Purchase this article with an account.
Geoffrey B. Simpson, Mohammad Dastjerdi, Dawei W. Dong; Generating identical retinal input with and without eye movements during viewing of natural time-varying images. Journal of Vision 2004;4(8):653. doi: https://doi.org/10.1167/4.8.653.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Objective: Our objective is to understand the fundamental principals of how the nervous system codes and uses information. Eye movements are an example of an integration of sensory and motor systems that serves to modify the information extracted from the environment. In this study, we introduce and analyze a technique to replicate the effects of these eye movements on the retinal input in order to isolate the effect that they have on neural activity. Method: We generate two different sets of short (30 sec) movies of natural scenes, both with the same retinal input. In the first series, we allowed subjects to naturally view the images, and we record their eye movements with an eye-tracking device. For the second series, we used the same movies, but altered them such that the center of the screen corresponded to the eye position recorded during the natural viewing. While these new films are presented to the subjects, we superimpose a small flickering cross on the center of the screen for the subject to fixate on. Thus in the first series of movies, the eyes are moving, and in the second series, the movies themselves are moving. We again record their eye movements as they view the second set of films. We then analyze their ability to successfully fixate on the cross in the center of the screen during the viewing. In addition, we analyze the statistics of the images between the two conditions. Results: We found that subjects were able to focus on the cross during the second session, as their eye movements deviated only slightly from the center of the cross. These slight deviations did not affect the spatial power spectrum nor the velocity distribution of the scenes to a significant degree. Conclusion: This method is successful in replicating the effects of eye movements on the retinal input. It can be used in conjunction with neuroimaging techniques to isolate or eliminate the effects that eye movements have upon the visual input.
FAU RIA-25, NIMH 1151-019-42
This PDF is available to Subscribers Only