September 2005
Volume 5, Issue 8
Vision Sciences Society Annual Meeting Abstract  |   September 2005
Person recognition across multiple viewpoints
Author Affiliations
  • Sapna Prasad
    Rutgers University
  • Fani Loula
    Rutgers University
  • Maggie Shiffrar
    Rutgers University
Journal of Vision September 2005, Vol.5, 941. doi:10.1167/5.8.941
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sapna Prasad, Fani Loula, Maggie Shiffrar; Person recognition across multiple viewpoints. Journal of Vision 2005;5(8):941. doi: 10.1167/5.8.941.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Individuals' identities can be determined from point-light displays of their actions (Cutting, 1977). Interestingly, observers demonstrate the greatest visual sensitivity to their own movements (Loula, Prasad, & Shiffrar, 2005). Does enhanced visual sensitivity to self-produced motion reflect a lifetime of experience observing one's own limbs move? If so, then visual sensitivity to actor identity should be viewpoint dependent. To test this hypothesis, naïve participants viewed point-light movies of themselves, friends, and strangers performing various actions. Viewpoint varied across condition. Actors were matched for gender, age, and body size so that these cues could not be used for actor identification. Each trial consisted of two different sequentially presented point-light movies of actors performing two different actions. Participants performed a 2AFC identity discrimination task and reported whether each trial showed the same actor or two different actors. In Experiment 1, head mounted cameras were used to create head-centered, axial displays of the point-light actors. In Experiment 2, front and rear views of the point-light actors were created. Consistent with Bulthoff et al (1997), performance was uniformly poor with axial depictions since these point-light displays are difficult to organize. This finding is interesting because observers have extensive visual experience with the head-centered, axial perspective on their own body. Yet, this experience was not reflected with a performance increment. In Experiment 2, identity discrimination performance with self-motion was superior to that with friend and stranger motion across both the front and back views. Since observers have substantially more experience viewing the front, as compared to the back, of their own bodies, these findings suggest that identity discrimination is not defined by view specific experience.

Prasad, S. Loula, F. Shiffrar, M. (2005). Person recognition across multiple viewpoints [Abstract]. Journal of Vision, 5(8):941, 941a,, doi:10.1167/5.8.941. [CrossRef]
 Acknowledgement: Supported by NIH EY12300

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.