Search tips
Search criteria 


Logo of schbulOxford JournalsContact UsMy BasketMy AccountSchizophrenia BulletinAbout this JournalContact this JournalSubscriptionsCurrent IssueArchiveSearch
Schizophr Bull. 2009 September; 35(5): 919–930.
Published online 2008 June 11. doi:  10.1093/schbul/sbn058
PMCID: PMC2728815

Shared Neurocognitive Dysfunctions in Young Offspring at Extreme Risk for Schizophrenia or Bipolar Disorder in Eastern Quebec Multigenerational Families


Background: Adult patients having schizophrenia (SZ) or bipolar disorder (BP) may have in common neurocognitive deficits. Former evidence suggests impairments in several neuropsychological functions in young offspring at genetic risk for SZ or BP. Moreover, a dose-response relation may exist between the degree of familial loading and cognitive impairments. This study examines the cognitive functioning of high-risk (HR) offspring of parents having schizophrenia (HRSZ) and high-risk offspring of parents having bipolar disorder (HRBP) descending from densely affected kindreds. Methods: The sample consisted of 45 young offspring (mean age of 17.3 years) born to a parent having SZ or BP descending from large multigenerational families of Eastern Québec that are densely affected by SZ or BP and followed up since 1989. The offspring were administered a lifetime best-estimate diagnostic procedure (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [DSM-IV]) and an extensive standard neuropsychological battery. Raw scores were compared with age- and gender-matched controls. Results: The offspring displayed differences in memory and executive functions when compared with controls. Moderate to large effect sizes (Cohen d) ranging from 0.65 to 1.25 (for IQ and memory) were observed. Several of the cognitive dysfunctions were present in both HRSZ and HRBP, even when considering DSM-IV clinical status. Conclusions: HRSZ and HRBP shared several aspects of their cognitive impairment. Our data suggest that the extremely high genetic and familial loading of these HRs may have contributed to a quantitatively increased magnitude of the cognitive impairments in both HR subgroups, especially in memory. These offspring at heightened risk present difficulties in processing information that warrant preventive research.

Keywords: high risk, schizophrenia, bipolar disorder, genetics, developmental precursor, cognition


The identification of susceptibility genes for schizophrenia (SZ) and bipolar disorders (BPs) has been complicated by the role of environmental factors, genetic heterogeneity, and uncertainty about the Diagnostic and Statistical Manual of Mental Disorders (DSM) phenotype.1 Endophenotypes, defined as measurable traits (biochemical, neuroanatomical, cognitive) related to an illness, may be helpful in identifying susceptibility genes because they are thought to have a simpler genetic etiology than the DSM categories.13

Several studies have observed significant neuropsychological deficits in nonaffected adult relatives (NAARs) that tended to be similar to those seen in SZ patients but with smaller effect sizes.46 For instance, the cognitive functions showing a decreased performance in NAARs span attention,79 episodic memory,1012 and executive functions1316 as well as language and spatial abilities.46 Fewer studies have investigated NAARs of BP patients17 and some reported that NAARs of BP may perform worse than controls on memory,18,19 in executive functions,19 in executive inhibitory processes,20 and in psychomotor performance speed,21 suggesting that some cognitive impairments might be shared by NAARs of SZ and those of BP. As regards the direct comparison of NAARs of SZ to those of BP, greater impairments in memory have been observed among the former while intellectual function, executive functions, and psychomotor performance were found similar.22 Zalla et al20 investigated executive functions and attention and found no difference between NAARs of SZ and NAARs of BP while Toulopoulou et al23 observed that relatives of BP showed greater verbal-performance IQ discrepancy scores compared with the SZ relatives.

This would be congruent with former findings showing similar neuropsychological deficits in patients with SZ or BP,20,2430 particularly the recent report of Schretlen et al24 who administered the same extensive battery to a large sample of SZ and BP patients. They obtained Cohen d mean effect sizes of 0.97 across tests for SZ patients and of 0.59 for BP patients, then suggesting that SZ had more severe global cognitive impairments than BP and that both disorders presented qualitatively similar profiles.

The study of neuropsychological functioning in young high-risk (HR) offspring (ie, offspring of an affected parent) is particularly appealing. Indeed, extensive reviews3133 have confirmed the presence of a poorer performance in HRs in several tasks encompassing attention, memory, and executive functions, which may be congruent with the dysfunctions observed in NAARs. Byrne et al,34 using an extended battery, observed in high-risk offspring of parents having schizophrenia (HRSZ) a poorer performance in intellectual level, executive functions, and memory when compared with controls. Less attention has been given to cognitive functioning in high-risk offspring of parents having bipolar disorder (HRBP).35 Recently, Klimes-Dougan et al36 investigated 43 adolescent offspring (average age of 15 years) of BP mothers from an affluent, high-achieving milieu, who were compared with offspring of mothers with major depression and with controls. They found deficits mainly in executive functions and attention which may be seen as somewhat overlapping those observed in HRSZ studies, even though much work remains to be done to investigate such commonalities.

