基礎比率の無視を生み出す直観的確率判断過程のモデルの検討
数量的および言語的確率表現を用いて
1998 年 69 巻 5 号 p. 384-392
詳細
1998 年 69 巻 5 号 p. 384-392
Two models-the weighted averaging model and the anchoring and adjustment model-have been proposed to explain the base-rate neglect. In an attempt to test these two models, the cab problem (Bar-Hillel, 1980) was presented in two conditions. In one condition, the probability in case information was presented numerically, and in another, it was presented verbally. According to the weighted averaging model, subjects' estimates under verbal expression condition were predicted to be equal to or lower than the probability implied by the verbal expression, because legibility of verbal expression made subjects afford to consider base-rate. According to the anchoring and adjustment model, on the other hand, subjects' estimates under verbal expression condition were predicted to be higher than the probability implied by that expression (counter base-rate effect), because under a directional hypothesis that “blue cab was in accident”, in anchoring process, ambiguity of verbal expression was dragged. Results of three experiments were found to be consistent with the prediction from the anchoring and adjustment model.
