Elliot Hirshman

Modeling the conscious correlates of recognition memory : Reflections on the remember-know paradigm

Understanding how memory processes contribute to the conscious experience of memory is central to contemporary cognitive psychology. Recently, many investigators (e.g., Gardiner, 1988) have examined theremember-know paradigm to understand the conscious correlates of recognition memory. A variety of studies have demonstrated that variables have different effects on remember and know responses, and these findings have been interpreted in the context of dual-process models of recognition memory. This paper presents a single-process model of the remember-know paradigm, emphasizing the dependence of remember and know judgments on a set of common underlying processes (e.g., criterion setting). We use this model to demonstrate how a single-process model can give rise to the functional dissociations presented in the remember-know literature. We close by detailing procedures for testing our model and describing how those tests may facilitate the development of dual-process models.

Decision processes in recognition memory: criterion shifts and the list-strength paradigm.

This article focuses on decision processes in recognition memory. It begins with investigation of the hypothesis that the measured criterion increases systematically with the memorability of old items. Three experiments using the list-strength paradigm, and a review of the prior literature, present results consistent with this hypothesis. Several psychological models of criterion placement are examined, generating different predictions about the relative sizes of criterion shifts for strong and weak items. A range model, in which criterion placement depends on the estimated range of the old and new distributions, predicts that criterion shifts should be larger for weak items; this result emerges in a reanalysis of prior studies. The general discussion elaborates on how a focus on criterion placement can explain the mirror effect (Glanzer, Adams, Iverson, & Kim, 1993) and provides a framework for testing Shiffrin, Ratcliff, and Clarks (1990) claims about why null effects of list strength occur with repetition.

The Role of Decision Processes in Conscious Recollection

Dual-process models of recognition memory posit a rapid retrieval process that produces a general sense of familiarity and a slower retrieval process that produces conscious recollections of prior experience. The remember/know paradigm has been used to study the subjective correlates of these two processes, with remember judgments assumed to index conscious recollection and know judgments assumed to index familiarity. Recently, a two-criterion signal detection model has been proposed as an alternative account of this paradigm. This model assumes only a single memory process with a criterion separating remember from know responses. This report presents an empirical test of the models critical prediction that manipulations that influence criterion placement should influence both remember and know judgments. An experiment confirmed this prediction, demonstrating that subjects who were told that 70% of the test items were study items produced more remember and know responses than subjects who were told that 30% of the test items were study items.

Combined Pharmacological and Electrophysiological Dissociation of Familiarity and Recollection

Dual-process theories of recognition memory hypothesize separate underlying familiarity and recollection processes, but the necessity of multiple processes is debated. Previous research has suggested that scalp-recorded event-related brain potentials (ERPs) may index the activity of separate familiarity and recollection processes. Other research indicates that the amnestic drug midazolam impairs recollection more than familiarity. Here, we used a convergent pharmacological and electrophysiological approach to manipulate and monitor human brain activity and provide evidence for separate processes. Midazolam selectively influenced the putative ERP-correlate of recollection but not the putative ERP-correlate of familiarity. Under control conditions (saline), subjects’ accuracy correlated with the recollection-related but not the familiarity-related ERP component, suggesting that recollection was dominant in driving memory. The opposite pattern was observed under midazolam administration, suggesting that when recollection fails, subjects may leverage familiarity to compensate. Thus, in contrast to perspectives holding that familiarity represents the default process, these results suggest that recollection was dominant until its impairment unveiled the influence of familiarity.

Midazolam amnesia and dual-process models of the word-frequency mirror effect

The word-frequency mirror effect (Glanzer & Adams, 1985) is the finding that subjects are more accurate on low frequency words than high frequency words for old and new items in recognition memory. Recently, several theorists (Guttentag & Carroll, 1997; Joordens & Hockley, 2000; Reder et al., 2000) have proposed dual-process accounts of the word-frequency mirror effect. These accounts hypothesize that the low frequency advantage on old items arises from increased recollection of these items, while the low frequency advantage on new items arises from reduced familiarity of these items. We tested these views using midazolam amnesia. Midazolam is a safe, fast-acting benzodiazepine that produces a dense anterograde amnesia. Based on the hypothesis that midazolam amnesia should have larger effects on recollection than familiarity, we predicted that: (1) old high frequency words should have an advantage over old low frequency words in a midazolam condition (i.e., the traditional effect should reverse); and (2) the traditional advantage of new low frequency words over new high frequency words should replicate in a midazolam condition. Both predictions of the dual-process approach were confirmed. Additional analyses demonstrated that a single-process signal detection model could not account for the current results and that midazolam amnesia produces a larger effect on recollection processes than on familiarity processes.

