Purpose. To study the dynamics nf orientation tuning in macaque VI. Methods. Cells were stimulated with a last sequence of images. Each image was a sinusoidal grating having a. fixed spatial frequency (optimal for the cell) but random orientations and spatial phases. Given a fixed time-delay parameter τ. we compute the probability τr,(θ) that an orientation angle θ was present in the image sequence τ msec before a nerve impulse was generated by the cell. For each τ. the function rτ (θ) is a probability distribution on the orientation domain. We study the dynamics of orientation tuning by looking at how the distributions rτ(B) evolve with the time-delay τ. Results. Two main patterns of responses are observed; "unimodal" and "multimodal". The broad orientation tuning seen in the input layers 4Cα and -1Cβ is associated with responses that are unimodal in orientation. However, sharper orientation tuning in supragranular and infragranular layeis is accompanied bv dvnamical responses that are multimodal in orientation (with modes often corresponding to orthogonal orientations). Simulations of feedforward networks yield unimodal responses. The dynamical features observed outside layer 4C arise naturally in feedback models. Conclusions. The relatively broad orientation bias seen in 4Cα and 4Cβ may be computed by a feedforward network. Cortical feedback is responsible for sharpening orientationselectivity and causing multimodal responses outside layer 4C. The multimodal dynamics observed might reflect, other aspects of cortical processing, such as the detection of coiners and junctions.
|Original language||English (US)|
|Journal||Investigative Ophthalmology and Visual Science|
|State||Published - 1997|
ASJC Scopus subject areas
- Sensory Systems
- Cellular and Molecular Neuroscience