Research
With almost 33 000 described species in 515 families (Nelson 2006), fish represent the largest extant vertebrate group, totaling more species than all other vertebrate groups (amphibians, reptiles, birds and mammals) combined. While new species are still discovered almost daily (Reid et al. 2013), knowledge about many species or even entire families is scarce or even lacking. Specifically the cognitive abilities of fish have only 'more recently' become a topic of research interest, with an upsurge in behavioral cognition studies over the last 20 years. However, considering fish diversity and the immense ecological and physiological breadth of fish, fish as a group are still severely understudied.
Sharks, rays and chimaeras comprise the class Chondrichthyes (cartilaginous fishes), which represents the oldest extant jawed vertebrates. Today, there are more than 1200 known species, which inhabit almost every aquatic environment and hold a key phylogenetic position to understanding brain evolution in jawed vertebrates. While this group was formerly known as ‘primitive fish with primitive brains’, research over the last few decades has provided increasing amount of evidence that sharks and rays show sophisticated behavior, have a complex biology and are equipped with sensory systems that are perfectly adapted to life underwater. Our research over the last 15 years has shown that elasmobranchs have cognitive abilities that closely match those of many teleosts and other vertebrates.
In the last three years we have made a significant head start looking at the cognitive abilities of Branderhorst’s snapping turtle, a species that is highly motivated in behavioural learning tasks. Topics are conducted concurrently to those in fish, to allow for interspecific comparison.
1. Behavior
Behavioural topics include visual discriminations and categorizations of 2D and 3D structures, matching-to-sample and same-from-oddity tasks, perception of illusionary contours and amodal completion, symmetry perception and categorization, form constancy, (serial-) reversal learning, memory retention capabilities, avoidance behavior, the perception of movement (including biological motion) and color, assessment of numerical abilities as well as experiments on visual resolution thresholds. Several extensive and ongoing experimental series deal with spatial orientation, spatial memory and numerical abilities in elasmobranchs as well as abstract concept learning. There is large inter- and intraspecific variation regarding performance of fish, with some reoccurring trends (e.g. in general sharks are better at abstract concept learning while cichlids are better at movement perception) which may be related to different ecologies and lifestyles. Other studies have included experiments on sex differences in regards to solving cognitive tasks in sharks and stingrays. We have also been looking at acoustic discrimination abilities in bamboo sharks using alternative forced choice and go/no go procedures. Sharks differentiated between two low frequency sounds in the presence or absence of a secondary reinforcer (light). Transfer tests elucidated that in situations where both cues are available, vision is more important than sound; however, in the absence of visual cues, acoustic cues are sufficient to solve a task. Frequencies close to training frequencies elicit similar responses as these, while unfamiliar frequencies lead to random swimming behavior. Overall, there is large inter- and intraspecific variation regarding performance of fish and reptiles.
2. Neurobiology
Comparatively few studies have successfully identified the neural substrates involved in the processing of cognitive information in fish. Brains of social species living in complex habitats, such as coral reef fish or cichlids feature several brain areas or nuclei not found in more basal teleost fish as for example the osteoglossomorpha (e.g. African weakly electric fish), elopomorpha (e.g. eels) and clupeomorpha (e.g. herrings). The development of such structures found in the telencephalon or diencephalon of modern teleosts may quite likely be linked to more complex social hierarchies and habitats and thereby to more elaborate cognitive abilities. For this reason, we selectively lesion some of these structures in cichlids and goldfish, and assess performance in spontaneous behavioural tasks as well as specifically designed training assays. Additionally, we work to identify some of the neural structures underlying visual discrimination learning by assessing and comparing the expression patterms of different neural activity markers (c-fos, egr-1 and pS6) in the brains of differently trained or treated fish.
In bamboo sharks, we used selective lesion experiments (telencephalon ablation) to successfully identify relevant neural substrates for particular aspects of spatial orientation (Fuss, Bleckmann, Schluessel 2014a,b) and avoidance learning (Schwarze, Bleckmann, Schluessel 2013). Additionally, a three-year pilot study using immediate early genes (egr-1, c-fos) to identify neural substrates relevant to visual discrimination learning in elasmobranchs was successfully conducted (Fuss and Schluessel 2018).
3. Cognition in turtles
We are looking at the cognitive abilities of Branderhorst’s snapping turtle, a species that is highly motivated in behavioural learning tasks. So far, experiments on visual discrimination of stationary and moving objects, categorization, symmetry perception, matching-to-sample and same-different learning as well as memory retention, numerical abilities and serial reversal learning have been successfully investigated. Other studies have included mirror recognition and brain lateralization tests.