Some limitations in existing HR studies may also affect the interpretation of the results. First, the offspring's selection criterion was most often based on an affected parent without documenting the family history. Hence, the HRs’ parents in most former studies may either present the sporadic or the familial form of illness. Focusing on parents with the familial form may be particularly fruitful given the evidence that cognitive dysfunction in relatives of SZ probands may depend on their degree of genetic relatedness with the probands. For example, cognitive dysfunction in SZ may covary with genetic relatedness such as in the affected monozygotic and dizygotic cotwins of SZ patients.37 Congruently, in family studies of NAARs of SZ patients, more severe cognitive dysfunctions were associated with a higher family genetic loading,38,39 ie, the larger the number of relatives affected by the disease, the more severe the cognitive dysfunctions in NAARs. A similar familial trend was also observed for executive functions in the HRSZ study of Byrne et al34 This body of data supports the assessment of the nature and magnitude of cognitive dysfunctions in HR offspring having a high familial loading. A second limitation is that the diagnosis of the spouse of the index patient was often not assessed. Third, only 2 of the prospective studies of HRSZ, started between 1952 and 1973, included an extensive battery and completed a longitudinal follow-up until mid-adulthood.33,34,40 Fourth, few studies have examined the specificity of the findings by comparing HRSZ to HRBP within the same study.

The current study addresses some of these limitations. We had the opportunity to focus on offspring at extreme risk having affected parents descending from densely affected multigenerational families. These adult family members (ie, including spouses and relatives of affected parents) were extensively and directly diagnosed, followed up longitudinally, and had been formerly submitted to diverse family genetic studies.4143 HRSZ and HRBP were assessed with similar methods.

We administered an extensive neuropsychological battery to cover several areas of cognitive functioning. Our hypotheses were that (1) these young HR offspring would display cognitive underperformances in memory and executive functions, when compared with age- and gender-matched normal controls, and that the expected effect sizes would be considerable given the high familial loading of the affected parent and (2) some of the expected dysfunctions would be shared by the HRSZ and the HRBP. We also aimed to test the effect of the presence of a DSM nonpsychotic diagnosis on the cognitive differences. We have indeed just reported that these young offspring, having not reached the age of incidence of psychosis, displayed high frequency of behavior disorders as indexed by nonpsychotic DSM-IV diagnoses.44 The DSM defined behavior disorders were mostly shared by the HRSZ and the HRBP and, overall, resembled the clinical profile reported for other HR offspring cohorts having an absent or a lesser familial loading.


Ascertainment of Kindreds

We targeted all the multigenerational families densely affected by SZ or BP in the Eastern Québec (Canada) catchment area. Families were selected if there was at least 1 first-degree relative affected with the same disorder as the proband and if there were at least 4 affected individuals with the same disorder. The first wave of kindred enrolment comprised 21 multigenerational families: 6 had 30–50 members, 5 had 20–29, 7 had 10–20, and 3 had less than 10 family members. The average number per kindred was 26 members. The families had an average number of 6 members affected by SZ or BP. The final sample consisted of 7 SZ kindreds (at least 85% of ill members affected by SZ or an SZ spectrum disorder, the remaining 15% having a BP spectrum disorder), 6 BP kindreds (at least 85% of ill members affected by BP or a BP spectrum disorder, the remaining 15% having an SZ spectrum disorder), and 8 mixed kindreds, ie, affected almost equally by both major psychoses. The high rate of mixed pedigrees may have resulted from our using a blind best-estimate diagnosis (BED). We have indeed demonstrated that unblind diagnosis had greater continuity with the most predominant diagnosis in a kindred than did blind diagnosis.45 The mean age of onset of adult family members was 25.4 (SD = 8.5) years for SZ and 28.8 (SD = 10.3) years for BP. The mean age at evaluation was 43.8 and 56.4 years, respectively.41

Sample of Offspring

The ascertainment of kindreds was done in 2 waves, and the present sample of 54 offspring was drawn from the first wave of assessment and led to the identification of offspring who belonged to the most proximal generations.

Inclusion and Exclusion Criteria.

The inclusion criteria were (1) having a parent with a definite DSM-IV SZ or BP disorder and (2) having a neuropsychological evaluation before 23 years old. In the present cognitive study, the HRs’ exclusion criteria were the presence of a diagnosis of DSM-IV psychotic disorder, BP or major depression, and brain and metabolic disorders known to cause neuropsychological impairments. Out of the 54 offspring, 1 HR was excluded for a recurrent major depression and 8 others were excluded because they could not have their neuropsychological assessment done by the age of 22, which left 45 subjects (mean age = 17.33, SD = 4.31, table 1).

Table 1.
Offspring Sample Characteristics

Sample of Control Subjects

We selected, by advertisements in local newspapers and in the population, the 45 unrelated normal controls (mean = 17.32, SD = 4.30) from the same population. They were matched for gender and age. The exclusion criteria were the same as those in HRs with the addition of any axis I DSM diagnosis or a positive family history of SZ or BP spectrum disorders. One volunteer was excluded because of an attention-deficit hyperactivity disorder (ADHD) diagnosis; another one was excluded because of an attention-deficit hyperactivity disorder (ADHD) diagnosis and because of family history of depression and 3 because of substance use. The study was explained and a signed consent was obtained, as reviewed by our University Ethics Committee.

Assessment of Clinical Diagnosis in the Affected Parent and in the Parent's Relatives

The affected parent and the adult relatives had the same direct BED. This lifetime BED was based on information from an interview with the subjects (Structured Clinical Interview for DSM-III-R [SCID]), from other family informants, and from all available medical records across lifetime. Based on this information, a consensus BED was derived by a panel of 4 research psychiatrists who were blind to diagnoses in relatives.45,46 The Global Assessment Scale (GAS)47 in the affected parents was used to measure the severity and the social functioning during the intervals between acute episodes across the entire life according to a method we reported elsewhere.45


The 45 HR offspring and 45 controls were administered clinical and neuropsychological measures. The detailed description of the DSM-IV lifetime BED procedure in the offspring in this sample is detailed elsewhere.44 In summary, the methods consisted in a direct interview with the parents and the children Kiddie-Schedule for Affective Disorders and Schizophrenia48 for subjects under 18 or the SCID49 for those over 18). A lifetime best-estimate DSM-IV diagnosis used all available information concerning the offspring,44 as had been formerly made in parents and relatives.45,46 A summary description of diagnoses among HRs is provided in table 2.

