CHAPTER 28 The Cognitive Neuroscience of Memory and...

P1: JzG0521857430c28 CUFX049/Zelazo 0 521 85743 0 printer: cupusbw November 6, 2006 15 :58

C H A P T E R 2 8

The Cognitive Neuroscienceof Memory and Consciousness

Scott D. Slotnick and Daniel L. Schacter

Abstract

In this chapter, we delineate the neuralactivity associated with conscious memo-ries characterized by different degrees of‘retrieval content’ (i.e. sensory/contextualdetail). Based primarily on neuroimagingevidence, we identify the neural regionsthat are associated most consistently withthe following conscious memory processes:retrieval success versus retrieval attempt,remembering versus knowing, and truerecognition versus false recognition. A num-ber of patterns emerge from the compari-son of memories with high retrieval con-tent (i.e., retrieval success, remembering,and true recognition) and memories withlow retrieval content (i.e., retrieval attempt,knowing, and false recognition). Memorieswith both high and low retrieval content areassociated with activity in the prefrontal andparietal cortex, indicating that these regionsare generally associated with retrieval. Thereis also evidence that memories with lowretrieval content are associated with activityin the prefrontal cortex to a greater degreethan memories with high retrieval con-

tent, suggesting that low retrieval contentmemories are associated with greater post-retrieval monitoring (although this activ-ity does not necessarily reflect differentialretrieval content per se). Finally, memo-ries with high retrieval content, to a greaterdegree than memories with low retrievalcontent, are associated with activity in theparietal cortex and sensory cortex (alongwith the medial temporal lobe for retrievalsuccess > attempt and remembering > kno-wing). This increased activity in sensory cor-tex (and medial temporal lobe) for mem-ories with high retrieval content indicatesthat conscious memories are constructedby reactivation of encoded item features atretrieval.

Introduction

In 1985 , Endel Tulving lamented the lackof interest in the topic of memory andconsciousness shown by past and presentmemory researchers: “One can read articleafter article on memory, or consult bookafter book, without encountering the term

807

slotnick

Text Box

Reference Slotnick, S. D., & Schacter, D. L. (2007). The cognitive neuroscience of memory and consciousness. In P.D. Zelazo, M. Moscovitch, & E. Thompson (Eds.), Cambridge handbook of consciousness (pp. 809-827). New York: Cambridge University Press.


808 the cambridge handbook of consciousness

‘consciousness.’ Such a state of affairs mustbe regarded as rather curious. One mightthink that memory should have somethingto do with remembering, and rememberingis a conscious experience (Tulving, 1985b,p. 11).”

Though Tulving provided an accurateassessment of the field at the time, the yearin which he voiced his complaint provedto be a kind of turning point in researchon memory and consciousness. Tulving’s(1985b) own article focused on the impor-tant distinction between remembering, whichinvolves specific recollections of past experi-ences, and knowing, which involves a generalsense of familiarity without specific recollec-tion, and introduced seminal techniques forexperimentally assessing these two forms ofmemory.

In a different paper published that sameyear, Tulving (1985a) argued that eachof three dissociable memory systems isuniquely associated with a particular typeof consciousness. Specifically, he contendedthat procedural memory (learning of motor,perceptual, and cognitive skills) is associatedwith anoetic or “nonknowing” consciousness,which entails simple awareness of exter-nal stimuli; semantic memory (general fac-tual knowledge) is associated with noetic or“knowing” consciousness; and episodic mem-ory (recollection of personal experiences) isassociated with autonoetic or “self-knowing”consciousness.

Finally, in 1985 Graf and Schacterintroduced the related distinction betweenimplicit and explicit memory. According toGraf and Schacter (1985), explicit memoryrefers to the conscious recollection of pre-vious experiences, as revealed by standardtests of recall and recognition that requireintentional retrieval of previously acquiredinformation. Implicit memory, by contrast,refers to non-conscious effects of past expe-riences on subsequent behavior and perfor-mance, such as priming or skill learning, thatare revealed by tests that do not requireconscious recollection of previous experi-ences (for precursors, see also Cermak, 1982 ;Moscovitch, 1984).

During the 20 years that have elapsedsince the publication of these papers, avast amount of research has been pub-lished on the distinctions that they intro-duced. For example, many cognitive stud-ies have used the techniques introducedby Tulving (1985b) to delineate the func-tional and phenomenological characteristicsof remembering and knowing (for reviewsand contrasting perspectives, see Dunn,2004 ; Gardiner, Ramponi, & Richardson-Klavehn, 2002). Likewise, cognitive studieshave also explored numerous aspects of therelation between implicit and explicit formsof memory (for reviews, see Roediger &McDermott, 1993 ; Schacter, 1987; Schacter& Curran, 2000).

Although purely cognitive studies haveplayed a significant role in advancing ourunderstanding of memory and conscious-ness, cognitive neuroscience studies – whichattempt to elucidate the nature of, andrelations between, the brain systems andprocesses that support various forms ofmemory – have also been critically impor-tant. Indeed, much of the impetus for thedistinction between implicit and explicitmemory was provided initially by neu-ropsychological studies of amnesic patients,who exhibit severe impairment of explicitmemory for previous experiences as aresult of damage to the hippocampus andrelated structures in the medial temporallobe (MTL; e.g., Moscovitch, Vriezen, &Goshen-Gottstein, 1993 ; Nadel & Moscov-itch, 1997; Squire, 1992 ; Squire, Stark, &Clark, 2004). Nonetheless, it has beendemonstrated repeatedly that conditionsexist in which amnesics can exhibit robustand sometimes normal implicit memory foraspects of prior experiences, as exempli-fied by such phenomena as preserved prim-ing and skill learning (for recent reviews,see Gooding, Mayes, & van Eijk, 2000;Schacter, Dobbins & Schnyer, 2004). Stud-ies of other neuropsychological syndromeshave likewise revealed dissociations betweenimplicit and explicit forms of percep-tion, language, and related cognitive andmotor processes (e.g., Goodale & Westwood,


the cognitive neuroscience of memory and consciousness 809

2004 ; Guzeldere, Flanagan, & Hardcastle,2000; Kohler & Moscovitch, 1997; Schacter,McAndrews, & Moscovitch, 1988; Warring-ton & Weiskrantz, 1974 ; Young, 1994).

Although neuropsychological studieshave been crucial to advancing our under-standing of the relation between memoryand consciousness, during the past decadecognitive neuroscience analyses havefocused increasingly on research usingfunctional neuroimaging techniques, suchas positron emission tomography (PET)and functional magnetic resonance imaging(fMRI). A vast amount of neuroimagingresearch has been published, and much ofit is well beyond the scope of this chapter.However, we believe that several linesof research concerned with elucidatingthe neural correlates of explicit memoryprocesses do provide useful insights intothe cognitive neuroscience of memory andconsciousness (for reviews of neuroimagingstudies concerning the neural substratesassociated with implicit memory, seeHenson, 2003 ; Schacter & Buckner, 1998;Schacter et al., 2004 ; Wiggs & Martin, 1998).

Many neuroimaging studies of explicitmemory have used a recognition paradigm,where items such as words or objects arestudied, and then on a subsequent test,these old items are randomly intermixedwith new items, and participants decidewhether each item is “old” or “new.” Itemrecognition has been associated most consis-tently with activity in three neural regions:(1) prefrontal cortex (anterior and dorso-lateral), (2) parietal cortex, and (3) theMTL (for reviews, see Buckner & Schacter,2004 ; Buckner & Wheeler, 2001; Slotnick,Moo, Segal, & Hart, 2003 ; Tulving, Kapur,Craik, Moscovitch, & Houle, 1994). Thefunctional role(s) subserved by each of theseregions is currently an active area of inves-tigation, as we discuss later in this chap-ter. At a very general level, prefrontal cor-tex has been associated with the controlof retrieval (e.g., increases in activity thatcorrelate with retrieval demands; Velanovaet al., 2003 : Wheeler & Buckner, 2003)and in addition has been associated with

post-retrieval monitoring (for a review, seeSchacter & Slotnick, 2004). Parietal cortex,particularly in Brodmann Area (BA) 39/40,has recently been associated with the ten-dency to make “old” responses (Velanovaet al., 2003 ; Wheeler & Buckner, 2003).The MTL, as mentioned previously, is nec-essary for explicit memory, with the hip-pocampus proper possibly serving the roleof binding together information from dis-parate cortical regions (Squire, 1992). Thatis, the hippocampus may serve a centralrole in combining disparate features to con-struct a unitary memory (e.g., Moscovitch,1994 ; Schacter, Norman, & Koutstaal, 1998a;Squire, 1992).

Providing support for this constructiveview of memory, recent neuroimaging evi-dence indicates that explicit memory evokesactivity in the appropriate domain-specificprocessing regions (i.e., retrieval-relatedreactivation of processing regions associ-ated with memorial encoding). Specifically,memory for actions activates motor pro-cessing regions (Nyberg, Petersson, Nilsson,Sandblom, Åberg, & Ingvar, 2001), mem-ory for sounds activates auditory process-ing regions (Nyberg, Habib, McIntosh, &Tulving, 2000; Wheeler & Buckner, 20003 ;Wheeler, Petersen, & Buckner, 2003), mem-ory for odors activates olfactory process-ing regions (Gottfried, Smith, Rugg, &Dolan, 2004), and memory for visual stimuli(e.g., shapes or objects) activates occipital-temporal regions in the ventral visual pro-cessing stream (Moscovitch, Kapur, Kohler,& Houle, 1995 ; Slotnick et al., 2003 ; Vaidya,Zhao, Desmond, & Gabrieli, 2002 ; Wheeler& Buckner, 2003 ; Wheeler et al., 2000).Such domain-specific sensory reactivationis typically taken as evidence for the con-scious re-experiencing of sensory attributesof items from the study episode.

