neurology_prefrontal cortex and long-term memory encoding an integrative review of findings from...

http://nro.sagepub.comThe Neuroscientist

DOI: 10.1177/1073858407299290 2007; 13; 280 Neuroscientist

Robert S. Blumenfeld and Charan Ranganath Neuropsychology and Neuroimaging

Prefrontal Cortex and Long-Term Memory Encoding: An Integrative Review of Findings from

http://nro.sagepub.com/cgi/content/abstract/13/3/280 The online version of this article can be found at:

Published by:

http://www.sagepublications.com

can be found at:The Neuroscientist Additional services and information for

http://nro.sagepub.com/cgi/alerts Email Alerts:

http://nro.sagepub.com/subscriptions Subscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://nro.sagepub.com/cgi/content/refs/13/3/280SAGE Journals Online and HighWire Press platforms):

(this article cites 124 articles hosted on the Citations

2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by Favio Vega on November 18, 2007 http://nro.sagepub.comDownloaded from

Memorizing phone numbers, remembering past conversa-tions, finding where you left your keysthese are thekinds of daily activities that depend on the ability to formcoherent episodic memories. Recent research in neu-ropsychology and neuroimaging has highlighted theimportance of the prefrontal cortex (PFC) in promotingsuccessful episodic long-term memory (LTM) formation,although its precise role remains poorly understood. Here,we will review this evidence and present a theoreticalframework for understanding how different regions of thePFC contribute to episodic memory encoding. Central tothis framework is the idea that different regions in the PFCimplement different control processes that augment mem-ory by enhancing or attenuating certain aspects of a par-ticular item or event.

Our review will have three main sections. First, we willbegin by discussing results from studies of memory inpatients with focal PFC lesions that converge on the ideathat specific deficits in cognitive control underlie the LTMimpairments in PFC patients. Next, we will review resultsfrom neuroimaging studies of cognitive control, whichsuggest that these different cognitive control processes are performed by distinct PFC regions. Finally, we willreview the neuroimaging literature on episodic memory

encoding and consider how our framework can help toclarify the role of the PFC in LTM encoding.

Effects of Prefrontal Lesions on LTM Encoding

Clinicians have long noted that focal prefrontal lesions inhumans produce subtle but noticeable memory deficits,and this impression accords well with results from neu-ropsychological studies (Milner 1962; Stuss and Benson1986; Shimamura 1995; Ranganath and Knight 2003).

In general, patients with PFC lesions show impairmentson a wide range of memory tasks that tax executive con-trol during encoding and retrieval. For instance, PFCpatients exhibit impaired performance on unconstrainedmemory tests such as free-recall (Jetter and others 1986;McAndrews and Milner 1991; Eslinger and Grattan 1994;Stuss and others 1994; Moscovitch and Winocur 1995;Wheeler and others 1995; Dimitrov and others 1999; seeBox 1). In contrast to healthy control participants, PFCpatients tend not to spontaneously cluster or group recalloutput according to the semantic relationships within acategorized word list (della Rocchetta 1986; Hirst andVolpe 1988; della Rocchetta and Milner 1993; Stuss andothers 1994; Gershberg and Shimamura 1995). When pre-sented with several study-recall blocks of the same wordlist, PFC patients show less consistency of recall outputorder from trial to trial (i.e., subjective organization; seeBox 1) compared to controls (Stuss and others 1994;Gershberg and Shimamura 1995; Alexander and others

Prefrontal Cortex and Long-Term MemoryEncoding: An Integrative Review of Findings from Neuropsychology and NeuroimagingROBERT S. BLUMENFELD and CHARAN RANGANATHCenter for Neuroscience and Department of Psychology, University of California at Davis

Recent findings have led to a growing appreciation of the role of the lateral prefrontal cortex (PFC) inepisodic long-term memory (LTM). Here, the authors will review results from neuropsychological and neu-roimaging studies of humans and present a framework to explain how different regions of the PFC con-tribute to successful LTM formation. Central to this framework is the idea that different regions within thePFC implement different control processes that augment memory by enhancing or attenuating memory forcertain aspects of a particular item or event. Evidence reviewed here suggests that ventrolateral regions ofthe PFC contribute to the ability to select goal-relevant item information, and that this processing strength-ens the representation of goal-relevant features of items during LTM encoding. Dorsolateral regions of thePFC may contribute to the ability to organize multiple pieces of information in working memory, therebyenhancing memory for associations among items in LTM. Thus, dorsolateral and ventrolateral regions of thePFC may implement different control processes that support LTM formation in a complementary fashion.NEUROSCIENTIST 13(3):280291, 2007. DOI: 10.1177/1073858407299290

KEY WORDS Working, Organization, Selection, Frontal, Lobes, Associative, Episodic, Memory

280 THE NEUROSCIENTIST Prefrontal Cortex and Memory Encoding

This work was supported by PHS grant 1R01MH068721.