Table 2.
Proportions of Lifetime DSM-IV Diagnoses in the Total HR Offspring Sample and in HRSZ and HRBP Parents at Time of Neuropsychological Testing

The standard neuropsychological tests of our battery met the following criteria: (a) well-known psychometric properties; (b) high heritability of the performance or function; (c) previous use in studies of HRs, NAARs, and patients, for the sake of future comparability; (d) available French versions when the test had a significant verbal component; and (e) possible use in children, adolescents, and adults.

Based on this, we assessed the following 10 cognitive domains: (1) Intelligence: A full standard intelligence scale (Wechsler Intelligence Scale for Children, 3rd ed.-Wechsler Adult Intelligence Scale, 3rd ed. [WISC-III/WAIS-III] for HR and Wechsler Abbreviated Scale of Intelligence for controls) was completed to assess global intellectual level.5052 The Continuous Performance Test-II (CPT-II)53 assesses different aspects of attention. Subjects had to press the space bar each time they were shown a letter on the screen (all except the X) for an uninterrupted period of 14 min. (2) Sustained attention was assessed with some CPT variables “hit reaction time block change” (increased speed over time) and “hit standard error block change” (variability of performance). (3) Selective attention was assessed with index of Omissions (number of targets missed), of commissions (number of Xs identified as targets), and of detectability d′. The Stroop Test54 was added to more specifically assess inhibition in subjects with the use of the provided formula taking into account the individual's speed of processing in measuring resistance to interference. (4) Verbal episodic memory (VEM) was assessed with the California Verbal Learning Test,55 in which subjects had to learn a series of words presented orally over 5 trials and to immediately recall them after each presentation (total recall of 5 trials) or with a 20-min delay (delayed recall). They were also asked to recognize target words between distracters (recognition). (5) Visual episodic memory was assessed with the Rey Complex Figure.56 Subjects had to copy a figure and then recall after 3 (immediate recall), and 30 min (delayed recall). They were then presented with 24 items (targets and distracters) and asked to identify each item that was included in the initial figure (recognition). (6) Working memory was assessed first with the digit span (WISC-III or WAIS-III subtest), in which subjects had to recall serially sequences of digits, and second, with the Corsi,57 in which the subjects had to recall series of blocks. (7) Executive functions/problem-solving abilities were assessed with the Wisconsin Card Sorting Test-128 cards (WCST:CV4) in which participants had to classify series of cards into 3 categories, after having found the experimenter's classification rule (color, number, or forms).58 (8) Executive functions/initiation executive ability was assessed with the Verbal Fluency Test,59 in which subjects had to produce a maximum number of words during a 1-min interval. The first condition consisted in producing words from a phonological cue (ie, words beginning with the letter “p”), and for the second condition, subjects were asked to produce as many words as possible from the same semantic category (ie, “animals”). (9) Executive functions/planning was evaluated with the Tower of London (TOLDX),60 where subjects had to plan and reproduce models with 3 balls of different colors while respecting explicit rules. (10) Motor coordination: In the Purdue Pegboard,61 subjects were asked to insert as many pins as possible into small holes simultaneously with both hands, in a 1-min period.

In terms of procedure, subjects were individually assessed in a quiet room for a period of 3–4 h. Pauses were offered when needed. Clinical assessments were conducted by an experienced nurse and supervised by a research psychiatrist (Maziade and Roy). The neuropsychological assessment was made by certified psychologists or graduate PhD students supervised by a senior neuropsychologist (Rouleau). Depending on the age of the subject, the children or adolescent/adult version of those tests was used. The tests were administered in the same order in all subjects and analyzed and corrected blind to parents’ diagnosis.

Socioeconomic Status

To characterize family socioeconomic status (SES), we used the Blishen index62 based on the highest SES of the 2 parents at time of offspring upbringing. This index is based on education and income and on a Canadian census of 514 occupational categories according to the Canadian Classification and Dictionary of Occupations. An advantage is that norms were available from the Canadian general population: mean was 42.74 (SD = 13.3) and range from 17.81 (low SES) to 101.74 (high SES). In our offspring sample, mean was 41.34 (SD = 15.50) and range from 22.08 to 70.19, showing no difference with general population (1-sample t test for the mean, P = .55). Family SES was not correlated with IQ (r = .18, P = .24, NS) nor with the cognitive domains of table 3 (median r of 0.12, NS).

Table 3.
Comparisons of HR to Controls on the Neuropsychological Tests

Statistical Analysis

The analyses were done in 3 phases all including age and gender as covariables. We did so despite the matching between HR and control subjects on these variables because such matching was lost for several of the comparisons reported below, such as the analyses comparing HRSZ to HRBP and those excluding HR subjects having specific DSM diagnoses.

First, we compared the total sample of HRs (N = 45) with the controls (N = 45) on each of the cognitive tests by means of analyses of covariance (ANCOVAs) (when we used an analysis of variance, without entering age and gender as covariables, the results remained the same). The effect sizes coming out of the comparisons between the 2 groups were calculated by computing the Cohen d.63 We also tested for the impact of a DSM-IV diagnosis of ADHD or substance use in the offspring by excluding subjects presenting those diagnoses. To address the possible effect of lack of independence of observations due to some of the 45 offspring who were siblings (9 two- and 3 three-sibships), multilevel modeling64 was done using as the random effect, the sibships nested in the group.

