Fragile X syndrome (FXS) is the most common hereditary cause of intellectual disability in males, and is the leading single genetic cause of autism. The disorder is caused by methylation of an expanded trinucleotide CGG (>200) repeat in the promoter of the fragile X mental retardation 1 (FMR1) gene, which is located on the q27.3 site of the X chromosome , and is associated with low or absent levels of FMR1 mRNA and protein (FMRP). FMRP binds to as much as 4% of all mRNA in mammalian brains , and is thought to play a crucial role in synapse development and plasticity. Indicative of the importance of FMRP, the level of cognitive ability correlates with the level of FMRP in males with FXS . Carriers of the fragile X premutation (fXPCs) are defined by the presence of 55 to 200 CGG repeats within the FMR1 gene, which results in a three- to eight-fold increase in FMR1 mRNA levels in leukocytes , but little or no reduction in FMRP. It is estimated that 1 in 260–813 males and 1 in 113–259 females carry the premutation allele , and approximately 40% of male fXPCs and 8-16% of female fXPCs develop fragile X-associated tremor ataxia syndrome (FXTAS) . FXTAS is a late-onset (>50 years old) neurodegenerative disorder associated with tremors, gait ataxia, parkinsonism, and impairments in short-term memory and executive functions . It is thought to result from RNA toxicity due to elevated FMR1 mRNA levels .
Recent indications that the fragile X premutation allele negatively affects neurodevelopment during an individual’s life span shifted the research field away from evaluating the premutation solely as a risk factor for neurodegeneration (that is, FXTAS), and towards elucidating the consequences of this altered neurodevelopment. Some studies suggest that fXPCs younger than 50 years of age are largely cognitively unaffected by the mutation [9–12]; however, adult female fXPCs tend to have faster oral and manual motor psychomotor speed than adults not carrying the premutation [13, 14]. This suggests that a lack of difference in performance between female fXPCs and controls in cognitively demanding, non-standardized tasks may actually represent slowing of cognitively modulated performance times, which is masked by enhanced simple reaction time.
Studies reporting no differences between male fXPCs and controls have used tasks such as the Controlled Oral Word Generation Task, Stroop Color-Word Task, sections of the Behavioral Dyscontrol Scale, and certain selective attention tasks [12, 15–20], and some of these controlled for manual motor performance using independent assessment by the Purdue Pegboard Test or CATSYS system (http://www.catsys.dk/purchase.htm) as covariates [15–17]. Because male fXPCs are at increased risk for developing FXTAS , which is characterized by motor impairment, any reaction time differences between fXPCs and controls could be due to cognitive slowing or motor slowing. Thus, it is particularly important to control for psychomotor speed of male fXPCs when assessing performance on cognitive tasks.
Because fXPCs are at increased risk of developing FXTAS, which is a neurodegenerative disorder often accompanied by attentional control impairments, it may be prudent to determine whether similar cognitive impairments are observable in fXPCs asymptomatic for FXTAS. If cognitive impairments precede motor impairments, cognitive impairments might be used as a biomarker for risk of disease progression. To understand the cognitive phenotype of fXPCs, we may look to the similarities between fXPCs and individuals with FXSc, because one way to view the effects of all FMR1 mutations is to view them as existing on a phenotypic spectrum that is modulated by FMR1 dosage in terms of CGG repeats and gender. Specifically, FMR1 dosage increases with CGG repeat length, and males have a higher FMR1 ‘dose’ than females, because the premutation allele is expressed in all of their cells, and they lack a second, unaffected FMR1 allele. In support of a genetically modulated phenotypic spectrum are the findings that individuals with FXS or the premutation (with or without FXTAS) share symptoms of executive function impairments [21–23] that are modulated by CGG repeat length [18, 19, 24–26], FMR1 mRNA , and FMRP [28–32]. Additionally, CGG repeat length relates to age of onset of FXTAS  and to degree of brain atrophy .
Individuals with FXS exhibit difficulties in understanding space, time, and numbers, which result in characteristic quantitative and numerical impairments [35–37]. Functional brain activation during arithmetic processing in females with FXS was found to be related to FMRP expression, suggesting that decreased FMRP expression underlies impairments in mathematics performance in individuals with FXS . Similarly, adult female fXPCs also have arithmetic impairments , and impairments in judging relative magnitude and enumeration that are modulated by CGG and age [40, 41]. Positron emission tomography imaging in adult female fXPCs indicates hypometabolism of the right parietal, temporal, and occipital association areas , suggesting that these impairments may be due to abnormal functioning of brain regions involved in visuospatial attention. Additionally, girls with FXS have impaired performance on ‘where’ tasks , and men with FXS and adult fXPCs have specific magnocellular (M) pathway impairment [43–45]. CGG knock-in mice, a model for fXPCs, also show similar spatiotemporal processing impairments [46–48]. Thus, although individuals with the full mutation or premutation are thought to be affected via different mechanisms (FMRP deficiency and RNA toxicity, respectively), evidence suggests that they share a common impairment of spatial and temporal processing that affects higher-level processing (for example, numerical thinking and arithmetic).
The purpose of this study was to determine whether young adult male fXPCs, asymptomatic for FXTAS, exhibit impairments in numerical visuospatial tasks. We used two psychomotor speed tasks (manual and oral), which allowed us to control for baseline differences in response time. We also used two visuospatial tasks: magnitude comparison and enumeration. These tasks allowed us to examine judgments of relative magnitude (numerical distance effect) and the ability to indicate the number of presented items. Our previous work with adult female fXPCs in the same age range reported impairments that were modulated by CGG repeat length and age, indicating that these tasks are sensitive to FMR1 allele variants [40, 41]. Because females have a second, unaffected FMR1 allele that is expressed randomly in 50% of their cells, they should be less affected than males. Thus, we hypothesized that male fXPCs would be impaired relative to HCs and would be more cognitively affected than female fXPCs.