"choice is contingent"

"The notions of both rationality and limited processing capacity are consistent with the rowing belief among decision researchers that preferences for options of any complexity or novelty are often constructed, not merely revealed, in making a decision. People often do not have well defined preferences; instead, they may construct them on the spot when needed, such as when they must make a choice. Thus, consumer preference formation may be more like architecture, building some defensible set of values, rather than like archaeology, uncovering values that are already there.
.
The idea of constructive preferences denies that individuals simply refer to a master list of preferences in memory when making a choice and also asserts that preferences are not necessarily generated by applying some invariant algorithm such as a weighted adding model. Rather than one invariant approach to solving choice problems, consumers appear to utilize a wide variety of approaches, often developed on the spot. Consumers may also develop problem representations on the spot by structuring or restruchuing the available information. One important property of this constructive viewpoint is that preferences will often be highly context dependent. In addition. because decision approaches are developed on the fly, processing will be highly sensitive to the local problem structure. This implies that processing approaches may change as consumers learn more about problem structure during the course of making a decision.
.
Why are preferences constructive? One reason individuals may construct preferences is that they lack the cognitive resources to generate well-defined preferences for many situations. A second important reason is that consumers often bring multiple goals to a given decision problem. Preferences are not always constructed; people do have firm and stable preferences for some objects.
...
choice is contingent"

Trechos retirados de "Constructive Consumer Choice Processes" de James R. Bettman, Mary Frances Luce, e John W. Payne, publicado por Journal of Consumer Research, Vol. 25, No. 3 (December 1998), pp. 187-217