Modeling flexible behavior with remapping-based hippocampal sequence learning

Neuroscientists Yoshiki Ito and Taro Toyoizumi have published a groundbreaking computational model that explains how the brain's hippocampus enables flexible behavior. Their novel, biologically plausible reinforcement learning framework centers on a mechanism called 'Context selector,' which drives the formation of context-dependent sequential neural activity. This process, known as remapping, allows animals and humans to swiftly adapt their actions based on changing environmental cues. The model successfully reproduces a wide array of experimental data, including neural activity patterns, results from optogenetic inactivation studies, human fMRI scans, and clinical research observations.

Beyond explaining normal brain function, the model makes a significant predictive leap into neuropsychiatry. It proposes that an imbalance in the ratio between sensory and contextual representations within the Context selector mechanism can lead to maladaptive behaviors. Specifically, the researchers suggest this computational dysfunction underlies symptom profiles observed in schizophrenia (SZ) and autism spectrum disorder (ASD). This provides a unifying theoretical bridge between cellular-level hippocampal activity and complex behavioral disorders, offering a new target for computational psychiatry and potential therapeutic strategies focused on neural circuit dynamics.