Our objective is to investigate whether handedness is associated with performance on verbal and visual neuropsychological tests in people with severe mental disorders. A recent study, applying a continuous scale of hand preference, reports that handedness is not associated with test performance in people with schizophrenia disorders. Conversely, in a recent large meta-analysis where handedness was applied as a dichotomous variable, right-handers had better performance in spatial ability (but not verbal ability) compared to left-handers, irrespective of gender or health status. We hypothesize that a dichotomous classification of handedness will reveal an advantage of right-handedness on tests of visuospatial functions—but not verbal functions—in people with severe mental disorders. We expect that gender will not be associated with the neuropsychological test results.

Data from a sample of 385 patients with severe mental disorders, mainly within the schizophrenia spectrum, were analyzed. All participants had Norwegian as their first language. Their mean age was 24.8 years (SD=6.2) and 153 (39.7%) were women. Handedness was evaluated by observation of preferred hand in writing and drawing during neuropsychological assessment. Chi-square tests were used to compare proportions of cases with reported frequencies of handedness in the general population and comparable clinical samples. Raw scores on Semantic Fluency and Line Orientation from the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) were analyzed using Mann-Whitney U tests, and possible effects of gender with two-way ANOVA. Frequencies of low scores were analyzed using frequency analyses.

Overall prevalence of left-handers was 10.4% compared to 10.6% in the general population (χ2=.018, p=.893). Observed prevalence for women was 9.2% compared to expected prevalence of 9.5% (χ2=.026, p=.873) and for men 10.7% and 11.6%, respectively (χ2=.039, p=.844). There was a significant difference in Line Orientation scores (Mdright-handers=18, Mdleft-handers=17; U=5268.0, p=.013) but not Semantic Fluency scores (Mdright-handers=17, Mdleft-handers=18.5; U=7568.5, p=.315). Right-handed men had higher scores on Line Orientation but there was no handedness by gender interaction (F(1)=1.69, p=.194). For Semantic Fluency, left-handed men had higher scores and a gender by handedness interaction was found (F(1)=7.21, p=.008). Using scores corresponding to <5th percentile, 15% of left-handers and 8% of right-handers had scores <5th percentile on Line Orientation, as opposed to 15% and 14% on Semantic Fluency.

Right-handers had significantly better performance on a test measuring visuospatial function, irrespective of gender. Left-handers had about twice the number of scores in the impaired range (i.e., <5th percentile) compared to right-handers. Left-handed men had better performance on a test of verbal functions, which was unexpected. A recent study reported no right-hand associated advantage on visuospatial tests in people with schizophrenia disorders when measuring handedness on a continuous scale. This suggests that the classification of handedness as either a dichotomous or as a continuous variable is important in studies of handedness and cognitive functions.