Hippocampal serotonin has been implicated in fear learning and memory. But how the dynamics of serotonin in the hippocampus during fear conditioning in vivo has not been much explored. This project explores this using the hippocampal serotonin signals collected by fiber photometry during contextual fear conditioning (Fig. 12A). The model in the project is designed to predict freezing behavior (freezing or not) from time-windowed serotonin signals. Specifically, the time windows are set to be 1.5s and 5s over the 5-min recording. Using FDA as a classifier, feezing behavior is successfully predicted from hippocampal signals during fear conditioning (F1-score > 0.6) (Fig. 12B). Next, the project trains the FDA model using the conditioned fear data and tests it on the general fear data and shows that it can significantly predicts general freezing behavior. Minseok also identifies that long-timescale serotonin dynamics initially contributes to prediction, but later short-timescale dynamics become more predictive. Cross-context predictability (from conditioned to general fear context) also suggests a common serotonin-behavior relationship.