PROJECT SUMMARY Brain oscillations are thought to be critical for cognitive functions and are disrupted in all major neurological and psychiatric disorders, such as Alzheimer’s disease, epilepsy, depression, and schizophrenia. There has been increasing interest in understanding the relation between neural computations and oscillatory patterns in the healthy and diseased brain because oscillation patterns can be targeted for treatment with noninvasive and invasive stimulation devices. While most brain rhythms are generated by either neuronal pacemakers or local circuits, breathing generates rhythmic brain activity by an external loop. Nasal air flow stimulates olfactory sensory neurons which generate respiration-related oscillations (RROs) in the olfactory bulb (OB). RROs have been shown to widely propagate to cortical areas, including piriform cortex (PC), prefrontal cortical areas, lateral entorhinal cortex (lEC), and hippocampus (HC). In HC, RROs can be detected in parallel with pacemaker- generated theta oscillations, which are critical for memory function and overlap in frequency. It is not clear to what extent the two types of oscillations are merely parallel phenomena or functionally coordinated to support neural computations. We hypothesize that RROs and theta oscillations do not globally couple, but that subpopulations of neurons across brain regions are synchronized with each oscillation pattern across different memory phases. To address this question, we will perform recordings and manipulations of local field potentials (LFPs) and neuronal firing patterns in odor- guided working memory tasks, where both types of oscillations are prominent. In Aim 1, we will determine whether RROs, theta oscillations, and oscillations at higher frequencies are coordinated across brain regions in an olfactory working memory task by recording LFPs and/or single-units in the OB, anterior PC, lEC, ventral HC, dorsal HC, and medial prefrontal cortex (mPFC). These brain regions are included because RROs and canonical theta oscillations have been reported in all of these regions and because these brain regions are thought to be critical for working memory performance. In Aim 2, we will then use optogenetics to change the frequency of RROs and canonical theta oscillations to determine during which phases of working memory they are critical for task performance. Finally, in Aim 3, we will use a match/non-match version of the odor-guided working memory task to determine how the initial sensory code is transformed into activity patterns that remain informative over the retention interval. Taken together, our aims will reveal how neuronal activity patterns are coordinated by two types of oscillations during working memory. A mechanistic understanding of brain oscillations that support memory computations is foundational for devising and applying brain stimulation therapies to improve memory in neurological and psychiatric diseases.