Thought-intrusions, automatic inferences, and other unintended thought are beginning to play an important role in the study of psychiatric disease as well as normal thought processes. We examine one method for the study of task-unrelated imagery and thought (TUIT). TUIT likelihood was shown to be reliably measured over a wide range of vigilance tasks, to have high short-term and long-term test-retest reliability, and to be sensitive to information processing demands. Likelihood of TUITs was shown to be different as a function of aging, hyperactivity, time of the day, and level of depression. Thus, we now can reliably measure the influence of endogenous and exogenous influences on TUITs. In addition, TUIT measurement was proposed as a minimally interfering and natural second task for determining resource utilization in a primary task. Finally, this method was offered as a reliable approach to quantification of such mental states as obsessions and drug craving and addiction.