Chun-I Yeh: “Processing of artificial and natural stimuli in macaque primary visual cortex”

August 2, 2016  16:00 〜 17:00

CiNet 1F Conference Room

Chun-I Yeh

Assistant Professor
National Taiwan University

Host : Kei Watanabe (Suzuki group)


Processing of artificial and natural stimuli in macaque primary visual cortex Chun-I Yeh Department of Psychology, National Taiwan University, Taiwan

Using natural stimuli to study receptive field properties has become an important branch of visual research. In contrast with artificial stimuli that have simple and well characterized properties, natural stimuli frequently have very complex dynamics.
However, a systematic investigation of the contextual effect of the receptive field with both stimulus types has not yet been accomplished.
Here we used both artificial and natural stimuli to study how spatial receptive-field properties may change with the characteristics of the visual stimulus ensembles (i.e. with or without spatial correlation).
Artificial stimuli include both sparse noise and dense noise (binary checkerboard and subspace gratings), and natural stimulus is a movie clip with three different versions (original, original with 90-degree rotation, and randomized). A multi-electrode matrix (8×8 array,
Neuronexus) was used to simultaneously record from multiple neurons in different layers of macaque monkey V1. Here we report some of the main findings in our simple cell population (modulation ratio f1/f0 >1).
First, the contextual effect is significantly larger for neurons in the superficial layers 2/3 than for those in the input layer 4c of V1.
Second, the spatial correlation in artificial images may contribute to the increase of the subregion number and the aspect ratio of the receptive field. Third, the orientation bias in natural images may contribute to the change of the preferred orientation axis of the receptive field. Overall, these results indicate that macaque monkey V1 is highly adaptive and dynamic: the receptive field of many V1 neurons may change accordingly with the statistics of the visual scene.