Facing Our Standards: On the validity of standardised faces in face perception research

Abstract

Many studies in face perception research use standardised facial stimuli: digitally manipulated facial images used in eye tracking experiments studying facial viewing behaviour and emotion perception. With technological innovations in eye tracking equipment inviting us to study faces out in the wild, the present study investigated this seeming trend in research literature, identifying a predilection for stimulus control as an underlying line of reasoning for standardisation, alongside a potential problem if viewing behaviour showing visual preference for the eyes found in standardised stimuli, does not generalise to viewing behaviour showing visual preference for the eyes in non-standardised, “real” faces. As typical viewing behaviour in the literature shows the eyes have a strong attention maintaining capacity, a data analysis using a range of unusual non-standardised stimuli was conducted, estimating relative dwell time to the eyes. Results showed that while viewing behaviour differed between standardised and non-standardised stimuli, these differences were minor and did not substantially differ from results found in face perception literature. If additional studies confirm this generalisability, increasing the use of non-standardised stimuli could prove useful in in bridging the transition of face perception from the lab to the wild.