A theoretical account of the mirror effect for word frequency and of dissociations in the pattern of responding Remember vs. Know (R vs. K) for low- and high-frequency words was tested both empirically and computationally by comparing predicted with observed data theory in 3 experiments. The SAC (Source of Activation Confusion) theory of memory makes the novel prediction of more K responses for high- than for low-frequency words, for both old and new items. Two experiments used a continuous presentation and judgment paradigm that presented words up to 10 times. The computer simulation closely modeled the pattern of results, fitting new Know and Remember patterns of responding at each level of experimental presentation and for both levels of word frequency for each participant. Experiment 3 required list discrimination after each R response (Group 1) or after an R or K response (Group 2). List accuracy was better following R responses. All experiments were modeled using the same parameter values.

This article reviews the major empirical results and theoretical issues from over 20 years of research on people's acceptance of false information about recently experienced events (see, e.g., Loftus, 1975). Several theoretical perspectives are assessed in terms of their ability to account for the various and sometimes conflicting results in the literature. Theoretical perspectives reviewed include the trace alteration hypothesis, the blocking hypothesis, the task demands/strategic effects hypothesis, source monitoring, and an activation-based semantic memory account. On the basis of its ability to account for the reviewed data and other cognitive phenomena, an activation-based semantic network model of memory is suggested for understanding the data and planning future research in the area.

The word-frequency mirror effect (more hits and fewer false alarms for low-frequency than for high-frequency words) has intrigued memory researchers, and multiple accounts have been offered to explain the result. In this study, participants were differentially familiarized to various pseudowords in a familiarization phase that spanned multiple weeks. Recognition tests given during the first week of familiarization replicated a result of W. T. Maddox and W. K. Estes (1997) that failed to show the classic word-frequency mirror effect for pseudowords; however, recognition tests given toward the end of training showed the classic mirror pattern. In addition, a stimulus-frequency mirror effect for "remember" vs. "know" judgments was obtained. These data are consistent with an account of the mirror effect that posits the involvement of dual processes for episodic recognition.

Previous research has demonstrated that people have enormous difficultyin detecting distortions in such questions as, "How many animals of each kind did Moses take in the Ark?" Reder and Kusbit (1991) argued that the locus of the effect must be the existence of a partial-match process. Other research has suggested that this partial-match process operates at the word level and that, with adequate focus on the relevant word, the Moses illusion is greatly diminished. The present experimental results argue that those conclusions were based on a shift in response criterion with no concomitant change in ability to detect distortions. Furthermore, the data suggest that the matching process operates below the word level, at the level of distinctive features.Understanding the nature of access to memory-how we parse questions, query memory, and decide that the requisite information has been found-is one of the central issues in understanding memory. An important premise in our work is that the matching process in memory must be a partial-matching process. This may not seem obvious until one considers, for example, that information is often not queried in exactly the same form as originally presented or encoded, and that people do not look exactly the same from one occasion to the next; thus, person recognition must be flexible. Given that we believe that the memory-match process must be a partial-match process, the questions become: How much must the memory probe overlap with the memory trace to be accepted as a match? What portions of the memory probe are used to match to memory?One fruitful approach to these questions is to look at instances in which this partial-matching system leads people astray. For example, when asked, "How many animals ofeach kind did Moses take in the Ark?" most people immediately respond with "Two." This confident answer comes even from those who know that Noah, not Moses, built the Ark (Erickson & Mattson, 1981). When a term in a sentence or question (the "critical" term) is replaced with a semantically similar but incorrect term (the "distorted" term), people sometimes respond as if this distortion were not present. This tendency to overlook distortions in statements is known as the Moses illusion. Studying when the Moses illusion occurs and what factors influence it can shed light on the underlying memory processes in question answering and text comprehension.The work reported here was sponsored by Grant BNS-8908030 from the National Science Foundation to L.M.R. and by a National Science Foundation Graduate Research Fellowship to E.N.K. We thank G. Kusbit for her assistance in conducting Experiments I and 3 and Jason Wyse for comments on the manuscript. Correspondence concerning this paper should be addressed to L. M. Reder, Department of Psychology, Carnegie Mellon University, Pittsburgh, PA 15213 (e-mail: reder+@cmu.edu). This paper continues the line of research begun by Reder and Kusbit (1991), in which they concluded that the best explanation for the Moses illusion was imperfect matching...

No abstract