Second, we examined the main effect of the group membership (HRSZ, HRBP, or controls) using ANCOVAs. When the main effect was found statistically significant, we examined the differences between HRBP and controls, HRSZ and controls, and HRBP and HRSZ, using post hoc least significant difference (LSD) procedure based on Student t test.

Third, given that (1) general intelligence may affect the performance at each cognitive test level and that, in turn, a cognitive impairment in a specific domain may affect the general IQ performance and (2) intelligence quotient was modestly associated with most neuropsychological variables in our sample (correlations of .30–.43), we performed the analyses (ANCOVAs and effect sizes) by controlling for IQ in addition to age and gender.

To examine whether the neuropsychological difference between the offspring and controls might be due to the presence of a DSM behavior disorder, an ANCOVA was conducted among HRs with a DSM nonpsychotic diagnosis, HRs without such a diagnosis, and controls. The group with a DSM behavior disorder was composed of HRs with a nonpsychotic DSM clinical disorder, excluding DSM developmental disorder (language disorder, learning disorder), and axis II disorders. We excluded language and learning disorders because the interest here was in disturbed behavior and not in diagnosed developmental disorder already associated with cognitive functioning. The effect sizes coming out of the comparisons between the groups were also calculated.

Multiple Testing.

To take account of multiple testing, we set our criterion for detecting a significant overall F test for ANCOVA at P = .005, while a P value found between .05 and .005 was interpreted as a possible tendency. This stringent criterion is the result of the usual threshold for significance of .05 divided by 10, given that 10 neuropsychological domains were assessed and given that a principal component analysis confirmed that 10 independent factors explained more than 80% of the variance (results not shown). The α = .05 significance level was used for post hoc analyses adjusting for IQ and for comparing groups via LSD procedures. All analyses were performed using SAS (version 9.1; PROC GLM).


Analysis in Total HR Sample and Intervening Factors

Compared with controls (table 3), the HR group (N = 45) showed a lower global IQ (d = −1.25), poorer verbal and visual episodic memory (d = −.77 to −.95), lower working memory (d = −.66), and poorer performance on executive functions such as initiation and planning (d = −.88 to −.75). As mentioned earlier, the ANCOVAs’ results were interpreted when the statistical threshold of .005 was reached. When the 7 HRs with a substance-use diagnosis were withdrawn from the analysis, the effect sizes remained in the same range (d = −.66 to −1.25) as well as when the 6 HRs with ADHD were withdrawn (d = −.58 to −1.18). To address the possible effect of lack of independence of observations due to some of the 45 offspring who were siblings, the analysis was redone using multilevel modeling and yielded consistent results in table 3, suggesting that the number of related offspring had not affected the results. Finally, to assess the possibility that the cognitive differences might be mainly due to a subgroup being in a prodromal state, we stratified the HR sample into those under age 17 (N = 19) and those aged 17 and over (N = 26). Most of the differences were similar in each of the 2 age strata, respectively, for IQ: d = −1.03, PANCOVA = .003 and d = −1.40, PANCOVA < .0001, and for visual episodic memory: d ≤ −1.16, PANCOVA ≤ .002 and d ≤ −.68, PANCOVA ≤ .02.

Comparisons of HRSZ, HRBP and Controls

The ANCOVAs adjusted for age and gender revealed group effects significant at the .005 level for IQ, verbal and visual episodic memory, and executive functions (problem solving, initiation, and planning, table 4). Post hoc analyses revealed, in all but one of these instances, a significantly poorer performance in HRBP and HRSZ compared with controls. For the WCST total errors, the HRSZ did not differ from controls, whereas the HRBP did. Other potential trends (P = .01) for differences across the 3 groups were found such as for working memory and sustained attention.

Table 4.
Comparison of HRSZ, HRBP, and Controls on Neuropsychological Tests

Cognitive Differences Adjusted for IQ

Tables 3 and and44 also provide the ANCOVA's results when adjusted for IQ, in addition to age and gender. Several of the differences with controls remained significant in the total HR group and in the HRSZ and HRBP subgroups, and most of the Cohen d remained moderate to large.

Overall Neuropsychological Profiles

Both HRSZ and HRBP had a lower IQ than controls, respectively, 94.1, 99.6, and 108.3 (d = −1.56, P < .0001 for HRSZ and d = −.99, P =.0002 for HRBP, table 4). Figure 1 illustrates the general cognitive profiles observed in HRSZ and HRBP. Overall, the effect sizes for the different cognitive domains suggest that (1) both HRSZ and HRBP showed a decreased cognitive functioning particularly in memory, (2) effect sizes were of moderate to high magnitude, in the 0.5–1.0 range, in measures like episodic memory and global IQ, and (3) the cognitive profiles of HRSZ and HRBP were to a certain degree superposed. As some HR subjects came from families (or kindreds) that were more homogeneous or that were mixed in terms of family history, we investigated whether family history could have influenced cognitive differences. When we regrouped the HRs into those having a SZ parent from a SZ kindred (n = 8 offspring), those having a BP parent from a BP kindred (n = 17 offspring), excluding those with a parent from a mixed kindred (n = 20 offspring), the same pattern of lower IQ was observed for both subgroups when compared with controls (mean IQ of HRSZ from SZ kindred = 96.1, P = .002; mean IQ of HRBP from BP kindred = 101.8, P = .006, data not shown).

