The role of language and culture in face and scene processing and description strategies

• Main Author: Lee, Ai-Suan
• Published: University of Nottingham 2018
• Subjects: QP Physiology
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.748350

Face perception is important in a variety of human social interactions, allowing us to keep track of individuals’ identities, recognise emotional expressions and intentions and make judgements about variables such as age, ethnicity and health. While early research assumed that face recognition strategies were universal, more recent studies have shown that East Asian and White Caucasian observers use different looking strategies to recognise faces, with East Asian participants focusing more on the centre of the face, which has been interpreted as representing a configural processing strategy, while White Caucasian observers fixate more on the eyes and mouth, which has been interpreted as representing a more featural processing strategy. Debate continues over the reasons behind this difference, with some researchers arguing that it represents an extension of more holistic cognition in the more collectivist East Asian cultures, and more analytic cognition in individualist Western cultures. The Sapir-Whorf hypothesis suggests that cognition is bound by language, and there have been studies showing changes in response patterns in tasks conducted by bilingual participants in their different languages. Others argue that these differences in face processing are driven instead by different salient diagnostic features of faces of different ethnicity. In this thesis, I present the results of five studies examining the role of culture and facial appearance in determining the looking strategy of East Asian and White Caucasian observers. In Chapter 2, we attempted to use a Navon task to prime featural or configural processing in Malaysian Chinese observers engaged in a face recognition and description task of East Asian and White Caucasian faces. While the Navon task failed to elicit a change in either looking or description strategy, it was noted that the features fixated on most were not the features described most frequently. Further, the race of the face impacted on the looking strategy used to recognise faces, with participants fixating more on Caucasian hair than Asian hair, suggesting that the different diagnostic features may drive differences in looking strategies. It was also casually observed that observers with stronger Asian accents made more configural descriptions. In Chapter 3, I investigate the strategies used by Malaysian Chinese and White Caucasian observers when recognising and describing East Asian and White Caucasian faces. A linguistic/cultural priming paradigm was used in an attempt to induce featural or configural processing in observers. In Study 1, the East Asian observers’ eye movements were impacted by the race of the faces, making more fixations on Caucasian hair and eyes than on Asian hair and eyes. Again, patterns of looking and description were very different. Also, the description patterns differed by language, with participants making more descriptions of hair when speaking English and more descriptions of noses when speaking Chinese, suggesting that descriptions may be constrained by language. In Study 2, White Austrian Caucasian observers again showed very different description and fixation patterns. Observers again showed different fixation patterns for Asian and Caucasian faces, fixating more on Caucasian hair than Asian hair, suggesting that fixation pattern may be driven by the diagnostic features of the faces. Observers made more descriptions in German than in English, but did not show a difference in the pattern of describing different facial features depending on either the language spoken or the race of face, suggesting that the more similar German and English languages have similar constraints. Asian observers have been previously shown to direct more attention to contextual information in images of scenes than Caucasian observers, possibly due to a more holistic/configural cognitive style. Since it is known that faces are processed in a different way to other stimuli, in Chapter 4, I report the results of two studies investigating the impact of linguistic/cultural priming on participants’ eye movements and descriptions when describing street scenes. The Malaysian Chinese participants made more fixations on, and descriptions of, nonfocal than focal objects in Asian street scenes and when speaking Chinese, but not when describing European scenes in English. The White Austrian Caucasian observers did not show any difference in fixation or description patterns depending on linguistic condition, other than making more descriptions overall in German than in English. This suggests that, in a non-face description task, linguistic/cultural priming was successful in eliciting cultural “frame shifting” in Malaysian Chinese participants speaking English and Chinese, but not in Austrian Caucasian participants speaking the culturally more similar English and German. We conclude that culture/language does impact on description patterns in face and scene stimuli, possibly reflecting the constraints of different languages. Further, an impact of linguistic/cultural priming was found on fixation patterns in street scene stimuli. However, in face perception tasks, race of face, but not cultural/linguistic condition, impacted on fixation patterns. We conclude that, while language and culture may have an impact on cognition, and place constraints on descriptions, the diagnostic features of faces appear to primarily determine the fixation patterns on face stimuli.