Cortical Underpinnings of Cognitive Trajectories Associated with Psychosis Risk Restricted; Files Only

Abstract

Retrospective cohort studies demonstrate that people who later develop schizophrenia, on average, present with mild cognitive deficits in childhood and endure a cognitive decline in adolescence and adulthood. Yet, tremendous heterogeneity exists in cognitive functioning during the course of psychotic disorders, including the prodromal period prior to clinical onset. Individuals who manifest subthreshold symptoms similar to those associated with the psychosis prodrome are designated as being at clinical high-risk for psychosis (CHR-P) and are at heightened risk for subsequently developing psychosis. Prospective studies of CHR-P samples have revealed cognitive deficits can be longitudinally stable, ameliorate over time, or worsen in cognitive performance. Additionally, it is unclear how variation in cortical thickness and surface area—two indices that comprise volume and represent divergent characteristics of the cortex—relate to cognitive changes observed in CHR-P samples. The present studies were conducted with participants at CHR-P from the second cohort of the North American Prodromal Longitudinal Study (ages 12-35) who completed longitudinal follow-ups of cognitive functioning and/or structural neuroimaging. The first set of analyses parsed out several latent trajectories within visual learning and reasoning and problem-solving domains using latent cluster modeling. Trajectories are predominantly characterized as intact performance across time, mild deficits, and substantial deficits; thus, there was no evidence of a decline using this bottom-up approach. Secondly, longitudinal changes in cortical morphometry were delineated both within and between cognitive subgroups based on identified latent trajectories. Within latent classes, cortical thickness predominantly in right frontal regions decreased over time within classes that demonstrated intact performances or mildly impaired performances. Findings will aid in identifying which individuals at CHR-P are most likely to benefit from cognitive remediation and also shed light on the cortical substrates of cognitive trajectories based on clinically meaningful subtypes.

Table of Contents

General Introduction 1

References 16

Appendix 32

Study 1: Identification of Latent Trajectories of Cognitive Functioning among Clinical High Risk for Psychosis 33

Methods 37

Results 43

Discussion 49

References 56

Table 1. Summary of Demographic and Cognitive Variables at Baseline for the Current Subsample 64

Table 2. Summary of Demographic and Cognitive Variables at Baseline for Participants Without Cognitive Variables 65

Table 3. Domains and Their Corresponding Subtests for the MATRICS Consensus Cognitive Battery and the Wechsler Abbreviated Scale of Intelligence 66

Table 4. Count for Number of Participants for Each Cognitive Domain 68

Table 5. Class Membership and Fit Indices for MATRICS Consensus Cognitive Battery Domains and Wechsler Abbreviated Scale of Intelligence IQ without Covariates 69

Table 6. Latent Class Mixed Model Results for Each MATRICS Consensus Cognitive Battery Domain and WASI IQ without Controlling for Covariates (i.e., Age, Sex) 72

Table 7. Class Membership and Fit Indices for MATRICS Consensus Cognitive Battery Domains and Wechsler Abbreviated Scale of Intelligence IQ with Covariates (i.e., Age, Sex) 73

Figure 1. Longitudinal Trajectory of One-Class Solution for Intellectual Functioning) 76

Figure 2. Longitudinal Trajectory of One-Class Solution for the Processing Speed Domain of the MATICS Cognitive Consensus Battery 77

Figure 3. Longitudinal Trajectory of One-Class Solution for the Verbal Learning Domain of the MATICS Cognitive Consensus Battery 78

Figure 4. Longitudinal Trajectory of Three-Class Solution for the Visual Learning Domain of the MATICS Cognitive Consensus Battery. 79

Figure 5. Longitudinal Trajectory of One-Class Solution for the Attention/Vigilance Domain of the MATICS Cognitive Consensus Battery 80

Figure 6. Longitudinal Trajectory of Three-Class Solution for the Reasoning and Problem-Solving Domain of the MATICS Cognitive Consensus Battery 81

Figure 7. Longitudinal Trajectory of One-Class Solution for the Working Memory Domain of the MATICS Cognitive Consensus Battery 82

Table 8. Latent Class Mixed Model Results for Each MATRICS Consensus Cognitive Battery Domain and WASI IQ with Covariates (i.e., Age, Sex). 83

Table 9. Demographic and Clinical Characteristics by Latent Trajectories in Visual Learning Domain from the MATRICS Consensus Cognitive Battery 85

Table 10. Demographic and Clinical Characteristics by Latent Trajectories in Reasoning and Problem-Solving Domain from the MATRICS Consensus Cognitive Battery 89

Table 11. Correlation Matrix of WASI IQ and T-Scores from MATRICS Consensus Cognitive Battery Domains for all observations in the subsample at CHR-P 93

Study 2: Cortical Trajectories among Latent Classes of Clinical High Risk for Psychosis 94

Methods 99

Results 103

Discussion 108

References 116

Table 1. Demographic and Baseline Clinical Characteristics for Participants at CHR-P with and without Longitudinal MRI Data Available 130

Table 2. Demographic and Clinical Characteristics for Participants at CHR-P by Latent Class in Visual Learning Domain 133

Table 3. Demographic and Clinical Characteristics for Participants at CHR-P by Latent Class in Reasoning and Problem-Solving Domain 136

Figure 1. Statistical Maps of Rate of Change in Cortical Thickness in Frontal Regions Across Time by Latent Classes for Visual Learning 139

Figure 2. Statistical Maps of Rate of Change in Cortical Thickness Across Time by Latent Classes for Visual Learning 140

Figure 3. Statistical Maps of Rate of Change in Cortical Thickness in Frontal Regions Across Time by Latent Classes for Reasoning and Problem-Solving 141

Figure 4. Statistical Maps of Rate of Change in Cortical Thickness Across Time by Latent Classes for Reasoning and Problem-Solving 142

Table 4. Results for Latent Class × Time Interaction Term for Linear Mixed Effects Models for Visual Learning Domain 143

Table 5. Results for Latent Class × Time Interaction Term for Linear Mixed Effects Models for Reasoning and Problem-Solving Domain 144

Table 6. Summary for Changes in Cortical Thickness and Surface Area for Latent Class Based on Verbal Learning 145

Table 7. Summary for Changes in Cortical Thickness and Surface Area for Latent Class Based on Reasoning and Problem-Solving 146

Table 8. Correlation Table of Cognitive Measures and Mean Cortical Thickness and Total Surface Area by Lobe Among Participants at Clinical High Risk for Psychosis 147

Table 9. Correlation Table of Cognitive Measures and Regional Cortical Thickness and Surface Area within the Frontal Lobe Among Participants at Clinical High Risk for Psychosis 149

General Discussion 152

References 159

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Psychology
Subfield / Discipline	Clinical Psychology
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Psychology, Clinical
Keyword	clinical high risk for psychosis schizophrenia neuropsychology
Committee Chair / Thesis Advisor	Elaine F. Walker, Emory University
Committee Members	Rohan H.C. Palmer, Emory University Patricia Brennan, Emory University Joseph Cubells, Emory University Brianna L. Yamasaki, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 21 August 2031	2025-07-10 11:49:23 -0400	File download under embargo until 21 August 2031

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions