Our first experiments involved prosimian primates, which retain many primitive features from when they diverged from our own ancestors some 50 million years ago. We implemented a mobile gaze-tracking system and recorded gaze behavior in ringtailed lemurs at the Duke Lemur Center. The small size and weight of the system allowed the lemur to socialize and move freely about his enclosure. (It is probably similar to a heavy backpack on a person.) I wrote up some crude Matlab code to help analyze video data aquired through this or similar systems, available here. Skriatok VIDEOSCORE provides a GUI interface for manual video tracking of point-of-regard and other target locations relative to each other and to static environmental markers, while Skriatok VIDEOSKETCH attempts to extend these abilities by confirming and correcting gaze calibration, by displaying gaze and scene statistics in terms of world, head, or eye-centered coordinates, and by exporting marked frames or video. (Shepherd & Platt 2006) The code is not under active development, but if you are interested in taking it over, please contact me and I'll help however I can. With this technology we were able to show that lemurs preferentially attend toward their social group when at rest, when vision isn't being used for other tasks, such as planning movement. Furthermore, lemurs appear to use one another as social cues to their environment: when our subjects saw another lemur, they tended to look next in the direction the observed lemur gazed. (Shepherd & Platt 2008) To see annotated data recorded during this project, please see these example video clips.

        I next looked at shifts in attention as rhesus macaques respond to the observed gaze of others. My observations suggest that gaze-following is not a single process but involves both automated and cognitive components. The first is characterized by a rapid (<200ms) shift of attention, followed later (>600ms) by inhibition of return, and appears not to be modulated by social context. This pattern predominates among low-status / low-testosterone male macaques. Among high-status / high-testosterone male macaques, however, a different pattern was dominant. These macaques also follow gaze, but the effect develops only after longer cue exposures (>100ms), and then only in response to cue images from likewise dominant macaques. This second process thus suggests a more cognitive, volitional type of social attention. (Shepherd et al 2006)

        Our studies in humans observed a slightly different breakdown. Looking closely at contextual effects on gaze following, we replicated earlier findings that human females follow gaze more strongly than human males. Furthermore, we found that human females were modulating their gaze-following behavior depending on the identity of the gaze-cue: females followed gaze much more strongly for familiar individuals than unfamiliar. Several caveats apply: most importantly, because only male cue images were used, we cannot distinguish between female advantage and opposite-sex advantage. We do, however, believe that these results strongly support our claim that gaze-following cannot operate via an informationally-encapsulated module: though gaze following is quite fast, it is also quite well integrated with other sorts of social processing. (Deaner et al 2006)

        My final project in the Platt lab was an investigation into the neural pathways through which gaze-following operates. We have looked in the parietal lobe of primates, a region located in the upper rear of the one's head. We recorded single unit activity in the lateral intraparietal sulcus (area LIP) of rhesus macaques, and observed that subpopulations within this region act both to encourage and suppress behavioral responses to irrelevant gaze cues. (Shepherd et al, in press)

Relevant Publications:

"Neuroethology of Social Attention in Primates."
Shepherd SV.
Dissertation, Duke University, 2008.
[pdf]

"Mirroring of attention by neurons in macaque parietal cortex."
Shepherd SV, Klein JT, Deaner RO, Platt ML.
Proceedings of the National Academy of Science, in press
[pdf] [press]

"Neuroethology of Attention in Primates."
Shepherd SV, Platt ML. In Cognitive Biology: Evolutionary and Developmental Perspectives on Mind, Brain and Behavior (eds. L. Tommasi, M. Peterson & L. Nadel), MIT Press, 2009.
[contact for preprint]

"Spontaneous social orienting and gaze-following in ringtailed lemurs (Lemur catta)."
Shepherd SV, Platt ML.
Animal Cognition 11(1):13-20, 2008.
[pdf]

"Familiarity accentuates gaze-cuing in women but not men."
Deaner RO, Shepherd SV, Platt ML.
Biology Letters 3:64-7, 2006.
[pdf]

"Social status gates social attention in monkeys."
Shepherd SV, Deaner RO, Platt ML.
Current Biology 16(4):R119-R120, 2006.
[pdf] [press] [audio]

"Noninvasive telemetric gaze tracking in freely-moving socially-housed prosimian primates."
Shepherd SV, Platt ML.
Methods 38(3):185-194, 2006.
[pdf]&[Skriatok scripts for Matlab]