Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




Research has indicated that Continuous Performance Tests (CPTs) can differentiate Attention-Deficit/Hyperactivity (ADHD) subjects from controls without any psychiatric illness. However, CPTs have neither accurately differentiated ADHD children from those with other psychiatric disorders—nor differentiated subtypes of ADHD from each other. The Test of Variables of Attention (TOVA), a new CPT, has several advantages over its predecessors which may allow the TOVA to be more effective in this differentiation process.

Data from ADHD subjects was selected from children who were administered the TOVA as part of their evaluation for ADHD at Lakeland Mental Health Center in Moorhead, MN, the Child Evaluation and Treatment Program in Grand Forks, ND, and the Behavioral Health Clinic at the St. Cloud Hospital in St. Cloud, MN. Learning Disordered subjects' data was obtained from a previous study by Clay et al. (1996). Children with no history of psychiatric illness were recruited by offering research participation credit to University of North Dakota students who agreed to have their children participate in this study. Results were evaluated by using a group (ADHD-C, ADHD-I, Learning Disordered, and non-patient control) by TOVA quartile (1,2,3,4) mixed ANOVA on all TOVA variables (using age-corrected standard scores). In addition, I computed the Positive Predictive Power (PPP), Negative Predictive Power (NPP), and Sensitivity of the TOVA variables in order to determine the diagnostic utility of these measures. Finally, to test a theory that “high consistency” ADHD children might outperform controls, each group was divided into halves based upon the group's response time variability scores (by a simple median split). A group by consistency (high variability vs. low variability) ANOVA was conducted on the remaining TOVA variables (errors of omission, errors of commission, and response time).

Results of the PPP/NPP analyses suggested that some TOVA variables are useful in differentiating ADHD children from non-patient controls, but not useful in differentiating ADHD from LD children. Also, TOVA data do not appear to be able to differentiate ADHD subtypes from each other. Finally, the data provided little support for the theory that a subgroup of “high consistency” ADHD children would outperform controls on other TOVA variables.

Included in

Psychology Commons