Thus, adult zebrafish can be an attractive animal model for investigating the evolutionarily conserved and universal mechanisms of behavioral control by the telencephalon. Further, the use of pigment-deficient mutant strains 18, 19 enables observation of the telencephalic neural activity without opening the skull. The zebrafish brain is very small (3 mm 3) 15 as compared to that of mice (509 mm 3) 16 or humans (1400 cm 3) 17, allowing us to observe neural activity in a relatively wide brain region 8. These regions include the isocortex, hippocampus, amygdala, and the cortico-basal ganglia circuit, which is implicated in behavioral selection 11, 12, 13, 14. However, there has been little experimental data to confirm whether active inference indeed acts to correct behaviors.Īdult zebrafish have the ability to learn various adaptive behaviors, and their telencephalon has regions and neural circuits that are evolutionarily homologous to those of other vertebrates, including mammals 10. Another control process of this goal-directed behavior could underlie the principle of minimization of surprise (i.e., the prediction error between the real perceived state and the predicted state) and has been theoretically formulated as “active inference” 2, 3. The goal of reinforcement learning is to maximize the predicted reward under a given spatial distribution of utility (reward or punishment associated with spatial subregions) 6, 7, 8, 9. How these mechanisms are actually adopted by animals and are reflected in their behavior remains unknown 4, 5Īctive avoidance has been regarded as the most typical model-free decision-making behavior based solely on the basic principle of reinforcement learning. In addition to this, adaptive behavior requires animals to generate an internal model of their environment and to take actions to minimize surprise (i.e., improbability) about the state they encounter in comparison with the state predicted from the internal model 2, 3. One prevailing model underlying this behavioral process is based on the idea that the ultimate aim of choice is to maximize utility or reward 1. Making optimal decisions according to the current sensory input is essential for animals. Our results suggest that zebrafish can use both principles of goal-directed behavior, but with different behavioral consequences depending on the repertoire of the adopted principles. The fish with the latter ensemble escape more efficiently than the fish with the former ensembles alone, even though both fish have successfully learned to escape, consistent with the hypothesis that the latter ensemble guides zebrafish to take action to minimize this prediction error. Additionally, one third of fish generated another ensemble that becomes activated only when the real perceived scenery shows discrepancy from the predicted favorable scenery. Analysis of the neural activity of the dorsal pallium during training revealed neural ensembles assigning rules to the colors of the surrounding walls. We addressed this question using a closed-loop virtual reality system to train adult zebrafish for active avoidance. It is still unclear how these principles are represented in the brain and are reflected in behavior. Animals make decisions under the principle of reward value maximization and surprise minimization.
0 kommentar(er)
