Chromatic and achromatic vision: parameter choice and limitations for reliable model predictions. (21st October 2017)

• Record Type: Journal Article
• Authors: Olsson, Peter; Lind, Olle; Kelber, Almut
• Editors: Simmons, Leigh
• Is Part Of: Behavioral ecology. Volume 29:Number 2(2018)
• Page Start: 273
• Page End: 282
• Publication Date: 2017-10-21
• Subjects: achromatic contrast sensitivity -- color contrast -- intensity contrast -- visual modeling -- Weber fraction; Animal behavior -- Periodicals; Behavior evolution -- Periodicals; Ecology -- Periodicals; Psychology, Comparative -- Periodicals; 591.5
• DOI: 10.1093/beheco/arx133
• Languages: English
• ISSNs: 1045-2249

Abstract

Abstract : The use of visual models in behavioural ecology can be improved when it is based on observations of animal behavioural performance. The reliability of these mathematical models depends on how parameters and assumptions are chosen and vetted. From a literature review, we outline and explain possibilities and limitations from these procedures, and we point out promising directions in future research. Abstract: Many animals use vision to detect, discriminate, or recognize important objects such as prey, predators, homes, or mates. These objects may differ in color and brightness—having chromatic and achromatic contrast to the background or to other objects. Visual models are powerful tools to investigate contrast detection, but need to be calibrated by experimental data to provide robust predictions. The most critical parameter of current models—receptor noise—is usually estimated from a small number of behavioral tests on chromatic contrast thresholds, while equivalent tests of achromatic thresholds in a wide range of animals have often been ignored. We suggest that both chromatic and achromatic contrasts in studies of visual ecology should be examined using calibrated model parameters, and we provide a compilation of what is currently known on visual thresholds and corresponding noise estimates. Besides the need for careful parameter estimation, we discuss how the robustness of model predictions depends on assumptions about overall light intensity, background color and brightness, object size, and behavioral context.