This chapter considers three lines ofresearch that have examined aspects of thismemory-related sensory/contextual activityand the associated subjective experience(or phenomenal consciousness; see Block,1995 ; also referred to as the ‘contents ofconsciousness’ or more simply ‘retrieval



content’; Wheeler & Buckner, 2003). Webelieve that each of these lines of researchhas provided new information regarding theneural underpinnings of conscious experi-ences of remembering. First, we considerattempts to separate explicit retrieval intoseparate components that can be groupedbroadly into two categories: retrieval suc-cess, which involves the recovery of informa-tion presented during a prior study episode,and retrieval attempt, which refers to strate-gic processes involved in explicit retrievalthat operate even when recovery is not suc-cessful. Evidence from recent neuroimag-ing studies points toward different neu-ral substrates subserving these two broadclasses of conscious memory processes. Sec-ond, we discuss imaging experiments con-cerned with the distinction between remem-bering and knowing (Tulving, 1985b) thatexamine how neural activity correlates withdiffering degrees or types of conscious expe-riences. Third, we consider recent work con-cerned with delineating the neural substratesof true versus false memories, where therole of sensory reactivation in the consciousexperience of remembering has been exam-ined in the context of questions concerningthe accuracy of explicit retrieval. Althoughwe focus on neuroimaging studies in eachof the three lines of research, we also dis-cuss, when relevant, complementary datafrom neuropsychological studies of brain-damaged patients.

Neural Substrates of Retrieval SuccessVersus Attempt

When a brain region shows changes in activ-ity during explicit retrieval, the changes arenot necessarily associated with the consciousexperience of successfully recovering pre-viously studied information. Such changescould instead reflect, entirely or in part,conscious processes involved in the deploy-ment of attention or effort when individ-uals attempt to retrieve the target mate-rial, independent of whether retrieval issuccessful. Once neuroimaging studies ofepisodic memory had demonstrated that

explicit retrieval is accompanied by activa-tion in specific brain regions – most promi-nently, regions within prefrontal cortex, butalso within the MTL (e.g., Schacter, Alpert,Savage, Rauch, & Albert, 1996a; Squire et al.,1992 ; Tulving et al., 1994) – it became impor-tant to specify further the nature of theobserved activity. Early PET studies adoptedtwo main experimental approaches to thisissue: (1) producing high and low levels ofsuccessful retrieval by manipulating studyconditions and (2) manipulating the num-ber of previously studied items that appearduring a particular test. We briefly sum-marize studies that have used each type ofapproach.

In a PET study by Schacter et al. (1996a),subjects studied some words four timesand judged the number of meanings associ-ated with each item (high-recall condition);they studied other words once and judgedthe number of t-junctions in each item(low-recall condition). Subjects were thenscanned during an explicit retrieval task(stem-cued recall, e.g., tab for table), withseparate scans for high-recall words and low-recall words. The logic underlying the exper-iment is that regions that are selectivelyactivated during the high-recall condition,when subjects correctly recall a large pro-portion of the study list words, are pref-erentially associated with successful con-scious recollection; by contrast, regions thatare activated during the low-recall condi-tion, when subjects retrieve only a fewstudy lists words, are preferentially asso-ciated with retrieval attempt. Analysis ofPET data revealed blood flow increasesin the hippocampal formation during thehigh-recall but not the low-recall condi-tion, and a significant difference betweenthe two conditions, thereby suggesting thathippocampal activation is associated withsome aspect of the successful consciousrecall of a previously studied word, ratherthan retrieval attempt (see also, Nyberg,McIntosh, Houle, Nilsson, & Tulving, 1996;Rugg, Fletcher, Frith, Frackowiak, & Dolan,1997). Schacter et al. (1996a) also found thatanterior/dorsolateral areas within prefrontalcortex were preferentially activated in the



low-recall condition, thus raising the possi-bility that blood flow increases in anteriorprefrontal cortex during stem-cued recall areassociated with retrieval orientation effects(cf., Nyberg et al., 1995). Such findingsaccord well with theoretical proposals thathave linked the MTL/hippocampal regionwith the automatic recovery of stored infor-mation and regions within prefrontal cor-tex with strategic aspects of retrieval (e.g.,Moscovitch, 1994).

In a related PET study by Rugg et al.(1997), subjects studied word lists andeither generated sentences for each word(deep encoding) or made judgments aboutthe letters in each word (shallow encod-ing). Following each type of encoding task,they were given either an old-new recog-nition test (intentional retrieval) or an ani-mate/inanimate decision task (unintentionalretrieval). Deep encoding produced moreaccurate memory on the intentional retrievaltask. Performance was at ceiling levels onthe unintentional task, but subjects reportedspontaneously noticing that test words camefrom the study list more often after deepthan shallow encoding, perhaps providing arough index of unintentional conscious rec-ollection. There was greater activation inleft MTL areas after deep encoding thanafter shallow encoding during both inten-tional and unintentional retrieval. Thus,these data suggest that hippocampal activityduring retrieval is observed with high lev-els of conscious recollection, regardless ofwhether subjects voluntarily try to remem-ber the study list items. By contrast, therewas greater right prefrontal activation dur-ing intentional retrieval than during uninten-tional retrieval after both deep and shallowencoding.

Several PET studies have attemptedto separate retrieval success and retrievalattempt by manipulating the proportion ofold items presented to subjects during a par-ticular scan. The reasoning here is that pre-senting large numbers of old items duringa particular scan will produce more suc-cessful retrieval than presenting only a fewold items. In general, these studies focusedon issues concerning the characterization of

retrieval-related activation observed in rightanterior prefrontal cortex. However, resultsfrom these studies were inconclusive, withsome evidence linking right prefrontal acti-vation with retrieval attempt (e.g., Kapur,Craik, Jones, Brown, Houle, & Tulving, 1995)and others reporting evidence for retrievalsuccess effects (e.g., Rugg, Fletcher, Frith,Frackowiak, & Dolan, 1996; for an attemptto reconcile some of these conflicting earlyresults, see Wagner, Desmond, Glover, &Gabrieli, 1998).

The development of event-related fMRIin the late 1990s provided a more directmeans of examining brain activations asso-ciated with retrieval success and retrievalattempt. The PET studies reviewed aboveused blocked designs in which items fromdifferent conditions were presented in sepa-rate blocks, and data concerning brain activ-ity were collapsed across subjects’ behavioralresponses. Taking advantage of the superiortemporal resolution of fMRI compared withPET, event-related fMRI allows intermix-ing of items from different conditions and,more importantly, permits analysis of brainactivity conditional on subjects’ responses(Dale & Buckner, 1997). Thus, for exam-ple, in a recognition memory task, “old” and“new” responses can be analyzed separatelyfor old and new items. Thus, retrieval suc-cess should be maximal when subjects make“old” responses to old items (hits) and mini-mized when subjects make “new” responsesto new items (correct rejections).

A number of studies have used event-related fMRI to examine retrieval successversus retrieval attempt with an old-newrecognition test for previously studied itemsintermixed with new, non-studied items.The critical comparison involves a contrastof brain activity during hits and correctrejections. A number of early studies usingthese procedures failed to reveal clear evi-dence of brain activation differences dur-ing hit versus correct rejection trials (e.g.,Buckner, Koutstaal, Schacter, Dale, Rotte, &Rosen, 1998; Schacter, Buckner, Koutstaal,Dale, & Rosen, 1997a). However, as dis-cussed by Konishi, Wheeler, Donaldson, andBuckner (2000), these failures to observe



evidence for retrieval success effects likelyreflected technical limitations of early event-related fMRI procedures, such as low statis-tical power resulting from the use of longintertrial intervals and a correspondingly lownumber of items per experimental condi-tion. Consistent with this possibility, studiesusing more powerful event-related meth-ods revealed evidence for greater activa-tion during hits than during correct rejec-tions in a number of cortical regions, mostconsistently in prefrontal and parietal cor-tices (e.g., Konishi et al., 2000; McDermott,Jones, Petersen, Lageman, & Rodeiger, 2000;Nolde, Johnson, & D’Esposito, 1998) but alsoin the MTL (as is discussed below).

The results of the foregoing studies areconsistent with the conclusion that regionswithin prefrontal and parietal cortices arespecifically related to successful consciousrecollection of some aspects of a pre-vious experience. However, this conclu-sion depends critically on the assumptionthat comparing hits with correct rejectionsisolates successful retrieval. Although theassumption appears straightforward enough,the comparison between hits and correctrejections necessarily confounds subjects’responses (“old” or “new”) and item type(old and new). It is conceivable, therefore,that hit greater than correct rejection-relatedbrain activations do not exclusively reflectdifferences in conscious experience relatedto subjects’ responses (e.g., calling an item“old” versus calling it “new”), but insteadreflect differences in responses to old versusnew items, irrespective of subjects’ experi-ences. For example, differential responses toold and new items might reflect the occur-rence of priming or related processes thatcan occur independently of conscious mem-ory (Schacter & Buckner, 1998; Schacteret al., 2004 ; Wiggs & Martin, 1998).

