| People often base their judgments and decisions on various pieces of information, or cues. Once decision-makers have identified the relevant cues, they must consider the best way to weight (i.e., assign importance to) and aggregate the cues. Typical models of cue weighting have focused on how decision-makers weight cues individually. The current research explores a different strategy where decision-makers consider groups or clusters of information when aggregating cues. Consider a case where a person must make a judgment about the performance of financial markets based on data for different stocks. This person can assign weights to the data for each individual stock. Or this person might notice that there are multiple stocks from each financial sector. By recognizing clusters of data points, the person might draw inferences about sector-level data and use the sector-level evaluations to inform their judgment. Stated broadly, decision-makers can weight cues individually (i.e., at the cue-level) or as part of groups (i.e., at the group-level). Across a series of experiments, I demonstrate how cue-level and group-level aggregation of the same information can yield substantially different judgments. These studies show how people spontaneously pack information into cue groups and also how group-level interpretations are malleable. Different groups might emerge depending on how people are primed, for instance. Finally, these studies suggest that group-level aggregation is more effortful than simple cue-level aggregation strategies. |
