Computational exploration of the relationship between holistic processing and right hemisphere lateralization in featural and configural recognition tasks

Holistic processing has long been considered as a property of right hemisphere (RH) processing. Nevertheless, recent studies showed reduced holistic processing and increased RH lateralization in Chinese character recognition expertise, suggesting that these two effects may separate. Through computational modeling, in which we implement a theory of hemispheric asymmetry in perception that posits a low frequency bias in the RH and a high frequency bias in the left hemisphere, we show that when the recognition task relies purely on featural information, holistic processing increases whereas RH lateralization decreases with increasing stimulus similarity; there is a negative correlation between them. In contrast, when the task relies purely on configural information, although RH lateralization negatively correlates with stimulus similarity, holistic processing does not correlate with stimulus similarity; there is a positive correlation between them. This suggests that holistic processing and RH lateralization do not always go together, depending on the task requirements.