Marian Annett

The Binomial Distribution of Right, Mixed and Left Handedness:

Right, mixed and left handers are found in binomial proportions in seven samples of varied subjects whose lateral preferences were ascertained by several methods. These proportions have been obtained in previous studies of humans and animals when the performance of several actions has been recorded in complete samples and when consistent right and left subjects have been separated from those of mixed usage.

Handedness in families

The distribution of left-handedness in young adults and their relatives is described in this paper in two ways. First, the incidence of sinistrality is given in families grouped according to parental handedness as in previous reports of similar material. Secondly, the incidence is given in the relatives of rightand left-handed individuals. Left handers can then be treated as propositi in estimating heritability (Falconer, 1965). Falconer’s method is suitable for characters which are discrete in the sense that a disease can be said to be present or absent in an individual but the occurrence of which can be assumed to depend on an underlying continuous distribution of liability. Handedness is probably such a character. Most individuals can be classified readily right or left handed on the basis of the writing hand but there are many different patterns of preference when several actions are considered (Annett, 1970) and these are related to a continuous normal distribution of differences between the hands in strength (Woo & Pearson, 1927) and speed (Annett, 1970). Estimates of the incidence of left-handedness vary widely (Wile, 1934). This variation depends largely on whether the criterion is left preference for all or for any of several actions which ranged from about 4 to 35 % in undergraduates questioned about 12 unimanual tasks (Annett, 1970). Part of the variation depends on differences between sexes and between generations. Sex differences are usually found and when present in normal samples are always in the direction of greater sinistrality in males. A difference between generations was found by Chamberlain (1928) and by Rife (1940) and is marked in the present sample. Sex and generation differences often affect the incidence of disease and can be taken into account in estimating heritabilities. The data of Chamberlain (1928) and Rife (1940) are re-analysed to permit comparisons of the findings of the three samples. Information as to the handedness of grandparents in the present sample is also analysed for heritability using parents as propositi.

The control of movement in the preferred and non-preferred hands.

The nature of the difference in skill between the preferred and non-preferred hands was investigated using a peg-board task. The first experiment examined the effects of varying movement amplitude and target tolerance on performance. The difference between hands was found to be related to tolerance rather than movement amplitude. The second study analysed a film record of well-practised subjects, confirming the hypothesis that most of the difference between hands is due to relative slowness of the non-preferred hand in the positioning phase involving small corrective movements. Analysis of the type and number of errors further suggested that this result is not due to differences in duration of movements but to their increased frequency, implying greater accuracy of aiming with the preferred hand. Thus whilst the initial gross analysis implicated feedback processing in skill differences the more detailed analysis suggests that motor output of the nonpreferred hand is simply more variable.

Genetic and nongenetic influences on handedness.

A brief review of theories of the causes of handedness shows why the major determinant is probably accidental variation, modified in man by a genetic factor inducing dextral bias. Levys (1976, 1977) criticisms of Annett (1973) are answered by a detailed consideration of Chamberlains (1928) report and by the presentation of an alternative analysis of the Hull family data using a criterion more similar to that of Rife (1940). The revised analysis is incompatible with the Levy-Nagylaki (1972) model of the genetics of handedness. New data on family preferences in two further student samples support the Hull findings. Data on manual skill in families are more consistent with an accidental than with a polygenic origin of human variability in handedness.

Hand preference and the laterality of cerebral speech.

A theory of the origins of hand preferences (Annett, 1972) is outlined and its implications for the distribution of cerebral speech in handedness groups are examined. Five studies of dysphasia in right and left handers with unilateral cerebral lesions are compared and found surprisingly consistent. Data from three series found alike in all essential respects are combined to give a substantial number of cases with which to test the expectations of the theory. The agreement is close enough to justify a reconsideration of the model in the light of the evidence and also an analysis of possible reasons for the main differences between series. In spite of the many uncertainties involved in inferences from dysphasia to cerebral speech laterality, it is concluded that the evidence is strong enough to offer a useful test of hypotheses as to the distribution of cerebral speech in handedness groups. Other questions bearing on handedness and cerebral speech are briefly considered in order to show the explanatory range and power of the theory.

