Abstract

Investigating individual differences in cognition requires addressing questions not often thought about in standard experimental designs, especially those regarding the psychometrics of a task. The purpose of the present study is to use the AX-CPT cognitive control task as a representative case study example to address four concerns that may impact the ability to answer questions related to individual differences. First, the importance of a task's true score variance for evaluating potential failures to replicate predicted individual differences effects is demonstrated. Second, evidence is provided that Internet-based studies (e.g., MTurk) can exhibit comparable, or even higher true score variance than those conducted in the laboratory, suggesting the potential advantages of such data. Third, the need to evaluate and assess psychometrics between theoretically-driven and raw behavioral measures, and how they may show different correlation patterns with an individual difference outcome measure is shown for both internal consistency reliability and test-retest reliability. Finally, the need to restrict generalizations of psychometrics across samples or populations is highlighted by demonstrating differences in true score variance and their consequences in a schizophrenia cohort compared to matched controls.

Committee Chair

Todd Braver

Committee Members

Deanna Barch, Joshua Jackson

Comments

Permanent URL: https://doi.org/10.7936/K7MS3R6P

Degree

Master of Arts (AM/MA)

Author's Department

Psychology

Author's School

Graduate School of Arts and Sciences

Document Type

Thesis

Date of Award

Winter 12-2016

Language

English (en)

Download

Included in

Psychology Commons

Share

COinS