Dobbins, Rice, Wagner, and Schacter(2003) approached this issue within the con-text of the theoretical distinction betweenrecollection (i.e., memory for the contextualdetails of a prior encounter) and familiar-ity (i.e., recognition without recollection ofcontextual details). Both recollection andfamiliarity can, in principle, operate on a par-

ticular memory test (for a contrasting view,see Slotnick & Dodson, 2005). Moreover,each of the two processes are potentially sep-arable into the two components on which wehave focused in this section of the chapter,retrieval success and retrieval attempt Dob-bins et al. used the closely related phrase,retrieval orientation to refer to the extent towhich subjects recruit each process duringparticular retrieval tasks). With respect tothe issues raised in the preceding paragraph,Dobbins et al. (2003) noted that, whenpresented with new items, subjects couldrely entirely on familiarity-based processes,rejecting new items when they are not famil-iar, and might not even attempt to engage inrecollection-based retrieval. Thus, it is con-ceivable that previous findings of prefrontaland parietal activations associated with hitsgreater than correct rejections might reflectattempted recollective retrieval, rather thansuccessful conscious recollection.

To address this issue, Dobbins et al.(2003) used a different type of experimen-tal design in which all items had been pre-sented previously, and task demands werevaried to require differential reliance onrecollection and familiarity. Prior to scan-ning, subjects were presented visually witha long list of nouns, and then they alter-nated between two semantic encoding tasks(pleasant/unpleasant and concrete/abstractjudgments). Subjects were then scannedduring two different two-alternative forced-choice tests: a source memory test and arecency memory test. During source mem-ory, subjects selected the member of thepair previously associated with a particu-lar encoding task; that is, they had to rec-ollect some type of detail associated withthe particular encoding judgment performedearlier. In contrast, the recency judgmentrequired subjects to select the most recentlyencountered item of the pair, regardlessof how it had been encoded. The sourcememory test is assumed to rely on recol-lection, whereas recency decisions can relyon a familiarity signal. Furthermore, suc-cessful and unsuccessful trials within eachretrieval task were contrasted to determinewhether retrieval success effects occurred in



Table 2 8.1. Neural regions associated most consistently with conscious memory processes

Region Prefrontal Cortex Parietal Cortex MTL Sensory Cortex

Retrieval Success & Attempt X XRetrieval Success > Attempt XRetrieval Attempt > Success XRemembering & Knowing X XRemembering > Knowing X X XKnowing > Remembering XTrue & False Recognition X X XTrue > False Recognition X XFalse > True Recognition X

Note: Regions of common (&) and differential (>) activity were identified via review of the neuroimagingliterature.

overlapping or dissimilar brain regions com-pared to those associated with each retrievalorientation.

Results revealed left lateral prefrontaland parietal activations that distinguishedattempted source recollection from judg-ments of relative recency; these retrievalattempt or orientation effects were largelyindependent of retrieval success. Impor-tantly, these activations occurred largely inthe same left prefrontal and parietal regionsthat had been previously identified withretrieval success. Because these regions werenot associated with successful retrieval in theDobbins et al. (2003) design, which con-trolled for old-new item differences presentin previous studies, it is plausible that theprefrontal and parietal activations in ear-lier studies reflect attempted, rather thansuccessful, conscious recollection (for fur-ther relevant analyses, see Dobbins, Foley,Schacter, & Wagner, 2002). In contrast,Dobbins et al. (2003) found that MTL struc-tures (hippocampus and parahippocampalgyrus) were differentially more active dur-ing successful recollection, showing similarlyreduced responses during failed source rec-ollection and judgments of recency. Thesefindings complement previous data link-ing MTL regions with successful consciousrecollection (e.g., Maril, Simons, Schwartz,Mitchell, & Schacter, 2003 ; Rugg et al., 1997;Schacter et al., 1996a; but see, Buckner et al.,1998; Rugg, Henson, & Robb, 2003), as wellas other results from related paradigms con-sidered later in the chapter.

Kahn, Davachi, and Wagner (2004) haveprovided converging evidence on the fore-going conclusions using an old-new recog-nition test for previously presented wordsand new words in which subjects also madea source memory judgment (whether theyhad read a word at study or imagineda scene related to the word). They con-cluded that left prefrontal/parietal regionsare related to attempted recollection ofsource information, but not to success-ful recollection of that information; bycontrast, MTL activation (in the parahip-pocampal region) was related to successfulsource recollection. Importantly, Kahn et al.(2004) also provided evidence indicatingthat left frontal/parietal activity is related tofamiliarity-based retrieval success. Thus, thegeneral distinction between retrieval successand retrieval attempt (or orientation) maybe too coarse to prove useful theoretically.Instead, it may be necessary to specify aparticular form of retrieval to make senseof neuroimaging data concerning the neu-ral correlates of successful and attemptedretrieval (e.g., recollection versus familiarity,or remembering versus knowing, which areconsidered in detail below).

The results summarized in this sec-tion indicate that neuroimaging studiesare beginning to dissociate componentsof conscious retrieval that are related toactivity in particular brain regions. In par-ticular, three patterns of results can beobserved (Table 28.1, left column). First,both retrieval success and retrieval attempt



were associated with activity in the pre-frontal cortex and parietal cortex. Sec-ond, retrieval success to a greater degreethan retrieval attempt was associated withactivity in the MTL. Third, there is someevidence that retrieval attempt may be asso-ciated with greater prefrontal cortex activ-ity than with retrieval success. As researchin this area progresses, increasingly finer dis-tinctions will be made regarding the neuralsubstrates associated with particular aspectsof conscious memorial experience. Of note,the fact that retrieval success – which canbe assumed to reflect greater retrieval con-tent than retrieval attempt – is preferentiallyassociated with the MTL suggests this regionplays a role during conscious remembering.

Neural Activity Associated withRemembering and Knowing

We reviewed research in the preceding sec-tion that attempts to dissociate recollec-tion and familiarity by manipulating taskdemands. However, as noted earlier, rec-ollection and familiarity can be assesseddirectly by asking participants about theirsubjective experiences during a memorytask; that is, to classify “old” responses basedon the associated memorial experienceof remembering or knowing. Rememberresponses indicate recollection of specificcontextual detail associated with a previousexperience, whereas know responses referto a sense of familiarity without contextualdetail (Tulving, 1985b). Comparing the neu-ral activity associated with remember andknow responses is thus expected to pro-vide additional insight into the substrates ofspecific types of conscious experiences con-sidered under the general rubric of explicitmemory.

In an event-related fMRI study ofremembering and knowing (Henson, Rugg,Shallice, Josephs, & Dolan, 1999), subjectsfirst studied a list of words. For each itemon a subsequent recognition test, partici-pants responded “remember” or “know” toitems they judged to be “old” and other-wise responded “new.” Both correct remem-

ber responses and correct know responses,relative to new-correct rejections, were asso-ciated with activity in prefrontal cortex (dor-solateral and medial) and medial parietalcortex (precuneus). Relative to new-correctrejections, remember judgments (but notknow judgments) were also associated withadditional activity in parietal cortex (supe-rior parietal lobule and inferior parietal lob-ule) and the MTL (parahippocampal gyrus).The direct contrast between remember andknow responses complemented these resultsby showing activity in the parietal cortex(superior parietal lobule and inferior pari-etal lobule). Although the MTL activationdid not survive this direct contrast, it shouldbe noted that only remember responses (ver-sus new-correct rejections) evoked activityin the MTL, providing some indication thatthis region is preferentially associated withremembering. The reverse contrast betweenknow and remember was associated withactivity in the prefrontal cortex (dorsolat-eral and medial), albeit to a less extensivedegree than that associated with remem-ber or know responses (versus new-correctrejections), and medial parietal cortex(precuneus).

A subsequent event-related fMRI recog-nition memory study, also using words asstudy and test materials, replicated andextended the previous pattern of resultsby focusing on differential neural activ-ity associated with remember and knowresponses (Eldrige, Knowlton, Furmanski,Bookheimer, & Engel, 2000). The contrastbetween remember and know responses wasassociated with activity in the prefrontal cor-tex (dorsolateral), the parietal cortex (infe-rior parietal lobule), the MTL (both hip-pocampus and parahippocampal gyrus), andthe fusiform gyrus. This fusiform gyrus activ-ity (coupled with the MTL activity) likelyreflects a greater degree of sensory reacti-vation associated with remember as com-pared to know responses. The know greaterthan remember contrast was associated witha distinct region in the (anterior) prefrontalcortex.

