Access Type

Open Access Dissertation

Date of Award

January 2016

Degree Type


Degree Name




First Advisor

Lisa J. Rapport


Purposeful presentation of neurocognitive impairment (i.e., dissimulation) in assessment of brain injury is a primary pitfall to accurate psychological assessment, especially among individuals seeking compensation. Current methods used to evaluate effort test failure (EFT; Webb et al., 2012) and dissimulation in brain injury assessment has advanced over the past few decades, but remains unacceptably inaccurate. In diagnostic decision-making, current methods identify obvious cases of purposefully poor performance, but they are considerably less accurate in subtle cases typically seen clinically; more important, they are vulnerable to coaching. Oculomotor behavior during visual tasks may be a promising avenue in the assessment of performance validity. Oculomotor patterns observed after brain injury have been well documented, and patterns characteristic of normal decision-making have been studied in healthy adults, but findings from these endeavors have not been applied to performance validity assessment. Accordingly, this study evaluated contributions of oculomotor patterns to detection of purposeful poor performance using state-of-the-science eye-tracking equipment by studying the predictive ability of a gold-standard performance validity test: The Test of Memory Malingering (TOMM). The study examined 39 adults with moderate to severe traumatic brain injury (TBI), 42 healthy adults coached to simulate memory impairment (SIM), and 50 healthy adults providing full effort (HC). The results supported the main hypothesis: One index derived using oculomotor patterns of performance provided a reliable increase to the predicative accuracy of the TOMM in differentiating bona fide TBI from simulated TBI. Numerous other oculomotor indexes showed promise, both in their relationships to key cognitive constructs and in their ability to differentiate dissimulation from healthy adults and bona fide TBI. The predicative ability of these measures was insignificant, however, due to an underpowered sample size and violations of the assumptions of pivotal statistical models. As such, future research is needed to replicate these findings and should strive to increase sample sizes to more accurately assess those visual patterns that showed predictive potential.

Included in

Psychology Commons