Brain morphology correlates of learning and cognitive flexibility in a fish species ( Poecilia reticulata )

Abstract

Determining how variation in brain morphology affects cognitive abilities is important to understand inter-individual variation in cognition and, ultimately, cognitive evolution. Yet, despite many decades of research in this area, there is surprisingly little experimental data available from assays that quantify cognitive abilities and brain morphology in the same individuals. Here, we tested female guppies ( Poecilia reticulata ) in two tasks, colour discrimination and reversal learning, to evaluate their learning abilities and cognitive flexibility. We then estimated the size of five brain regions (telencephalon, optic tectum, hypothalamus, cerebellum and dorsal medulla), in addition to relative brain size. We found that optic tectum relative size, in relation to the rest of the brain, correlated positively with discrimination learning performance, while relative telencephalon size correlated positively with reversal learning performance. The other brain measures were not associated with performance in either task. By evaluating how fast learning occurs and how fast an animal adjusts its learning rules to changing conditions, we find support for that different brain regions have distinct functional correlations at the individual level. Importantly, telencephalon size emerges as an important neural correlate of higher executive functions such as cognitive flexibility. This is rare evidence supporting the theory that more neural tissue in key brain regions confers cognitive benefits.

Experimental expansion of relative telencephalon size improves the main executive function abilities in guppy

“…Based on artificial selection, the results show how telencephalon expansion through mosaic brain evolution can improve executive functions. In both the reversal learning and detour tasks, a relatively larger telencephalon appeared to correlate positively with performance in these two tasks (see also Triki et al ( 34 , 44 )). This suggests that relative telencephalon size facilitated flexible learning.…”

“…Interestingly, we detect no differences in our nonexecutive function assay, the color discrimination learning. This might be because this task does not require complex processing but rather basic association formation through operant conditioning ( 44 ). This is consistent with the findings from fish whose entire telencephalon being ablated can still perform successfully in simple associative learning tasks ( 57 ), while they failed in more complex tasks like reversal learning ( 58 ).…”

“…We used survival analyses with the Cox proportional hazards models to evaluate learning performance in the color discrimination and reversal learning tests ( Coxph function from R package survival ). For this, we replaced “death” in the classic survival analyses with “success” in the learning tests ( 44 ). These types of Coxph models simultaneously test both the rate of success and failure and the time to succeed.…”

Relative telencephalon size does not affect collective motion in the guppy (Poecilia reticulata)

“…The lack of any effects of telencephalon size could be because we have not yet been able to generate enough divergence between the large and small telencephalon size selected lines. However, we have previously found that the existing divergence in telencephalon size affects several aspects of cognition ( Triki et al 2022 , 2023 ). This includes several advanced cognitive abilities such as detour learning, reversal learning and working memory that were enhanced in the large compared to the small telencephalon size selection lines ( Triki et al 2022 , 2023 ).…”

Toward Real-Time Animal Tracking with Integrated Stimulus Control for Automated Conditioning in Aquatic Eco-Neurotoxicology

“…The study of neurotoxic effects due to exposure of animals to anthropogenic contaminants primarily involves analyzing alterations in their behavioral phenotypes. − Behaviors are increasingly recognized as sensitive and early indicators of sublethal chemical risks in exposed animals, representing highly integrative endpoints that encompass a wide range of complex developmental and physiological processes. , Despite the increasing use of behavioral assays in aquatic eco-neurotoxicology, one of the least studied aspects is the impact of pollutants on cognitive behaviors. − Memory and learning represent the highest-level neurological functions that have been reported in a wide range of taxa, including larval stages of fish and amphibians. , Moreover, recent studies have demonstrated that aquatic invertebrates, such as planarians, mollusks, and arthropods, also exhibit various levels of memory and learning despite some controversies related to historical methodological errors. − As such, there has been a notable resurgence of interest in cognitive research using several promising aquatic models, such as planarians, zebrafish ( Danio rerio ), guppy fish ( Poecilia reticulata ), and tadpoles of the African clawed frog ( Xenopus laevis ). ,,− …”