ABSTRACT
Recent meta-analyses have demonstrated that data from female rodents, tested without regard for estrous stage, is no more variable than male data across a range of traits. Nonetheless, widespread use of male-only samples persists in preclinical studies of anxiety disorders, despite this condition being twice more prevalent amongst women relative to men. We conducted a meta-analysis of over 4900 data points obtained from 263 articles assessing behavioural measures of fear and anxiety in rodents. We found no evidence for greater female variability on any measure. Overall, males had greater variability than unstaged females, which was predominantly driven by studies of learned fear. Compared to unstaged females, staged, but not ovariectomised, females showed reduced variability. Experiments using individual housing and rats were associated with greater variability relative to those using group housing and mice; these effects were not moderated by sex. These results illustrate that the estrous cycle does not inflate variability in females beyond that of males, despite being a female-specific modulator of fear and anxiety behaviour.