Mathematical Ability and Lateral Asymmetry

The hypothesis that special ability in mathematics is associated with a reduction in bias to dextral preference and skill was examined in several samples of students and in 97 male and 27 female teachers of mathematics, mainly in Universities and Polytechnics. The math students and math teachers differed from controls, both general and academic, in the direction predicted and several comparisons were statistically significant. Differences were in most cases clearer for males than females. An analysis of the findings in relation to the right shift (RS) theory of handedness (Annett, 1972, 1978) suggests that the incidence of left preference and skill is slightly raised in mathematicians not because of any intrinsic advantage of left preference but rather because extreme bias to the right, as expected in those carrying a hypothesised rs++ genotype, is disadvantageous for mathematical thinking. If the role of mathematics can be regarded as one of developing languages to describe those aspects of human experience which otherwise could be understood only in visuo-spatial images, it can be seen to require a coordination of those aspects of human intelligence which have been distinguished as depending differentially on the left and right cerebral hemispheres. The present findings suggest that this process might be impeded by a double dose of a gene which promotes left hemisphere language specialisation.

Five Tests of Hand Skill

Four group tests of hand skill, square marking (SQUARES), dotting between targets (DOTS), line drawing between targets (LINES) and punching holes through targets (HOLES), were given to samples of undergraduates and schoolchildren, most of whom were also tested individually on a peg moving task (PEGS). Findings for PEGS were shown to be comparable to those of previous samples. Each new task was compared with PEGS for a standard measure of hand skill asymmetry (R-L%) using 4 criteria of comparison. For 3 criteria there was good agreement: correlations were statistically significant, hand preference subgroups were similar for rank order and the extent of asymmetry was more strongly associated with left hand scores than with right hand scores. This last finding was true for all of 12 comparisons for differing tests and samples, indicating that the finding is unlikely to be an artefact of score transformations as suggested by Bishop (1990a, 1990b). The fourth criterion of comparison, the distribution of R-L% scores, differed between the tasks. All of the new tasks found larger differences between the hands than PEGS. Whereas the shape of the distribution was unimodal for PEGS, as in previous studies, the distributions for DOTS and HOLES were clearly bimodal. Possible reasons for the larger between hand differences for some tasks are considered. It is suggested that the main difference between PEGS (and also to a lesser extent LINES) in comparison with DOTS and HOLES lies in the demand characteristics of the testing situation. Tasks which merely invite subjects to mark as many targets as possible in a unit of time may underestimate the skill of the nonpreferred hand.

Family handedness in three generations predicted by the right shift theory.

The hand preferences of Open University (OU) students and their relatives, including children, are described. As in earlier series, estimates of heritability are higher for mothers than fathers. There is no evidence of smaller heritability for paternal than maternal grandparents. The distribution of left-handedness in families is examined in the light of predictions of the right shift theory and on the assumption that the shift depends on a single gene. Good agreement is found between the observed and expected numbers of R x R, L x R and L x L families. Predictions are successful for both strict and generous criteria of sinistrality. Generation differences are found between OU students and their parents and between the students and their children. These are discussed from the viewpoint of a possible heterozygote advantage in intelligence. The higher proportion of sinistral children born to sinistral mothers than fathers can be partly accounted for by supposing that the right shift is more effective in females than males.

Atypical cerebral dominance: Predictions and tests of the right shift theory

Alexander and Annett (Brain and Language, in press) described new cases of atypical cerebral specialization, and suggested that these observations and others in the literature could be explained by the right shift (RS) theory. The theory generates specific predictions as to the prevalence of different patterns of cerebral dominance and their distribution among right-handers and left-handers. Predictions differ between strict and generous criteria of sinistrality, as between left writers and non-right-handers. Tests of the predictions against reports in the literature reveal good fits for most data. New studies will test the RS theory if their design permits examination of the present predictions.

Handedness as a continuous variable with dextral shift: Sex, generation, and family handedness in subgroups of left- and right-handers.

Percentages of left-handed relatives were compared between subgroups of left-handers and of right-handers to test the prediction that the subgroups are ordered for probability of carrying thers+/gene (hypothesized by the right-shift theory of handedness). Data included families of undergraduates and Open University students who described their children as well as other relatives. Linear relationships were found between subgroup order and percentage of left-handed relatives. In agreement with previous evidence that some right-writers with weak sinistral preferences (class 2) are in fact more dextral than consistent right-handers (class 1), class 2 tended to have fewer left-handed relatives than class 1 in both samples. Left writers with weak dextral tendencies (class 7) tended to have more left-handed relatives that consistent left-handers (class 8). Comparisons of the distribution of subgroup handedness in undergraduates and their parents revealed strong effects for sex and for generation. In both cases, the main contrasts were not between left-handers and right-handers but between left-handers plus right-handers with weak dextrality and right-handers with strong dextrality. The findings are consistent with the theory that the relevant factor is not handedness as such but, rather, the absence or presence of thers+/gene.