An investigation of paradoxical memory effects

Abstract The expectation-violation effect ( Hirshman, 1988 , J. Memory Language 27, 40–58) occurs when weakly related word pairs are better remembered than strongly-related word pairs following intentional learning. Hirshman (op. cit.) claimed that surprise responses to weakly related pairs caused the expectation-violation effect. This paper specifies this hypothesis further and demonstrates that the bizarre-imagery effect ( McDaniel & Einstein, 1986 , J. Experimental Psychology: Learning Memory and Cognition, 12, 54–65) is also a member of the class of expectation-violation effects. Specifically, bizarre sentences are better remembered than normal sentences because surprise responses to bizarre sentences increase the association of items in bizarre sentences to general contextual cues. The utility of the current theory is in explaining two seemingly unrelated effects with a simple theoretical mechanism. Four initial experiments demonstrate that the bizarre-imagery effect and the expectation-violation effect obey the same empirical regularities. Two further implications of the claim that the bizarre-imagery effect is a member of the class of expectation-violation effects are then tested. Two experiments demonstrate that the bizarre-imagery effect can occur without imaginal encoding instructions and that it disappears when subjects are informed of the presence of bizarre sentences in the study list.

Investigations of the testing effect

This paper focuses on methodological problems arising in the study of the testing effect. These problems arise because processes that are correlated with, but logically independent of, the process of testing often differ across the study and test conditions. Carrier and Pashler (1992) recently reviewed these problems and proposed a paired-associate procedure for alleviating them. In this procedure, robust testing effects occurred, suggesting that the processes underlying testing may differ from those in a standard study condition. One potential problem with Carrier and Pashlers procedure is that opportunities to restudy previously tested items may be contributing to the testing effect. We use a modified Brown-Peterson paradigm (Peterson & Peterson, 1959) to provide converging evidence for Carrier and Pashlers (1992) conclusions. The results of Experiment 1 demonstrate testing effects even when there are no opportunities to restudy previously tested items. Experiment 2 examines whether, as suggested by Carrier and Pashler (1992), stronger interitem associations might be producing the testing effect arising from an intervening free recall test. The results demonstrate that testing effects occur even when single item lists are used, ruling out the view that testing effects are due solely to stronger interitem associations.

Perceptual interference improves explicit memory but does not enhance data-driven processing

Nairne (1988) has recently demonstrated that interfering with the perceptual processing of an item at study improves later memory for that item. Nairne hypothesized that interfering with perceptual processes induces a data-driven generation process that enhances the representation of visual information. Using a variant of Nairnes procedure, we both replicated his original findings and tested his hypothesis that enhanced data-driven processing causes the current effect. The results of studies using free recall and perceptual identification tests were inconsistent with Nairnes hypothesis. We consider several alternative interpretations in the General Discussion.

Theoretical implications of the mnemonic benefits of perceptual interference

How should interfering with the perception of items during study affect memory for those items? Recent research by Nairne (1988) and Hirshman and Mulligan (1991) has demonstrated that backward pattern masking during study enhances later memory. This article examines whether traditional explanations of encoding benefits, including rehearsal, visual distinctiveness, and encoding effort, can account for this result. No evidence was found for any of these hypotheses. An explanation that focuses on the compensatory processing of higher level perceptual representations is proposed. This explanation provides a plausible explanation of the results of 7 experiments. The article concludes with a discussion of the implications of the explanation for perceptual priming and other manipulations of perceptual interference. How does interfering with the perception of items during study affect memory for those items? Nairne (1988) and Hirshman and Mulligan (1991) recently investigated this question using visual masking. These investigators examined memory performance on items presented in two study conditions. In the control or intact condition items were presented for 2.5 s. In the experimental or interference condition items were presented briefly (i.e., 100 ms) and then followed with a backward pattern mask (i.e., a row of Xs) presented for 2.4 s. The mask obscures the visual features of the study item, interfering with its perception. Surprisingly, these studies have demonstrated that interfering with perception using visual masking can actually improve memory performance. Nairne (1988) reported an advantage for the interference condition in recognition memory. Hirshman and Mulligan (1991) replicated Nairnes recognition memory results and demonstrated that this effect also occurs in free recall.1 These results seem surprising from several perspectives. First, many classical (Atkinson & Shiffrin, 1968; Waugh &

Evidence that androgenic and estrogenic metabolites contribute to the effects of dehydroepiandrosterone on cognition in postmenopausal women

Prior studies of the effects of dehydroepiandrosterone (DHEA) on cognition have produced complex and inconsistent results. We hypothesize that these results may arise, in part, because of DHEAs metabolism into estrogens and androgens that produce opposing effects on cognition. Our study administered 50 mg of oral DHEA daily for 4 weeks in a placebo-controlled crossover design to six postmenopausal women. We measured blood levels of androgens (total testosterone, free testosterone, DHEA, DHEAS), estrogens (estradiol, estrone), and cognitive performance on recognition memory, perceptual identification, digit span memory, and visual attentional vigilance under both drug and placebo conditions. Multiple regression models incorporating the factors of age and body mass index (BMI) were used to ascertain the relation between sex steroids and cognitive performance. Our results demonstrated that estrogens produced a positive effect on recognition memory, while androgens produced a negative effect. This pattern reversed in perceptual identification with estrogens producing a negative effect and androgens producing a positive effect. In addition, BMI produced a negative effect on digit span memory, age produced a negative effect on perceptual identification, and androgens produced a negative effect on visual attentional vigilance. These results help, in part, to explain DHEAs complex effects on cognition. The diverse effects of sex steroids across tasks underscore the importance of identifying the specific cognitive mechanisms influenced by sex steroids and emphasizes that one should not expect sex steroids to produce homogeneous effects across cognitive tasks.