In an event-related fMRI remember-know paradigm conducted by Wheeler &



Buckner (2004 ; adapted from a paradigmoriginally designed to investigate memory-related domain specific sensory reactivation;see Wheeler & Buckner, 2003 ; Wheeleret al., 2000), words were paired with eithersounds or pictures at study. On the sub-sequent recognition test, old words (thosepreviously paired with pictures, the onlytype of old items considered in the analysis)and new words were presented. Participantsresponded “remember,” “know,” or “new.”Correct remember and know responses wereassociated with the same degree of activ-ity in one subregion of the parietal cor-tex, whereas another subregion was asso-ciated with greater activity for rememberthan know responses (both regions were inthe inferior parietal lobule). The contrastof remember versus know was also asso-ciated with activity in the prefrontal cor-tex (medial), the MTL (hippocampus), andthe fusiform cortex. Because subjects wereremembering previously studied pictures,the fusiform cortex activity in this studylikely reflects memory-related sensory reac-tivation. Know versus remember responseswere associated with activity in the (dorso-lateral) prefrontal cortex.

Across the studies reviewed, a numberof patterns emerge (Table 28.1, middle col-umn). First, remembering and knowing, ascompared to new-correct rejections, wereassociated with activity in prefrontal cor-tex and parietal cortex. Second, remem-bering evoked greater activity than know-ing most consistently in parietal cortex, theMTL, and sensory cortex. Third, know-ing evoked greater activity than remember-ing in the prefrontal cortex. These findingsare largely consistent with evidence fromremember-know ERP studies, which haveshown greater remember than know activityat parietal scalp electrodes (approximately400–800 ms from stimulus onset) in addi-tion to similar remember and know activ-ity (both greater than new) at frontal scalpelectrodes (approximately 1000–1600 msfrom stimulus onset; Curran, 2004 ; Duarte,Ranganath, Winward, Hayward, & Knight,2004 ; Duzel, Yonelinas, Mangun, Heinze, &Tulving, 1997; Smith, 1993 ; Trott, Friedman,

Ritter, Fabiani, & Snodgrass, 1999; see alsoWilding & Rugg, 1996).

Neuropsychological evidence convergesto some extent with the neuroimaging (andERP) findings. In a study by Knowlton andSquire (1995), amnesic patients with MTLdamage studied a list of unrelated words andthen made “remember,” “know,” or “new”judgments on a subsequent recognition test.Amnesic patients showed a large decrementin remember responses as compared to con-trol participants and a more modest but stillsignificant decline in know responses (at a10-minute delay between study and test).Subsequent studies showed a similar patternof results, where amnesic patients showeda severe impairment in remembering alongwith more modest trends for impair-ments in knowing (Schacter, Verfaellie, &Anes, 1997b; Schacter, Verfaellie, & Pradere,1996c; Yonelinas, Kroll, Dobbins, Lazzara, &Knight, 1998) and unilateral temporal lobec-tomy patients have been shown to only beimpaired in remembering (Moscovitch &McAndrews, 2002).

Although these group studies includepatients with damage to a variety ofMTL structures, more recent studies haveattempted to distinguish between patientswith damage restricted to the hippocam-pal formation and those with more exten-sive MTL damage. Yonelinas et al. (2002)found deficits in both remembering andknowing in patients with damage to boththe hippocampus and surrounding parahip-pocampal gyrus. By contrast, they foundimpairments of remembering – but notknowing – in patients who developed mem-ory deficits as a result of hypoxia, whichis known to produce damage restrictedto the hippocampal formation in patientswhose deficits are restricted to memory(see Yonelinas et al., 2002). Note, how-ever, that anatomical information was notprovided concerning the precise lesion sitesof the hypoxic patients included in theYonelinas et al. (2002) study, so the anatom-ical implications of these findings are uncer-tain. Manns, Hopkins, Reed, Kitchener,and Squire (2003) reported significant andcomparable deficits of remembering and



knowing in amnesics with restricted hip-pocampal damage, compared with controls.By contrast, a recent case study of patientB.E., who has selective bilateral hippocam-pal damage, suggests that damage to the hip-pocampal region alone can result in a specificdeficit in remembering with relative spar-ing of knowing (Holdstock, Mayes, Gong,Roberts, & Kapur, 2005 ; see also Holdstocket al., 2002).

In summary, although all neuropsycho-logical studies of amnesic patients with MTLor restricted hippocampal damage revealsevere deficits of remembering, the evi-dence is mixed concerning the role of theMTL generally, and of hippocampus specif-ically, in knowing. Given current controver-sies in the interpretation of remember/knowdata (cf., Dunn, 2004 ; Gardiner et al.,2002 ; Rotello, Macmillan, & Reeder, 2004 ;Wixted & Stretch, 2004), it is perhaps notentirely surprising that clarification of therelative status of remembering and know-ing in amnesic patients will require furtherstudy. Nonetheless, these neuropsychologi-cal studies complement imaging data by pro-viding evidence that the MTL is criticallyinvolved in remembering, which reflects arich form of conscious recollective experi-ence (see discussion in Moscovitch, 1995 ,2000).

Sensory Reactivation in Trueand False Memory

In the type of recognition memory para-digms we have considered thus far, analy-ses of cognitive and brain activity typicallyfocus on accurate responses: “old” responsesto studied items (old-hits) or “new” respon-ses to non-studied items (new correct rejec-tions). False alarms to new items in suchparadigms are usually too few to allow mean-ingful analysis. However, cognitive psycho-logists have developed a number of para-digms that yield much larger numbers offalse alarms, thus allowing comparison of thecognitive and neural properties of true mem-ories and false memories (for a recent review,see Schacter & Slotnick, 2004). For example,

Roediger and McDermott (1995), extend-ing earlier work by Deese (1959), reporteda paradigm that produces extremely highlevels of false memories (now commonlyreferred to as the DRM paradigm). In theDRM paradigm, participants are presentedwith lists of associated words (e.g., fly, bug,insect, web, and other related words) thatare related to a non-studied lure item (e.g.,spider). Roediger and McDermott’s (1995)study showed that subjects falsely recog-nized a high proportion of these relatedlure items and often claimed to specifically“remember” (versus “know”) that the lureitems appeared on the study list. A simi-lar paradigm has been used to study falsememory for visual shapes, in which subjectsstudy physically related shapes and later pro-duce high levels of false alarms to perceptu-ally similar shapes that had not been pre-viously seen (Koutstaal, Schacter, Verfaellie,Brenner, & Jackson, 1999). From the per-spective of the present chapter, the develop-ment of such paradigms allows us to exam-ine the similarities and differences in theneural correlates of conscious experiencesassociated with accurate and inaccuratememories.

In the first neuroimaging study to com-pare true and false memory (Schacter,Reiman, Curran, Yun, Bandy, McDermott, &Roediger, 1996b), participants heard DRM-associated lists followed by a recognition test(consisting of studied/old words, lures/non-studied related words, and non-studied unre-lated/new words). Each item type at test(old, related, and new), in addition to abaseline passive fixation condition, was pre-sented in a separate PET scanning block.Both true and false recognition, comparedto baseline fixation, were associated withactivity that included anterior/dorsolateralprefrontal cortex (BA 10/46), precuneus(medial parietal cortex), and parahippocam-pal gyrus (within the MTL). The directcontrast between true and false recognitionwas associated with activity in a left tempo-ral parietal cortex, a region linked to audi-tory processing. This latter finding can betaken as evidence for greater sensory reac-tivation (i.e., auditory cortex activation dur-



ing memory for previously spoken words)during true memory as compared to falsememory.

A similar experiment was conductedusing event-related fMRI (Schacter et al.1997a), where event types during the recog-nition test were intermixed, and it showedthat both true and false recognition (com-pared to baseline fixation) were associatedwith similar patterns of activity includingthe prefrontal cortex, parietal cortex, andthe MTL. However, unlike the previousstudy, the true greater than false recog-nition contrast did not reveal activity inany region. At the same time, an ERPexperiment suggested that true greater thanfalse recognition-related activity could beattributed to differences in blocked ver-sus event-related designs (Johnson, Nolde,Mather, Kounios, Schacter, & Curran, 1997).Although this latter finding suggests com-mon neural substrates underlying true andfalse recognition, subsequent fMRI studieshave shown more convincing evidence oftrue/false differences in brain activity andhave begun to elucidate the nature of thatactivity.

In an event-related fMRI study conductedby Cabeza, Rao, Wagner, Mayer, and Schac-ter (2001), a male or female (on videotape,a relatively rich contextual environment)spoke words from DRM lists of seman-tic associates or similar categorized lists(e.g., onion, cucumber, and pea are exem-plars of the category ‘vegetable’). Partici-pants were instructed to remember eachword and whether it was spoken by themale or female. At test, old words, relatedwords (non-presented associates and cate-gories), or new words were presented, andparticipants made an old-new recognitiondecision. True recognition and false recog-nition, as compared to new items, wereassociated with activity in the dorsolateralprefrontal cortex, parietal cortex (medialand inferior parietal lobule), and the MTL,specifically the hippocampus. The contrastof false recognition versus true recognitionwas associated with greater activity in ven-tromedial prefrontal cortex. True recogni-tion versus false recognition was associated

with greater activity in the parietal cor-tex (inferior parietal lobule) and anotherregion of the MTL, the parahippocampalgyrus. This parahippocampal gyrus activity(which has also been reported in a true/falserecognition paradigm by Okado and Stark,2003) may reflect greater true than falserecognition-related contextual reactivation(possibly reflecting memory for the video-taped speakers), because this region has beenassociated with processing visual context(Bar & Aminoff, 2003 ; Epstein & Kanwisher,1998).

Slotnick and Schacter (2004) used event-related fMRI to investigate the neural sub-strates of true and false recognition forabstract visual shapes. During the studyphase, participants viewed sets of exemplarshapes that were similar to a non-presentedprototype shape (analogous to DRM wordlists). At test, old shapes, related shapes(e.g., non-studied but similar shapes), ornew shapes were presented, and participantsmade an old-new recognition judgment.True recognition and false recognition, ascompared to new-correct rejections (i.e.,responding “new” to unrelated new items),were associated with activity in the ante-rior/dorsolateral and medial prefrontal cor-tex, parietal cortex (superior parietal lobule,inferior parietal lobule, and precuneus), theMTL (hippocampus), and ventral occipital-temporal visual processing regions (BA17/18/19/37). Although the true greater thanfalse recognition contrast and the reversecontrast were each associated with activityin different regions of the dorsolateral pre-frontal cortex and parietal cortex (includ-ing precuneus and inferior parietal lobule),only the true greater than false recogni-tion contrast was associated with activity invisual processing regions, specifically in BA17 and BA 18. These latter regions may reflectgreater visual sensory reactivation associ-ated with true recognition as compared tofalse recognition. The results of ERP stud-ies investigating the neural basis of trueand false visual spatial memory are consis-tent with these findings (Fabiani, Stadler, &Wessels, 2000; Gratton, Corballis, & Jain,1997).



Thus, across a number of studies, cor-tical activity that is likely associated withsensory/contextual processing is greater fortrue than false recognition. Such differen-tial activity might be taken as reflectingconscious recollection of sensory/contextualdetails that are remembered during truebut not false recognition, an idea thathas received some support from behavioralstudies (e.g., Mather, Henkel, & Johnson,1997; Norman & Schacter, 1997). Slotnickand Schacter (2004) attempted to iden-tify visual processing regions that reflectconscious memory by contrasting old-hits(responding “old” to old items) and old-misses (responding “new” to old items; seealso, Wheeler & Buckner, 2003 , 2004). Ifactivity within such regions reflects con-scious memory, then brain activity should begreater for old-hits than for old-misses. Con-versely, regions that reflect non-consciousmemory should respond equivalently dur-ing old-hits and old-misses, but in both casesto a greater degree than during new cor-rect rejections (Rugg, Mark, Walla, Schlo-erscheidt, Birch, & Allan, 1998). Slotnickand Schacter found that conscious mem-ory, as identified by the old-hits greaterthan old-misses contrast, was associated withactivity in later visual processing regions(BA 19/37), whereas non-conscious mem-ory – identified by contrasting both old-hits and old-misses each with new-correctrejections – was associated with activity inearlier visual processing regions (BA 17/18).The same functional-anatomic dichotomywas also observed in a follow-up experiment.Both the true greater than false recogni-tion activity and the old-hits and old-missesgreater than new-correct rejections resultsprovide convergent evidence that activity inBA17/18 reflects nonconscious memory, atleast in the paradigm used by Slotnick andSchacter.

We have proposed that this early visualarea activity may reflect the influence ofpriming, which as noted earlier is a non-conscious form of memory (Slotnick &Schacter, 2004). One possible problem withthis idea is that neuroimaging studies haveoften shown that priming is associated withdecreases in activity following repetition of

familiar items, such as words and pictures ofcommon objects (for reviews, see Henson,2003 ; Schacter & Buckner, 1998; Schacteret al., 2004 ; Wiggs & Martin, 1998). How-ever, it has also been found that repetition ofnovel (or masked) faces, objects, or shapeselicits increases in regional brain activity(e.g., Henson, Shallice, & Dolan, 2000;James, Humphrey, Gati, Menon, & Goodale,2000; Schacter, Reiman, Uecker, Polster,Yun, & Cooper, 1995 ; Uecker, Reiman,Schacter, Polster, Cooper, Yun, & Chen,1997). Because novel abstract shapes servedas materials in the Slotnick and Schac-ter study, the priming hypothesis remainsviable. Note also that Slotnick and Schac-ter used the identical shapes at study andtest, thus allowing for repetition primingto occur (see also, Slotnick et al., 2003),whereas other visual memory studies thatfailed to observe memory-related activityin early visual regions (but found memory-related activity in late visual regions BA19/37) did not use the identical stimuli atstudy and test, thus reducing the possibilityof repetition priming (Vaidya et al., 2002 ;Wheeler & Buckner, 2003 ; Wheeler et al.,2000).

The overall pattern of results thus sug-gests that memory-related activity in BA17/18 may be non-conscious. This observa-tion has ramifications for interpreting activ-ity associated with performance on explicitmemory tests. Typically, activation associ-ated with explicit memory tests such asold-new recognition is attributed to con-scious processing; however, the present anal-ysis indicates this is not always the case.Rather, additional analyses (such as the old-hits versus old-misses contrast) appear nec-essary to investigate and characterize thenature of activity associated with explicitmemory.

The idea that activity in early visualregions that distinguishes between trueand false recognition reflects non-consciousmemory processes may also help explainwhy false recognition occurs at high lev-els, even though brain activity can distin-guish between true and false memories. Ifthe activity in early visual regions that dis-tinguished between true and false memories



had been consciously accessible, participantsshould have used this activity to avoid mak-ing false alarms to the related shapes. Thefact that there was nonetheless a high rate offalse recognition makes sense if the activitywithin these regions reflects a non-consciousform of memory.

Across the true and false recognition stud-ies reviewed, a number of patterns can beobserved (Table 28.1, right column). First,consistent with regions previously associatedwith explicit retrieval, true and false recog-nition (versus new-correct rejections) wereboth associated with activity in prefrontalcortex, parietal cortex, and the MTL (mostconsistently within the hippocampus). Sec-ond, true greater than false recognition wasassociated with activity in the parietal cortexand sensory/contextual processing regions.Third, false greater than true recognitionwas associated with activity in the prefrontalcortex (distinct from the commonly activeregions).

Neuropsychological studies have pro-vided convergent evidence, particularlyregarding the role of the MTL in bothtrue and false recognition. In a study bySchacter et al. (1996c), amnesic patients(with MTL damage) took part in a recog-nition memory paradigm that used associa-tive word lists. As expected, these patientsshowed lower levels of true recognition(and higher levels of false alarms to newwords) as compared to control participants;in addition, the patients had lower levelsof false recognition (i.e., a reduced rate offalse alarms to semantically related words;see also, Melo, Winocur, & Moscovitch,1999; Schacter, Verfaellie, Anes, & Racine,1998b). Similarly, reduced levels of both trueand false recognition in amnesic patientshave also been shown in recognition mem-ory paradigms that have employed concep-tually related words (e.g., “twister,” “fun-nel”) and perceptually related words (e.g.,“hate,” “mate”; Schacter et al., 1997b), orabstract visual patterns (Koutstaal et al.,1999; similar to those used by Slotnick &Schacter, 2004). Furthermore, Alzheimer’sdisease patients (with neuropathology thatincludes, but is not limited to, the MTLregions) also have lower levels of false recog-

nition as compared to control participants(Balota, Watson, Duchek, & Ferraro, 1999;Budson, Daffner, Desikan, & Schacter, 2000;Budson, Desikan, Daffner, & Schacter,2001; Budson, Sullivan, Daffner, & Schacter,2003). These neuropsychological studiesindicate that the MTL is critically involvedin both true and false recognition.

Concluding Comments

In this chapter we have reviewed cog-nitive neuroscience evidence concerningthree distinctions that illuminate differentaspects of the relation between memoryand consciousness: retrieval success versusattempt, remembering versus knowing,and true versus false recognition. Retrievalsuccess involves memory of a previouslyexperienced item or event, whereas retrievalattempt refers to the effort associated withremembering (without success). As such,successful retrieval (based on the asso-ciated memorial experience/details) canbe said to reflect high retrieval content.whereas retrieval attempt can be said toreflect low retrieval content. By definition,remember-know studies are used to studydistinctions between contextual differencesin explicit memory: Remember responsesare associated with greater sensory/contextual detail (i.e., high retrieval con-tent), whereas know responses are notassociated with sensory/contextual detail(i.e., low retrieval content). True recog-nition has been associated with access togreater sensory/contextual detail as com-pared to false recognition (Mather et al.,1997; Norman & Schacter, 1997; Schooler,Gerhard, & Loftus, 1986). Accordingly,retrieval content can be considered greaterduring true as compared to false memory(although not to such a degree as to precludethe occurrence of false memories). Thus,although both true and false recognition areforms of explicit memory, where commonneural substrates likely reflect mechanismsof general retrieval, regions differentiallyassociated with true and false recognitioncan be assumed to reflect high and lowretrieval content, respectively.


