This repository describes two visual stimuli used in functional descriptions of cortical circuits: Monet and Trippy.
Both stimuli were designed to simultaneously measure linear receptive fields and motion tuning of visually responsive neurons.
Some of the results from using these stimuli were included in our SfN 2018 poster: SfN 2018 Poster
The Monet stimulus comprises smoothened Gaussian noise with coherent orientation and motion in various directions. A sample of the stimulus can be found here: https://vimeo.com/249686342
The stimulus is generated by:
- Generating an i.i.d. (independent identically distributed) Gaussian noise movie
- Low-pass filtering the movie in time. The Hamming filter with 4 Hz bandwidth was used in most experiments.
- Low-pass filtering the movie frames spatially. This was accomplished by a radially symmetric Gaussian filter with the bandwith of 0.5 cycles/degree atthe closest point of the screen.
- Adding orientation bias and coherent motion to the movie. Each 15-second movie interval was split into 16 equal intervals with 16 unique directions of motion of 20 degrees/s at the closest point on the screen. The directions were randomly shuffled in each movie. The orientation bias was added as the radial bandpass filter passing only frequencies within 15 degrees of the selected orientation.
The code for generating the stimulus is available at https://github.com/cajal/trippy-monet
The Trippy stimulus comprises irregular fluid ripples wriggling randomly across the visual field; they are synthesized as the cosine of a smooth Guassian noise movie.
A sample of the stimulus can be found here here: https://vimeo.com/249683927