Probability errors in adults' and children's decision-making

Abstract

Three studies evaluated Tversky and Kahneman's (1983) proposal that the conjunction fallacy (judging the probability of a conjunction of two events to be higher than that of its component events) arises due to the representativeness heuristic. Since such heuristic thinking is not innate and depends upon the individual learning the extent to which situations are likely to occur, our evaluation adopted a developmental approach. Study 1 (N = 82 adults; N = 71 4- to 5-year-olds), Study 2 (N = 130 adults; N = 148 4- to 11-year-olds), and Study 3 (N = 76 adults) assessed objective probability judgements by asking participants to determine whether a single player or a two-player team would win based on assigned poker chip (adults) or building block (children) distributions. Social judgements were based on descriptions of individuals. All three studies showed that adults' conjunction fallacies in objective probability judgements were (a) influenced by the likelihood of winning, and (b) positively correlated with conjunction fallacies in judging social characteristics. Children's conjunction fallacies in objective probability judgements were not influenced by manipulating the probabilities assigned to either team, and did not differ as a function of children's age. Fallacies on the objective and social judgement tasks were positively correlated in 10- and 11-year-olds, but not in younger children. Study 3 showed a “thinking aloud” procedure (to facilitate rational, non-heuristic decision-making) reduced adults' fallacies on the social judgement, but not the objective probability task. Findings are discussed in relation to developmental changes in decision-making, and common versus distinct cognitive processes associated with objective and social judgement errors.