High-level visual function in non-human primates

点击翻译按钮获取中文摘要

The objective of this grant proposal is to seek funding to permit face recognition studies on non-human primates. We have recently found that faces that are defined by dynamic cues are recognized above chance by human observers. This was surprising because although form-from-motion is a well-known visual phenomenon, veridical face discrimination is unexpected because it is known to be a very high-level process.This finding has significant implications for the way that the two dominant visual streams-dorsal and ventral pathways-interact with each other. A face stimulus defined by dynamic cues must, according to our current understanding, arise in dorsal stream structures. However, face processing leading to recognition is widely accepted to be a ventral stream function. Our results imply that a significant degree of information relay must occur at the highest levels between the two streams to permit face recognition with dynamic cues.The specific goal of the proposed study is to assess visual function in non-human primates using similar display protocols as with our study on humans. We wish to undertake this study in both an Old World and a New World species (vervet and marmoset monkeys, respectively). Face recognition performance has not been well documented in either species and therefore the planned experiments will break new ground in this area. Furthermore, a finding that vervets and marmosets are capable of recognizing faces defined by dynamic cues would imply that the interaction between the two streams generalizes across primates and pave the way for more detailed anatomical and physiological explorations of high-level visual function in these species.