[1] B. W. Wildemuth, L. Freund, and E. G. Toms. Untangling search task complexity and di culty in the context of interactive information retrieval studies. Journal of Documentation, 70(6):1118{1140, 2014.
In a recent review of over one hundred interactive information retrieval studies, Wildemuth et al. [1] observed that despite widespread usage of the concepts of task complexity and task di culty, \clear and consistent de nitions of these attributes are lacking and there is no consensus on how to distinguish levels of complexity or di culty within a set of search tasks" (p. 1119). Importantly, the authors called for the creation of richer conceptual models and more precise and valid measurement practices. In this talk, I will present a theory of search task di culty that arose from results of a recent empirical study that evaluated a set of search tasks that were constructed using a cognitive complexity framework from educational theory. While results of this study showed participants engaged in signi cantly more search activity when completing more cognitively complex tasks, they did not evaluate more cognitively complex tasks as more di cult and were equally satis ed with their performances across tasks. These ndings have led to a new theory of search task di culty that is rooted in expectancy and appraisal theories, and posits that when searchers rst encounter a search task description, they rst appraise the task and its requirements in the context of their abilities, desires and other aspects of the search situation (e.g., system familiarity, time limits) and then make an estimate of how much e ort will be needed to complete the task to achieve their ideal level of performance. A search task, consequently, becomes \di cult" when expended e ort exceeds expected e ort. This theory di ers from many current notions of search task di culty that assume a linear relationship between search task di culty and search behavior. It also has implications for how search task di culty is measured. Biography