82 0 the cambridge handbook of consciousness

As reflected in our summaries at theconclusion of each section of the chapter,the patterns of results for retrieval successand attempt, for remembering and know-ing, and for true and false recognition showstriking parallels (Table 28.1). The patternsof results for retrieval success and attemptdiffered from the patterns for remember-ing and knowing only in that rememberinggreater than knowing (and not retrieval suc-cess greater than attempt) was associatedwith activity in parietal and sensory cor-tex (which may simply reflect general dif-ferences in the use of stimulus materials;e.g., pictures versus words). The patterns ofresults for true versus false recognition werelargely identical to the patterns of results forremembering versus knowing, except thattrue and false recognition were both asso-ciated with MTL activity, whereas somedata indicate remembering but not know-ing were associated with MTL activity (asnoted earlier, however, the neuropsycholog-ical evidence for this conclusion is uncertain,with some data indicating a link betweenknowing and MTL structures). That theMTL is associated with false recognitionmay provide some explanation why partic-ipants respond “old” despite the fact theremay be less contextual detail associated withthese items.

We now consider the common neuralactivity associated with high retrieval con-tent (i.e., retrieval success, remembering,and true recognition) and low retrievalcontent (i.e., retrieval attempt, knowing,and false recognition). Memories with bothhigh and low retrieval content were associ-ated with activity in the prefrontal cortexand parietal cortex, which indicates theseregions are generally associated with explicitretrieval. There was also some evidencethat memories with low retrieval content,to a greater degree than those with highretrieval content, may be associated withincreased prefrontal cortex activity; how-ever, this activity has been attributed togreater low retrieval content-related post-retrieval monitoring (Schacter & Slotnick,2004). That is, although there may bemore effortful conscious processing with

low retrieval content items (which can beconsidered access-consciousness; see Block,1995), and is perhaps attributable to greatertask difficulty, this is typically not the cen-tral focus in discussions of consciousnessand memory. Rather, high retrieval contentand low retrieval content refer to the sen-sory/contextual experience associated withretrieval of episodic memories. Relevant tothis point, high retrieval content memories,to a greater degree than low retrieval con-tent memories, were associated with activ-ity in the parietal cortex (most consistentlythe inferior parietal lobule) and sensoryprocessing regions (at least for remember-ing and true recognition, with a null resultfor retrieval success). The parietal activ-ity may reflect a greater degree of atten-tion during retrieval of memories with highretrieval content as compared to those withlow retrieval content (Corbetta & Shul-man, 2002 ; Hopfinger, Buonocore, & Man-gun, 2000). Critically, however, the greaterdegree of sensory activity associated withmemories with high versus low retrieval con-tent provides evidence that memories areconstructed by reactivation of features thatcomprised a previous item or event (Squire,1992 ; Schacter et al., 1998a).

The present chapter shows that a cog-nitive neuroscience approach can illumi-nate the relation between memory and con-sciousness, highlighting how explicit mem-ories with different degrees of retrieval con-tent can be linked to distinct neural sub-strates. Although we would be remiss notto point out that this area of research is inits infancy, we also believe that the field hasadvanced significantly since the publicationof Tulving’s (1985b) lament concerning thelack of interest in memory and conscious-ness. We suspect that advances during thenext 20 years will be even more impressivethan those of the past two decades.

Acknowledgments

Preparation of this chapter was sup-ported by grants NIA AG08441 and NIMHMH060951.


the cognitive neuroscience of memory and consciousness 82 1

References

Balota, D. A., Watson, J. M., Duchek, J. M., &Ferraro, F. R. (1999). Cross-modal semantic andhomograph priming in healthy young, healthyold, and Alzheimer’s disease individuals. Jour-nal of International Neuropsychological Society,5 , 626–640.

Bar, M., & Aminoff, E. (2003). Cortical analysisof visual context. Neuron, 38, 347–358.

Block, N. (1995). On a confusion about a functionof consciousness. Behavioral and Brain Sciences,18, 227–287.

Buckner, R. L., Koutstaal, W., Schacter, D. L.,Dale, A. M., Rotte, M., & Rosen, B. R.(1998). Functional-anatomic study of episodicretrieval: Selective averaging of event-relatedfMRI trials to test the retrieval success hypoth-esis. NeuroImage, 7 , 163–175 .

Buckner, R. L., & Schacter, D. L. (2004). Neuralcorrelates of memory’s successes and sins. In M.S. Gazzaniga (Ed.), The cognitive neurosciences(3 rd ed., pp. 739–752). Cambridge, MA: MITPress.

Buckner, R. L., & Wheeler, M. E. (2001).The cognitive neuroscience of remember-ing. Nature Reviews Neuroscience, 2 , 624–634 .

Budson, A. E., Daffner, K. R., Desikan, R., &Schacter, D. L. (2000). When false recogni-tion is unopposed by true recognition: Gist-based memory distortion in Alzheimer’s dis-ease. Neuropsychology, 14 , 277–287.

Budson, A. E., Desikan, R., Daffner, K. R., &Schacter, D. L. (2001). Perceptual false recog-nition in Alzheimer’s disease. Neuropsychology,15 , 230–243 .

Budson, A. E., Sullivan, A. L., Daffner, K. R.,& Schacter, D. L. (2003). Semantic versusphonological false recognition in aging andAlzheimer’s disease. Brain and Cognition, 51,251–261.

Cabeza, R., Rao, S. M., Wagner, A. D., Mayer,A. R., & Schacter, D. L. (2001). Can medialtemporal lobe regions distinguish true fromfalse? An event-related functional MRI studyof veridical and illusory recognition memory.Proceedings of the National Academy of SciencesUSA, 98, 4805–4810.

Cermak, L. S. (1982). Human memory and amne-sia. Hillsdale, NJ: Erlbaum.

Corbetta, M., & Shulman, G. L. (2002). Controlof goal-directed and stimulus-driven attention

in the brain. Nature Reviews Neuroscience, 3 ,201–215 .

Curran, T. (2004). Effects of attention and con-fidence on the hypothesized ERP correlates ofrecollection and familiarity. Neuropsychologia,42 , 1088–1106.

Dale, A. M., & Buckner, R. L. (1997). Selectiveaveraging of rapidly presented individual trialsusing fMRI. Human Brain Mapping, 5 , 329–340.

Deese, J. (1959). On the prediction of occurrenceof particular verbal intrusions in immediaterecall. Journal of Experimental Psychology, 58,17–22 .

Dobbins, Foley, Schacter, & Wagner, (2002).Dobbins, I. G., Rice, H. J., Wagner, A. D., &

Schacter, D. L. (2003). Memory orientationand success: Separable neurocognitive com-ponents underlying episodic recognition. Neu-ropsychologia, 41, 318–333 .

Dobbins, I. G., Schnyer, D. M., Verfaellie, M., &Schacter, D. L. (2004). Cortical activity reduc-tions during repetition priming can result fromrapid response learning. Nature, 42 8, 316–319.

Duarte, A., Ranganath, C., Winward, L., Hay-ward, D., & Knight, R. T. (2004). Dissociableneural correlates for familiarity and recollec-tion during the encoding and retrieval of pic-tures. Cognitive Brain Research, 18, 255–272 .

Dunn, J. C. (2004). Remember-know: A matterof confidence. Psychological Review, 111, 524–542 .

Duzel, E., Yonelinas, A. P., Mangun, G. R.,Heinze, H., & Tulving, E. (1997). Event-relatedbrain potential correlates of two states of con-scious awareness in memory. Proceedings of theNational Academy of Sciences USA, 94 , 5973–5978.

Eldridge, L. L., Knowlton, B. J., Furmanski, C. S.,Bookheimer, S. Y., & Engel, S. A. (2000).Remembering episodes: A selective role for thehippocampus during retrieval. Nature ReviewsNeuroscience, 3 , 1149–1152 .

Epstein, R., & Kanwisher, N. (1998). A corticalrepresentation of the local visual environment.Nature, 392 , 598–601.

Fabiani, M., Stadler, M. A., & Wessels, P. M.(2000). True but not false memories producea sensory signature in human lateralized brainpotentials. Journal of Cognitive Neuroscience, 12 ,941–949.

Gardiner, J. M., Ramponi, C., & Richardson-Klavehn, A. (2002). Recognition memory and



decision processes: A meta-analysis of remem-ber, know, and guess responses. Memory, 10,83–98.

Goodale, M. A., & Westwood, D. A. (2004).An evolving view of duplex vision: Separatebut interacting cortical pathways for percep-tion and action. Current Opinion in Neurobiol-ogy, 14 , 203–211.

Gooding, P. A., Mayes, A. R., & van Eijk, R.(2000). A meta-analysis of indirect memorytests for novel material in organic amnesics.Neuropsychologia, 38, 666–676.

Gottfried, J. A., Smith, A. P. R., Rugg, M. D.,& Dolan, R. J. (2004). Remembrance of odorspast: Human olfactory cortex in cross-modalrecognition memory. Neuron, 42 , 687–695 .

Graf, P., & Schacter, D. L. (1985). Implicit andexplicit memory for new associations in normalsubjects and amnesic patients. Journal of Exper-imental Psychology: Learning, Memory, and Cog-nition, 11, 501–518.

Gratton, G., Corballis, P. M., & Jain, S. (1997).Hemispheric organization of visual memories.Journal of Cognitive Neuroscience, 9, 92–104 .

Guzeldere, G., Flanagan, O., & Hardcastle,V. G. (2000). The nature and function ofconsciousness: Lessons from blindsight. In M.S. Gazzaniga (Ed.), The cognitive neurosciences(2nd ed., pp. 1277–1284). Cambridge, MA:MIT Press.

