The effect of facial expression and gaze direction on memory for unfamiliar faces

Abstract

We report data from an experiment that investigated the influence of gaze direction and facial expression on face memory. Participants were shown a set of unfamiliar faces with either happy or angry facial expressions, which were either gazing straight ahead or had their gaze averted to one side. Memory for faces that were initially shown with angry expressions was found to be poorer when these faces had averted as opposed to direct gaze, whereas memory for individuals shown with happy faces was unaffected by gaze direction. We suggest that memory for another individual's face partly depends on an evaluation of the behavioural intention of that individual.

Keywords:

• Face memory
• Facial expression
• Gaze direction

Acknowledgments

This research derives from a master's thesis submitted by SN to Kyoto University.

The authors would like to thank Yuko Morimoto, Keiko Nakamoto, Yukiko Uchida, and anonymous reviewers for their insightful comments on an earlier version of this manuscript.

Notes

¹This database includes standardised Japanese facial photographs. Posers were 122 adults (61 male and 61 female, aged from 19 to 29) who were asked to pose expressions by mimicking examples from the standard Ekman and Friesen (1976) set. All of the pictures were taken in a studio under similar lighting conditions with the same camera positioned at a fixed distance from the poser.

²We only used photos of faces with closed mouths (see Figure 1). Images of smiling faces with bared teeth will contain high-contrast areas around the mouth which might be expected to draw participants' attention from the eye-region, which is important in face recognition.

³It is not possible to compute independent estimates of the false alarm rate in each experimental condition because participants always responded to neutral faces in the test phase. Consequently we did not apply a signal detection analysis to the data.

⁴Participants in the D'Argembeau et al. (2003) study who, like our participants, learned direct gaze faces incidentally (i.e., without instructions to study the faces in preparation for a recognition test) showed no memory advantage for happy over angry faces. Similarly, our data indicate equivalent recognition memory performance for happy and angry faces that were initially encountered with direct gazes.