Gary Marsat

The paper shows cerebellar feedback in brown ghost knifefish filters both temporal and spatial features of natural electrosensory signals, adapting to recent input statistics to improve sensory representation efficiency.

The paper models how receptor density affects signal detection and localization in ghost knifefish, finding that higher density improves detection but not localization for diffuse signals.

The paper investigates how subpopulations of hindbrain pyramidal cells in weakly electric fish encode spatial information about conspecifics, finding that specific cell types provide more accurate spatial coding.

The paper compares weighted spike distance metrics to other methods for analyzing sensory neuron population responses and finds that the proposed WED method performs as well or better than neural networks and outperforms traditional spike distance metrics.

The paper quantifies differences in sodium channel density across segments of the electrosensory lateral line lobe and links these differences to burst coding and network regulation.

This study compares neural coding strategies in two closely related weakly electric fish species, finding differences in how they encode communication signals, which correlates with their behavioral use of these signals.

The paper identifies non-random co-expression patterns of neurotransmitters in local interneurons of the antennal lobe, suggesting some organizational principles but also significant neuron-level flexibility.

The paper shows that Apteronotus leptorhynchus uses two distinct neural coding strategies for detection and discrimination tasks, with behavioral performance matching coding strategies.

This study investigates the context-dependent effects of THC on social and behavioral traits in weakly electric fish using various behavioral paradigms.