Studying sex difference in the brain is controversial given the existence of societal gender-biases and a long history of biological data being used justify sexist ideas (1). However, the investigation on sex differences in the human brain is of great interest given the sex differences in the prevalence of certain diagnoses (2, 3). For example, there is a greater prevalence of depression, chronic pain and Alzheimer’s disease in females. On the other hand, males have higher rates of autism, dyslexia, and Parkinson’s disease. To date, many reviews of the literature have reported inconsistencies of finding across studies. This is not surprising based on the extensive variation in methodological and analytical decisions and strategies employed such as how brain regions are defined, the age of range of participants, and controlling for total brain size. Furthermore, a great majority of studies used small sample sizes and therefore lacked the statical power to detect differences even if they existed.
To address limitations due to sample sizes in the neuroscience of sex differences, Ritchie et al.2 utilize data from the UK Biobank to perform a comprehensive well powered study of sex differences in the adult human brain 5226 participants. The UK biobank (https://www.ukbiobank.ac.uk/) is a large-scale biomedical database and research resource that has enabled several scientific discoveries that have led to improvement in human health. Since 2006, UK Biobank has collected an unprecedented amount of biological and medical data on half a million people, aged between 40 and 69 years old and living in the United Kingdom.
Methods
This study examining brain imaging data from 2750 females and 2466 males from 47 -77 years of age with an average age of 61.7 y. Subjects had structural T1-magnetic resonance imaging (MRI, gray matter), functional resting-state MRI, and diffusion tension imaging (DTI, white matter) assessments. Analyses were performed to examine sex difference in the mean and variability of total and regional gray matter volume, cortical thickness, and surface area from T1 imaging, white matter microstructure from DTI, and the large-scale organization of functional brain networks from resting state fMRI. Importantly, differences in brain regions were determined before (raw analyses) and after (adjusted analyses) controlling for total brain size (, i.e., total gray matter volume for regional volumes, mean cortical thickness for regional thickness). The author used total gray matter volume instead of total intracranial volume as the later shows minimal change with age. All analyses controlled for age and ethnicity by removing their effects via regression before analyzing the data. For mean difference Welsch’s T-test was applied. To test for differences in variance, the Variance ratio F test was employed. Effect sizes were computed using Cohen’s d which reflect difference between in terms of standard deviation units. As a rule of thumb Cohen’d d is d=0.20 is considered small, d=0.50 moderate, and d=0.80, large. Estimation of effect sizes is important for research effects can be statistically significant even when the magnitude of those effects is negligible and meaningless. Based on this study’s sample size there was adequate power to detect an effect size as low as d=.08.
Results
The study reported several findings:
- Males generally had greater gray matter volumes and surface area compared to female that showed greater cortical thickness, with effect size differences as large as d=1.4.
- Sex differences in regional brain volumes after controlling for total brain volume. were small, ranging from d=.10 to d=.20
- Based on white matter properties, females showed lower directionality and higher white matter tract complexity
- Resting state analysis showed stronger connectivity for males in unimodal sensorimotor cortices. The sensorimotor cortices receive sensory input from the periphery and is important for body sensations awareness and generation of appropriate motor responses. Stronger connectivity for females in the default mode network. the default mode network is widely distributed brain network active during internally focused, non-task directed thought. It has been associated with sense of self, rumination, future planning, and introspective thought.
- There was generally greater male variance across the raw structural measures.
- There was considerable distributional overlap between the sexes for all brain measures even those that should differences.
Limitations
- The age range of the UK biobank, 44-77, is not representative of the life span
- The was no control for menopausal status or hormone replacement therapy that may affect the brain.
- No control of psychosocial experiences over development.
Conclusions
- This study supports the hypotheses that sex differences in brain structure and function exists.
- The neurobiological and other mechanisms that drive these differences remain to be determined.