Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session B: Wednesday, August 13, 1:00 – 4:00 pm, de Brug & E‑Hall

Would the same inferences about visual memory have been made with LFPs as compared to spikes?

Catrina M. Hacker¹, Simon E. Bohn, Brett L. Foster, Nicole C. Rust; ¹University of Pennsylvania

Presenter: Catrina M. Hacker

A central premise of basic neuroscience research is that insights about the healthy brain may eventually inform treatments for neurological and neuropsychiatric disorders. While much of the recent progress in systems neuroscience has relied on densely sampled, high spatial resolution measures of neural activity (like spikes), most neural recordings available in humans are field potentials. Consequently, bridging the divide between animal and human neuroscience requires understanding how neural representations compare in those different types of data. To fill this gap, we analyzed a dataset with established spiking representations of visual memory to answer a simple question: would the same inferences about the neural representations supporting memory have been made if measures were limited to field potentials? Using spike and local field potential (LFP) data simultaneously recorded in inferotemporal cortex (ITC) of four macaque monkeys performing a visual memory task, we show that the neural representations of three variables are aligned across spikes and high-gamma activity: memory, memorability, and contrast. In addition, we show that condition-specific, and in some cases image-specific, neural representations are matched across both measures. These results suggest that data can be meaningfully compared across animals and humans in support of translational work, such as in developing brain computer interfaces and closed-loop therapeutic stimulation devices.

Topic Area: Visual Processing & Computational Vision

Extended Abstract: Full Text PDF