Affiliation:
1. Department of Psychology, North Dakota State University, Fargo, ND 58102, USA
2. Center for Visual and Cognitive Neuroscience, North Dakota State University
Abstract
Observers are capable of distinguishing real faces from artificial faces of various types (eg dolls, computer-generated faces) relatively easily. While a number of diagnostic cues are potentially available to observers to accomplish this task, the appearance of the eyes has been shown to be critically important. However, eye appearance appears to interact with other cues, like the appearance of the skin, in some settings. The ‘uncanny’ appearance of some artificial faces appears to result from multiple visual features and their departure from typical face norms, for example, and recent results investigating how real and artificial features are perceived in chimeric faces also suggest that observers use multiple cues to measure face animacy. Presently, we examined the cues that support real-artificial face discrimination by using contrast negation and so-called ‘contrast chimeras’ to selectively disrupt the appearance of the eyes and the remainder of the face pattern. First, we demonstrate that, like other aspects of face perception, perceived animacy is significantly impacted by contrast negation. Second, by selectively manipulating the contrast of the eyes relative to the rest of the face, we demonstrate that these face regions are of approximately equal use to observers for animacy discrimination.
Subject
Artificial Intelligence,Sensory Systems,Experimental and Cognitive Psychology,Ophthalmology
Cited by
25 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献