Abstract Schizophrenia is a debilitating disorder whose treatment has stagnated over the last several decades. Despite some success in treating its positive symptoms (hallucinations and delusions), ~40% of patients are treatment-resistant and almost all exhibit cognitive dysfunction. Targeting cognition in schizophrenia is a huge unmet public health need, with a potential to restore around 1% of global population to the workforce. Abnormal prefrontal cortex (PFC) function is known to be central for the cognitive deficits in schizophrenia, but recent evidence suggests additional abnormalities in the interactions between the PFC and its major subcortical partner, the mediodorsal thalamus (MD). This idea is supported by studies showing reduced structural and functional connectivity between these two structures in patients and in relevant animal models. This proposal aims to build a basic science and translational foundation around the idea that rescuing MD- PFC interactions is a viable strategy for cognitive enhancement in schizophrenia. We are motivated by our generation of new mouse models based on the SCHEMA study, which offer a strong genetic link to the illness, and with the development of new tasks that can selectively access different funcitonal features of this circuitry in intact animals. In Aim I, we will characterize one SCHEMA model, the Grin2a halpoinsufficiency in two distinct switching tasks that rely on MD-PFC interactions, identifying funcitonal signatures of this networks dysfunction that can be targeted by rescue methodology. In Aim II, we will ask whether the preliminary deficits observed in Grin2a mice generalize to two other models, Gria3 and Setd1A, along with a broader characterization of structural and functional features of their MD-PFC network. In Aim III, we will perform deep molecular characterization of the MD, identifying novel cell types using scRNA sequencing and validation with MERFISH. Lastly, in Aim IV, we will use CELLREADR technology to drive opsins in promising cell types to identify future targets of intervention in these animal models.