Face encoding is not categorical

Consistent evidence across multiple types of contingent aftereffects

O. Scott Gwinn*, Kevin R. Brooks

*Corresponding author for this work

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Race-contingent aftereffects indicate that faces of different races are encoded via dissociable visual channels. Race-contingent aftereffects reflect perceptions of typicality, indicating a gradual transition of activity across channels as faces change from looking more typical of one race to the other We examine whether faces forming more discrete categories (sex: male/female; species: human/monkey) are encoded in a similar fashion, or whether they are instead encoded via more dichotomous categorical judgements. Curves representing the rate of change of aftereffect magnitude as stimuli changed from male to female produced shallow slopes, closely resembling ratings of typicality, but differing significantly from slopes for categorical judgements. For species, aftereffect slopes were significantly shallower than both ratings of typicality and categorical judgements. Overall, these results suggest that the encoding of facial properties such as these is not categorical, but instead involves a graded response as stimulus typicality varies. Aftereffect slopes were similar across the two experiments, raising the possibility of a common system that is recruited during contingent adaptation, regardless of the specific dimension involved or the categories used by the experimenter.

Original languageEnglish
Pages (from-to)867-893
Number of pages27
JournalVisual Cognition
Volume23
Issue number7
DOIs
Publication statusPublished - 9 Aug 2015

    Fingerprint

Keywords

  • Face perception
  • adaptation
  • encoding
  • sex
  • species

Cite this