Unveiling Cognitive Processes Underlying Depression-like Behaviors in the Force Swim and Tail Suspension Tests

The forced swim test (FST) and tail suspension test (TST) are widely used to assess depression-like behaviors in rodents, with immobility time serving as a primary metric. However, this measure oversimplifies the complex cognitive processes involved, obscuring the relationship between immobility and cognitive biases. Here, we employ a custom-developed tool to capture complete behavioral trajectories in mice and apply reinforcement learning models for systematic analysis. Our findings indicate that behaviors in both tests can be interpreted as reinforcement learning processes that encompass learning, consequence perceiving, and decision-making. Notably, the underlying cognitive processes differ between the two tests, challenging the assumption that the FST and TST are interchangeable for cross-validation. Additionally, we exam two methods for inducing depression-like behavior: chronic restraint stress and medial prefrontal cortex inhibition. Both approaches increase immobility by altering learning dynamics and consequence sensitivity. Regression analyses further reveal two distinct behavior phases: an early phase dominated by learning-related factors and a later phase influenced by consequence sensitivity. These results suggests that conventional metrics focus solely on the final minutes of testing may underestimate learning processes while overemphasize sensitivity to consequences. Our study underscores the importance of computational modeling in revealing the complex cognitive mechanisms underlying animal behavior.