Alan Richardson-Klavehn

Remembering and knowing

Remembering and knowing are two states of awareness that, respectively, entail conscious recollection or feelings of familiarity in the absence of any recollective experiences. This chapter reviews what has been learned about remembering and knowing from experimental investigations of participants’ reports of them. Remembering and knowing have been shown to be selectively affected by many different experimental manipulations and to differ systematically in different populations. They also have distinct neural correlates. Theoretically, remembering and knowing have, respectively, been identified with episodic and semantic memory systems or with different memory processes. The responses have also been modeled using signal detection methods.

Experiences of Remembering, Knowing, and Guessing

This article presents and discusses transcripts of some 270 explanations subjects provided subsequently for recognition memory decisions that had been associated with remember, know, or guess responses at the time the recognition decisions were made. Only transcripts for remember responses included reports of recollective experiences, which seemed mostly to reflect either effortful elaborative encoding or involuntary reminding at study, especially in relation to the self. Transcripts for know responses included claims of just knowing, and of feelings of familiarity. These transcripts indicated that subjects were often quite confident of the accuracy of their decisions, compared with those for guess responses. Transcripts for decisions associated with guess responses also expressed feelings of familiarity but additionally revealed various strategies and inferences that did not directly reflect memory for studied items. The article concludes with a historical and theoretical overview of some interpretations of the states of awareness measured by these responses.

How level of processing really influences awareness in recognition memory.

Abstract In yes/no and two - alternative forced - choice recognition tests, subjects reported one of three states of awareness when selecting each target: remembering, knowing, or guessing. A remember response indicated recollection of the targets occurrence in the study list. A know response indicated the target was familiar in the experimental context but not recollected. A guess response indicated the target was selected in the absence of either remembering or knowing. In Experiments 1 and 2, level of processing influenced remember responses but not know responses and, in Experiment 3, generating versus reading similarly influenced remember but not know responses. In each experiment, when subjects reported that they were guessing they showed no ability to discriminate targets from lures. These results show that remember/know findings generalize from yes/no to two - alternative forced - choice recognition and that knowing is dissociable from guessing. The results also provide no support for the hypothesis, based on an independence model of remembering and knowing, and some other apparently contradictory results, that variables that have large effects on remembering produce opposite effects on knowing. A meta - analysis of previous level - of - processing studies yielded evidence consistent with these conclusions. In recent studies we have been developing an experiential approach to memory and awareness (Gardiner & Java, 1993a, 1993b; Java, 1994; Richardson - Klavehn & Gardiner, 1995a; Richardson - Klavehn, Gardiner, & Java, 1994, 1996). The essence of this approach is that objective measures of performance are supplemented by subjective reports of states of awareness. Put another way, third - person accounts of memory and awareness are supplemented by first - person accounts (see Velmans, 1991). This approach allows inferences theorists make indirectly about subjective states of awareness to be checked directly against subjective reports about these states of awareness, and so it raises the rather general problem of how third - person and first - person accounts of memory and awareness are to be reconciled. At its most fundamental, this problem concerns relations between (1) hypothetical constructs used in theory, such as memory systems and processes; (2) memory performance; and (3) awareness, especially subjects awareness of what they are trying to do to meet the demands of the tasks they are given, and their awareness of memory for the particular items they may encounter when carrying out such tasks (Gardiner & Java, 1993a, 1993b; Richardson - Klavehn et al., 1996). Where recognition memory is concerned, subjects awareness of memory for particular items they may encounter includes at least two discrete states: recollection, and feelings of familiarity in the absence of recollection. These states of awareness have been measured by remember and know responses (Gardiner, 1988; Tulving, 1985a), which have revealed many functional dissociations within overall recognition memory performance (for reviews, see Gardiner & Java, 1993a, 1993b; Rajaram & Roediger, in press; Richardson - Klavehn et al., 1996). In Tulvings (1983, 1985a, 1985b) theory, remembering and knowing are associated respectively with retrieval from episodic and semantic memory systems, and it is assumed that the relation between these systems is one of inclusivity (or redundancy), in the sense that representation in the episodic system necessarily entails representation in the semantic system, but not vice versa. Remember and know responses are defined exclusively, but at the level of the responses the difference between exclusivity and inclusivity is only a matter of definition. In each case, know responses would be defined as familiarity in the absence of recollection, and, with inclusivity, remember responses would be defined as recollection and familiarity, not just recollection. But subjects would still experience the same two states of awareness, regardless of whether the responses define these states as having an inclusive or an exclusive relation. …