Fig. 1.
Effect Size of Cognitive Functioning Among High-Risk Offspring of Parents Having Bipolar Disorder and High-Risk Offspring of Parents Having Schizophrenia. Effect sizes were calculated according to the Cohen d index. HRBP values are represented by triangles, ...

DSM Nonpsychotic Behavior Disorders and Cognitive Functioning

Because 60% presented a DSM-IV nonpsychotic diagnosis warranting a psychiatric consultation44 (table 2), we examined whether the former differences in cognitive functioning between offspring and controls might be due to the presence or absence of a DSM behavior disorder. The difference with controls remained the same when taking into account clinical status: the effect sizes for the comparison of HRs with a behavior disorder to controls and for the comparison of HRs without a behavior disorder to controls were, respectively, for IQ (d = −1.14 and −1.15), for VEM (d = −1.02 and −0.95), for visual episodic memory (d = −0.88 and −0.84), and for executive function/planning (d = −0.78 and −0.67).


Methodological Issues

Our design had several strengths such as a definite high familiality of the parent's illness, a lifetime BED of parents, of relatives, and of offspring, and extensive neuropsychological battery with the same methods for both HRSZ and HRBP. However, our study also had limitations. The first one is whether such a very HR sample and their ill parents are representative of all parents with major psychosis and their offspring, thus calling for cautiousness before generalizing to the entirety of SZ and BP. In counterpart, however, our previous reports42,45 have shown that the clinical and epidemiological characteristics of the adults with SZ and BP in these kindreds were quite similar to those observed in general patient samples regarding severity, age of onset, factor structure of symptoms, response to treatments, and gender differences. The clinical profile of these HRs shows a rate and type of DSM nonpsychotic disorders that are similar to those reported in HRs having an affected parent with no or low family history.44 Second, the small size of the sample may generate type II error especially with regards to the HRSZ/HRBP comparisons. Third, despite the use of an extensive neuropsychological battery, potentially meaningful areas of cognition such as language or praxia were not covered by our study. Fourth, we did not include an instrument specifically designed for assessing prodromal symptoms because this HR study was started 10 years ago and also because the predictive value of these instruments is still debated.65 However, the hypothesis that our HRs could be in a prodromal state explaining their neurocognitive deficits remains unlikely, based on 2 sets of considerations: (1) the cognitive differences were present in preadolescence as much as in postadolescence years in our sample and (2) the HRs were administered an extensive lifetime BED procedure that would have probably detected most of the known prodromal symptoms (subdelusions, dysperceptual symptoms, etc.).44

Similarities with Findings in HRs, in NAARs, and in Patients

These young offspring at extreme risk for SZ or BP presented a dysfunction in several cognitive domains as indexed by the observed effect sizes. Most of the HRs’ differences with controls on this extensive neuropsychological battery appeared rather congruent with recent findings from other HR studies.31,32,34 The offspring underperformance in episodic memory and executive functions is also compatible with those showed in a meta-analysis of adult SZ patient studies66 and in meta-analyses of NAARs’ studies.46,12 As in other HR studies using extended batteries,32,36,67 we observed that differences in attentional processes might not be the main or only cognitive feature to focus on for future use as an endophenotype in genetic research.

Quantitative Differences in Cognition and Familial Loading

Our findings suggest that offspring with a very high familial/genetic loading have cognitive impairments that may be qualitatively similar to those observed in HRs with milder familial/genetic loading and in NAARs. However, in quantitative terms, the present effect sizes (Cohen d in the 0.6–1.0 range) approach those observed in adult patients, typically in the 0.8–1.2 range,24,66 and is higher than those usually reported for NAARs, which is in the 0.3–0.5 range.4 Our findings would then be congruent with (1) former observations that cognitive impairment in NAARs of SZ tends to increase with the number of affected members38 and (2) the preliminary observation by Byrne et al34 that some executive functions in HRSZ are more impaired if the familial loading is higher.

Interestingly, our findings differ slightly from those of Klimes-Dougan et al36 who found in their HRBP more dysfunctions in executive functions than in memory. However, the sample was from an affluent and highly educated milieu and Klimes-Dougan et al's affected mothers might have been sporadic or less familial cases than in our sample.

Shared and Specific Neurocognitive Characteristics

To our knowledge, this is the first study having investigated HRSZ and HRBP concurrently with the same sampling and assessment methods, allowing for the direct testing of early cognitive dysfunctions that could be shared by both risk groups. We observed that the HRs’ differences with controls in IQ, episodic memory, and executive functions were present in both subgroups. Figure 1 illustrates their largely superposed neuropsychological profiles. While visual inspection of effect sizes suggests some discrepancies, only one domain, namely executive functions/problem solving, reached our statistical threshold for a HRBP/HRSZ difference. This again appears globally in continuity with the accumulating evidence that adult SZ and BP patients have many phenotypic features in common,42,68 as well as numerous genomic susceptibility loci evidenced by linkage and association studies.41,69 Our findings in HRs are also much in continuity with the recent study of adult SZ and BP patients by Schretlen et al,24 who also observed that the neurocognitive profiles were largely shared by the 2 disorders when using a neuropsychological battery quite comparable to ours. As suggested by the present data, the early risk mechanisms of the 2 major psychoses would show several commonalities.70

Further studies are needed to investigate whether the type of family history, homogeneous or mixed, of the affected parent may affect the cognitive profile of the offspring. We did not have sufficient sample size to explore satisfactorily this issue. Also, it is still difficult at this point in time to have a definite interpretation of the meaning of the possible specific impairments presented by the HRBP in sustained attention and in executive functions such as planning and subtests of problem solving. More studies in different populations and further follow-up of our HRs may throw light on these patterns because specific cognitive dysfunctions may appear later in life.

