Apropiación Social del Conocimiento · 2025

The effect of context switching on predictive learning with funeral and marketing stimuli

In a trial-by-trial predictive judgment task, human participants were trained to predict the probability of a specific outcome value (discounts) based on various cues (funeral products) in different contexts (funeral parlor vs. home). The experiment used a mixed factorial design with two groups of participants. In Group Partial-One-Context, a cue (X) was paired with three distinct discount values: 50%, 30%, and 0% (partial reinforcement, PR). Cue Z was paired with a discount value of 30% for all trials (continuous reinforcement, CR). Both cues were trained in context A. In context B, cue Y was paired with a 0% discount. Testing was conducted in Context A and Context B. Participants in Group Partial-Both-Contexts received the same training as Group Partial-One-Context, except that in Context B, Cue Y was paired with three different discount values: 5%, 10%, and 0% (under PR). In Group Partial-One-Context, context-switching effects were found for Cues X and Y but not for Cue Z. In contrast, Group Partial-Both Contexts showed no context-switching effects. These findings are discussed in relation to the context-switching effects observed with emotionally charged stimuli, such as funeral products. In particular, the results have implications for understanding how consumers respond to marketing cues and promotional tactics, such as discounts, when dealing with emotionally sensitive products across varying contexts.