Abstract

The lateral geniculate nucleus (LGN) is conventionally regarded as a passive thalamic relay between the retina and primary visual cortex. However, growing evidence suggests that its neuronal populations engage in higher-order visual information processing. To explore this possibility, we recorded neural activity from the LGN of six adult female Albino rats. Animals were anesthetized, and a 4x4 mm craniotomy was performed over the right LGN. A 32-channel silicon microelectrode array recorded extracellular activity for approximately 26 minutes in a darkened room, with a 13-inch screen positioned tangentially to the left eye at a distance of ~15 cm. The visual stimulus was a 4 × 8-pixel checkerboard in which four randomly selected pixels flickered ON for 200 ms, followed by 300 ms OFF; 32 distinct patterns were presented in pseudorandom order, each repeated 100 times. Spike trains were extracted for each neuron, and their tuning curves were analyzed to estimate receptive fields, evaluating how individual neurons responded to the stimuli. Cross-correlation analysis was performed to infer functional connectivity, and trial-specific adjacency matrices were used to construct directed graphs depicting excitatory and inhibitory connections. We trained support vector machine (SVM) classifiers to distinguish between all possible pattern pairs using three representations of population activity: raw spike trains, spike counts, and connectivity matrices. Across animals, mean classification accuracy for the top 10 stimulus pairs was highest when using connectivity features (mean = 0.86), followed by spike counts (mean = 0.79) and spike trains (mean = 0.75). Effective connectivity outperformed the other representations in five out of six rats (Wilcoxon signed-rank test, p < 0.05), suggesting that incorporating network-based features enhances decoding performance. Statistical significance was corroborated using t-tests, yielding complementary results. Tuning curve and receptive field analyses revealed that individual neurons did not consistently prefer the most accurately classified patterns, suggesting that improved decoding arose from emergent population-level dynamics captured through connectivity. These findings support the hypothesis that interactions among LGN neurons encode information about visual stimuli, consistent with a computational role for the LGN beyond simple signal relay. They further align with recent anatomical findings and theoretical models implicating the LGN in thalamocortical coordination. Overall, our results underscore the value of connectivity-based representations in understanding sensory population coding and inform future applications in visual prosthetics.

School

School of Sciences and Engineering

Department

Biotechnology Program

Degree Name

MS in Biotechnology

Graduation Date

Winter 1-31-2026

Submission Date

9-18-2025

First Advisor

Seif Eldawlatly

Second Advisor

Ahmed Moustafa

Committee Member 1

Mohamed Salama

Committee Member 2

Bassem Abdullah

Committee Member 3

May Bakr

Extent

81 p.

Document Type

Master's Thesis

Institutional Review Board (IRB) Approval

Not necessary for this item

Disclosure of AI Use

Thesis text drafting; Code/algorithm generation and/or validation

Download

Available for download on Saturday, September 18, 2027

Share

COinS