Sex, Age, and Emotional Valence: Revealing Possible Biases in the ‘Reading the Mind in the Eyes’ Task

The ‘Reading the Mind in the Eyes’ test (RMET) assesses a specific socio-cognitive ability, i.e., the ability to identify mental states from gaze. The development of this ability in a lifespan perspective is of special interest. Whereas former investigations were limited mainly to childhood and adolescence, the focus has been shifted towards aging, and psychiatric and neurodegenerative diseases recently. Although the RMET is frequently applied in developmental psychology and clinical settings, stimulus characteristics have never been investigated with respect to potential effects on test performance. Here, we analyzed the RMET stimulus set with a special focus on interrelations between sex, age and emotional valence. Forty-three persons rated age and emotional valence of the RMET picture set. Differences in emotional valence and age ratings between male and female items were analyzed. The linear relation between age and emotional valence was tested over all items, and separately for male and female items. Male items were rated older and more negative than female stimuli. Regarding male RMET items, age predicted emotional valence: older age was associated with negative emotions. Contrary, age and valence were not linearly related in female pictures. All ratings were independent of rater characteristics. Our results demonstrate a strong confound between sex, age, and emotional valence in the RMET. Male items presented a greater variability in age ratings compared to female items. Age and emotional valence were negatively associated among male items, but no significant association was found among female stimuli. As personal attributes impact social information processing, our results may add a new perspective on the interpretation of previous findings on interindividual differences in RMET accuracy, particularly in the field of developmental psychology, and age-associated neuropsychiatric diseases. A revision of the RMET might be afforded to overcome confounds identified here.