Is there a relationship between the performance in a chronometric mental-rotations test and salivary testosterone and estradiol levels in children aged 9-14 years?

The consistent gender differences favoring males in some spatial abilities like mental rotation have raised the question of whether testosterone or other gonadal hormones contribute to these differences--especially because such gender differences seem to appear mainly from the age of puberty on. Studies generally suggest that spatial ability is facilitated by moderately high testosterone levels (i.e., levels that are relatively high for females and relatively low for males). However, the role of sex steroids for mental-rotation performance of (pre-) pubertal children has not been the focus of research, yet. In our study, the relationships between different aspects of mental-rotation performance (accuracy, reaction time, rotation speed) and salivary testosterone and estradiol levels were investigated. Subjects were 109 children (51 boys and 58 girls) aged between 9 and 14 years (M = 11.41, SD = 1.74). They performed a chronometric mental-rotations test, in which the stimuli consisted of three-dimensional drawings of Shepard and Metzler cube figures. In addition, saliva samples were gathered for the analysis of free testosterone and estradiol levels. Results showed a significant gender difference in reaction time and rotational speed in favor of boys, and a significant age, but no gender difference in testosterone and estradiol levels. We found no significant relationships between hormonal levels and any measure of mental-rotation performance.