Date of Award
Fall 2022
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Interdepartmental Neuroscience Program
First Advisor
Corlett, Philip
Abstract
Schizophrenia is a debilitating psychiatric illness affecting roughly 1 percent of the global population. Psychiatry classifies schizophrenia, along with several other diagnoses, as a psychotic illness because patients experience a loss of touch with reality (psychosis). The current diagnostic criteria involve broad domains of symptoms including: Hallucinations, delusions, disorganized speech, grossly disorganized or catatonic behavior and negative symptoms (defined broadly by a lack of mental functioning). If a person experiences symptoms in two or more of these domains, they can be diagnosed with schizophrenia. This expansive definition makes schizophrenia difficult to investigate and treat. When examining the cause of schizophrenia, case-control study designs in humans compare a highly heterogeneous group of participants with the same diagnosis to a highly heterogeneous group of people without that specific diagnosis. When we try to model the disease in animals to further elucidate its biological basis, we are attempting to capture this wide range of experiences which may have unique etiology, and, in some cases may be exclusive to humans. Our approach for addressing this heterogeneity in the experience of patients is two-pronged. First, we explicitly probe how these different symptoms relate to one another and develop a framework to organize this range of experiences. Then we use a targeted approach to try to relate specific symptoms to biology (using pharmacological models) and cognition (using behavioral models) in a way that allows for translation across species. Chapter 1 provides context for how we define schizophrenia, psychotic symptoms experienced by these patients, the stages and outcomes for those diagnosed and how these symptoms are most commonly modeled across species using pharmacological and behavioral models. In Chapter 2 we focus on understanding these symptoms which are used to define schizophrenia. Psychotic symptoms experienced by patients range from auditory-verbal hallucinations to incoherent speech and catatonic behavior. We applied dimensionality-reduction techniques to two detailed clinical measures of symptom severity to determine whether certain subsets of symptoms co-occur more than others. We did this in two unique samples of participants: one with a persistent psychotic illness (most commonly schizophrenia) and the other in their first-episode of psychosis that were initially seeking treatment. Consistent with previous work, we found that symptoms generally divide into reality distortion (e.g., hallucinations and delusions), thought disorder (e.g., disorganized thought and speech) and negative symptoms (e.g., amotivation and apathy). However, of interest to us, the reality distortion symptoms also showed subdivisions which were consistent across groups and subgroups of participants and was based on content of the symptom. These subdivisions were further replicated and delineated using an independent statistical technique in the same samples. Overall, this chapter of my dissertation provides a framework to study certain symptoms of psychosis and suggests that different types of hallucinations and delusions may have unique neurobiological and cognitive basis. In Chapter 3 we test biological effects of ketamine, a pharmacological agent known to induce certain psychotic symptoms. Ketamine or placebo was administered to human participants and functional magnetic resonance imaging scans were taken while the participant was not completing any task (often referred to as resting state). We tested for changes in functional connectivity between brain regions or networks known to consistently show altered connectivity in schizophrenia and to correlate with specific psychotic symptoms of interest in patients. There are many functional connectivity changes found in patients with schizophrenia as compared to controls, in part due to heterogeneity in patient populations. We wanted to determine whether some of these changes may relate to NMDA blockade to help elucidate a more mechanistic explanation. We found many consistencies between our findings and those observed in the first-episode phase of psychosis. The insula and cingulate gyrus appeared to be important hubs in psychosis. Next, in Chapters 4 and 5 we tested the effects of ketamine and another psychotomimetic agent, LSD, on a mediated learning task in rats. These tasks are designed to probe a hallucination-like phenotype based on perceptual expectation. Various psychotomimetic manipulations have been applied to mediated learning paradigms in rodents and shown to increase propensity towards mediated learning. However thus far, most of these manipulations have not been targeted to any specific timepoint in this task, making interpretation of the exact cognitive processes which may underlie this phenomenon and its relationship to psychosis unclear. Thus, we wanted to use pharmacological manipulations which can be targeted to the specific timepoints that we hypothesize are directly relevant to a hallucination-like phenotype. A related task in humans specifically tests whether eliciting learned perceptual expectations can cause false percepts, referred to as “conditioned hallucinations.” Proneness to conditioned hallucinations has been shown to be higher in people who hallucinate regularly. Thus, we wanted to target the timepoint in our mediated learning task when the expected cue is elicited to determine whether psychotomimetic manipulations similarly cause excess representation of this expected sensory cue in our paradigm. In Chapter 4 we used ketamine as our psychotomimetic manipulations, probing its effect when targeted to the point in the mediated learning paradigm when the representation of an absent but expected cue is elicited and its value can be updated accordingly, driving mediated learning. We found that a moderate dose of ketamine at this timepoint increased mediated learning as hypothesized. And in Chapter 5 we find that LSD administered during the same timepoint did not have effects on mediated learning, potentially confounded by some non-specific effects of LSD on behavior. Finally, in Chapter 6 we discuss these chapters together, draw conclusions from this dissertation work, propose neurobiological mechanisms which may underlie specific processes in mediated learning and hallucinations in patients with schizophrenia. We also suggest some important future directions to this work in this final chapter. Overall, this dissertation details my use of many different approaches to better understand symptoms of psychosis. A challenge with studying these symptoms and determining their biological and cognitive basis is that they span so many different experiences. I approached this challenge by first inspecting more closely how these symptoms may be organized in a data-driven way, then focusing in on pharmacological models which can recapitulate some of these symptoms through well-characterized neurochemical mechanisms. Finally, I focused on a particular behavior (mediated learning) which relates to specific cognitive functions (learning, prior beliefs, reality monitoring) believed to underlie subsets of symptoms. Linking phenomenology and biology more carefully will facilitate a deeper understanding and ultimately better treatments for psychosis.
Recommended Citation
Fleming, Leah, "Symptoms of Schizophrenia and Their Relationship With Pharmacology and Learning Processes Across Species" (2022). Yale Graduate School of Arts and Sciences Dissertations. 834.
https://elischolar.library.yale.edu/gsas_dissertations/834