Recognition memory and decision processes: A meta-analysis of remember, know, and guess responses

A meta-analysis of proportions of remember, know, and guess responses was carried out on observations from 86 experimental conditions in 23 different recognition memory experiments. Unlike remember and know responses, guess responses revealed no memory for the test items that elicited them. A signal detection analysis of these data showed that A′ estimates of the strength of the memory trace depended on response criteria. A′ estimates increased significantly when know responses were added to remember responses, and decreased significantly when guess responses were added to remember and know responses. It was guessing, rather than knowing, that was most strongly correlated with overall response criteria. Nor were remembering and knowing correlated significantly. These results do not support a quantitative trace strength model according to which these responses merely reflect different response criteria. They support theories that ascribe remembering and knowing to qualitatively distinct memory systems or processes.

Involuntary conscious memory and the method of opposition

Priming in an indirect test of stem completion should reflect involuntary memory, but can be accompanied by conscious awareness of the past (involuntary conscious memory) or unaccompanied by such awareness (involuntary unconscious memory). We adapted the method of opposition developed by Jacoby, Woloshyn, and Kelley (1989) to obtain a measure of stem-completion priming that should reflect only involuntary unconscious memory. Subjects completed stems with the first word coming to mind, but wrote down a different word if the word that came to mind first had been previously encountered. Facilitatory priming was expected only when involuntary unconscious influences outweighed inhibitory effects of involuntary conscious memory, or of intentional retrieval. We observed a facilitation effect for items processed graphemically at encoding, in conjunction with an inhibition effect for items processed semantically at encoding. In contrast, a standard indirect test showed similar levels of priming following graphemic and semantic encoding, whereas a direct test showed a strong advantage of semantic over graphemic encoding. We argue that the two encoding activities produced approximately equivalent involuntary influences of memory, but that items encoded semantically were associated with involuntary conscious memory to a greater extent than were items encoded graphemically. Comparing indirect and opposition test performance can provide a quantitative index of relative levels of involuntary conscious and involuntary unconscious memory.

Maintenance rehearsal affects knowing, not remembering; elaborative rehearsal affects remembering, not knowing.

In a directed-forgetting paradigm, each word in a study list was followed by a cue designating that word as eitherlearn orforget. This cue appeared after either a short or a long delay. It was assumed that a long delay would increase maintenance rehearsal of all the words, and that only the words followed by a learn cue would be rehearsed elaboratively. Moreover, because the interval between the words was constant, a short cue delay should allow more time for elaborative rehearsal. In a subsequent test, subjects maderemember orknow responses to indicate whether recognition of each word was accompanied by conscious recollection or by feelings of familiarity in the absence of conscious recollection. The hypothesis was that remembering depends on elaborative rehearsal, and knowing depends on maintenance rehearsal. In accord with this hypothesis, the learn-versus-forget designation influenced remember but not know responses, and there were more remember responses after the short cue delay; cue delay influenced know responses, regardless of word designation.

Cross-modality priming in stem completion reflects conscious memory, but not voluntary memory

A comparison of incidental and intentional stem-completion tests confirmed that cross-modality priming occurs when performance conforms completely to the retrieval intentionality criterion, indicating involuntary—not voluntary—retrieval in the incidental test. However, an on-line measure of awareness in the incidental test, and a process-dissociation analysis of the intentional test, indicated only within-modality, but not cross-modality, transfer of involuntary retrieval that is unaccompanied by memorial awareness. These results imply that conscious memory should not be equated with voluntary retrieval, and unconscious memory should not be equated with involuntary retrieval, because involuntary retrieval can be accompanied by memorial awareness.

Depth-of-processing effects on priming in stem completion: tests of the voluntary-contamination, conceptual-processing, and lexical-processing hypotheses.

Depth-of-processing effects on incidental perceptual memory tests could reflect (a) contamination by voluntary retrieval, (b) sensitivity of involuntary retrieval to prior conceptual processing, or (c) a deficit in lexical processing during graphemic study tasks that affects involuntary retrieval. The authors devised an extension of incidental test methodology--making conjunctive predictions about response times as well as response proportions--to discriminate among these alternatives. They used graphemic, phonemic, and semantic study tasks, and a word-stem completion test with incidental, intentional, and inclusion instructions. Semantic study processing was superior to phonemic study processing in the intentional and inclusion tests, but semantic and phonemic study processing produced equal priming in the incidental test, showing that priming was uncontaminated by voluntary retrieval--a conclusion reinforced by the response-time data--and that priming was insensitive to prior conceptual processing. The incidental test nevertheless showed a priming deficit following graphemic study processing, supporting the lexical-processing hypothesis. Adding a lexical decision to the 3 study tasks eliminated the priming deficit following graphemic study processing, but did not influence priming following phonemic and semantic processing. The results provide the first clear evidence that depth-of-processing effects on perceptual priming can reflect lexical processes, rather than voluntary contamination or conceptual processes.

Retrieval volition and memorial awareness in stem completion: an empirical analysis.

Facilitation in an incidental test of stem completion shows little influence of depth of processing at study, whereas facilitation in an opposition test (in which subjects give the first word coming to mind, but omit studied words) occurs following graphemic processing, but not following semantic processing. We argue that completions come to mind involuntarily in both tests. Involuntary conscious memory causes studied words to be omitted in an opposition test, but not in an incidental test, so that the difference in priming between tests is a measure of involuntary conscious memory. We obtained data consistent with this hypothesis by making overt the mental activities that occur covertly in an opposition test: (1) an on-line recognition measure in an incidental test showed a strong advantage of semantic over graphemic processing, even though depth of processing exerted little influence on priming; (2) conditionalizing on recognition failure resulted in accurate estimates of opposition performance; and (3) stems were completed much more rapidly in incidental and opposition tests than in an intentional test, in which voluntary retrieval was engaged. The data provide further evidence that retrieval volition (voluntary vs. involuntary) is dissociable from memorial state of awareness (conscious vs. unconscious). We contrast our approach with the process-dissociation approach, which confounds conscious awareness of the past with voluntary retrieval, overlooking involuntary conscious memory.

ERP evidence for successful voluntary avoidance of conscious recollection

We investigated neurocognitive processes of voluntarily avoiding conscious recollection by asking participants to either attempt to recollect (the Think condition) or to avoid recollecting (the No-Think condition) a previously exposed paired associate. Event-related potentials (ERPs) during Think and No-Think trials were separated on the basis of previous learning success versus failure. This separation yielded temporal and topographic dissociations between early ERP effects of a Think versus No-Think strategy, which were maximal between 200 and 300 ms after stimulus presentation and independent of learning status, and a later learning-specific ERP effect maximal between 500 and 800 ms after stimulus presentation. In this later time-window, Learned Think items elicited a larger late left parietal positivity than did Not Learned Think, Learned No-Think, and Not Learned No-Think items; moreover, Learned No-Think and Not Learned Think items did not differ in late left parietal positivity. Because the late left parietal positivity indexes conscious recollection, the results provide firm evidence that conscious recollection of recollectable information can be voluntarily avoided on an item-specific basis and help to clarify previous neural evidence from the Think/No-Think procedure, which could not separate item-specific from strategic processes.