ABSTRACT
In contrast to the classic view that facial expressions convey specific emotional states, recent theories have postulated that perception is a highly contextualised phenomenon. The present study utilised sequential effects as a probe to examine how the preceding context informs current facial expression perception, while participants performed a binary categorisation task on a sequence of expressions morphed from fearful to disgusted prototypes. We found that preceding stimuli/responses played differential roles in expression-based sequential effects. When preceding responses were analytically controlled for, the participants were biased to categorise the current targets as being either away from or close to the category of preceding expressions, yielding stimulus-related contrast or assimilation effects, depending on whether the stimulus similarity between the preceding and current expressions was small or large. When the stimulus similarity between successive expressions was analytically controlled for, preceding responses slanted current categorisation judgments toward response-related assimilative consequences. Distinct from stimulus-related contrast, both stimulus-related and response-related assimilations were robust since they could persist for expressions presented up to two trials back and when the perceptual quality of the stimuli was poor. We suggest that the criterion-shift account is generally compatible with the overall findings, revealing how facial expression perception is temporally contextualised.
Acknowledgements
This work was supported by the Ministry of Science and Technology of Taiwan (R.O.C.) under Grants 107-2420-H-002-031-RE2 and 108-2634-F-002-022. We thank Andy Young for commenting on an early version of the manuscript.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Recent research showing regrowing interest in sequential effects has tended to equate contrast effects with classical adaptation effects. However, as revealed in the early literature and our discussion on page 16, these two phenomena exhibit distinct characteristics. Given that this issue goes beyond the current scope, we discuss here contrast effects in their original form without referencing the recent literature (e.g., Pascucci et al. (Citation2019)).