analyses

The literature discussed above and the emergent hypotheses predict that where differences between groups occurs on lexico-semantic tasks they will be in the direction of a TS advantage, whereas where differences between groups occur on executive tasks they will be in the direction of TS impairment. One-tailed tests were therefore applied in line with those expectations. There was no explicit prediction in term of the overall CELF batteries and two-tailed tests were applied to these scores. As

Discussion

This study analysed the effects of the haploinsufficiency of the X chromosome upon lexical skills in school-starters and young adults with TS. Previous knowledge of language in these two age groups was sparse but evidence of their language skills is relevant to delineating the effects of the X chromosome on normal brain development. Based on previous research on the language skills of children with TS in middle childhood, it was hypothesised that if a difference between groups existed, it would

Conclusions

Haploinsufficiency of the X-chromosome drives mechanisms that affect the anatomical and neurochemical development of the brain. This study demonstrates that these result in enhanced temporal lobe lexico-semantic representation of language yet impaired frontal lobe executive language systems. This means not only that these modular elements of language can decouple but that their very independence is driven by mechanisms linked to the X-chromosome.

Acknowledgments

This research was supported by both ESRC Grant R000239496 and a studentship to the Developmental Neuropsychology Unit. Dr. Patricia Martin Sanfilippo provided assistance in the assessment of two cases and in some analyses of the child data.

References (104)

  • M.J. Marzelli et al.

    Neuroanatomical spatial patterns in Turner syndrome

    Neuroimage

    (2011)
  • M.M.M. Mazzocco

    The cognitive phenotype of Turner syndrome: Specific learning disabilities

    International Congress Series

    (2006)
  • D.G. Murphy et al.

    X-chromosome effects on female brain: A magnetic resonance imaging study of Turner’s syndrome

    Lancet

    (1993)
  • D.G.M. Murphy et al.

    A PET study of Turner’s syndrome: Effects of sex steroids and the X chromosome on the brain

    Biological Psychiatry

    (1997)
  • M.W.G. Nijhuis-van der Sanden et al.

    A review of neuropsychological and motor studies in Turner syndrome

    Neuroscience and Biobehavioral Reviews

    (2003)
  • B.F. Pennington et al.

    The neuropsychological phenotype in Turner syndrome

    Cortex

    (1985)
  • J.S. Peper et al.

    Sex steroids and brain structure in pubertal boys and girls

    Psychoneuroendocrinology

    (2009)
  • A. Raznahan et al.

    Cortical anatomy in human X monosomy

    Neuroimage

    (2010)
  • J.L. Ross et al.

    The Turner syndrome-associated neurocognitive phenotype maps to distal Xp

    American Journal of Human Genetics

    (2000)
  • T.J. Simon et al.

    Overlapping numerical cognition impairments in children with chromosome 22q11.2 deletion or Turner syndromes

    Neuropsychologia

    (2008)
  • C.M. Temple

    Oral fluency and narrative production in children with Turner’s syndrome

    Neuropsychologia

    (2002)
  • C.M. Temple et al.

    Lexical skills in William’s syndrome: A cognitive neuropsychological analysis

    Journal of Neurolinguistics

    (2002)
  • C.M. Temple et al.

    Patterns of spatial functioning in Turner’s syndrome

    Cortex

    (1995)
  • C.M. Temple et al.

    Reading skills in children with Turner’s syndrome: An analysis of hyperlexia

    Cortex

    (1996)
  • D. Alexander et al.

    Detective figure drawing, geometric and human, in Turner’s syndrome

    Journal of Nervous and Mental Disease

    (1966)
  • D. Alexander et al.

    Reading ability, object constancy and Turner’s syndrome

    Perceptual and Motor Skills

    (1965)
  • D. Alexander et al.

    Turner’s syndrome and Gerstmann’s syndrome. Neuropsychologic comparisons

    Neuropsychologia

    (1966)
  • D. Alexander et al.

    Studies in direction sense: 1 Turner’s syndrome

    Archives of General Psychiatry

    (1964)
  • A.P. Arnold et al.

    Gonadal steroid induction of structural sex differences in the CNS

    Annual Review of Neuroscience

    (1984)
  • E.A. Beaton et al.

    Atypical functional brain activation during a multiple object tracking task in girls with Turner syndrome: Neurocorrelates of reduced spatiotemporal resolution

    American Journal of Intellectual and Developmental Disabilities

    (2010)
  • B. Bender et al.

    Cognitive development of children with sex chromosome abnormalities

  • J.V. Borsel et al.

    Communication problems in Turner syndrome: A sample survey

    Journal of Communication Disorders

    (1999)
  • P.W. Burgess et al.

    The Hayling and Brixton tests

    (1997)
  • B. Butterworth et al.

    Language and the origins of number skills: Karyotypic differences in Turner’s syndrome

    Brain and Language

    (1999)
  • A.S. Clark et al.

    Gonadal hormones influence the emergence of cortical function in non-human primates

    Behavioral Neuroscience

    (1989)
  • C. Clark et al.

    Regional cerebral glucose metabolism in Turner syndrome

    Canadian Journal of Neurological Sciences

    (2002)
  • Coltheart, M. (1981)....
  • M. Coltheart et al.

    Phonological encoding in the lexical decision task

    Quarterly Journal of Experimental Psychology

    (1979)
  • V.H. Denenberg et al.

    Sex, brain, and learning differences in rats

    Behavioral and Brain Sciences

    (1988)
  • J. Downey et al.

    Cognitive ability and everyday functioning in women with Turner syndrome

    Journal of Learning Disabilities

    (1991)
  • J. Druks et al.

    An object and action naming battery

    (2000)
  • L.M. Dunn et al.

    British picture vocabulary scale

    (1997)
  • C. Elliot et al.

    British abilities scales, manual 4

    (1983)
  • T. Elliott et al.

    Positron emission tomography and neuropsychological correlations in children with Turner’s syndrome

    Developmental Neuropsychology

    (1996)
  • N. Frederickson et al.

    The phonological assessment battery

    (1997)
  • S.L. Fryer et al.

    Corpus callosum and posterior fossa development in monozygotic females: A morphometric MRI study of Turner syndrome

    Developmental Medicine and Child Neurology

    (2003)
  • S.E. Gathercole et al.

    The children’s test of nonword repetition: A test of phonological working memory

    Memory

    (1994)
  • C.D. Good et al.

    Dosage-sensitive X-linked locus influences the development of amygdale and orbitofrontal cortex, and fear recognition in humans

    Brain

    (2003)
  • R.W. Goy et al.

    Sexual differentiation of the brain

    (1980)
  • M.F. Haberecht et al.

    Functional neuroanatomy of visuo-spatial working memory in Turner syndrome

    Human Brain Mapping

    (2001)
  • Cited by (19)

    • Genetic Syndromes as Model Pathways to Mathematical Learning Difficulties: Fragile X, Turner, and 22q Deletion Syndromes

      2015, Development of Mathematical Cognition: Neural Substrates and Genetic Influences: Volume 2
    • Improving self-esteem in women diagnosed with turner syndrome: Results of a pilot intervention

      2014, Journal of Pediatric and Adolescent Gynecology
      Citation Excerpt :

      It is associated noticeably with primary amenorrhea, infertility, and shorter than typical stature. There are other variable effects including, for some individuals, subtle impairment in hearing3 and non-verbal learning.4 In the longer term, there are elevated risks of developing certain cardiovascular and metabolic diseases.5

    View all citing articles on Scopus
    View full text