Any method of multi-attribute decision making that is lexicographic inasmuch as the decision maker proceeds by first comparing the alternatives on the most important attribute, and if more than one alternative is best on the first attribute, the decision maker then compares them on the next most important attribute, and so on until just one alternative emerges as best or the list of attributes is exhausted without a definite first preference emerging. Suppose a decision maker has to choose between two travel agencies, namely A1 offering a tour only to destination D1 (Africa), and A2 offering tours to both D1 (Africa) and D2 (the Far East). The decision maker is equally attracted by Africa and the Far East and is indifferent between the travel agencies. According to the conventional linear value-maximization model of decision making, the three alternatives A1D1, A2D1, A2D2 are equally valued and are therefore equally likely to be chosen; but it is intuitively obvious that most human decision makers would choose first between destinations D1 and D2, because this attribute is most important, and only then between travel agencies A1 and A2, and consequently the probability of choosing A2D2 would be greater than that of the other two alternatives. Also called lexicographic decision making or lexical choice, although the latter is avoided in careful usage, because it causes confusion with lexical decision task. See also elimination by aspects, lexicographic semiorder.