Ease of Recall vs Recalled Evidence in Judgment: Experts vs Laymen

When major reasons of fault are presented in a pruned format, people fail to transfer the proportions of the omitted categories to the "all-other-problems" category. The present research investigated the underlying judgmental processes and the effect of domain knowledge on this phenomenon, known as pruning bias. In Experiment 1, although people with higher levels of domain knowledge failed to adjust completely for omission from the fault tree, they were more accurate and less biased than people with lower levels of domain knowledge. In Experiment 2, two hypothe ses are contrasted: The first, the availability explanation, sug gests that the underestimation of the "all-other-problems" category stems from its label being a less effective retrieval cue for specific fault reasons than the labels of major categories. The second hypothesis suggests that people judge by perceived ease of recall. The results suggest that people generating less failure reasons provide higher proportions to the "all-other-problems" category than people generating more reasons. The latter generation condition was perceived to be more difficult, signaling to people that the "all-other-problems" category is not so prevalent. The results of Experiments 3 and 4 indicate, as hypothesized, that the judgments of people with relatively low levels of domain knowledge are influenced by perceived ease of recall, whereas the judgments of people with higher levels of domain knowledge are influenced by recalled evidence. Copyright 2000 Academic Press.