A Statistical Parametric Method for the Extraction of Stimulus Dependent Activity from Intrinsic Optical Signals

The anterior part of inferotemporal cortex (IT) is thought to be critical for the object recognition and discrimination. To understand the mechanism on how objects are presented by the activation of neurons, and then how the information on the objects is processed in this brain area, the optical imaging technique based on intrinsic signals was applied to directly visualize the responsive neurons when viewing various object images. Instead of using just the ratio of the optical intensities, here we proposed a novel method based on a statistical parametric map for the extraction of the stimulus-dependent signals from the response images. Statistical parameter, t was introduced to evaluate the difference of means between the response and control images obtained with and without visual stimulation. t value for each pixel was computed by comparing the means of optical intensities at the particular pixel in response images and control images, and used to evaluate the significance of the response. The performance of this method was evaluated by the number of spiking neurons. In the region extracted with the proposed method, in as high as 81.8% of penetrations, neurons were responsive, this was significantly higher than 45.5% if in the region extracted by using the ratio of the optical intensities. The results demonstrate that the method proposed here is effective for the signal extraction in the optical imaging experiments.