Address correspondence to: Robert S. Blumenfeld, UC Davis Centerfor Neuroscience, 1544 Newton Ct, Davis, CA 95616 (e-mail: [email protected]).

Volume 13, Number 3, 2007Copyright 2007 Sage PublicationsISSN 1073-8584


Volume 13, Number 3, 2007 THE NEUROSCIENTIST 281

2003). In addition, patients with PFC lesions exhibitimpairments on tests of source memory (Janowsky andothers 1989b), memory for temporal order (Milner andothers 1985; Shimamura and others 1990; Kesner and oth-ers 1994), recency (Milner and others 1991), frequency(Stanhope and others 1998), and associative learning(Shimamura and others 1995; Swick and Knight 1996;Dimitrov and others 1999). Furthermore, PFC patientsoften lack insight into their own memory problems and failto spontaneously use common memory strategies (Hirstand Volpe 1988; Janowsky and others 1989a; Moscovitchand Melo 1997; Vilkki and others 1998).

In contrast to these deficits, patients with prefrontallesions can often perform at near-normal levels whengiven structured encoding tasks or tests that do not taxstrategic retrieval processes. For instance, PFC patientsperform better at cued-recall compared to free-recall andcan perform similarly to control subjects on item recogni-tion (Kesner and others 1994; Stuss and others 1994;Swick and Knight 1996; Dimitrov and others 1999;Alexander and others 2003; but see Wheeler and others1995). Moreover, patients can show marked improve-ments on a variety of recall measures if given sufficientpractice or environmental support at encoding or retrieval(Hirst and Volpe 1988; della Rocchetta and Milner 1993;Stuss and others 1994; Gershberg and Shimamura 1995).

In general, theoretical accounts have stressed that LTMimpairments associated with PFC damage arise as a con-sequence of deficits in control processing rather than froma primary deficit in memory storage (Ranganath andKnight 2003). Some theories emphasize the role of thePFC in selection processes that direct attention towardgoal-relevant information and task-appropriate responses

(for reviews, see Milner and others 1985; Stuss andBenson 1986; Miller and Cohen 2001). Thus, memorydeficits may occur in patients with prefrontal lesionsbecause they are unable to select relevant information orinhibit distracting or interfering items or responses duringencoding or retrieval (Luria 1966, 1973; Perret 1974;Shimamura and others 1995). One finding consistent withthis account comes from a study of paired-associate learn-ing (see Box 1) in patients with focal PFC lesions andmatched controls (Shimamura and others 1995). In thisstudy, participants learned a list of word pairs (A-B) andthen learned an overlapping list of word pairs (A-C)across several trials. Recall success on the second listrequired subjects to inhibit the well-learned A-B pairingand select the appropriate A-C pairing. Consistent withthe selection account, patients with PFC lesions showed asignificant and disproportionate decrement in recall per-formance between the first trial of A-B learning comparedto the first trial of A-C learning (i.e., a measure of proac-tive interference). Furthermore, during cued recall for theA-C list, PFC patients recalled fewer appropriate targetsand recalled significantly more words from the A-B list.These findings are consistent with the idea that PFClesions induce difficulties in resolving competitionbetween items in memory, possibly due to impairments ofcontrolled selection processes.

Other theories emphasize the role of the PFC in guidingspontaneous organization of information during theencoding and retrieval of memories (Petrides and Milner1982; Milner and others 1985; della Rocchetta and Milner1993; Gershberg and Shimamura 1995). Specifically,organization refers to memory strategies that emphasizecomparison or transformation of relations among items

Box 1. Memory Measures in Behavioral Research

In behavioral studies, episodic memory is usually assessed by testing memory for lists of items suchas words (Ebbinghaus 1885/1964). With this approach, memory can be tested in a number of ways.One is free-recall, in which participants are simply asked to recall as many items as they can remem-ber from the study list. Organizational strategies strongly impact recall performance, and this impactcan be measured by analyzing the characteristics of items that are recalled and the order in whichthey are recalled. For example, if participants are given lists of categorically related words to memo-rize, they tend to recall semantically related items together even if these items were distributed acrossthe study list. This phenomenon, called semantic clustering, can provide a quantitative index of organ-ization in memory. A related phenomenon, subjective organization, is seen in studies in which thesame list is repeatedly studied and tested. In this situation, participants will tend to recall sets of itemsin the same order across different recall trials. This tendency is thought to reflect the fact that partic-ipants form associations between items during encoding, and subsequently make use of these asso-ciations to guide retrieval (Sternberg and Tulving 1977).

