The ATB Framework: Quantifying and Classifying Epistemic Strategies in Tangible Problem-Solving Tasks

In task performance, pragmatic actions refer to behaviors that make direct progress, while epistemic actions involve altering the world so that cognitive processes are faster, more reliable or less taxing. Epistemic actions are frequently presented as a beneficial consequence of interacting with tangible systems. However, we currently lack tools to measure epistemic behaviors, making substantiating such claims highly challenging. This paper addresses this problem by presenting ATB, a video-coding framework that enables the identification and measurement of different epistemic actions during problem-solving tasks. The framework was developed through a systematic literature review of 78 papers, and analyzed through a study involving a jigsaw puzzle -- a classical spatial problem -- involving 60 participants. In order to assess the framework's value as a metric, we analyze the study with respect to its reliability, validity and predictive power. The broadly supportive results lead us to conclude that the ATB framework enables the use of observed epistemic behaviors as a performance metric for tangible systems. We believe that the development of metrics focused explicitly on the properties of tangible interaction are currently required to gain insight into the genuine and unique benefits of tangible interaction. The ATB framework is a step towards this goal.