Abstract One of the hallmark symptoms of anxiety disorders is fear generalization. Patients diagnosed with a disorder form this group are characterized by not only fearing stimuli and situations that are dangerous or closely resemble to the context in which the original trauma occurred but also fearing stimuli and situations that are objectively safe or just faintly resemble the original trauma context. Hippocampus as the part of the limbic system receives and integrates spatial and contextual information and internal states. Moreover, parts of the hippocampus, specifically the dentate gyrus (DG) are both one of the major targets of the effects of antidepressants and anxiolytics, and also play a role in encoding information about the environment by distinguishing between similar contexts, a process called pattern separation. This computation was thought to mostly performed by granule cells however recent experimental results suggest that mossy cells (MCs) the second major principal cell type of the DG may also participate in this process. Therefore, we hypothesized that fear generalization in a stress induced fear learning mouse model is reflected by impaired contextual discrimination function of MCs. We will test this hypothesis by recording the calcium activity of MCs in vivo in stressed mice during a head fixed pattern separation paradigm. SSRI treatment, as the first line of medication for patients diagnosed with anxiety disorders is thought to restore the low serotonin (5-HT) levels in the DG. Despite this notion there is much less known about the actual activity pattern of 5-HT axons under normal or pathological conditions and there is absolutely no data about the correlated activity of the principal cells and 5-HT axons. MCs express the excitatory 5-HT2A receptors which suggests that the activity of MCs may be controlled by 5-HT and indeed, chronic SSRI treatment has been shown to enhance the activity of MCs. Therefore, we hypothesized that decreased activity of 5-HT axons in stressed mice contributes to low levels of 5-HT in the DG which eventually leads to suppressed activity of MCs and disrupted patter separation. To test this hypothesis, first we will confirm the excitatory effect of 5-HT on MCs by combining optogenetical manipulation of 5-HT axons with two-photon imaging, then we will record the calcium activity of 5-HT axons and MCs simultaneously with dual color two-photon imaging following stress induced fear learning. Altogether, this proposal aims to better understand the role of MCs, a still “enigmatic” cell type in the DG in fear generalization and to provide direct evidence for the role of the 5-HT system in the pathogenesis of anxiety disorders. Furthermore, our experiments will lay the foundation for further studies aiming at understanding the mechanism of neuromodulation at the microcircuit level.