There is a Poster PDF for this presentation, but you must be a current member or registered to attend VSS 2023 to view it.
Please go to your Account Home page to register.
Holistic face processing is an influential account of the mechanisms of face encoding. However, there is no clear agreement across researchers as to what exactly holistic processing entails. It is also well established that attention is necessary for encoding of unfamiliar faces but precisely how it functions in face encoding is not well understood. This study investigates whether attention helps in binding features into a holistic representation by studying the illusory conjunction, which is to mistakenly combine features from one stimulus to another when focused attention is preoccupied. In Experiment 1a, participants were cued to learn 2 digits while 2 faces were presented simultaneously and parafoveally for 200 to 300ms. After reporting the 2 digits, participants identified the presented face amongst 2 novel faces, and 1 conjunction face constructed by conjoining different facial features from the 2 presented faces. Participants were able to reject the novel faces but were very poor at differentiating between the intact and the conjunction face. Reversing the primary task and cueing to the faces while leaving the rest of the paradigm unchanged (Experiment 1b), participants were better at rejecting not only the novel faces but also the conjunction face. Experiment 2 replaced 1 novel face with a “feature face” created by conjoining 1 novel facial feature with the rest of one of the presented faces. The feature face showed a similar response profile as a novel face, ruling out the possibility that the high response rate to the conjunction face was triggered merely by the similarity between the conjunction and studied faces. Our findings argue for a holistic face representation that is integrative in nature. Not all face processing mechanisms require attention. Featural information can be extracted preattentively, but that features are left unbound until attention is available to bind them into a holistic representation.
Acknowledgements: This research was supported by a Grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (HKU17608519) to W.G.H.