An Ishihara-style test of animal colour vision

Abstract

191. Colour vision mediates ecologically relevant tasks for many animals, such as mate choice, 20 foraging and predator avoidance. However, our understanding of animal colour perception is 21 largely derived from human psychophysics, even though animal visual systems differ from 22 our own. Behavioural tests of non-human animals are required to understand how colour 23 signals are perceived by them. 24 2. Here we introduce a novel test of colour vision in animals inspired by the Ishihara colour 25 charts, which are widely used to identify human colour deficiencies. These charts consist of 26 dots that vary in colour, brightness and size, and are designed so that a numeral or letter is 27 distinguishable from distractor dots for humans with normal colour vision. In our method, 28 distractor dots have a fixed chromaticity (hue and saturation) but vary in luminance. Animals 29 can be trained to find single target dots that differ from distractor dots in chromaticity. We 30 provide Matlab code for creating these stimuli, which can be modified for use with different 31 animals. 32 3. We demonstrate the success of this method with triggerfish, Rhinecanthus aculeatus, and 33 highlight behavioural parameters that can be measured, including success of finding the 34 target dot, time to detect dot and error rate. Triggerfish quickly learnt to select target dots that 35 differed from distractors dots regardless of the particular hue or saturation, and proved to use 36 acute colour vision. We measured discrimination thresholds by testing the detection of target 37 colours that were of increasing colour distances (∆S) from distractor dots in different 38 directions of colour space. At least for some colours, thresholds indicated better 39 discrimination than expected from the Receptor Noise Limited (RNL) model assuming 5% 40Weber fraction for the long-wavelength cone. 41 4. This methodology seems to be highly effective because it resembles natural foraging 42 behavior for the triggerfish and may well be adaptable to a range of other animals, including 43 mammals, birds, bees and freshwater fish. Other questions may be addressed using this 44 methodology, including luminance thresholds, sensory bias, effects of sensory noise in 45 detection tasks, colour categorization and saliency. 46 47

Colours and colour vision in reef fishes: Past, present and future research directions

“…Tests include wavelength discrimination, the characteristics of its colour space (Champ et al, 2016) and even, like in humans, its ability to see illusory colours (Simpson et al 2016). Various experiments are now examining the difference between chromatic and achromatic behavioural pathways, a critical factor to understand how reef fishes perceive pattern and luminance differences, as well as colour (Mitchell et al, 2017;Green et al, 2018;Cheney et al, 2018).…”

Colour discrimination thresholds vary throughout colour space in a reef fish (Rhinecanthus aculeatus)

“…The triggerfish, Rhinecanthus aculeatus, is a generalist omnivore, feeding primarily on molluscs and crustaceans on sub-tidal coral reef flats throughout the Indo-Pacific, which performs well in behavioral experiments [29]. For this study, fish ranged from 10 to 26 cm total length; sex could not be determined.…”

“…We conducted two experiments: Experiment 1 was carried out between Nov 2016-Jan 2017 (n = 9 fish) and Experiment 2 between July and September 2016 (n = 6 fish), with one fish used in both experiments (fish 'B'). A small number of these fish had been previously tested with the achromatic distractors in February -April 2016 [29] (Experiment 1: Fish B, N and O; Experiment 2: Fish B, C and E). Some fish were reused in different experiments to reduce the total number of individuals used in this study.…”

“…Here we compare behavioral thresholds across color space for a coral reef fish (Rhinecanthus aculeatus) with those predicted by the RNL model. We report results from four directions at each of four locations in color space, and compare them with five directions close to the achromatic point from a previous study using the same methodology [29]. Figure 1 outlines some ways in which color thresholds depart from RNL model predictions across color space, including non-linear responses from photoreceptors or retinal neurons [3,30], spectral opponent mechanisms [3,31], or higher-level processing including memory [32], learning [33] and color categorisation [34,35].…”

“…Figure 1 outlines some ways in which color thresholds depart from RNL model predictions across color space, including non-linear responses from photoreceptors or retinal neurons [3,30], spectral opponent mechanisms [3,31], or higher-level processing including memory [32], learning [33] and color categorisation [34,35]. To measure color thresholds, we use an Ishihara-style test [29], where the fish is challenged to find an 'odd-one-out' target dot within an array of distractor dots (Figure 2).…”