The term working memory describes the active part of our memory system. It is put to use, for example, when we ignore distractions in our environment in order to perform the mental calculations necessary for leaving a tip on a restaurant bill. An inability to carry out such tasks is one of the most devastating symptoms of schizophrenia. Indeed, not only is the severity of a patient's working memory impairment highly predictive for that individual's long-term prognosis, but current treatments also do little to ameliorate such deficits. In order to develop better treatments for these deficits, the neural mechanisms that support working memory must first be understood. An essential approach towards this end is the use of animal models, in which neural activity can be reversibly manipulated and causal relationships between activity and complex behavior established. Brain imaging studies have revealed that patients with schizophrenia have decreased levels of activity in the mediodorsal thalamus (MD) when performing working memory tasks. The MD shares a dense set of reciprocal connections with the prefrontal cortex (PFC), forming a closely-knit thalamo-cortical circuit. Dysfunction in this brain circuit has been hypothesized to underlie the working memory deficits of schizophrenia. To causally test the involvement of the MD in working memory we recently generated a mouse model with decreased neural activity in the MD. Our findings revealed that decreasing MD activity was sufficient to cause deficits in a working memory task. Interestingly, these deficits correlated with disruptions in synchronous activity between MD and the medial subnucleus of the PFC during working memory performance. While important findings, two limitations constrain the scope of interpretation. First, our experiments disrupted activity in all MD projections, including projections made to multiple PFC subnuclei. It is thus unclear whether only projections from the MD to the medial PFC are important for working memory. Second, while working memory processes in our task take place at a time-scale of seconds, our experiments disrupted MD activity at a time scale of hours. In order to understand how MD-PFC activity contributes to working memory it is necessary to know the precise time point during our task when MD-PFC activity increases and if it is tightly linked to proper performance. This information is essential for understanding whether MD-PFC activity is important for encoding information, retrieving information or executing behavior based on retrieved information. To obtain the spatial and temporal resolution necessary to address these questions, this proposal takes advantage of recently developed optogenetic tools in mice. By reversibly manipulating neural activity in defined MD-PFC projections with millisecond-time scale precision, the proposed experiments will reveal how this circuit supports working memory at a level of mechanistic detail not possible by our previous experiments, nor in human subjects. Findings from these experiments will lay the groundwork for understanding how disruption of this neuronal circuit can lead to working memory deficits in disorders like schizophrenia.

Public Health Relevance

Working memory-the ability to transiently hold information in mind to achieve short-term goals-is an essential cognitive faculty for the daily functioning of all mammals. While impairment in working memory is a devastating, core symptom of schizophrenia, current treatments are unable to relieve this deficit because little is known about the basic brain mechanisms that support working memory. This proposal takes advantage of novel techniques for precisely and reversibly manipulating neural activity in behaving mice in order to model aspects of brain pathology associated with working memory dysfunction in schizophrenia and illuminate the underlying brain circuits and neural mechanisms that support working memory processes.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Predoctoral Individual National Research Service Award (F31)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Rosemond, Erica K
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Columbia University (N.Y.)
Schools of Medicine
New York
United States
Zip Code
Padilla-Coreano, Nancy; Bolkan, Scott S; Pierce, Georgia M et al. (2016) Direct Ventral Hippocampal-Prefrontal Input Is Required for Anxiety-Related Neural Activity and Behavior. Neuron 89:857-66
Bolkan, S S; Carvalho Poyraz, F; Kellendonk, C (2016) Using human brain imaging studies as a guide toward animal models of schizophrenia. Neuroscience 321:77-98