Dr. Philip Corlett is testing whether an epilepsy medication might help schizophrenia patients improve their focus on reality, and testing innovative digital algorithms for predicting patients' prognoses and response to treatments.

People with schizophrenia tend to hear voices that are not real, to express unrealistic beliefs, and to lack interest in ordinarily pleasant activities; hallucinations, delusions and anhedonia are three of schizophrenia’s signature symptoms. Medications generally used to treat hallucination and delusions (i.e., psychosis) block dopamine receptors-but about a third of patients with psychosis are not helped, and some patients find their anhedonia worsened by antipsychotic therapy. What if, for some patients, a single neurobiological process underlies both psychosis and anhedonia? If this process can be pinpointed and addressed with an original therapy, a large proportion of people with schizophrenia could benefit.

Philip Corlett, Ph.D., Assistant Professor of Psychiatry at Yale University and 2013 One Mind / Janssen Rising Star Awardee, has an idea what that process might be-and a good idea for how to treat it. His knowledge of a field of cognitive neuroscience called Bayesian brain theory leads him to propose that hallucinations, delusions, and anhedonia are the product of a parallel underlying neurophysiology:  Individual neurons in a psychotic brain appear to make aberrant predictions about incoming stimuli based on previous stimuli, and repeatedly adjust their sensitivity to new stimuli unhealthily. It appears that as this process continues, the neuron’s voltage gated potassium channels, which enact healthy sensitivity, close down.

If his hypothesis is correct, then wouldn’t a drug which could open potassium channels be worth a try for treating psychosis and anhedonia? Is it possible that the drug Retigabine, a potassium-channel opener already in use to treat epilepsy, could also treat these symptoms?

Testing a Potential Targeted Schizophrenia Drug

Dr. Corlett is using his Rising Star Award funds to enact a pilot study of Retigabine treatment with a cohort of human patient volunteers who have not been helped by traditional antipsychotic therapy. He is observing their brain activity using functional MRI to understand why this treatment might help some of these patients better than others, in the hope to enable more personalized approaches to schizophrenia therapy.

Corlett’s team has begun to develop encouraging results. In comparing his patients’ performance before and after a Retigabine dose, he hopes to track more accurate learning, more innovative problem-solving, and healthier trust-building (an indicator of real-world social skill) from patients in these respective tests. In parallel, Corlett hopes these subjects’ fMRI data will show accompanying improvements in potassium channel performance in the prefrontal cortex, according to Corlett’s original, empirically-based computational model. If these predictions are met consistently, Dr. Corlett will have established good evidence for a novel mechanism of schizophrenia symptom formation. And moreover, he will have demonstrated an effective new schizophrenia drug.

Since 2014, Corlett has expanded the scope of his study, piloting an approach to understanding hallucinations, the same way he has explained delusions. The approach involves learning to predict when audio and visual stimuli will happen. Patients who hear voices might learn different predictions than those who do not hear voices.

Predicting and Measuring Psychotic Experience via Linguistic Analysis

Also, since 2015, Corlett’s lab has pioneered language analysis algorithms that suggest the ability to predict development and remission of psychosis based on the predictability and relative frequency of word use in speech and writing. Collaborating in this research with Guillermo Cecchi’s team at IBM, Corlett’s lab has also found that word use frequency varies in common patterns among persons with mental illness and persons with physical illness, in contrast with word use frequency in healthy individuals. This finding suggests that linguistic analysis could possibly be used to objectively gauge the degree of suffering in individuals with mental (or physical) illness. Corlett aims to confirm these effects in his study cohort by analyzing the speech of psychotic patients for changes that might accompany the potential therapeutic effects of Retigabine.

Dr. Corlett says he hopes that this study “will aid a more consilient explanation of psychosis in terms of cells, brain function and symptoms. More crucially, it will test a novel therapeutic approach to a pathophysiological mechanism implicated not by serendipity but by empirical data. Ultimately, the work addresses a severe unmet clinical need in a manner informed by translational neuroscience.”

We at One Mind are grateful that our donors have enabled us to fund this brilliantly innovative scientist.

Your donation helps One Mind launch promising early-career investigators and their innovative research through the Rising Star Awards. Make a donation today that will accelerate ground-breaking discoveries in brain illness.

Philip CorlettSchizophreniaTreatment Research