Implications for a Hypothetic Mechanistic Model of Vulnerability

Despite the methodological issues formerly raised, one can nevertheless assume that a putative model of vulnerability for major psychosis would have to reconcile the following empirical observations: (1) some of the childhood risk mechanisms for SZ and BP would be shared by the 2 major psychoses, and the commonalities of numerous phenotypic and epidemiological characteristics that the 2 disorders have in adulthood would originate in some early mechanisms, (2) the specificity of an orientation toward SZ or BP may not come from factors strongly related to neurocognitive functioning, even though cognitive differences between HRSZ and HRBP may exist, and (3) the very high genetic density in the former generations of these HRs could have affected the severity of the neurocognitive dysfunctions rather than the specificity of impairment in a particular domain. Congruent with a multifactorial oligogenic model that is largely accepted for the major psychoses,71 the greater the genetic loading in the extended family, the more serious the insult to brain functioning expressed in cognitive functioning.

In conclusion, the true test for revealing which of the developmental abnormalities better predicts the future appearance of a major psychosis and related disorders will rest on 2 complementary arms of evidence: on one hand, in the demonstration of a valid relationship between an early endophenotype and a genotype and, on the other hand, in the longitudinal demonstration that such an endophenotype-genotype match can predict the development of major psychosis when HR offspring are followed up beyond the age of incidence of the disease.


Canada Research Chair (#950-200810); a CIHR grant (#MOP-74430); the Fonds de la recherche en santé du Québec (M-A.R.).


We are grateful to our professional research assistants Linda René and Louise Bélanger and the family members, adults, and children, who have participated in this study. We also thank Institut interuniversitaire de recherches sur les populations for their collaboration regarding the BALSAC database.


