Endogenous testosterone and cortisol modulate neural responses during induced anger control

Abstract

Research with violent offenders and delinquent adolescents suggests that endogenous testosterone concentrations have the strongest positive correlations with violence among men who have low concentrations of cortisol. The present study tested the hypothesis that testosterone and cortisol would similarly interact to determine neural activation in regions supporting self-regulation in response to anger provocation. Nineteen healthy Asian male participants were insulted and asked to control their anger during functional magnetic resonance imaging (fMRI). When cortisol levels were low, testosterone positively correlated with activation in the dorsolateral prefrontal cortex (dlPFC) and thalamus, but not when cortisol levels were high. During induced anger control, functional connectivity was increased between the amygdala and a top-down prefrontal cortical control network. Moreover, the amygdala-PFC connectivity was strongest among those high in testosterone and low in cortisol. This research highlights a possible neural mechanism by which testosterone and cortisol may influence anger control.

Keywords:

• Prefrontal cortex
• Testosterone
• Cortisol
• Anger control
• Anger

Acknowledgments

This research was supported by the Australian Research Council's Discovery Projects funding scheme. Thanks to Lynette Roberts for help with data collection.

Notes

¹The recruitment of an Asian sample was unintentional. By chance, all participants who responded to the advertisement and passed safety screening were of Asian background. There is a large Asian student population at UNSW.

²As part of an unrelated study, participants also completed a Stroop and reversal learning task prior to the anagram task.

³Although some researchers suggest that GCM should not be used to determine effective connectivity, GCM is an acceptable method for identifying functional connectivity between neural regions across time (as is psychophysiological interaction analysis) (David et al., 2008; Friston, 2009).