Henson, R. N. (2003). Neuroimaging studies ofpriming. Progress in Neurobiology, 70, 53–81.

Henson, R. N. A., Rugg, M. D., Shallice, T.,Josephs, O., & Dolan, R. J. (1999). Recollec-tion and familiarity in recognition memory:An event-related functional magnetic reso-nance imaging study. Journal of Neuroscience,19, 3962–3972 .

Henson, R., Shallice, T., & Dolan, R. (2000).Neuroimaging evidence for dissociable formsof repetition priming. Science, 2 87 , 1269–1272 .

Holdstock, J. S., Mayes, A. R., Gong, Q. Y.,Roberts, N., & Kapur, N. (2005). Item recog-nition is less impaired than recall and asso-ciative recognition in a patient with selectivehippocampal damage. Hippocampus, 15 , 203–215 .

Holdstock, J. S., Mayes, A. R., Roberts, N., Ceza-yirli, E., Isaac, C. L., O’Reilly, R. C., & Norman,K. A. (2002). Under what conditions is recog-nition spared relative to recall after selectivehippocampal damage in humans? Hippocam-pus, 12 , 341–351.

Hopfinger, J. B., Buonocore, M. H., & Mangun,G. R. (2000). The neural mechanisms of top-down attentional control. Nature Neuroscience,3 , 284–291.

James, T. W., Humphrey, G. K., Gati, J. S.,Menon, R. S., & Goodale, M. A. (2000). Theeffects of visual object priming on brain activa-tion before and after recognition. Current Biol-ogy, 10, 1017–1024 .

Johnson, M. K., Nolde, S. F., Mather, M.,Kounios, J., Schacter, D. L., & Curran, T.(1997). The similarity of brain activity associ-ated with true and false recognition memorydepends on test format. Psychological Science,8, 250–257.

Kahn, I., Davachi, L., & Wagner, A. D.(2004). Functional-neuroanatomic correlatesof recollection: Implications for models ofrecognition memory. Journal of Neuroscience,2 4 , 4172–4180.

Kapur, S., Craik, F. I., Jones, C., Brown, G. M.,Houle, S., & Tulving, E. (1995). Functionalrole of the prefrontal cortex in retrieval ofmemories: A PET study. NeuroReport, 6, 1880–1884 .

Knowlton, B. J., & Squire, L. R. (1995). Remem-bering and knowing: Two different expressionsof declarative memory. Journal of ExperimentalPsychology: Learning, Memory, & Cognition, 2 1,699–710.

Kohler, S., & Moscovitch, M. (1997). Uncon-scious visual processing in neuropsychologicalsyndromes: A survey of the literature and eval-uation of models of consciousness. In M. D.Rugg (Ed.), Cognitive neuroscience (pp. 305–373). Cambridge, MA: MIT Press.

Konishi, S., Wheeler, M. E., Donaldson, D. I.,& Buckner, R. L. (2000). Neural correlates ofepisodic retrieval success. NeuroImage, 12 , 276–286.

Koutstaal, W., Schacter, D. L., Verfaellie, M.,Brenner, C., & Jackson, E. M. (1999). Perceptu-ally based false recognition of novel objects inamnesia: Effects of category size and similarityto category prototypes. Cognitive Neuropsychol-ogy, 16, 317–341.

Manns, J. R., Hopkins, R. O., Reed, J. M., Kitch-ener, E. G., & Squire, L. R. (2003). Recognitionmemory and the human hippocampus. Neu-ron, 37 , 171–180.

Maril, A., Simons, J. S., Mitchell, J. P., Schwartz,B. L., & Schacter, D. L. (2003). Feeling-of-knowing in episodic memory: An event-relatedfMRI study. NeuroImage, 18, 827–836.



Mather, M., Henkel, L. A., & Johnson, M. K.(1997). Evaluating characteristics of falsememories: Remember/know judgments andmemory characteristics questionnaire com-pared. Memory & Cognition, 2 5 , 826–837.

McDermott, K. B., Jones, T. C., Petersen, S. E.,Lageman, S. K., & Roediger, H. L. III. (2000).Retrieval success is accompanied by enhancedactivation in anterior prefrontal cortex duringrecognition memory: An event-related fMRIstudy. Journal of Cognitive Neuroscience, 12 ,965–976.

Melo, B., Winocur, G., & Moscovitch, M.(1999). False recall and false recognition:An examination of the effects of selectiveand combined lesions to the medial tempo-ral lobe/diencephalon and frontal lobe struc-tures. Cognitive Neuropsychology, 16, 343–359.

Moscovitch, M. (1984). The sufficient conditionsfor demonstrating preserved amnesia: A taskanalysis. In L. R. Squire & N. Butters (Eds.),Neuropsychology of memory (pp. 104–114). NewYork: Guilford Press.

Moscovitch, M. (1994). Memory and workingwith memory: Evaluation of a component pro-cess model and comparisons with other models.In D. L. Schacter & E. Tulving (Eds.) Memorysystems 1994 (pp. 269–331). Cambridge, MA:MIT Press.

Moscovitch, M. (1995). Recovered con-scious-ness: A hypothesis concerning modularity andepisodic memory. Journal of Clinical and Exper-imental Neuropsychology, 17 , 276–290.

Moscovitch, M. (2000). Theories of memory andconsciousness. In E. Tulving & F. I. M. Craik(Eds.), Oxford handbook of memory (pp. 609–625). London: Oxford University Press.

Moscovitch, M., Kapur, S., Kohler, S., & Houle,S. (1995). Distinct neural correlates of visuallong-term memory for spatial location andobject identity: A positron emission tomog-raphy study in humans. Proceedings of theNational Academy of Sciences USA, 92 , 3721–3725 .

Moscovitch, M., & McAndrews, M. P. (2002).Material-specific deficits in “remembering” inpatients with unilateral temporal lobe epilepsyand excisions. Neuropsychologia, 40, 1335–1342 .

Moscovitch, M., Vriezen, E., & Goshen-Gottstein, Y. (1993). Implicit tests of memoryin patients with focal lesions or degenerativebrain disorders. In F. Boller & J. Grafman (Eds.),

Handbook of neuropsychology (Vol. 8, pp. 133–173). Amsterdam: Elsevier.

Nadel, L., & Moscovitch, M. (1997). Memoryconsolidation, retrograde amnesia and the hip-pocampal complex. Current Opinion in Neuro-biology, 7 , 217–227.

Nolde, S. F., Johnson, M. K., & D’Esposito,M. (1998). Left prefrontal activation duringepisodic remembering: An event-related fMRIstudy. NeuroReport, 9, 3509–3514 .

Norman, K. A., & Schacter, D. L. (1997).False recognition in younger and older adults:Exploring the characteristics of illusory mem-ories. Memory & Cognition, 2 5 , 838–848.

Nyberg, L., Habib, R., McIntosh, A. R., & Tul-ving, E. (2000). Reactivation of encoding-related brain activity during memory retrieval.Proceedings of the National Academy of SciencesUSA, 97 , 11120–11124 .

Nyberg, L., McIntosh, A. R., Houle, S., Nils-son, L. G., & Tulving, E. (1996). Activationof medial temporal structures during episodicmemory retrieval. Nature, 380, 715–717.

Nyberg, L., Petersson, K. M., Nilsson, L., Sand-blom, J., Åberg, C., & Ingvar, M. (2001). Reac-tivation of motor brain areas during explicitmemory for actions. NeuroImage, 14 , 521–528.

Nyberg, L., Tulving, E., Habib, R., Nilsson, L. G.,Kapur, S., Houle, S., Cabeza, R., & McIntosh,A. R. (1995). Functional brain maps of retrievalmode and recovery of episodic information.NeuroReport, 7 , 249–252 .

Okado, Y., & Stark, C. (2003). Neural process-ing associated with true and false memoryretrieval. Cognitive, Affective, & Behavioral Neu-roscience, 3 , 323–334 .

Roediger, H. L. III, & McDermott, K. B. (1993).Implicit memory in normal human subjects. InH. Spinnler & F. Boller (Eds.), Handbook of neu-ropsychology (Vol. 8, pp. 63–13 1). Amsterdam:Elsevier.

Roediger, H. L.III, & McDermott, K. B. (1995).Creating false memories: Remembering wordsnot presented in lists. Journal of ExperimentalPsychology: Learning, Memory, and Cognition,2 1, 803–814 .

Rotello, C. M., Macmillan N. A., & Reeder, J. A.(2004). Sum-difference theory of remember-ing and knowing: A two-dimensional signal-detection model. Psychological Review, 111, 588–616.

Rugg, M. D., Fletcher, P. C., Frith, C. D., Frack-owiak, R. S., & Dolan, R. J. (1996). Differential



activation of the prefrontal cortex in successfuland unsuccessful memory retrieval. Brain, 119,2073–2083 .

Rugg, M. D., Fletcher, P. C., Frith, C. D., Frack-owiak, R. S., & Dolan, R. J. (1997). Brainregions supporting intentional and incidentalmemory: A PET study. NeuroReport, 2 4 , 1283–1287.

Rugg, M. D., Henson, R. N. A., & Robb, W. G. K.(2003). Neural correlates of retrieval process-ing in the prefrontal cortex during recognitionand exclusion tasks. Neuropsychologia, 41, 40–52 .

