Abstract

Faces spontaneously capture attention. However, which special attributes of a face underlie this effect is unclear. To address this question, we investigate how gist information, specific visual properties and differing amounts of experience with faces affect the time required to detect a face. Three visual search experiments were conducted investigating the rapidness of human observers to detect Mooney face images. Mooney images are two-toned, ambiguous images. They were used in order to have stimuli that maintain gist information but limit low-level image properties. Results from the experiments show: (1) Although upright Mooney faces were searched inefficiently, they were detected more rapidly than inverted Mooney face targets, demonstrating the important role of gist information in guiding attention toward a face. (2) Several specific Mooney face identities were searched efficiently while others were not, suggesting the involvement of specific visual properties in face detection. (3) By providing participants with unambiguous gray-scale versions of the Mooney face targets prior to the visual search task, the targets were detected significantly more efficiently, suggesting that prior experience with Mooney faces improves the ability to extract gist information for rapid face detection. However, a week of training with Mooney face categorization did not lead to even more efficient visual search of Mooney face targets. In summary, these results reveal that specific local image properties cannot account for how faces capture attention. On the other hand, gist information alone cannot account for how faces capture attention either. Prior experience facilitates the effect of gist on visual search of faces; making faces a special object category for guiding attention.

Highlights

  • IntroductionFaces are efficiently detected in visual search tasks (Hershler and Hochstein, 2005; Williams et al, 2005; cf VanRullen, 2006; Doi and Ueda, 2007)

  • The three-way interaction between inversion, set size and target presence was not significant [F(2,54) = 1.06, p = 0.35, η2p = 0.04]. These results demonstrate that gist information contributes significantly to rapid face detection but does not fully explain how faces capture attention

  • Our results suggest that while upright gist information allows for more rapid face detection, Mooney images do not provide enough information on their own to facilitate pre-attentive detection

Read more

Summary

Introduction

Faces are efficiently detected in visual search tasks (Hershler and Hochstein, 2005; Williams et al, 2005; cf VanRullen, 2006; Doi and Ueda, 2007). Because faces are detected so quickly, they may contain an emergent property that guides our attention in order to process informative social cues. Prior research investigating which properties of a face capture our attention has focused on facial expressions (Williams et al, 2005) and direction of eye gaze (Doi and Ueda, 2007) amongst distractors with neutral facial expression or opposing eye gaze. We evaluate the effects of gist information, individual features and amount of prior experience with the target faces on efficiency of detection

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call