The long-term objective of this research is to better understand the neural code for object shape by quantifying shape representation at the single cell and neural population levels. This objective is addressed by behavioral and neurophysiological studies of shape representation in the ventral (object-related) pathway of macaque visual cortex, specifically at intermediate and higher levels (V4, PIT/TEO, CIT/TE). The experiments utilize large, parametric shape stimulus sets to define neural tuning functions in shape-related dimensions such as contour curvature. The same stimuli are used for all cells to enable reconstruction of population responses.
Three specific aims are addressed: 1. To test whether shape-tuning functions can predict responses to realistic images of random objects. Photographic images of natural and man-made objects will be used to test the predictive value of tuning functions based on parametric shape stimuli. Preliminary results show that these tuning functions are highly predictive, suggesting that they successfully capture the coding functions of ventral pathway neurons. 2. To reconstruct shape representation at the neural population level in the ventral pathway. Shape is so complex that it can only be represented by a distributed neural code--i.e., by patterns of activity in large neural populations. Standard sets of both parametric and realistic shape stimuli will be used to study large numbers of neurons so that the population-level representations of these stimuli can be reconstructed. Population reconstruction is essential for understanding how neurons cooperate to encode shape information in a distributed fashion. 3. To test for correlation between neural responses and shape perception as revealed at the behavioral level. Neural/behavioral correlations provide the clearest evidence for causal relationships between neural responses and perceptual/cognitive events. Ambiguous shape stimuli will be created by parametric morphing between behavioral target stimuli. Behavioral decisions about ambiguous stimuli will be compared with neural responses during the same trials. Microstimulation will be used to attempt to bias behavioral responses in the direction indicated by local shape tuning. ? ?
| Sasikumar, Dennis; Emeric, Erik; Stuphorn, Veit et al. (2018) First-Pass Processing of Value Cues in the Ventral Visual Pathway. Curr Biol 28:538-548.e3 |
| Connor, Charles E; Knierim, James J (2017) Integration of objects and space in perception and memory. Nat Neurosci 20:1493-1503 |
| Yau, Jeffrey M; Pasupathy, Anitha; Brincat, Scott L et al. (2013) Curvature processing dynamics in macaque area V4. Cereb Cortex 23:198-209 |
| Carlson, Eric T; Rasquinha, Russell J; Zhang, Kechen et al. (2011) A sparse object coding scheme in area V4. Curr Biol 21:288-93 |
