Attentional Theory Is a Viable Explanation of the Inverse Base Rate Effect: A Reply to Winman, Wennerholm, and Juslin (2003).

Abstract

demonstrated the shortcomings of eliminative inference as an explanation of the inverse base rate effect, but they raise criticisms of Kruschke’s attentionally based explanation. First, Winman et al. pointed out that attentional shifting does not improve learning performance in Kruschke’s (1996) ADIT model, contrary to the claims that attentional shifting accelerates learning. This reply demonstrates that the deceleration of learning is a natural consequence when attentional shifts are not learned, as is the case in ADIT; however, when attentional shifts are learned, as was assumed by the underlying theory and as is the case in the EXIT model (Kruschke, 2001a, 2001b), then performance is indeed accelerated by attentional shifts. Second, Winman et al. pointed out that, whereas EXIT captures essentially all of the notable effects in the transfer data, it fails to capture a small effect [viz., p(C�PC) � p(R�PR)]. This reply demonstrates that when this trend in the data is merely weighted more heavily in the model fitting, then the EXIT model accommodates it. EXIT accomplishes this by emphasizing base rate learning more strongly. Thus, the EXIT model, and attentional theory more generally, remains a viable explanation of the inverse base rate effect. The inverse base rate effect refers to a perplexing behavior, first reported by Medin and Edelson (1988), in which people make

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.