Abstract:

When discussing vision science subjects, particular visual stimuli are necessary to show a certain visual effect or to perform an experiment. Often, neither teachers nor students possess the necessary computing programming knowledge and skills to generate these stimuli on a computer-controlled screen, the usual visualization device in this kind of experiments.

The students of the Degree in Optics and Optometry and of the Master of Advanced Optometry and Vision Sciences at the University of Valencia suffer this problem: they must understand, design and perform vision experiments and tests with visual stimuli favouring the response of a given visual mechanism thanks to precisely defined characteristics. The representation spaces where these characteristics are defined, though theoretically known of the student, are not always intuitive. Being able to manipulate easily these stimuli, greatly facilitates the learning process. We have been using a problem-solving methodology in color science with good results. The students described, generated and operated with chromatic stimuli using a Matlab library, COLORLAB, developed by our research team. Although some knowledge of the Matlab environment was required, the results were good and the students were even able to apply the knowledge so acquired to the design and implementation of tests for visual problems.

The structure of the new study syllabus has forced a change in our methodology, to keep its advantages without having to train the student to use the software. We developed a Graphic User Interface (GUIDE) for Matlab, allowing the characterization, generation and processing of simple visual stimuli, which does not require any programming skills, only some basic knowledge of vision science, to determine the stimulus’ characteristics. This tool has already been used in two subjects of the new study plan at the UV, both in laboratory sessions and as a tool for the student.
The GUIDE at present only works in Matlab, and having access to this program is the only difficulty for the user. Scenes consist of a stimulus on a background, the characteristics of which can be either freely chosen or selected from a menu. The user decides the best domain to define the stimulus (spatial or frequency domain), its shape (some simple geometrical shapes and significant functions for vision experiments, such as gratings and gaussians, optotypes or user-defined images, are available) and the parameters defining that shape. The chromaticities of the scene can be defined and represented in several usual colorimetric spaces. Accurate color reproduction of the stimulus can be ensured if the user uses a calibrated screen, otherwise approximate color reproduction is obtained with colorimetric data from a standard monitor. Once the characteristics of the scene are defined, the software generates the stimulus image and additional information in the shape of color diagrams, and spatial information both in the spatial and in the Fourier domain, which can be saved in any of the formats allowed by Matlab. Basic processing may also be performed on the images: the user may explore the effect of adding different kinds of noise and of filtering the frequency content of the images, both processes being quite usual and useful in vision science.

Keywords:

Educational software, visual stimuli, visual perception, matlab.