My current research focuses on animal and human behaviour in the adaptive and evolutionary perspective. How cognition, behaviour and personality have evolved through adaptation and natural selection? In my work I try to integrate both proximate and ultimate causation and use both experimental and modelling approaches.
The current work concerns developing a large scale simulation model that implements a general decision-making architecture in evolutionary agents. Each agent is programmed as a whole virtual organism including the genome, rudimentary physiology, the hormonal system, a cognitive architecture and behavioural repertoire. They "live" in a stochastic spatially explicit virtual 3-D environment with physical gradients, predators and prey. The primary aim of the whole modelling machinery is to understand the evolution of decision making, personality, emotion and behavioural plasticity within a realistic ecological framework.
I believe that understanding and modelling complex adaptive behaviour requires both extraneous factors and stimuli as well as endogeneous architectural mechanisms (genetic, hormonal, cognitive etc.) that produce the behaviour. Explicit proximate representation of the motivation and emotion systems, prediction-oriented cognition provides a better approach to understand the behaviour, adaptation and evolution of the whole organism. Ultimately, such an approach can help us understand the evolutionary emergence of consciousness and complex cognition.
For more details, links to source codes etc. see The AHA Model web page and this paper: Budaev, S., Jørgensen, C., Mangel, M., Eliassen, S., & Giske, J. (2019). Decision-making from the animal perspective: Bridging ecology and subjective cognition. Frontiers in Ecology and Evolution, 7, 164. doi:10.3389/fevo.2019.00164.
The cognitive machinery evolved in animals can provide a key for also understanding animal wellbeing and welfare. This requires consideration of subjective phenomena and sentience, a challenging topic, since these private properties cannot be observed directly. But computational models can help here. We propose that computational animal welfare science emerges at the intersection of animal behaviour, welfare and computational cognition. A modelling framework for subjective phenomena and animal wellbeing is described in this paper: Budaev, S., Kristiansen, T. S., Giske, J., & Eliassen, S. (2020). Computational animal welfare: Towards cognitive architecture models of animal sentience, emotion and wellbeing. Royal Society Open Science 7, 201886 doi:10.1098/rsos.201886.
I currently develop a process-based digital twin model of salmon appetite, feeding and digestion. It incorporates biological and environmental information to improve predictions of biological response to situations and optimise feeding with predictive capacity. Multilevel process-based simulation models based on complex software systems require the ability to accommodate experimental data and to perform sensitivity analysis and proper validation. The model will predict voluntary food intake, feeding and related behaviour.
This model is developed as a multifunctional application that allows running as a normal desktop program, command driven program, server-based application and a library that can be embedded into other systems with real-time data exchange with AI-based fish farm control system.
Although I am interested in any species, most of my work so far has been conducted on fish. Using series of tests we have shown that individual fish of several species have consistent personality traits that translate to a variety of different adaptive contexts. Individual fish with different personalities, such as shy and bold, may behave quite differently in their natural environment, e.g. prefer different social strategies (school or not to school) and different local habitats. Shy and bold fish choose their mates based on personality, personality also significantly affects their parental care tactics. Personality in fish can be linked to the operant learning performance. For example, shy fish may be more susceptible to the development of the conditioned fear, providing a link between emotion and personality in such "lower" vertebrates. It is also possible to trace the development and the appearance of consistent personality traits during the ontogeny. Certain environmental effects, like exposure to light, acting early in the ontogeny could significantly affect fish personality via the involvement of specific brain structures, such as the photosensitive habenula. Personality in fish and other species could be linked with lateral asymmetries via the involvement of the morphologically asymmetrical habenula. I am also interested in the adaptive and evolutionary mechanisms that bring about patterns of consistent personality and alternative strategies. We have shown that gender differences in personality follow from sex-related adaptive strategies in humans. In another study we have shown how a trade-off between parental food provisioning and the fry's own individual experience of searching for cryptic food creates a range of parental strategies in a cichlid fish. This reflects my specific interest in the evolution of mate choice and parental care in fish, and their potential role in sympatric speciation. Currently, we are developing models linking emotion and decision making to understand the proximate and ultimate factors governing the evolution of consistent personality. For more information read Budaev, S. & Brown, C. (2011) Personality traits and behavior, in Fish Cognition and Behavior, 2nd edn (ed. by C. Brown, K. Laland & J. Krause). London: Wiley-Blackwell, pp. 135–165.
I am also interested in complex biological interactions at various levels, e.g. competitive interactions between multiple cladoceran species and their predators, and relationships between various associates and the host within a symbiotic community. The former is closely linked with conservation and species invasion. We have developed a model that allows to predict the population dynamics and the invasion success among freshwater cladocerans in various conditions. I took part in several conservation projects, ranging from coral reef and freshwater conservation in Vietnam to fish monitoring and protection in subarctic Siberian rivers and optimising sturgeon hatcheries. We have developed a series of quick low-technology tests for rapid assessment of the coral reef health. Additionally, we have developed hydroacoustic methods for the assessment of the fish populations in very shallow water bodies, such as large Siberian floodplains.
Some of the code can be found at the University of Bergen GitLab server: https://git.app.uib.no/Sergey.Budaev
Department of Biological Sciences
University of Bergen
P.O. Box 7803