Behavior is organized across multiple spatial and temporal scales, ranging from sub-second motor commands over multi-second movement plans to long term foraging patterns. Currently it is unclear how the brain solves this coordination of multiple intertwined temporal demands. While classical neuroscience experiments typically look at or engage a fixed temporal scale or horizon, ethological studies have long focused on the analysis of naturalistic behavior across freely elicited temporal scales. Here we will use novel developed deep learning pose estimation approaches to study the behavior and associated single unit physiology of foraging behavior in freely moving rhesus macaques. First we will establish that timescales of macaque pose behavior encode cognitive variables such as reward expectation. Second we will establish the link between neural and behavioral timescale coordinating in the decision to action axis of the medial prefrontal wall leading to the anterior cingulate cortex by recording multi-region wireless electrophysiology in freely moving rhesus macaques. Third, using embedding and connectivity analysis we will uncover the mechanisms for inter and intra areal timescale coordination to understand how the brain balances temporal scale demands.