Neuroimaging studies have documented modulation of the activity of the amygdala – a key node in the neural network underlying emotion perception and processing, and one that has also been associated with regulating aggression – by exogenous testosterone. However, results on the impact of normal range testosterone levels on explicit emotion recognition as a prerequisite for social interaction and amygdala activation in healthy young males are missing.
Hence, we performed functional MRI at 3 T in a group of 21 healthy males during explicit emotion recognition with a protocol specifically optimized to reliably detect amygdala activation. We observed similar amygdala activation to all emotions presented without any effect of gender of poser or laterality. Reaction times to fearful male faces were found negatively correlated to testosterone concentration, while no significant effects emerged for other emotions and neutral expressions. Correlation analyses revealed a significant positive association between testosterone levels and amygdala response to fearful and angry facial expressions, but not to other expressions. Hence, our results demonstrate that testosterone levels affect amygdala activation and also behavioral responses particularly to threat-related emotions in healthy young males. We conclude that these findings add to our understanding of emotion processing and its modulation by neuroendocrine factors.
Converging evidence has shown that testosterone levels are associated with basic social abilities, e.g., facial mimicry as one component of empathic behavior (Hermans et al., 2006a), mood and selective attention to angry faces (Van Honk et al., 1999), and also moderation of the reinforcing qualities of angry faces (Wirth and Schultheiss, 2007). While animal research has unequivocally demonstrated the connection between elevated testosterone levels and increased aggressiveness (e.g., Lumia et al., 1994 and Melloni et al., 1997) in human studies only correlational evidence has been reported (e.g., Dabbs et al., 1995 and Archer, 2006). Regarding the possible anxiolytic effect of testosterone, this phenomenon has been replicated many times and in a large variety of animal species (e.g., Aikey et al., 2002). Although there is evidence for antidepressant effects in hypogonadal depressive patients after testosterone treatment (e.g., Wang et al., 1996), clear reductions in fear after testosterone administration or treatment have rarely been reported for humans (Van Honk et al., 2005 and Hermans et al., 2006b).
The special role of the amygdala in the processing of threat-related stimuli, in particular anger and fear is well documented (e.g., Adolphs, 2002). Consequently, it has been argued to be strongly involved in the pathways controlling aggression, and most neuroimaging studies have consistently observed amygdala activation to angry facial expressions (e.g., Whalen et al., 2001).
Recently, several neuroimaging studies have shown a modulating effect of testosterone on amygdala activation: Van Wingen et al. (2008a) used an emotion matching task and administered a single dose of testosterone to middle-aged healthy females. This modulated amygdala activation, leading to a higher reactivity comparable with the one of young, healthy females. Also, applying exogenous testosterone to healthy young females who were presented with angry faces in a passive viewing task, Hermans et al. (2008) observed stronger amygdala activation in subjects to whom higher testosterone doses had been administered.
Studies investigating possible association of testosterone levels with explicit emotion recognition, another basic prerequisite for social interaction, and the underlying amygdala activation are totally missing. Hence, we investigated whether normally distributed testosterone levels in healthy young males exert an influence on explicit emotion recognition and amygdala reactivity. We performed functional magnetic resonance imaging (fMRI) using an explicit emotion recognition paradigm in healthy young males. Based on previous results from our group (e.g., Habel et al., 2007 and Derntl et al., 2008a), we hypothesized amygdala activation to all emotions presented. Moreover, we hypothesized a significant positive association between testosterone levels and amygdala reactivity in healthy males especially to threat-related stimuli as shown by previous fMRI studies in healthy females (Hermans et al., 2008 and Van Wingen et al., 2008a) and behavioral data (Van Honk et al., 1999 and Wirth and Schultheiss, 2007). In light of assumptions that males respond more strongly to male faces (cf. Mazurski et al., 1996), we also investigated this aspect by presenting both female and male faces.
