Searching for Neural Signatures of Escape Decisions in Running Crabs Using Tetrode Recordings

Escape responses to imminent threats are essential for survival, yet the same stimulus does not invariably elicit escape, suggesting that escape expression involves a decision-making process. Previous intracellular and anatomical studies in the crab Neohelice granulata identified several classes of Lobula Giant (LG) neurons involved in visually evoked escape. Here, we investigated the relationship between LG activity and escape behavior using tetrode recordings from crabs running on an air-supported ball. We presented looming stimuli with different temporal dynamics to generate variable behavioral responses. Behavioral analyses revealed substantial variability not only across stimuli and individuals but also across trials within individuals. Thus, the same crab may display a vigorous run, a weak run, or fail to run in response to the same stimulus. We exploit this variability to identify neuronal activity associated with different escape outcomes. Preliminary analyses revealed distinct groups of units that were excited or inhibited by looming stimuli. We are using guided automation to classify neuronal responses and relate them to behavioral outcomes. Our ongoing analyses indicate that both the decision to run or not to run and the velocity of the escape response can be predicted from LG neuronal activity, providing potential neural signatures of escape decisions.