Previous research has shown that individuals endorsing inattention and hyperactivity-impulsivity have deficient performance on tasks tapping different aspects of attention. Although there is empirical evidence suggesting that the behavioral domains of inattention and hyperactivity-impulsivity are linked to functioning of independent and separate brain areas and neurotransmitter systems, cognitive characterization of adults presenting with problems within these domains is not complete. The aim for this study was to identify the cognitive correlates of the core behavioral domains that define the diagnosis of AD/HD (i.e., inattention and hyperactivity-impulsivity) in a sample of college students, utilizing a computerized attention task, the Attention Network Test (ANT). Different ANT task components have been found to activate separate brain areas linked to the functioning of alerting, orienting and executive attention, and have the potential to provide an indication of the efficiency of these brain networks. In addition to completing the ANT, the participants filled out questionnaires covering common symptoms of adult AD/HD, anxiety and depression. Hierarchical regression analyses revealed that there were no reliable relationships between self-reported symptoms of current inattention and hyperactivity-impulsivity and ANT performance. Further, self-reported depression and/or anxiety did not seem to impact the efficiency of attention networks to a significant degree in this study sample. Gender proved to be the most consistent predictor of ANT performance. Female gender was related to poorer executive attention efficiency. An exploratory ANCOVA revealed that individuals reporting high levels of impulsivity and emotional lability had poorer executive attention efficiency in comparison to those reporting these behaviors and problems to a lesser extent. Future research is needed in order to further explore the relationship between ANT performance and behavioral expressions of adult AD/HD and other neurological and psychiatric conditions.

Using orthogonal subtractions of performance in selected conditions the attentional network test (ANT) measures the efficacy of three isolable components of attention: alerting, orienting, and executive control. This dissertation evaluated: 1) the relationship between these attention networks and absentmindedness measured by the Cognitive Failures Questionnaire (CFQ) and 2) stability, isolability, robustness, and reliability of the two versions of the ANT (Fan et al., 2002 and Callejas et al., 2005) with young adults and older adults and of the child version of the ANT (Rueda et al., 2004) with young children when tested over 10 sessions. A greater degree of absentmindedness as measured with CFQ was associated with a greater alerting network score in RT and with a greater orienting network scores in error rate when the ANT-I was used. However, a greater degree of absentmindedness was associated with a smaller orienting network score in error rate when the ANT was used. These results suggest that the alerting and the orienting networks are related to absentmindedness. However, the orienting networks in the two ANTs were related to absentmindedness differently which supports the proposal (Klein, 2009) that there are fundamental differences between attention when controlled endogenously (ANT) as opposed to exogenously (ANT-I). For young adults and older adults, all network scores in RT remained robust even after nine previous sessions despite some practice effects especially in the executive network both with the ANT and the ANT-I. There was some evidence that the networks do not operate independently in all situations. As expected, reliability increased as more data are added. For young children, only the alerting network scores remained robust over time. Learning effects were observed only with the executive network. The reliability was poor even when more data were added. This made it difficult to assess the isolability of the network scores. The ANT and the ANT-I were associated to the CFQ scores in a limited way. The ANT and the ANT-I can be used for applications requiring repeated testing, but the child ANT may not be suitable for such purpose. / This is a thesis that is manuscript-based.