Cait Newport

Colour vision and response bias in a coral reef fish

SUMMARY Animals use coloured signals for a variety of communication purposes, including to attract potential mates, recognize individuals, defend territories and warn predators of secondary defences (aposematism). To understand the mechanisms that drive the evolution and design of such visual signals, it is important to understand the visual systems and potential response biases of signal receivers. Here, we provide raw data on the spectral capabilities of a coral reef fish, the Picasso triggerfish Rhinecanthus aculeatus, which is potentially trichromatic with three cone sensitivities of 413 nm (single cone), 480 nm (double cone, medium sensitivity) and 528 nm (double cone, long sensitivity), and a rod sensitivity of 498 nm. The ocular media have a 50% transmission cut off at 405 nm. Behavioural experiments confirmed colour vision over their spectral range; triggerfish were significantly more likely to choose coloured stimuli over grey distractors, irrespective of luminance. We then examined whether response biases existed towards coloured and patterned stimuli to provide insight into how visual signals – in particular, aposematic colouration – may evolve. Triggerfish showed a preferential foraging response bias to red and green stimuli, in contrast to blue and yellow, irrespective of pattern. There was no response bias to patterned over monochromatic non-patterned stimuli. A foraging response bias towards red in fish differs from that of avian predators, who often avoid red food items. Red is frequently associated with warning colouration in terrestrial environments (ladybirds, snakes, frogs), whilst blue is used in aquatic environments (blue-ringed octopus, nudibranchs); whether the design of warning (aposematic) displays is a cause or consequence of response biases is unclear.

Discrimination of human faces by archerfish ( Toxotes chatareus )

Two rival theories of how humans recognize faces exist: (i) recognition is innate, relying on specialized neocortical circuitry, and (ii) recognition is a learned expertise, relying on general object recognition pathways. Here, we explore whether animals without a neocortex, can learn to recognize human faces. Human facial recognition has previously been demonstrated for birds, however they are now known to possess neocortex-like structures. Also, with much of the work done in domesticated pigeons, one cannot rule out the possibility that they have developed adaptations for human face recognition. Fish do not appear to possess neocortex-like cells, and given their lack of direct exposure to humans, are unlikely to have evolved any specialized capabilities for human facial recognition. Using a two-alternative forced-choice procedure, we show that archerfish (Toxotes chatareus) can learn to discriminate a large number of human face images (Experiment 1, 44 faces), even after controlling for colour, head-shape and brightness (Experiment 2, 18 faces). This study not only demonstrates that archerfish have impressive pattern discrimination abilities, but also provides evidence that a vertebrate lacking a neocortex and without an evolutionary prerogative to discriminate human faces, can nonetheless do so to a high degree of accuracy.

Concept learning and the use of three common psychophysical paradigms in the archerfish (Toxotes chatareus).

Archerfish are well known for their specialized hunting technique of spitting water at prey located above the water line. This unique ability has made them a popular focus of study as researchers try to understand the mechanisms involved in targeting and spitting. In more recent years, archerfish have also become an increasingly popular model for studying visual discrimination and learning in general. Until now, only the alternative forced-choice (AFC) task has been used with archerfish, however, they may be capable of learning other classical discrimination tasks. As well as providing alternative, and potentially more efficient, means for testing their visual capabilities, these other tasks may also provide deeper insight into the extent to which an organism with no cortex can grasp the concepts underlying these tasks. In this paper, we consider both the matched-to-sample (MTS) and the odd-one-out (OOO) tasks as they require the subject to learn relatively sophisticated concepts rather than a straight, stimulus-reward relationship, of the kind underlying AFC tasks. A variety of line drawings displayed on a monitor were used as stimuli. We first determined if archerfish could complete the MTS and OOO test and then evaluated their ability to be retrained to new stimuli using a 4-AFC test. We found that archerfish were unable to learn the MTS and had only a limited capacity for learning the OOO task. We conclude that the MTS and OOO are impractical as paradigms for behavioral experiments with archerfish. However, the archerfish could rapidly learn to complete an AFC test and select the conditioned stimulus with a high degree of accuracy when faced with four stimuli, making this a powerful test for behavioral studies testing visual discrimination. In addition, the fish were able to learn the concept of oddity under particular training circumstances. This paper adds to the growing evidence that animals without a cortex are capable of learning some higher order concepts.

Complex, context-dependent decision strategies of archerfish (Toxotes chatareus).

Operant conditioning is fundamental to many animal behaviour experiments, including the alternative forced-choice (AFC) task, a powerful and popular paradigm for establishing learning and discrimination abilities across a wide array of species. One frequently overlooked aspect of this paradigm is the relative importance of the positive (S+ or ‘target’) and negative stimulus (S− or ‘distractor’). It is often assumed that subjects learn to associate S+ with a reward. However, it is equally possible that they learn to associate S− with the absence of a reward. The rule learned may have repercussions for the design of experiments, and may also provide information about the decision strategies employed by a particular animal or species. Archerfish are becoming increasingly popular as a model for testing cognition in lower vertebrates. We conducted three experiments to explore how archerfish learn to complete a common type of forced-choice test. In the first experiment, we showed that archerfish were able to discriminate a range of shapes in an AFC task. In the second experiment, we found that replacing S− with novel stimuli caused a greater disruption in performance than replacing S+. In the final experiment we showed that archerfish: (1) were probably using complex decision rules; (2) made context-dependent choices; and (3) examined some or all of the stimuli before making a decision when faced with multiple stimuli. Our results show that archerfish have a flexible learning strategy which may facilitate exploitation of novel food sources. The discovery that archerfish avoid negative stimuli fits with their generalist feeding ecology in which they may learn to avoid distasteful/toxic prey. At the same time the fish demonstrate a willingness to try new prey, which have the potential to offer a greater reward than familiar target prey.