Another paradigm used in many memory studies is paired-associate learning. In these studies,pairs of items are studied together, and then memory for these associations is tested. For example,memory for a studied word pair (A-B) can be tested via cued recall (A-?) or associative recognition (A-B or A-D?). Relevant to the studies discussed in our review, learning of a previous association (A-B)makes it more difficult to learn a new, overlapping association (A-C) (Barnes and Underwood 1959).This phenomenon is called proactive interference, and it may reflect competition between the pre-viously learned association and the new association. Accordingly, overcoming proactive interferencerequires the engagement of controlled selection processes to resolve this competition.


that are to be learned (Bower 1970; Hunt and Einstein1981). Some common organizational strategies duringencoding include categorizing words in a list according tosemantic features (Hunt and Einstein 1981), imaginingtwo or more items interacting (Bower 1970), or forming asentence out of two or more words. Organization duringencoding does not facilitate LTM by enhancing featuresof specific items in memory, but instead promotes mem-ory for associations among items (Bower 1970; Begg1979; McGee 1980). Accordingly, a number of measurescan be used to estimate the impact of organization onLTM performance (see Box 1).

In support of the notion that PFC is critical for organi-zation in memory, studies have demonstrated that patientswith prefrontal lesions do not spontaneously organizeinformation or make use of effective organizational strate-gies during encoding or retrieval (Hirst and Volpe 1988;della Rocchetta and Milner 1993; Stuss and others 1994;Gershberg and Shimamura 1995; Alexander and others2003) and that they show deficits on many memory meas-ures that are sensitive to organization. For instance,Gershberg and Shimamura (1995) compared patients withrelatively focal lesions to the dorsolateral PFC and healthycontrols on learning of lists of words that were eithersemantically related or unrelated. Relative to controls,patients were less likely to exhibit subjective organizationor semantic clustering (see Box 1) during recall of lists ofsemantically related words. Interestingly, recall and clus-tering performance increased for semantically relatedword lists when patients were explicitly asked to make acategory judgment during encoding, or when they wereprovided with the category names at test, or both. In con-trast, healthy controls performed at similar levels regard-less of whether they were given encoding instructions orretrieval cues. This pattern of results suggests that patientswere capable of using semantic information to guideencoding and retrieval, but they lacked the ability to spon-taneously use semantic organizational strategies. In con-trast, controls were spontaneously using organizationalstrategies during encoding and/or retrieval. These findings,and others (Hirst and Volpe 1988; della Rocchetta andMilner 1993; Stuss and others 1994; Alexander and others2003), are consistent with the idea that LTM deficits fol-lowing prefrontal lesions may result from a failure toorganize information during encoding and capitalize onorganizational structure during retrieval.

Many researchers have suggested that both selectionand organizational processes depend on the functioning ofthe PFC, and there are several studies that reported find-ings consistent with both the selection and organizationalaccounts (Hirst and Volpe 1988; Stuss and others 1994;Gershberg and Shimamura 1995; Shimamura and others1995; Alexander and others 2003). One question that can-not be addressed by the neuropsychological evidence iswhether selection and organization depend upon the sameregions of the PFC, because these studies typically usedsubject groups that have significant heterogeneity inlesion size and location. There are hints in the extant neu-ropsychological literature to suggest that patients withdeficits in subjective organization and recall clusteringtend to have damage to the dorsolateral PFC (Eslinger and

Grattan 1994; Gershberg and Shimamura 1995; Alexanderand others 2003; but see Stuss and others 1994).Nonetheless, the neuropsychological evidence cannot con-clusively resolve this issue because studies have typicallyused patient groups that have significant heterogeneity inlesion size and location.

