Neural activity during generalized flash suppression in visual cortex

The link between the activity of neurons in the nervous system and visual perception remains one of the most significant and puzzling problems in neuroscience. Generalized flash suppression (GFS) in which a salient visual stimulus can be rendered invisible despite continuous retinal input has provided a powerful means to study neural processes directly related to perception rather than to the external stimulus. However, the mechanisms underlying such perceptual suppression remain poorly understood. The goal of this work was to study how different neural signals represent the perception during visual suppression, how reliable the population responses can be used to determine the perceptual states, and what the roles of different visual cortical areas are during perceptual suppression.This work was carried out by using advance computational techniques to analyze multi-electrode recordings from visual cortex including primary visual cortex and extrastriate visual cortical areas, using ocular configurations of suppression and the paradigm of GFS. About 40 gigabytes of high quality data set was analyzed in this work, consisting of simultaneous multi-electrode multi-unit activity (MUA) and local field potential (LFP) recorded from visual cortical areas V1, V2 and V4 collected from three monkeys performing the GFS task.The first part of the thesis is dedicated to identify perception-related information revealed by spectral analysis performed on LFP signals, based on multivariate autoregressive modeling. The modulation revealed by power indicated the degree of synchrony and time of different visual cortex involved in suppression. Functional network analysis, as measured by coherence, revealed significant reduction of network connectivity. The second part of the thesis is dedicated to predict the trial by trial fluctuations of monkey’s perceptual state (visible vs. invisible), from different neural signals, spike and LFP in different visual cortex, based on decoding methods including a variety of computational and statistical techniques such as linear discriminant analysis, logistic regression, and support vector machine. In conclusion, LFP was closely related to perceptual suppression in V1, V2, V4, and the perceptual visibility was also reflected by integrated spiking activity. Multi-electrode recordings carried complimentary information while correlation limited the benefit of population activity. Different perception may be induced by different network and during suppression reduced connection strength was observed. Top-down information from V4 to V2 and then to V1 during perceptual suppression was indicated.Ph.D., Biomedical Engineering -- Drexel University, 201