A criterion-shift model for enhanced discriminability in perceptual identification: a note on the counter model

Performance benefits and costs in forced choice perceptual identification in amnesia: Effects of prior exposure and word frequency

Accuracy in identifying a perceptually degraded word (e.g., stake) can be either enhanced by recent exposure to the same stimulus or reduced by recent exposure to a similar stimulus (e.g., stare). In the present study, we explored the mechanisms underlying these benefits and costs by examining the performance of amnesic and control groups in a forced choice perceptual identification (FCPI) task in which briefly flashed words (that were identical to studied words, similar to studied words, or new) had to be identified, and two response choices were provided that differed from each other by one letter. Control participants showed a performance benefit and cost in FCPI with both high- and low-frequency words. Amnesic participants showed a benefit (but no cost) with high-frequency words and a benefit and a cost with low-frequency words. The benefit/cost pattern with low-frequency words in amnesia was obtained even when the to-be-identified stimulus in the FCPI task was eliminated (Experiment 2), suggesting that this effect was driven by processes operating at the level of the response choices. Our findings suggest that implicit memory effects in FCPI reflect the operation of multiple mechanisms, the relative contributions of which may vary with the frequency of the test stimuli. The results also highlight the need for caution in interpreting results from normal participants in the FCPI task, since those findings may reflect a contribution of explicit memory processes.

The Bayesian reader: explaining word recognition as an optimal Bayesian decision process

This article presents a theory of visual word recognition that assumes that, in the tasks of word identification, lexical decision, and semantic categorization, human readers behave as optimal Bayesian decision makers. This leads to the development of a computational model of word recognition, the Bayesian reader. The Bayesian reader successfully simulates some of the most significant data on human reading. The model accounts for the nature of the function relating word frequency to reaction time and identification threshold, the effects of neighborhood density and its interaction with frequency, and the variation in the pattern of neighborhood density effects seen in different experimental tasks. Both the general behavior of the model and the way the model predicts different patterns of results in different tasks follow entirely from the assumption that human readers approximate optimal Bayesian decision makers.

A model for evidence accumulation in the lexical decision task

We present a new model for lexical decision, REM-LD, that is based on REM theory (e.g., ). REM-LD uses a principled (i.e., Bayes' rule) decision process that simultaneously considers the diagnosticity of the evidence for the 'WORD' response and the 'NONWORD' response. The model calculates the odds ratio that the presented stimulus is a word or a nonword by averaging likelihood ratios for lexical entries from a small neighborhood of similar words. We report two experiments that used a signal-to-respond paradigm to obtain information about the time course of lexical processing. Experiment 1 verified the prediction of the model that the frequency of the word stimuli affects performance for nonword stimuli. Experiment 2 was done to study the effects of nonword lexicality, word frequency, and repetition priming and to demonstrate how REM-LD can account for the observed results. We discuss how REM-LD could be extended to account for effects of phonology such as the pseudohomophone effect, and how REM-LD can predict response times in the traditional 'respond-when-ready' paradigm.

Physiological evidence that a masked unrelated intervening item disrupts semantic priming: implications for theories of semantic representation and retrieval models of semantic priming

Event-related potentials were recorded in a paradigm where an unrelated word was interposed between two related words. In one condition, the intervening item was masked and in another condition it was not. The N400 component indicated that priming of the related word was disrupted by the intervening item whether it was masked or not. The data are interpreted to be inconsistent with retrieval models of priming.