Preliminary investigation of the temporal specificity of sequence learning in the primary visual cortex through predictive coding

The primary visual cortex (V1) has been classically viewed as an immutable feature detector, with robust responses to low-level characteristics of objects in the visual field. Recent studies have shown the capacity of this cortical area to perform more complex computations. Nominally, the phenomenon of sequence learning relies on the ability of V1 to encode the serial order and temporal frequency of a spatiotemporal visual sequence. Investigating the mechanisms driving this phenomenon through the lens of predictive coding will further the understanding of how V1 operates locally to encode time and learns to predict the future based on minimal sensory information. Through in vivo multi-unit recordings from awake mice, this study sought to isolate neural evidence for predictive processing within the paradigm of sequence learning. Seventy unique units were isolated from forty-two mice subjected to experimentation. Preliminary analyses revealed a significant effect that agrees with the initial report on sequence learning but contradicts predictive processing theory. Further investigation is required to draw more robust conclusions about the predictive computations that occur during sequence learning. Increased sample size and refinement of data analysis will likely lead to interesting results

Author dc:creator

• Khoudary, Anthony Amin

Advisors dc:contributor.advisor

• Gavornik, Jeffrey
• Davison, Ian

• Neurosciences

Language dc:language.iso

en_US

Handle dc:identifier.uri

https://hdl.handle.net/2144/44040

OAI identifier oai:identifier

oai:open.bu.edu:2144/44040