A theory of comparative response times and “difference” judgments

Abstract

In two experiments, subjects performed three tasks: First, they learned associations between names of hypothetical persons and adjectives that described them. Second, they judged "differences" in likeableness between pairs of names. Third, they pressed one of two keys to indicate which person of each pair was more (or less) likeable. Response times showed the traditional distance effect, end effect, and semantic congruity effect. A simple model was developed to describe response times and "difference" ratings, using a single scale of subjective magnitude. The second experiment applied this model to measure the likeableness of persons described by two or four adjectives. This experiment indicated that adjectives do not combine by a parallel-averaging model. Instead, "difference" ratings and comparative response times were consistent with Bimbaum's contigUral-weight theory, in which the weight that a stimulus receives depends on its relation to the other stimuli to be combined. In the case of impression formation, the less favorable information receives greater weight. Results are consistent with the hypothesis that ratings of "differences" and choice response times are mediated by the same scale of subjective value. Q 1990 Academic PESS, IK.The time required to compare two stimuli on a given dimension depends on the subjective difference between the stimuli on that dimension. The farther apart the stimuli are, the smaller the time required to decide which is greater. For example, as Dashiell (1937, p. 57) put it, "it should take a person less time to choose between something he likes very much and something he likes very little than for him to choose between two things he likes equally well." This phenomenon, sometimes called the "distance effect," has been observed in a large number of experiments involving a variety

Effects of instruction presentation mode in comparative judgments

“…Alternatively, according to differential bias theories of the SCE (Birnbaum & Jou, 1990;Link, 1990Link, , 1992Schwarz & Stein, 1998), this effect rises as a consequence of a dynamic, strategic adjustment of decisional criteria within an evidence accrual decision process (i.e., a random walk). For example, if the instruction is to choose the smaller stimulus and the stimulus pair contains a very small stimulus, bias theories assume that individuals recognize that this stimulus is likely to be the correct choice and lower the decision criterion associated with the accumulation of evidence for that stimulus (or conversely, recognize that this stimulus is unlikely to be the correct choice when the instruction is to choose the larger and raise that same decision criterion; see also Link, 1992, pp.…”

New paradoxes of risky decision making.

“…In addition, the authors found a contextual effect; this change of process occurs only when the gambles are of the form (x, p; 0), where all x Ͼ 0 or all x Ͻ 0 and few values were at or near zero. Birnbaum and Jou (1990) presented a model of choice response times to describe three major phenomena of choice response times: the distance effect, the end effect, and the semantic congruity effect. The distance effect refers to the fact that it takes less time to choose between two stimuli that are farther apart in subjective value.…”

“…Both models correctly imply changes of rank order between buying and selling prices. As noted by Leth-Steensen and Marley (2000) in reference to Birnbaum and Jou (1990), it can be difficult to distinguish effects of bias parameter from those attributed to the evaluation process. A critical test between these interpretations has not yet been conducted.…”

The discovery and comparison of symbolic magnitudes

“…Indeed, semantic congruity effects have been observed not only in magnitude comparisons involving objects, but also in judgments of preferential choice. For example, Birnbaum and Jou (1990) found that judging which individual is ''liked more'' for generally likeable individuals took less time than judging between unlikeable individuals, whereas judging which individual is ''liked less'' for likeable individuals took more time than judging between unlikeable individuals (also Nagpal & Krishnamurthy, 2008). The mechanisms instantiated in the BARTlet model may well prove applicable to decision making in areas such as consumer choice and social judgment.…”