Lab Seminar

Interacting with objects in three-dimensional (3D) space often requires the brain to determine their spatial pose (position and orientation) from visual signals. In 3D space, an object’s pose has six degrees of freedom – three specifying position and three specifying orientation. Conversely, retinal images are 2D and thus have only three degrees of freedom – two positional and one orientational. The missing three degrees of freedom must be inferred by the brain, making the construction of 3D visual representations a complex, but fundamental task. In this talk, I will discuss experiments we are conducting in the caudal intraparietal area (CIP) of the macaque monkey to elucidate the neural coding of 3D object pose.
Using novel analytical tools, I will first show that CIP neurons encode the 3D orientation of a planar surface (Rosenberg et al., 2013). I will then discuss experiments examining how texture and disparity cues contribute to CIP orientation selectivity. We find that the convergence of these cues depends on their reliabilities, suggesting that CIP plays an important role in creating robust, multimodal visual representations. Lastly, I will discuss ongoing work showing that individual CIP neurons jointly encode an object’s 3D position and orientation. Together, these results reveal a sophisticated representation of 3D visual spatial information, and suggest that CIP can answer two fundamental questions faced whenever we interact with an object: Where is it? and How is it oriented?

Rosenberg A, Cowan NJ, Angelaki DE (2013) The Visual Representation of 3D Object Orientation in Parietal Cortex. J Neurosci 33:19352-19361.