Rugg, M. D., Mark, R. E., Walla, P., Schloer-scheidt, A. M., Birch, C. S., & Allan, K. (1998).Dissociation of the neural correlates of implicitand explicit memory. Nature, 392 , 595–598.

Schacter, D. L. (1987). Implicit memory: His-tory and current status. Journal of ExperimentalPsychology: Learning, Memory, and Cognition,13 , 501–518.

Schacter, D. L., Alpert, N. M., Savage, C. R.,Rauch, S. L., & Albert, M. S. (1996a). Con-scious recollection and the human hippocam-pal formation: Evidence from positron emis-sion tomography. Proceedings of the NationalAcademy of Sciences USA, 93 , 321–325 .

Schacter, D. L., & Buckner, R. L. (1998). Primingand the brain. Neuron, 2 0, 185–195 .

Schacter, D. L., Buckner, R. L., Koutstaal, W.,Dale, A. M., & Rosen, B. R. (1997a). Late onsetof anterior prefrontal activity during true andfalse recognition: An event-related fMRI study.NeuroImage, 6, 259–269.

Schacter, D. L., & Curran, T. (2000). Memorywithout remembering and remembering with-out memory: Implicit and false memories. InM. S. Gazzaniga (Ed.), The cognitive neuro-sciences (2nd ed., pp. 829–840). Cambridge,MA: MIT Press.

Schacter, D. L., Dobbins, I. G., & Schnyer,D. M. (2004). Specificity of priming: A cogni-tive neuroscience perspective. Nature ReviewsNeuroscience, 5 , 853–862 .

Schacter, D. L., McAndrews, M. P., & Moscov-itch, M. (1988). Access to consciousness: Dis-sociations between implicit and explicit knowl-edge. In L. Weiskrantz (Ed.), Thought with-out language (pp. 242–278). New York: OxfordUniversity Press.

Schacter, D. L., Norman, K. A., & Koutstaal, W.(1998a). The cognitive neuroscience of con-

structive memory. Annual Review of Psychology,49, 289–318.

Schacter, D. L., Reiman, E., Curran, T., Yun,L. S., Bandy, D., McDermott, K. B., & Roediger,H. L. III. (1996b). Neuroanatomical correlatesof veridical and illusory recognition memory:Evidence from positron emission tomography.Neuron, 17 , 267–274 .

Schacter, D. L., Reiman, E., Uecker, A., Polster,M. R., Yun, L. S., & Cooper, L. A. (1995). Brainregions associated with retrieval of structurallycoherent visual information. Nature, 376, 587–590.

Schacter, D. L., & Slotnick, S. D. (2004). Thecognitive neuroscience of memory distortion.Neuron, 44 , 149–160.

Schacter, D. L., Verfaellie, M., & Anes, M. D.(1997b). Illusory memories in amnesic patients:Conceptual and perceptual false recognition.Neuropsychology, 11, 331–342 .

Schacter, D. L., Verfaellie, M., Anes, M. D.,& Racine, C. (1998b). When true recognitionsuppresses false recognition: Evidence fromamnesic patients. Journal of Cognitive Neuro-science, 10, 668–679.

Schacter, D. L., Verfaellie, M., & Pradere,D. (1996c). The neuropsychology of mem-ory illusions: False recall and recognition inamnesic patients. Journal of Memory and Lan-guage, 35 , 319–334 .

Schooler, J. W., Gerhard, D., & Loftus, E. F.(1986). Qualities of the unreal. Journal of Exper-imental Psychology: Learning, Memory, and Cog-nition, 12 , 171–181.

Slotnick, S. D., & Dodson, C. S. (2005). Sup-port for a continuous (single-process) model ofrecognition memory and source memory. Mem-ory & Cognition, 33 , 15 1–170.

Slotnick, S. D., Moo, L. R., Segal, J. B., & Hart,J. Jr. (2003). Distinct prefrontal cortex activityassociated with item memory and source mem-ory for visual shapes. Cognitive Brain Research,17 , 75–82 .

Slotnick, S. D., & Schacter, D. L. (2004). Asensory signature that distinguishes true fromfalse memories. Nature Neuroscience, 7 , 664–672 .

Smith, M. E. (1993). Neurophysiological man-ifestations of recollective experience duringrecognition memory judgments. Journal of Cog-nitive Neuroscience, 5 , 1–13 .

Squire, L. R. (1992). Memory and the hippo-campus: A synthesis from findings with rats,



monkeys, and humans. Psychological Review,99, 195–231.

Squire, L. R., Ojemann, J. G., Miezin, F. M.,Petersen, S. E., Videen, T. O., & Raichle, M. E.(1992). Activation of the hippocampus in nor-mal humans: A functional anatomical study ofmemory. Proceedings of the National Academyof Sciences USA, 89, 1837–1841.

Squire, L. R., Stark, C. E. L., & Clark, R. E.(2004). The medial temporal lobe. AnnualReview of Neuroscience, 2 7 , 279–306.

Trott, C. T., Friedman, D., Ritter, W., Fabiani,M., & Snodgrass, J. G. (1999). Episodic prim-ing and memory for temporal source: Event-related potentials reveal age-related differencesin prefrontal functioning. Psychology and Aging,14 , 390–413 .

Tulving, E. (1985a). How many memory systemsare there? American Psychologist, 40, 385–398.

Tulving, E. (1985b). Memory and consciousness.Canadian Psychology, 2 6, 1–12 .

Tulving, E., Kapur, S., Craik, F. I. M., Moscovitch,M., & Houle, S. (1994). Hemispheric encod-ing/retrieval asymmetry in episodic memory:Positron emission tomography findings. Pro-ceedings of the National Academy of SciencesUSA, 91, 2016–2020.

Uecker, A., Reiman, E. M., Schacter, D. L., Pol-ster, M. R., Cooper, L. A., Yun, L. S., &Chen, K. (1997). Neuroanatomical correlatesof implicit and explicit memory for structurallypossible and impossible visual objects. Learning& Memory, 4 , 337–355 .

Vaidya, C. J., Zhao, M., Desmond, J. E., &Gabrieli, J. D. E. (2002). Evidence for corti-cal encoding specificity in episodic memory:Memory-induced re-activation of picture pro-cessing areas. Neuropsychologia, 40, 2136–2143 .

Velanova, K., Jacoby, L. L., Wheeler, M. E.,McAvoy, M. P., Petersen, S. E., & Buckner,R. L. (2003). Functional-anatomic correlates ofsustained and transient processing componentsengaged during controlled retrieval. Journal ofNeuroscience, 2 3 , 8460–8470.

Wagner, A. D., Desmond, J. E., Glover, G. H.,& Gabrieli, J. D. E. (1998). Prefrontal cor-tex and recognition memory: Functional-MRI

evidence for context-dependent retrieval pro-cesses. Brain, 12 1, 1985–2002 .

Warrington, E. K., & Weiskrantz, L. (1974). Theeffect of prior learning on subsequent retentionin amnesic patients. Neuropsychologia, 12 , 419–428.

Wheeler, M. E., & Buckner, R. L. (2003).Functional dissociation among components ofremembering: Control, perceived oldness, andcontent. Journal of Neuroscience, 2 3 , 3869–3880.

Wheeler, M. E., & Buckner, R. L. (2004).Functional-anatomic correlates of remember-ing and knowing. NeuroImage, 2 1, 1337–1349.

Wheeler, M. E., Petersen, S. E., & Buckner, R. L.(2000). Memory’s echo: Vivid rememberingreactivates sensory-specific cortex. Proceedingsof the National Academy of Sciences USA, 97 ,11125–11129.

Wiggs, C. L., & Martin, A. (1998). Propertiesand mechanisms of perceptual priming. Cur-rent Opinion in Neurobiology, 8, 227–233 .

Wilding, E. L., & Rugg, M. D. (1996). An event-related potential study of recognition memorywith and without retrieval of source. Brain, 119,889–905 .

Wixted, J. T., & Stretch, V. (2004). In defense ofthe signal detection interpretation of remem-ber/know judgments. Psychonomic Bulletin &Review, 11, 616–641.

Yonelinas, A. P., Kroll, N. E. A., Dobbins, I.,Lazzara, M., & Knight, R. T. (1998). Recollec-tion and familiarity deficits in amnesia: Con-vergence of remember-know, process dissocia-tion, and receiver operating characteristic data.Neuropsychology 12 , 323–339.

Yonelinas, A. P., Kroll, N. E., Quamme, J. R., Laz-zara, M. M., Sauve, M. J., Widarman, K. F., &Knight, R. T. (2002). Effects of extensive tem-poral lobe damage or mild hypoxia on recol-lection and familiarity. Nature Neuroscience, 5 ,1236–1241.

Young, A. W. (1994). Conscious and nonconsiousrecognition of familiar faces. In C. Umilta & M.Moscovitch (Eds.), Attention and performance15 : Conscious and nonconscious information pro-cessing (pp. 153–178). Cambridge, MA: MITPress.


826

CHAPTER 28 The Cognitive Neuroscience of Memory and...

Documents

Transcript of CHAPTER 28 The Cognitive Neuroscience of Memory and...