1. Szatmari P, Maziade M, Zwaigenbaum LZ, et al. Informative phenotypes for genetic studies of psychiatric disorders. Am J Med Genet B Neuropsychiatr Genet. 2007;144:581–588. [PubMed]
2. Cannon TD. The inheritance of intermediate phenotypes for schizophrenia. Curr Opin Psychiatry. 2005;18:135–140. [PubMed]
3. Braff DL, Freedman R, Schork NJ, Gottesman I. Deconstructing schizophrenia: an overview of the use of endophenotypes in order to understand a complex disorder. Schizophr Bull. 2007;33:21–32. [PMC free article] [PubMed]
4. Snitz BE, Macdonald AW, 3rd, Carter CS. Cognitive deficits in unaffected first-degree relatives of schizophrenia patients: a meta-analytic review of putative endophenotypes. Schizophr Bull. 2006;32:179–194. [PMC free article] [PubMed]
5. Sitskoorn MM, Aleman A, Ebisch SJH, Appels MCM, Kahn RS. Cognitive deficits in relatives of patients with schizophrenia: a meta-analysis. Schizophr Res. 2004;71:285–295. [PubMed]
6. Szoke A, Schurhoff F, Mathieu F, Meary A, Ionescu S, Leboyer M. Tests of executive functions in first-degree relatives of schizophrenic patients: a meta-analysis. Psychol Med. 2005;35:771–782. [PubMed]
7. Cornblatt BA, Keilp JG. Impaired attention, genetics, and the pathophysiology of schizophrenia. Schizophr Bull. 1994;20:31–46. [PubMed]
8. Chen WJ, Faraone SV. Sustained attention deficits as markers of genetic susceptibility to schizophrenia. Am J Med Genet. 2000;97:52–57. [PubMed]
9. Mirsky AF, Lochhead SJ, Jones BP, Kugelmass S, Walsh D, Kendler KS. On familial factors in the attentional deficit in schizophrenia: a review and report of two new subject samples. J Psychiatr Res. 1992;26:383–404. [PubMed]
10. Park S, Holzman PS, Goldman-Rakic PS. Spatial working memory deficits in the relatives of schizophrenic patients. Arch Gen Psychiatry. 1995;52:821–828. [PubMed]
11. Conklin HM, Curtis CE, Katsanis J, Iacono WG. Verbal working memory impairment in schizophrenia patients and their first-degree relatives: evidence from the digit span task. Am J Psychiatry. 2000;157:275–277. [PubMed]
12. Faraone SV, Seidman LJ, Kremen WS, Pepple JR, Lyons MJ, Tsuang MT. Neuropsychological functioning among the nonpsychotic relatives of schizophrenic patients: a diagnostic efficiency analysis. J Abnorm Psychol. 1995;104:286–304. [PubMed]
13. Gilvarry CM, Russell A, Jones P, Sham PC, Hemsley D, Murray RM. Verbal fluency in patients with schizophrenia and affective psychoses and their first-degree relatives. Psychol Med. 2001;31:695–704. [PubMed]
14. Chen YLR, Eric YH, Chen FM. Semantic verbal fluency deficit as a familial trait marker in schizophrenia. Psychiatry Res. 2000;95:133–148. [PubMed]
15. Laurent A, d'Amato T, Naegele B, et al. Fonctions exécutives et mnésiques d'un groupe d'apparentés de premier degré de patients schizophréniques. L'Encéphale. 2000;26:67–74. [PubMed]
16. Laurent A, Biloa-Tang M, Bougerol T, et al. Executive/attentional performance and measures of schizotypy in patients with schizophrenia and their non psychotic first-degree relatives. Schizophr Res. 2000;46:269–283. [PubMed]
17. Murphy FC, Sahakian BJ. Neuropsychology of bipolar disorder. Br J Psychiatry. 2001;178:s120–s127. [PubMed]
18. Keri S, Kelemen O, Benedek G, Janka Z. Different trait markers for schizophrenia and bipolar disorder: a neurocognitive approach. Psychol Med. 2001;31:915–922. [PubMed]
19. Robinson LJ, Ferrier IN. Evolution of cognitive impairment in bipolar disorder: a systematic review of cross-sectional evidence. Bipolar disord. 2006;8:103–116. [PubMed]
20. Zalla T, Joyce C, Szöke A, et al. Executive dysfunctions as potential markers of familial vulnerability to bipolar disorder and schizophrenia. Psychiatry Res. 2004;121:207–217. [PubMed]
21. Antila M, Tuulio-Henriksson A, Kieseppa T, Eerola M, Partonen T, Lonnqvist J. Cognitive functioning in patients with familial bipolar I disorder and their unaffected relatives. Psychol Med. 2007;37:679–687. [PubMed]
22. McIntosh AM, Harrison LK, Forrester K, Lawrie SM, Johnstone EC. Neuropsychological impairments in people with schizophrenia or bipolar disorder and their unaffected relatives. Br J Psychiatry. 2005;186:378–385. [PubMed]
23. Toulopoulou T, Quraishi S, McDonald C, Murray RM. The Maudsley Family Study: premorbid and current general intellectual function levels in familial bipolar I disorder and schizophrenia. J Clin Exp Neuropsychol. 2006;28:243–259. [PubMed]
24. Schretlen DJ, Cascella NG, Meyer SM, et al. Neuropsychological functioning in bipolar disorder and schizophrenia. Biol Psychiatry. 2007;62:179–186. [PMC free article] [PubMed]
25. Morice R. Cognitive inflexibility and pre-frontal dysfunction in schizophrenia and mania. Br J Psychiatry. 1990;157:50–54. [PubMed]
26. Goldberg TE, Gold JM, Greenberg RD, et al. Contrasts between patients with affective disorders and patients with schizophrenia on a neuropsychological test battery. Am J Psychiatry. 1993;150:1355–1362. [PubMed]
27. McGrath J, Scheldt S, Welham J, Clair A. Performance on tests sensitive to impaired executive ability in schizophrenia, mania and well controls: acute and subacute phases. Schizophr Res. 1997;26:127–137. [PubMed]
28. Albus M, Hubmann W, Ehrenberg C, et al. Neuropsychological impairment in first-episode and chronic schizophrenia patients. Eur Arch Psychiatry Clin Neurosci. 1996;246:249–255. [PubMed]
29. Albus M, Hubmann W, Wahlheim C, Sobizack N, Franz U, Mohr F. Contrasts in neuropsychological test profile between patients with first-episode schizophrenia and first-episode affective disorders. Acta Psychiatr Scand. 1996;94:87–93. [PubMed]
30. Addington J, Addington D. Attentional vulnerability indicators in schizophrenia and bipolar disorder. Schizophr Res. 1997;23:197–204. [PubMed]
31. Niemi LT, Suvisaari JM, Tuulio-Henriksson A, Lönnqvist JK. Childhood developmental abnormalities in schizophrenia: evidence from high-risk studies. Schizophr Res. 2003;60:239–258. [PubMed]
32. Owens DG, Johnstone EC. Precursors and prodromata of schizophrenia: findings from the Edinburgh High Risk Study and their literature context. Psychol Med. 2006;36:1501–1514. [PubMed]
33. Erlenmeyer-Kimling L. Neurobehavioral deficits in offspring of schizophrenic parents: liability indicators and predictors of illness. Am J Med Genet. 2000;97:65–71. [PubMed]
34. Byrne M, Hodges A, Grant E, Owens DC, Johnstone EC. Neuropsychological assessment of young people at high genetic risk for developing schizophrenia compared with controls: preliminary findings of the Edinburgh High Risk Study (EHRS) Psychol Med. 1999;29:1161–1173. [PubMed]
