Holistic processing for left–right composite faces in Chinese and Caucasian observers

Abstract

In Caucasian individuals, holistic processing and face recognition is lateralized to the right hemisphere, whereas part-based processing and word recognition is lateralized to the left hemisphere. Whether this hemispheric complementarity holds more generally is unclear. We compare the hemispheric basis of holistic processing of faces in Caucasian and Chinese observers (who, as readers of logographic script, may have different hemispheric organization). Participants made same/different judgements about the left/right halves of two sequentially presented composite faces (comprised of the left half of one face and the right half of another face) when the halves were aligned or were misaligned. There was a larger congruency effect for aligned than misaligned faces, reflecting significant holistic processing, and this was equivalent for face halves judged in the right and left visual fields and for Caucasian and Chinese observers. This same result was replicated in a second study with Caucasian observers, in which we presented the cue simultaneous with the study face, rather than simultaneous with the test face. These findings reflect equal participation of both hemispheres in holistic face perception and suggest that orthographic experience does not necessarily affect the hemispheric basis of holistic processing.

Keywords:

• Holistic processing
• Composite face effect
• Face lateralization
• Lateralization of language processing
• Other race effect

This research was supported by the National Science Foundation to MB and DP [grant number BCS0923763], Temporal Dynamics of Learning Center [grant number SBE0542013; G. Cottrell: Co-PI: M. Behrmann], and Hong Kong Research Grants Council [grant number HKU744512H] to WGH; and Ungerleider Carnegie Prize Fellowship to TTL. The authors thank Jim Tanaka for providing the stimuli used to create the composite faces and Ryan Egan for processing the Caucasian face stimuli and helping with data collection.

This research was supported by the National Science Foundation to MB and DP [grant number BCS0923763], Temporal Dynamics of Learning Center [grant number SBE0542013; G. Cottrell: Co-PI: M. Behrmann], and Hong Kong Research Grants Council [grant number HKU744512H] to WGH; and Ungerleider Carnegie Prize Fellowship to TTL. The authors thank Jim Tanaka for providing the stimuli used to create the composite faces and Ryan Egan for processing the Caucasian face stimuli and helping with data collection.

Supplementary material

Supplemental material for this article is available via the supplemental tab on the article's online page at (http://dx.doi.org/10.1080/13506285.2014.944613).

Notes

¹ We refer to the observers (and the stimuli) in this paper as being “Chinese” or “Caucasian”. We realize there is some inconsistency in our use of these terms. The term “Chinese” can be taken as an ethnic description of the dominant ethnic group in greater China, overseas Chinese, or as umbrella term of ethnic groups in China (Zhonghua minzu), whereas the term “Caucasian” only refers to the Caucasian race. To name the two groups of observers and stimuli at the same level, one ought to call them by the race: Caucasian and Asian. We have elected to use the term “Chinese” for the observers and the displays because “Asian” captures a large group of different observers and faces and there is even a near-race effect within the class of Asian faces: Taiwanese adults show a near-race effect for Philippine faces (Wang, Chien, Hsu, & Chiu, 2011). We realize our terms are not equated but wished to convey the particular subgroup of observers and faces adopted in this study, i.e., Chinese and Caucasian.

² We printed all the 46 Chinese male faces in our database on paper and asked three naïve observers to group them in a way that maximizes face similarity within each group. Each group contained up to five faces (mostly four) and each face was used only once. The Caucasian faces were rated in a similar fashion by a naïve observer at CMU.

³ The black brackets in the figure are for illustration purposes only. In both Experiments 1 and 2, the cue was a yellow frame overlaid on top of the black outline of either the left or the right face half.