Language processing in both sexes: evidence from brain studies.

Thank you for the interesting paper by Frost et al. (Frost et al., 1999). As they suggest, understanding sex differences in the human brain is not trivial. Many studies of disease groups are conducted on one sex only and extrapolated across genders. Indeed, the basis of the Talairach atlas is one example of this approach. Therefore, determining whether clear sex differences exist in language-associated brain is important for future work. The work by Frost et al. adds important data to this confusing field. They report no differences between men and women on a semantic monitoring task (identify if an animal is ‘found in the United States’ and ‘used by human’) and then report no differences in functional MRI brain activation. The authors should be commended for their tight statistical design with a large number of subjects, an aspect of the work which should be used in future studies. However, it is important to note that such a semantic monitoring task would not be expected to show sex differences. These differences have been identified and replicated clearly in many studies in which females excel in tasks of verbal fluency and verbal memory types of language skills (Bradshaw, 1989; Mann et al., 1990; Halpern, 1992; Kimura, 1996). In defence of the expectation of Frost et al. that their semantic–lexical task would show sex differences, the authors argue that their task is an auditory equivalent of Pugh’s task on sex differences in aspects of reading. However, the task of Pugh et al. differs significantly from these auditory tasks due to the involvement of grapheme-based visuospatial aspects that change the task from a simple lexical–semantic task by adding in aspects of reading (Pugh et al., 1996). Unfortunately, Frost et al. have suggested that their results lead to a conclusion that there are no sex differences in language processing or in the neural organization of such processing [apart from those shown by Witelson et al. (1995)]. Although they discuss the data of Shaywitz et al. (1995) on functional language differences (Shaywitz et al., 1995) and our data on large proportional size differences in the planum temporale and Broca’s area (Harasty et al., 1997), the authors prefer to obscurate these data. Indeed Frost et al. suggest that our data are contradicted by those of Abiotiz et al. and by those of Witelson and Kigar