Analysis of response properties of multiple neurons recorded at single sites in the cat striate cortex

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical and Chemical Engineering


Bart Farell


Striate cortex, Tuning curves, Electrode drift, Spike waveforms

Subject Categories

Life Sciences | Molecular and Cellular Neuroscience | Neuroscience and Neurobiology


The primary visual cortex (V1) shows columnar organization, such that under a given location on the surface of V1 nearby neurons share common tuning parameters. Using tetrodes, this investigation assesses local diversity in orientation, direction and spatial frequency tuning properties. Specifically, the tendency for nearby cells to share similar tuning preference is studied in addition to how tightly nearby cells constrain their response for that preference - selectivity . The methodology used was first validated by studying diversity of tuning preference and comparing that to the rich literature provided by previous studies. These techniques were then applied to selectivity diversity concluding that orientation selectivity clusters significantly in the visual cortex; however, no such clustering was found for direction selectivity or spatial frequency selectivity.

Much work suggests that the major contributions of orientation selectivity for layer 4 V1 cells are the feedforward input -- input from the lateral geniculate nucleus (LGN), which contains input from cells with spatially aligned receptive fields and an untuned component from non-aligned cells -- and untuned inhibition, which suppresses responses to the untuned component of excitation. The width of the peak of the tuning curve about the preferred orientation provides a measure of the aligned feedforward input to the cell while a measure orthogonal to the preferred, reflects the overall degree to which untuned inhibition suppresses untuned excitation. Measures such as the circular variance have previously been proposed as a single measure that reflects both aspects of the tuning curve. The data presented here indicate that this measure has very little correlation with the stimulus-induced response at the null.

A particular challenge for experiments studying the cortical microcircuit is the length of time required to characterize each of the isolated cells. Electrode drift and varying neuronal responses routinely cause changes in the spike waveform. A technique is presented that compares the cell's four channel waveform shape within a file to waveform shapes across files and determines if the similarity exceeds a calculated acceptance threshold, which is defined as the optimal value separating within-file similarities and between-file similarities. Various examples are provided as a proof of concept.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.