Neural Mechanisms of Motivation and Reward
National Institute Of Mental Health
Investigators
Linked publications, trials & patents
Abstract
Neural signals encoding reward value play a central role in assigning context-dependent salience to sensory stimuli during decision-making. It is widely assumed that this process relies on a distributed value normalization signal that coordinates priorities across sensory and reward-processing pathways. To identify neural populations that may distribute such a signal, we used retrogradely transported lentiviruses encoding fluorescent or epitope-tagged reporters to label projection neurons across multiple pathways involved in value and sensory processing. Three macaques received intra-hemispheric injections of molecular color tagging genes into the orbital frontal cortex (OFC) and rostromedial caudate (rmCD), two areas implicated in reward-value processing. In a fourth monkey, these vectors were injected into anterior and posterior inferotemporal cortex (TE and TEO), while a third molecular vector was injected into the posterior putamen/tail of the caudate (pPU/tCD), regions associated with sensory and visuomotor functions. Across all four animals, we observed clusters of co-projecting neurons in several regions potentially involved in value averaging across the background, i.e., normalization, including the prefrontal cortex, temporal lobe, insular cortex, dorsomedial thalamus, and claustrum. The intermediate division of the basolateral amygdala (BLA) consistently contained the highest density of co-projecting neurons across all injection sites. For example, over 60% of BLA neurons projecting to the posterior putamen also projected to inferotemporal cortex, and up to 45% of those projecting to the OFC also sent axons to the ventromedial caudate. Further, the magnocellular BLA showed high co-projection to pPU/tCD and IT cortex, while the parvocellular BLA showed co-projection to OFC and rmCD. These results suggest that the BLA, particularly its intermediate division, is anatomically well positioned to broadcast a common signal via bifurcating axons to both reward and sensory pathways. This signal may serve as a candidate for normalizing value across context-dependent decision-making circuits.
View original record on NIH RePORTER →