Fish and chips - Using machine learning to estimate the effects of basal cortisol on fish foraging behavior

Aug 5, 2026

Foraging is a crucial survival behavior requiring both learning and decision-making, yet no effective mathematical framework has quantified performance while accounting for individual variability. This brief research report evaluates foraging performance within multi-armed bandit (MAB) problems using a biological model and a machine learning algorithm. Siamese fighting fish were tested in a four-arm cross-maze over 21 trials, showing basal cortisol modulates average rewards with an optimal intermediate level. An epsilon-greedy approach captures exploration-exploitation tradeoffs and links normalized cortisol levels to a tuning parameter, supporting physiology-behavior relationships.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Foraging, fitness, and individual variability

Coping styles and cortisol profiles

Multi-armed bandits and reinforcement learning framework

TYPE Brief Research Report PUBLISHED 08 February 2023 DOI 10. 3389/fnbeh.2023.1028190 OPEN ACCESS EDITED BY Tom V. Smulders, Newcastle University, United Kingdom REVIEWED BY Marcos Antonio Lopez-Patiño, University of Vigo, Spain V. Anne Smith, University of St Andrews, United Kingdom Emiliano V. Rodriguez, University of St Andrews, United Kingdom in collaboration with reviewer VS *CORRESPONDENCE Wallace M. Bessa wmobes@utu.ﬁ SPECIALTY SECTION This article was submitted to Motivation and Reward, a section of the journal Frontiers in Behavioral Neuroscience RECEIVED 25 August 2022 ACCEPTED 23 January 2023 PUBLISHED 08 February 2023 CITATION Bessa WM, Cadengue LS and Luchiari AC (2023) Fish and chips: Using machine learning to estimate the e􀀀ects of basal cortisol on ﬁsh foraging behavior. Front. Behav. Neurosci. 17:1028190 . doi: 10.3389/fnbeh.2023.1028190 COPYRIGHT © 2023 Bessa, Cadengue and Luchiari. This isan open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Fish and chips: Using machine learning to estimate the e􀀀ects of basal cortisol on ﬁsh foraging behavior Wallace M. Bessa1*, Lucas S. Cadengue2 and Ana C. Luchiari3 1 Department of Mechanical and Materials Engineering, University of Turku, Turku, Finland, 2 Programa de Pós-Graduação em Engenharia Elétrica e de Computação, Universidade Federal do Rio Grande do Norte, Natal, Brazil, 3 Departamento de Fisiologia, Centro de Biociências, Universidade Federal do Rio Grande do Norte, Natal, Brazil Foraging is an essential behavior for animal survival and requires both learning and decision-making skills. However, despite its relevance and ubiquity, there is still no e􀀀ective mathematical framework to adequately estimate foraging performance that also takes interindividual variability into account. In this work, foraging performance is evaluated in the context of multi-armed bandit (MAB) problems by means of a biological model and a machine learning algorithm. Siamese ﬁghting ﬁsh (Betta splendens) were used as a biological model and their ability to forage was assessed in a four-arm cross-maze over 21 trials. It was observed that ﬁsh performance varies according to their basal cortisol levels, i.e., a reduced average reward is associated with low and high levels of basal cortisol, while the optimal level maximizes foraging performance. In addition, we suggest the adoption of the epsilon-greedy algorithm to deal with the exploration-exploitation tradeo􀀀 and simulate foraging decisions. The algorithm provided results closely related to the biological model and allowed the normalized basal cortisol levels to be correlated with a corresponding tuning parameter. The obtained results indicate that machine learning, by helping to shed light on the intrinsic relationships between physiological parameters and animal behavior, can be a powerful tool for studying animal cognition and behavioral sciences. KEYWORDS foraging, ﬁsh, multi-armed bandit, epsilon-greedy, cortisol 1. Introduction Foraging plays an essential role in animal’s 􀀂tness and its e􀀔cacy has been shaped by evolutionary processes (Pearson et al., 2014) . However, in addition to its phylogenetic basis, foraging behavior is also largely in􀀃uenced by ontogeny (Hughes et al., 1992; Grecian et al., 2018) . In fact, both intrinsic and extrinsic factors that modulate behavior confer di􀀓erent consequences on 􀀂tness (Dingemanse et al., 2004; Brown et al., 2007) and the huge variation between individuals’ life history leads to several di􀀓erences in the way they deal with stressful situati

FAQs

How was fish foraging performance evaluated in the study?

Foraging was modeled as a multi-armed bandit (MAB) problem and assessed using Siamese fighting fish in a four-arm cross-maze over 21 trials.

What relationship did basal cortisol show with foraging rewards?

Fish performance varied with basal cortisol: reduced average reward occurred at low and high levels, while an intermediate optimal level maximized foraging performance.

Why was the epsilon-greedy algorithm used?

It was adopted to manage the exploration–exploitation tradeoff and to simulate foraging decisions, producing results closely aligned with the biological model.

Aug 5, 2026

Upload by Theodora