Functional Neuroimaging of Cognitive Control

Several functional magnetic resonance imaging (fMRI)studies have found support for the notion that selectionmechanisms and organizational control mechanisms areprocessed by different regions of the PFC (DEsposito,Postle, Ballard, and others 1999; Postle and others 1999;Petrides 2000a; Wagner, Maril, and others 2001; Glahnand others 2002; for review, see Fletcher and Henson2001). Specifically, neuroimaging studies of workingmemory (WM) and cognitive control have suggested thatregions within the ventrolateral PFC (VLPFC: BA,44,45,47,47/12) and regions in the dorsolateral PFC(DLPFC: BA 9,46) may implement control processes thatsupport memory encoding in a complementary fashion.

VLPFC and Selection Processes

Results from neuroimaging studies of proactive interfer-ence resolution (Jonides and Nee 2006), response inhi-bition (Aron and others 2004), verb generation (Buckner2003; Thompson-Schill and others 2005), and semanticand nonsemantic classification (Wagner and Davachi2001; Badre and others 2005) are consistent with theidea that VLPFC might be particularly important forimplementing selection processes that direct attentiontoward goal-relevant information or inhibit the influenceof irrelevant information on behavior.

Some support for this view has come from studies thatused the recent-probes task (Monsell 1978) to examinethe role of VLPFC in resolving proactive interference. Inthis task, participants must maintain a set of four lettersover a short delay and subsequently evaluate whether atest probe was in the memory set. Proactive interferenceis manipulated by varying the familiarity of the probesacross trials. For instance, on some trials, the probe letteris not in the current memory set but was in the memoryset on the previous trial. To perform accurately on theserecent negative trials, participants must inhibit the highfamiliarity of the probe to make the correct negativeresponse. Imaging studies have consistently shown thatactivation in the left mid-VLPFC (BA 44, 45; see Fig. 1)is increased on these trials, relative to trials with nonre-cent negative or positive probes (e.g., Jonides and others1998; DEsposito, Postle, Jonides, and others 1999;Bunge and others 2001; Mecklinger and others 2003;e.g., Nelson and others 2003; Postle and Brush 2004;Postle and others 2004; Badre and Wagner 2005; seeJonides and Nee 2006 for review).

Another line of evidence consistent with the selectionaccount has come from fMRI studies of verb generation andsemantic classification (Buckner 2003; Thompson-Schilland others 2005). For instance, one study (Thompson-Schill and others 1997) examined PFC activity associated




with selection processing across three different semantictasks. Within each task, selection demands were variedwithin each of these conditions by increasing the competi-tion among response alternatives. For example, in the highselection condition of the comparison task, participants hadto compare a target noun (e.g., tooth) to two highly relatedprobe nouns (e.g., tongue and bone) and decide which probewas most similar to the cue along a specific dimension (i.e.,color). However, in the low-selection condition, participantscould make this comparison based on global similaritybetween the target and probe words. Across all three tasks,a common region of the left mid-VLPFC (BA 44, 45) exhib-ited increased activation during blocks of high-selection tri-als, suggesting a role for this region in the selection ofsemantic features of items.

Results from some fMRI studies have shown that moreactivation in relatively more anterior and ventral regions ofthe VLPFC (BA 47/12, 45; see Fig. 1) is enhanced duringselection of semantic information, whereas activation inthe posterior and dorsal VLPFC regions (BA 44, 6; see Fig.1) is enhanced during processing of semantic, phonologi-cal, or orthographic attributes of words (e.g., Poldrack andothers 1999; Bokde and others 2001; Roskies and others2001; Wagner, Pare-Blagoev, and others 2001; Dobbinsand others 2002; McDermott and others 2003; Badre andothers 2005; Bunge and others 2005; Dobbins and Wagner2005; but see Otten and Rugg 2001b and Gold andBuckner 2002). Not all studies, however, have observedthis type of material-specific dissociation within the PFC(e.g., Barde and Thompson-Schill 2002). An alternativeview that has been proposed is that the extent and locationof VLPFC activity depends on the degree of controlledselection that must be engaged (Buckner 2003). Researchconsistent with this notion has found that posterior regionsof the VLPFC show recruitment during selection tasks that

require less control and that the anterior regions of theVLPFC are additionally recruited during selection tasksthat require considerably more control (Gold and Buckner2002; Gold and others 2005; but see Badre and others2005). Interestingly, some studies have found that the acti-vation of VLPFC subregions are influenced both by mate-rials and by the amount of controlled selection (Badre andothers 2005; Gold and others 2005). This suggests thatalthough both of these accounts have some explanatorypower, more research will be required to find the best wayto characterize the functions of VLPFC subregions.

