| Popis: |
In this project, we aim to understand better how memory processes are involved in preferential choice. Previous work suggests that value-based decisions require memory retrieval (Weber & Johnson, 2006). People sample evidence from memory and use this information to construct an option's value (Shadlen & Shohamy, 2016). Indeed, participants show greater activation in the hippocampus (a region associated with memory retrieval) for more extended deliberations during value-based choice (Bakkour et al., 2019). Nevertheless, it is still unclear to what degree and under what circumstances people require memory processes for value-based decisions. To better understand the specific role of memory in guiding preferential choice, we will compare two different types of multi-attribute binary choice: choices within the same domain (within-domain choices) and choices across different domains (across-domain choices). In within-domain choices, participants decide between two options from the same domain (e.g., two smartphones). In across-domain choices, participants decide between two options from different domains (e.g., a smartphone vs. a hotel stay). The critical difference between these two types of multi-attribute choice is that only in within-domain decisions, people can compare the options' attribute values directly with each other. This allows them to construct their preferences in the moment of choice (Slovic, 1995). In contrast, in across-domain decisions, people are required to represent both options' value on a common currency (Levy & Glimcher, 2011, 2012), which – as we hypothesize – requires sampling of evidence from memory (Shadlen & Shohamy, 2016). We expect these different underlying information processing mechanisms of choices across and within domains to be reflected in both the behavioral and the eye-tracking data. Within domains, participants can compare the attributes and select the option with a higher number of winning attributes (Hypothesis 1a). However, across domains, participants may compare the two alternatives as a whole, showing a slight preference for one or the other domain (Hypothesis 1b). Concerning visual attention, we assume more fixation transitions within single attributes across both options in within-domain choices and more fixation transitions within single alternatives across the alternatives' attributes in across-domain choices (Hypothesis 2). Previous research investigating the role of attention allocation in multi-attribute decisions (for example, Russo & Dosher, 1983) showed that in within-domain choices, within-attribute transitions are more frequent than within-alternative transitions. |