35. DelBello MP, Geller B. Review of studies of child and adolescent offspring of bipolar parents. Bipolar disord. 2001;3:325–334. [PubMed]
36. Klimes-Dougan B, Ronsaville D, Wiggs EA, Martinez PE. Neuropsychological functioning in adolescent children of mothers with a history of bipolar or major depressive disorders. Biol Psychiatry. 2006;60:957–965. [PubMed]
37. Cannon TD, Huttunen MO, Lonnqvist J, et al. The inheritance of neuropsychological dysfunction in twins discordant for schizophrenia. Am J Hum Genet. 2000;67:369–382. [PubMed]
38. Faraone SV, Seidman LJ, Kremen WS, Toomey R, Pepple JR, Tsuang MT. Neuropsychologic functioning among the nonpsychotic relatives of schizophrenic patients: the effect of genetic loading. Biol Psychiatry. 2000;48:120–126. [PubMed]
39. Tuulio-Henriksson A, Arajärvi R, Partonen T, et al. Familial loading associates with impairment in visual span among healthy siblings of schizophrenia patients. Biol Psychiatry. 2003;54:623–628. [PubMed]
40. Erlenmeyer-Kimling L, Rock D, Roberts SA, et al. Attention, memory, and motor skills as childhood predictors of schizophrenia-related psychoses: the New-York High-Risk Project. Am J Psychiatry. 2000;157:1416–1422. [PubMed]
41. Maziade M, Roy M-A, Chagnon Y, et al. Shared and specific susceptibility loci for schizophrenia and bipolar disorder: a dense genome scan in Eastern Quebec families. Mol Psychiatry. 2005;10:486–499. [PubMed]
42. Maziade M, Roy M-A, Martinez M, et al. Negative, psychoticism, and the disorganized dimensions in patients with familial schizophrenia or bipolar disorder: continuity and discontinuity between the major psychoses. Am J Psychiatry. 1995;152:1458–1463. [PubMed]
43. Mérette C, Roy M-A, Bureau A, et al. Replication of linkage with bipolar disorder on chromosome 16p in the Eastern Quebec population. Am J Med Genet B Neuropsychiatr Genet. In press. [PubMed]
44. Maziade M, Gingras N, Rouleau N, et al. Clinical diagnoses in offspring from Eastern Québec multigenerational families densely affected by schizophrenia or bipolar disorder. Acta Psychiatr Scand. 2008;117:118–126. [PMC free article] [PubMed]
45. Roy M-A, Lanctôt G, Mérette C, et al. Clinical and methodological factors related to reliability of the best-estimate diagnostic procedure. Am J Psychiatry. 1997;154:1726–1733. [PubMed]
46. Maziade M, Roy MA, Fournier J-P, et al. Reliability of best-estimate diagnosis in genetic linkage studies of major psychoses: results from the Québec pedigree studies. Am J Psychiatry. 1992;149:1674–1686. [PubMed]
47. Endicott J, Spitzer RL, Fleiss JL, Cohen J. The global assessment scale: a procedure for measuring overall severity of psychiatric disturbances. Arch Gen Psychiatry. 1976;33:766–771. [PubMed]
48. Kaufman J, Birmaher B, Brent D, et al. Schedule for affective disorders and schizophrenia for school-age children—present and lifetime version (K-SADS-PL): initial reliability and validity data. J Am Acad Child Adolesc Psychiatry. 1997;36:980–988. [PubMed]
49. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID) I: history, rationale and description. Arch Gen Psychiatry. 1992;49:624–629. [PubMed]
50. Weschler D. Échelle d'intelligence de Weschler pour adultes. Troisième édition Recherche. Version pour francophones du Canada. Toronto, Canada: The Psychological Corporation, Harcourt Assessment, Inc; 2005.
51. Weschler D. Weschler Abbreviated Scale of Intelligence. Toronto, Canada: The Psychological Corporation; 1999.
52. Weschler D. Échelle d'intelligence de Weschler pour enfants. Troisième édition. Version pour francophones du Canada. Toronto, Canada: The Psychological Corporation, Harcourt Assessment, Inc; 1997.
53. Conners K. Continuous Performance Test II. Odessa, FL: Psychological Assessment Resources; 1999.
54. Golden C. Stroop Color and Word Test. Tampa, FL: Phychological Assessment Resources: 1976.
55. Delis D, Kramer J, Kaplan E, Ober B. California Verbal Learning Test Manual. San Antonio, TX: Phychological Corporation: 1987.
56. Meyers JE, Meyers KR. Rey Complex Figure Test and Recognition Trial (RCFT) Odessa, FL: Phychological Assessment Resources: 1995.
57. Weschler D. WMS-III: Weschler Memory Scale. San Antonio, TX: Harcourt Assessment Inc; 1997.
58. Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G. Wisconsin Card Sorting Test: Manual Revised and Expanded. Odessa, FL: Psychological Assessment Resources; 1993.
59. Delis DC, Kaplan E, Kramer JH. Delis-Kaplan Executive Function System. San Antonio, TX: The Psychological Corporation; 2001.
60. Culbertson W, Zillmer E. Tower of London—TOL DX. Tampa, FL: Psychological Assessment Resources; 2001;
61. Tiffin J. Purdue Pegboard Test. Tampa, FL: Psychological Assessment Resources: 1948.
62. Blishen B, Carroll W, Moore C. The 1981 socioeconomic index for occupations in Canada. Canad Rev Sociol Anthropol. 1987;24:465–488.
63. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
64. Goldstein H. Multilevel models. In: Armitage P, Colton T, editors. Encyclopedia of Biostatistics. Chichester, West Sussex, England: John Wiley and Sons Ltd; 1998. pp. 2725–2731.
65. Olsen KA, Rosenbaum B. Prospective investigations of the prodromal state of schizophrenia: review of studies. Acta Psychiatr Scand. 2006;113:247–272. [PubMed]
66. Heinrichs RW, Zakzanis KK. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology. 1998;12:426–445. [PubMed]
67. Cornblatt B, Obuchowski M, Roberts S, Pollack S, Erlenmeyer-Kimling L. Cognitive and behavioral precursors of schizophrenia. Dev Psychopathol. 1999;11:487–508. [PubMed]
68. Maier W, Zobel A, Wagner M. Schizophrenia and bipolar disorder: differences and overlaps. Curr Opin Psychiatry. 2006;19:165–170. [PubMed]
69. Berrettini W. Evidence for shared susceptibility in bipolar disorder and schizophrenia. Am J Med Genet. 2003;123C:59–64. [PubMed]
70. Murray RM, Sham P, Van Os J, Zanelli J, Cannon M, McDonald C. A developmental model for similarities and dissimilarities between schizophrenia and bipolar disorder. Schizophr Res. 2004;71:405–416. [PubMed]
71. Gottesman II, Shields J. Schizophrenia and genetics: a twin study vantage point. New York, NY: Academic Press; 1972.
72. Shaffer D, Gould MS, Brasic J, et al. A children's global assessment scale (CGAS) Arch Gen Psychiatry. 1983;40:1228–1231. [PubMed]

Articles from Schizophrenia Bulletin are provided here courtesy of Oxford University Press