Abstract
Introduction: The just-noticeable difference (JND) has been used in psychophysics for over a century to determine the minimum threshold required to detect a change between two stimuli. However, little work has examined the JND for assessing sensitivity to changes in social stimuli. Here we attempt to characterize the JNDs at which a neutral face becomes expressive. We hypothesized that expression would interact with sex/gender; specifically, that happy and fearful expressions on females, and angry expressions on males, would have lower JNDs (i.e., a more precise visual discrimination) than opposite pairings due to sex/gender stereotypes for each expression, respectively. Methods: We used an adaptive staircase paradigm to locate the JND between a neutral face and target expressions (anger, fear, joy). Each trial in the staircase procedure consisted of an actor's neutral face displayed randomly in the LVF or RVF, along with a morph of the same actor expressing an emotion in the opposite VF at an intensity determined by the staircase procedure. Results: Male faces consistently had a JND that was lower than female faces. There was also an actor sex/gender by emotion interaction. Consistent with our hypothesis, female happy expressions had a lower JND than male happy expressions. However, both fear (stereotypically feminine) and anger (stereotypically masculine) expressions were lower for male rather than female faces, suggesting that the present results cannot be attributed entirely to stereotype congruency. Conclusion: Given the obtained results, the happiness advantage witnessed for female faces might be linked to a decreased salience of negative emotions and increased salience of positive emotions in female, but not male, faces. These results suggest an evaluative, rather than a stereotypic, happiness advantage for female faces (Hugenberg & Sczesny, 2006). Our findings offer one mechanism to help elucidate previous findings showing increased attention to female happy expressions.
Meeting abstract presented at VSS 2017