Abstract:

Introduction:
Seminar work in the Degree in Optics and Optometry (DOO) in the University of Valencia takes usually one of two shapes: elaborating bibliographical reviews that are subsequently presented in the classroom or solving collections of problems. We seldom aim to train other skills, such as the ability to design and perform experiments, either in clinical or in basic sciences settings, in part for the difficulty of providing the students with lab space and equipment.
A solution is to use software to create a virtual lab, and working along this line, we have tested along the years different strategies, from open environments, consisting in Matlab-Octave libraries [1] , very flexible but requiring programming skills, to target-directed apps [2], more limited in scope, but easier to use. We illustrate the use of this software in seminar work by one task proposed in Clinical Exploration Methods, a compulsory subject of the 4th year of the DOO.

Methodology:
We ask the students to produce an arrangement colour test, inspired in the Dichotomous Colour Test [3], consisting of 15 coloured samples, differing only in hue, which the patient must arrange by perceptual proximity. The patient’s arrangement pattern shows damage in a particular chromatic mechanism, because samples differing only in the response they would elicit in that mechanism, if normal, are perceived as equal.
The students work in pairs or small groups and are provided with both open software [1] and a close app [2]. They can solve the task with any of these tools or by combining both, depending on their programming skills.
The samples are generated in a constant lightness and chroma locus of a colour vision model. Different procedures can be devised to present the samples to the patient, although inserting them in a Power Point or Open Office document is a favourite. The patient arranges the samples and the resulting sequence is plotted in a polar diagram and analysed. The test should be administered to subjects with normal and variant colour vision. If subjects with abnormal colour vision are not available, they can be simulated with filters or with a colour vision model [4].

Results:
Solving the problem of generating the samples implies the use of colour vision models or colour databases, the understanding of the limitations of colour generating devices and of how chromatic and spatial characteristics of the stimulus determines the sensitivity and specificity of the test.
Simulating with a colour vision model how a dichromatic subject sees the samples and asking a normal subject to order the simulated samples illustrates the limitations of dichromatic vision and the rationale of the test: the arrangement patterns of the three kinds of dichromats are different enough to allow both detection and classification of colour vision defects.

References:
[1] Malo J and Luque MJ. "COLORLAB: a color processing toolbox for Matlab", http://isp.uv.es/colorlab.html, Last visited September 24th, 2015.
[2] Luque MJ, De Fez D, Malo J, MC García‐Domene, Díez‐Ajenjo A, Capilla P, Pons A, Demo_estimuls: instalador web. http://roderic.uv.es/handle/10550/45602, Last visited Mars 28th, 2016.
[3] Birch J, Diagnosis of Defective Colour Vision, 2nd Ed. Boston: Butterworth-Heinemann, 2001, Chap. 7
[4] Capilla P, Díez-Ajenjo MA, Luque MJ, Malo J. Corresponding-pair procedure: a new approach to simulation of dichromatic color perception, J Opt Soc Am. 2004;21(2):176-86.

Keywords:

Educational software, seminar work, optometry, vision sciences, clinical psychophysics.