“School of Cognitive Sciences”
Back to Papers HomeBack to Papers of School of Cognitive Sciences
| Paper IPM / Cognitive Sciences / 13876 |
|
||||||||||||
| Abstract: | |||||||||||||
|
In primary visual cortex, the variance of neuron's firing rate is directly proportional to its mean firing rate (Tolhurst et aI.,1983) and the firing rate is generally higher in complex cells than simple cells (Schiller et aI., 1976). Here, we introduce a new method for measuring response variability based on the mean variance of the frequency response amplitude (MVFRA). We measured the relation between response variability in the time domain and the frequency domain and quantified the mean variance of the frequency response amplitude in simple cells and complex cells. To measure the response variability of neurons, we used chronically implanted ultra-thin multi-electrode array with impedances of 1-3 MO (Swadlow et ai, 2005; Lashgari et ai, 2012) in primary visual cortex (V1 area) and measured the neuronal responses to gratings drifting at 2 Hz for 2-3 seconds. The gratings could have different orientations (8 orientations with 16 directions) or luminance contrasts (8 contrast). Preliminary results indicate that the stimulus that generate the
maximum firing rate not always generate the lowest response variability and that responses with the highest firing rate can be just as variable as those with lower firing rate. Our results suggest that the preferred stimulus of a V1 neuron is better described by the lowest response variability than by the maximum firing rate. These results may have
implications to understand how stimuli are best represented by neuronal responses and suggest that responses with low variability may be more reliable for higher brain functions such as attention and decision making.
Download TeX format |
|||||||||||||
| back to top | |||||||||||||
