Sequential Hypothesis Generation

Abstract

This paper examined how decision makers generate and evaluate hypotheses when data are presented sequentially. Hypothesis generation occurs in many judgment and decision making tasks, but no research has yet examined the underlying processes of hypothesis generation when data occur sequentially. In a series of three experiments, participants learned the relationship between data and possible causes of the data in a virtual environment. Data were then presented iteratively and participants either generated hypotheses they thought caused the data or rated the probability of possible causes of the data. In a fourth experiment, participants generated hypotheses and made probability judgments based on previously-stored general knowledge. The four experiments examined whether different orders of data led decision makers to consider different sets of hypotheses. Findings revealed that participants weighted data presented later in a sequence more heavily than data presented early in a sequence when responding after each datum was presented. Future experimental directions are detailed and potential assumptions necessary for a model to account for sequential hypothesis generation behavior are discussed.