Gonadal steroids and cognitive functioning in middle-to-older aged males.

Abstract

The basis for sex differences in cognitive ability remains poorly defined and controversial both scientifically and politically. One of the biological hypotheses on sex differences, of particular relevance to this thesis, concerns the role of gonadal steroids, specifically testosterone (T) and oestrogen, and their relationship to individual differences in the performance of specific cognitive tasks. In addition, the role that age-related changes in these hormones play in relation to generalised and pathological cognitive ageing in males is studied. It is important to determine whether decreases in T levels that occur with ageing in males are associated with age-related decreases in cognitive performance because T levels can potentially be modified. Males have consistently been found to outperform females on measures of visuo-spatial function; performance on the Vandenberg and Kuse Mental Rotation Test (MRT) shows the largest and most robust of sex differences. Gonadal steroids have both organisational and activational effects which contribute to both withinsex variability and between-sex differences in visuo-spatial cognition. As males age, endogenous plasma T levels decline gradually yet variably between individuals. Studies in older males show improvement in visuo-spatial cognition following T supplementation; however, it remains to be resolved whether decreases in endogenous T levels with ageing are associated with poorer MRT performance. Some recent studies in older males have reported positive correlations between measures of plasma T levels and cognitive functioning, including processing speed and executive function measures. These data are inconsistent, however, and important questions remain concerning, for example: the age at which the effect is strongest; whether there are different effects at different ages; whether there is an optimal level at which T levels affect particular abilities; and which abilities show the strongest association with endogenous plasma T levels. Increased intra-individual variability in performance on Choice Reaction Time (RT) tasks has recently been shown to be a strong predictor of cognitive functioning in university students. Methodological advances in the analyses of RT distributions has allowed for the calculation of robust estimates of intra-individual RT variability. The association between these estimates and cognitive performance in middle and older aged males, however, remains to be determined. Further, the association between endogenous plasma T levels and intra-individual RT variability in aged males is unknown. The thesis addresses these issues; firstly, through cross-sectional analyses of the associations between different measures of plasma T levels, learning and memory, processing speed, and executive function performance in a large population based sample of 1046 men aged between 35 and 81 years. Secondly, further cross-sectional analyses are reported from a subsequent study in a healthy sub-sample of 96 of these men on the associations between endogenous plasma T levels, MRT performance, constituent abilities related to MRT performance, and performance on composite measures of both processing speed and executive function. In a third study, these data are re-analysed in relation to intra-individual variability in RT performance. In light of the results of these studies, the role that age-related declines in plasma T levels play in relation to generalised age-related cognitive decline in males is discussed.