Prenatal Testosterone, Empathy, Emotion Recognition, and Facial Mimicry in Women

Abstract

This study examined whether individual differences in facial mimicry, emotion

recognition, and empathy are related to prenatal testosterone (PT) exposure. Previous

studies have found relationships among facial mimicry, emotion recognition, and selfreport

empathy scores; further, imaging studies suggest that mimicry, emotion

recognition, and empathy have shared neural bases. Previous evidence also suggests that

the development of some of these shared brain regions is influenced by prenatal sex

hormones (Goldstein et al., 2001), and a recent study demonstrated that exogenous

testosterone administration decreases facial mimicry in women (Hermans, Putman, & van

Honk, 2006). The present study examined the relationships among PT exposure

(estimated using the 2D:4D ratio), facial mimicry, emotion recognition, and self-reported

empathy in 53 women between the ages of 18 and 25 years. Facial mimicry was induced

and emotion recognition was tested using a newly developed Facial Expression Viewing

and Recognition Task (FEVRT). During the first block of the FEVRT participants

passively viewed dynamic facial expressions of emotion presented on a computer screen;

in the second block they viewed and identified the emotion that was displayed. During

both blocks the participants’ spontaneous facial reactions were recorded covertly using a

hidden video camera. The video recordings were assessed to determine the extent of

participants’ facial mimicry of the FEVRT stimuli, using Ekman, Friesen, and Hager’s

Facial Action Coding System (FACS; 2002). The left 2D:4D ratio significantly predicted

facial mimicry intensity, emotional empathy, and emotion recognition accuracy scores.

For all of the relationships, higher estimated PT levels were associated with lower scores

on the emotion measures. No relationships were found between mimicry and empathy or between mimicry and emotion recognition accuracy. Mediating relationships were not

observed amongst the 2D:4D ratio and mimicry, empathy, or emotion recognition scores.

These findings suggest that mimicry, emotion recognition, and empathy are distinct

processes, which are each independently influenced by PT. The findings support the

hypothesis that prenatal sex hormones have an effect on adult emotion processing. By

understanding the role that the prenatal sex hormone environment has on the

development of emotion processing, we can gain a better understanding of how

individual differences in emotion processing arise.