Amos Tversky

Record year1979–1992

summary1979 presented models of reference dependence, the value function, and decision weights; in 1992 the same authors changed them into a cumulative representation, extending multi-outcome/uncertain choice and correcting the boundary of stochastic dominance.

M rationaleCASBS, SEP, and the 1992 original paper support the two authors' model evolution. M6.7 is assigned for original-model foundation, cumulative revision, and scope of extension; 1992 is not treated as an external successor, nor is M duplicated by two papers.

attribution rationaleTversky/Kahneman each .34, team .05, prior work .15, institutions .04, unresolved .08.

overlap boundaryThe original 1979 prospect theory and Tversky/Kahneman's 1992 cumulative prospect theory: reference points, the value function, probability weighting, and treating multi-outcome/uncertain choices by cumulative probabilities; do not claim financial markets, policy, or all behavioral economics.

Evidence gapsThe division of labor between the 1979/1992 authors, data inputs, and the actual predictive increment of the cumulative revision still require original-manuscript archives.

Record year1977

summaryChallenged metric/dimensional assumptions, using a contrast model of feature matching to explain asymmetry, context, and common/distinctive-feature effects.

M rationaleThe original paper's abstract and first page of text make the model, assumptions, experimental subjects, and scope of generalization explicit; M5.9 is assigned for an independent form and experimental outcome.

attribution rationaleTversky .50, research-discussion team .09, prior work .24, institutions .04, unresolved .13.

overlap boundaryThe set-feature representation, common/distinctive-feature weighting, and explanations of asymmetrical and contextual similarity in 1977 Features of Similarity; do not claim successor classification, machine learning, or consumer applications.

Evidence gapsSpecific discussion collaborators, experimental materials, and subsequent revisions remain unresolved.

Record year1970s–1980s

summaryUsed a series of experiments to show predictable departures in judgment due to simplified heuristics and presentation frames, and provided a testable classification.

M rationaleThe joint-paper bibliography confirms representative works on heuristics and framing. M6.3 is assigned for testable psychological mechanisms and a research agenda across tasks; differences in replication, task dependence, and the risk of mistaking description for universal causal prescription all limit M.

attribution rationaleTversky .31, Kahneman .31, research execution .07, prior work .16, institutions .04, unresolved .11.

overlap boundaryThe body of joint experiments and discussions on representativeness, availability, anchoring, and framing effects from the 1970s through the 1980s; it excludes prospect theory's formal risk model and does not claim all behavioral economics, policy applications, or successor replications.

Evidence gapsSample sizes, materials, research assistants, and replication boundaries for individual experiments remain unresolved.

Record year1969–1972

summaryDemonstrated that intransitive preferences can systematically occur under specific conditions, and provided a process theory of attribute-by-attribute elimination that can be tested by choice probabilities and extends existing models.

M rationaleThe 1969 abstract and the 1972 original-paper abstract support the experimental proposition and model outcome, assigning M5.9; M is not duplicated for two papers.

attribution rationaleTversky .48, research team .08, prior work .25, institutions .04, unresolved .15.

overlap boundaryThe predictable preference-intransitivity experiment of 1969 and the 1972 attribute-by-attribute elimination/choice-by-elimination model; the two are combined as a continuous theoretical family of ‘attribute-by-attribute choice without global utility,’ excluding risk prospect models.

Evidence gapsModel uptake between the two papers and personnel jointly carrying out the experiments remain unresolved.

Record year1967–1990

summaryExtended conjoint measurement to partial orders, finite structures, and general polynomial models, and systematized with coauthors the bases for additive, polynomial, and other representations/uniqueness.

M rationaleThe Michigan original-paper page makes the 1967 five types of extension explicit, and the bibliography of the three volumes makes four coauthors explicit. M5.9 is assigned for an independent formal system and is not duplicated because of the number of volumes.

attribution rationaleTversky/Krantz/Luce/Suppes each .17, prior work excluding Luce .15, publishing .04, unresolved .13.

overlap boundaryTversky's 1967 general polynomial conjoint measurement, and the representational system in the three 1971, 1989, and 1990 volumes of Foundations of Measurement by Krantz/Luce/Suppes/Tversky; do not claim the Luce–Tukey 1964 original invention or consumer applications.

Evidence gapsDivision of labor chapter by chapter across the three volumes, and the specific extent to which the 1967 paper was consumed by the first volume, remain unresolved.