Capturing Context-Sensitive Information Usage in Choice Models via Mixtures of Information Archetypes

Joffre Swait, Monica Popa, and Luming Wang
Article Snapshot
Current average rating    
Key Takeaways

​The economic rationality perspective of full information usage and the psychological bounded-rationality view of selective information use can co-exist in the same decision-making space; we jointly test them and find support for both.

This research offers a new modeling approach and model structure for testing how much product information gets used, and what particular pieces of information are being used, when making product choices.

Information usage and preference heterogeneity can be simultaneously and separately captured.

Article Snapshot​s: Executive Summaries from the Journal of Marketing Research​​​​​

Investigating if consumers use all product information available to them when making a choice, we find that sometimes individuals do behave rationally (according to economic theory) as full-information users, while in other contexts they use very little data, acting instead as selective information managers.


Explicit modeling of information usage is largely overlooked in choice models. Economic models have been criticized for assuming that consumers are always using full product information. Ignoring consumer psychology insights on information selectivity implies that the choice model may be substantively disconnected from the phenomenon it tries to represent. In marketing practice, neglecting information use can lead to inaccurate policy recommendations (as a result of discounting the mechanism behind consumers’ choices) and can thwart significant opportunities to be responsive to consumers.


We offer a conceptual and practical solution for modeling context-sensitive information use in consumer product choice. We test the proposed framework by directly extracting information usage insights from observed choice data, without relying on self-reports. Four studies of multiple choice categories (beauty products, juices, computers, and digital cameras), with a gamut that spanned from simple to intricate choice situations, allowed a comprehensive examination of consumers’ contextual adaptation. The studies relied on choice data from four representative panels of consumers.


The proposed model is able to detect situations in which consumers use all information, limited information, or none at all when making product choices. The model captures information management situations, when consumers adjust their usage depending on the choice context. The model reveals how much information gets used in a given task, which particular pieces of information get used, and how the change in one feature can alter the interpretation of the choice environment (modifying the use of information about other attributes).


Rather than arguing that, when making choices, consumers are full information users (economic rationality view) or limited information managers driven by constraints (bounded rationality), it is more interesting to imagine a mix between the two archetypes. Embracing the mixed perspective allows researchers to gauge the real impact of information use heterogeneity on choices (going beyond preference heterogeneity), allows managers to avoid costly consequences of forecasting choices without properly accounting for information usage, and allows consumers to better understand their own product selections.  


​Questions for the Classroom

  • Is the economic rationality assumption that consumers use full product information in decision-making ever applicable or realistic in choice data?
  • What are the implications of ignoring information usage patterns in choice modeling?
  • What contexts are more conducive to selective information use when individuals are making product choices?

Article Citation

Joffre Swait, Monica Popa, and Luming Wang (2016), “Capturing Context-Sensitive Information Usage in Choice Models via Mixtures of Information Archetypes,” Journal of Marketing Research, 53 (5), 646-664.

Joffre Swait is Research Professor and Director, Research of Institute for Choice, University of South Australia (e-mail:

Monica Popa is Assistant Professor of Marketing, Edwards School of Business, University of Saskatchewan (e-mail:

Luming Wang is Assistant Professor of Marketing, University of Manitoba (e-mail:

Author Bio:

Joffre Swait, Monica Popa, and Luming Wang
Add A Comment :

Become a Member
Access our innovative members-only resources and tools to further your marketing practice.

 Upcoming Conferences


 The AMA Journal Reader

Sign up to receive Article Curations from AMA Journals


 Upcoming Webcasts


 Marketing News