Special Session Summary the Dynamics of Preference

Christina L. Brown, New York University
[ to cite ]:
Christina L. Brown (1996) ,"Special Session Summary the Dynamics of Preference", in NA - Advances in Consumer Research Volume 23, eds. Kim P. Corfman and John G. Lynch Jr., Provo, UT : Association for Consumer Research, Pages: 10.

Advances in Consumer Research Volume 23, 1996      Page 10



Christina L. Brown, New York University

The economic approach to preference assumes that preferences are independent of transient endowments, and that tastes are stable over time. The three papers in this session looked at consumer preference formation dynamically to demonstrate whether the adaptation and development of preferences over time, which we call "preference dynamics," generate any exceptions to these assumptions. The topic of preference dynamics is of interest to consumer behavior researchers because: (1) it addresses the age old question of whether or not individual preferences are meaningful and stable; (2) it sheds light on issues associated with consumer learning and expertise development; and (3) it examines issues associated with "hedonic consumption." Drazen Prelec was the discussant. The papers identified several key conceptual and methodological issues to stimulate additional research in this area:

How does past experience or ownership of a product impact current utility?

The paper by Michal Strahilevitz, George Loewenstein, and Daniel Kahneman focused on the effects of previous ownership on present utility. The paper demonstrated a potential violation of the economic assumption of independence by showing that although previous ownership does not affect attractiveness of a product, it does affect willingness to trade or sell, thus implying that previous ownership caused subjects to value a product at a level greater than the utility (attractiveness) it provides. Reported results showed that items never possessed may be valued less than items previously possessed. Furthermore, the findings indicated that the value placed on a current possession will increase as a function of the amount of time that has elapsed since the object was acquired. Among the theoretical issues addressed was the notion of simultaneous multiple reference points which may include both what we currently have and what we have had in the past.

To what extent are consumers able to predict the impact of current consumption on future utility?

Utility gained from consuming a good should diminish as the rate of consumption increases. The paper by Barbara Kahn, Rebecca Ratner, and Daniel Kahneman concerned whether people actually act as though they understand thisCthat their current consumption behavior affects future utility. In an experiment involving repeated consumption of favorite popular songs over time, subjects not only demonstrated an implicit understanding of diminishing returns by avoiding overexposure to favorites, but overcompensated by decreasing consumption too much (i.e., before experienced utility began to decrease). This is contrary to Herrnstein's principle of melioration, which suggests that people tend to be myopic by choosing their favorite to consume immediately, without taking into account that the utility of a good diminishes as the rate of consumption increases. Thus consumers acted appropriately but were not well calibrated in predicting their own future responses.

What affects whether consumers' preferences develop to a meaningful, predictable state or remain unfocused and unpredictable? How might the development of consumer preferences over time be modeled?

Fischhoff (1991) has suggested that consumers often do not have well-developed preference functions except for the most basic and frequently-encountered products. The paper by Patricia M. West, Christina L. Brown, and Stephen J. Hoch, examined prerequisites for the development of meaningfully consistent preferences. They proposed that consumers have difficulty learning their own preferences without the aid of a skeletal category structure, or "consumption vocabulary." They demonstrated how a rudimentary vocabulary of product attributes can encourage learning by providing such a structure. The "vocabulary" effect was tested in an experiment in which subjects provided with a vocabulary 1) exhibited better-defined and more consistent preferences than control subjects; 2) showed improved cue discovery, by becoming more consistent with regard to "implicit" attributes visible in the stimuli but omitted from the provided vocabulary; and 3) showed learningCi.e., increases in consistency over time. Repeated exposures allowed the degree of preference formation to be tracked dynamically over the course of the experiment through a novel moving-window model estimation procedure.