Same/Different Abstract Concept Learning by Archerfish (Toxotes chatareus).

While several phylogenetically diverse species have proved capable of learning abstract concepts, previous attempts to teach fish have been unsuccessful. In this report, the ability of archerfish (Toxotes chatareus) to learn the concepts of sameness and difference using a simultaneous two-item discrimination task was tested. Six archerfish were trained to either select a pair of same or different stimuli which were presented simultaneously. Training consisted of a 2-phase approach. Training phase 1: the symbols in the same and different pair did not change, thereby allowing the fish to solve the test through direct association. The fish were trained consecutively with four different sets of stimuli to familiarize them with the general procedure before moving on to the next training phase. Training phase 2: six different symbols were used to form the same or different pairs. After acquisition, same/different concept learning was tested by presenting fish with six novel stimuli (transfer test). Five fish successfully completed the first training phase. Only one individual passed the second training phase, however, transfer performance was consistent with chance. This individual was given further training using 60 training exemplars but the individual was unable to reach the training criterion. We hypothesize that archerfish are able to solve a limited version of the same/different test by learning the response to each possible stimulus configuration or by developing a series of relatively simple choice contingencies. We conclude that the simultaneous two-item discrimination task we describe cannot be successfully used to test the concepts of same and different in archerfish. In addition, despite considerable effort training archerfish using several tests and training methods, there is still no evidence that fish can learn an abstract concept-based test.

Intraspecific cleaning behaviour of adult cleaner wrasse, Labroides dimidiatus(Perciformes: Labridae)

Here we report on intraspecific cleaning behaviour between two adult bluestreak cleaner wrasse Labroides dimidiatus of similar size on coral reefs surrounding Lizard Island, Great Barrier Reef, Australia. During a SCUBA dive, we observed these individuals posing and the resulting cleaning interactions. While aggression in this species is common between adults of similar size and social class, our observation suggests that these individuals may also cooperate and partially rely on conspecific individuals for cleaning.

Object recognition in fish: accurate discrimination across novel views of an unfamiliar object category (human faces)

Accurate visual object recognition is essential to survival for a wide range of species across a wide range of evolutionary histories and visual requirements. However this task is solved, it is a major achievement because object recognition is far from simple. The appearance of an object can alter almost completely as viewing conditions change, not least under variations in lighting and orientation. Determining the recognition limits of a species is important to understanding its visual ecology and can help identify conditions under which recognition may fail. In this study, we tested whether a species of fish can recognize objects from an unfamiliar object class (human faces) across changes in viewing direction. Using operant conditioning, we trained archerfish, Toxotes chatareus, to discriminate between two frontal views of standardized human faces and, critically, tested whether they could continue to do so as the orientation in depth of the faces changed. All fish learned the initial discrimination task and could also recognize rotated forms. These results represent the first conclusive evidence that a species of fish can generalize recognition across views, speaking against a strict image-matching process. This ability rather speaks to the capacity of relatively simple brains to tackle the hard problem of view invariance and provides insight into the mechanisms employed in more complex organisms such as humans. Although we speculate that other fish species may demonstrate similar abilities, a visual system capable of recognition across changes in viewpoint may be especially important to the unique hunting strategy of archerfish.

Comparison of functional and anatomical estimations of visual acuity in two species of coral reef fish

ABSTRACT The high-contrast, complex patterns typical of many reef fish serve several purposes, including providing disruptive camouflage and a basis for vision-based communication. In trying to understand the role of a specific pattern, it is important to first assess the extent to which an observer can resolve the pattern, itself determined, at least in part, by the observers visual acuity. Here, we studied the visual acuity of two species of reef fish – Pomacentrus amboinensis and Pseudochromis fuscus – using both anatomical and behavioural estimates. The two species share a common habitat but are members of different trophic levels (predator versus herbivore/omnivore) and perform different visual tasks. On the basis of the anatomical study, we estimated visual acuity to lie between 4.1 and 4.6 cycles deg−1 for P. amboinensis and 3.2 and 3.6 cycles deg−1 for P. fuscus. Behavioural acuity estimates were considerably lower, ranging between 1.29 and 1.36 cycles deg−1 for P. amboinensis and 1.61 and 1.71 cycles deg−1 for P. fuscus. Our results show that two species from the same habitat have only moderately divergent visual capabilities, despite differences in their general life histories. The difference between anatomical and behavioural estimates is an important finding as the majority of our current knowledge on the resolution capabilities of reef fish comes from anatomical measurements. Our findings suggest that anatomical estimates may represent the highest potential acuity of fish but are not indicative of actual performance, and that there is unlikely to be a simple scaling factor to link the two measures across all fish species. Summary: Comparison of behavioural and anatomical estimates of visual acuity in coral reef fish reveals higher values for anatomical estimates but these may not be indicative of actual performance.

Human facial recognition by archerfish

Using a two-alternative forced-choice procedure, we tested whether four archerfish (Toxotes chatareus) could learn to discriminate human faces.