Diagnostic judgments about learners are usually associated with uncertainty (because, for example, multiple causes for an error are possible and student behavior is not always consistent). Such uncertain judgments are to be understood as hypotheses that are supported or weakened by further information. Theoretically, such a multi-step diagnostic judgement can therefore be seen as a reduction of uncertainty through data-based inference: Based on new available data (e.g. further student solutions) the uncertain diagnostic hypotheses are updated. This process can be described using models of Bayesian inference (cf. Griffiths, Kemp & Tenenbaum, 2008). While Bayesian models describe ideal inferences in the presence of uncertainty, human inferences are affected by contextual conditions (e.g., Krolak-Schwerdt et al., 2012). For example, mindset theory (Gollwitzer, 2012; Gollwitzer & Keller, 2016) predicts that uncertainty is more strongly considered in so-called deliberate mindsets than in implemental mindsets. The deliberate mindset is characterized by weighing multiple decision aspects before making a decision. An implemental mindset, on the other hand, focuses on goal-oriented, rapid problem solving after a decision to act has been made. In the project, prospective teachers are placed in a deliberate or implemental mindset prior to the diagnostic situation. Subsequently, the (prospective) teachers make decisions on the basis of learner products (with systematically varied cue stimuli) about their latent characteristics (e.g. misconception, knowledge level etc.). The (non-verbalized) uncertainty of the decision is captured by having the subjects locate their decision in a geometric space (where the vertices are to be interpreted as high certainty, the spaces in between as uncertainty, see work program). We expect to gain insights into how well information processing in diagnostic judgments can be modeled as uncertainty reduction in terms of Bayesian inference and to what extent the deviation of human judgments from an ideal Bayesian judgment can be explained by the mindsets of the judging teacher.