Children's and adults' interpretation of communicated probabilities: studies on directionality and preciseness

Twelve experiments investigated how children and adults interpret verbal probabilities (e.g., it is likely). The experiments were designed to determine if and when children and adults use the directionality or the likelihood of verbal probabilities. In Experiments 1a and 1b, I showed that children use only the directionality of verbal probabilities to make decisions. However, they dismiss it when speakers are malevolent. In Experiment 2a, adults showed that they do not consider only the directionality or the likelihood when making decisions. Rather response times suggested that adults are sensitive to the potential conflict between the two features. In Experiment 2b, I showed that, given an unlimited time to decide, adults can show less preference for the positive directionality. However in Experiment 3a, I found that in conversational context, adults prefer the positive directionality even when given more time to decide. In Experiment 3b, adults used the directionality in different ways according to speakers' intentions. In contrast with children in Experiment 1b, they preferred the negative directionality when the speaker was malevolent, rather than dismissed the directionality overall. In Experiments 4a to 4e, counter to expectations, I did not find that speakers using more precise format to communicate probabilities are judged more responsible based on their predictions' accuracy. Instead the results suggest that listeners reward predictions that suggest that speakers' wish for the best outcome for listeners. Finally in Experiment 5, I found that the preference for receiving more precise probabilistic information is contingent on speakers' expertise. These results together support a pragmatic account of verbal probabilities. The directionality of verbal probabilities is a pragmatic cue that influences decision making by shaping listeners' assumptions.