DLPFC and Organization

Unlike VLPFC, DLPFC is not robustly recruited duringtasks that solely require selection of task-relevant informa-tion. Instead, evidence from neuroimaging studies suggeststhat DLPFC is involved in organizationthat is, the com-parison or transformation of relationships among items thatare active in memory (DEsposito, Postle, Ballard, and oth-ers 1999; Postle and others 1999; Wagner, Maril, and oth-ers 2001; Barde and Thompson-Schill 2002; Glahn andothers 2002; Bor and others 2003; Veltman and others2003; Cannon and others 2005; Blumenfeld and Ranganath2006; Crone and others 2006; Mohr and others 2006). Forexample, DLPFC activation is reported in manipulationtasks that involve sequencing of information that is beingmaintained in WM (DEsposito, Postle, Ballard, and others1999; Postle and others 1999; Postle and DEsposito 1999;Wagner, Maril, and others 2001; Blumenfeld andRanganath 2006) or monitoring of previous responseswhen selecting a future response (Petrides 2000a, 2000b;Stern and others 2000). DLPFC activation has also beenreported in studies of chunking (Bor and others 2003;Bor and others 2004), the process by which multiple sepa-rate pieces of information are organized into fewer units ofinformation (Miller and others 1960).

One study of verbal WM highlights the selective roleof DLPFC in organization (Postle and others 1999). Inthis study, participants were scanned during three tasksthat required maintenance of verbal information acrossan eight-second delay period. In two conditions, partici-pants had to maintain the identity and the serial positionof either two (forward 2 trials) or five (forward 5 tri-als) letters. In the third condition, participants wereprompted to rearrange a set of five letters in alphabeticalorder and to maintain the identity and serial position ofthe rearranged set of letters (alphabetize 5 trials).Whereas forward 5 and forward 2 trials differed prima-rily in terms of the amount of information to be main-tained, alphabetize 5 and forward 5 trials differedprimarily in the kind of processing that was engaged.Specifically, alphabetize 5 trials placed greater demandson organizational processing, because these trialsrequired participants to use the alphabetize rule totransform the serial order relationships among the items.Results from this study showed that, for each subject,DLPFC activation was increased during the delay periodof alphabetize 5 trials, as compared with forward 5 tri-als. DLPFC activation in this study could not be attrib-uted to WM maintenance demands, because it did not

Fig. 1. Subregions of the human prefrontal cortex (PFC).The dorsolateral prefrontal cortex (DLPFC) (BA 9, 46) isshown in green, and the ventrolateral prefrontal cortex(VLPFC) (BA 44, 45, and 47/12) is shown in red. The bluedashed line indicates the approximate borders betweenthe DLPFC and VLPFC. White dashed lines indicate theapproximate borders between the subregions of theVLPFC.


significantly differ between forward 2 and forward 5 tri-als. This pattern of results has been replicated in otherstudies with verbal (DEsposito, Postle, Ballard, and oth-ers 1999; Wagner, Maril, and others 2001; Barde andThompson-Schill 2002; Blumenfeld and Ranganath2006) and spatial (Glahn and others 2002; Mohr and oth-ers 2006) stimuli, suggesting that DLPFC activation ismore specifically increased during tasks that requireorganization of information that is active in WM.

Additional evidence for the organizational account ofDLPFC function comes from neuroimaging studies ofchunking (Bor and others 2003; Bor and others 2004).For example, in one study, participants were scannedwhile memorizing a sequence of four spatial locationswithin a 4 4 grid of red squares (Bor and others 2003).Critically, on some trials, the spatial arrays exhibited acoherent structure, in that an element in the sequencewas either on the same column, row, or diagonal as theelement preceding it. The resulting sequences thereforeappeared geometrically regular and were more readilyorganized into single shapes. In contrast, on other trials,the elements within the sequence did not exhibit anygeometric structure. Behavioral performance was higheron structured trials, presumably because participantswere able to exploit the structure to organize, or chunk,the spatial sequences. Critically, DLPFC activation wasincreased on structured, as compared to unstructured,trials even though the structured task was easier. Thisfinding suggests that, rather than simply reflecting non-specific demands related to task difficulty, DLPFC acti-vation is more specifically sensitive to demands fororganizational processing.

Functional Neuroimaging of LTM Encoding

We began this review by summarizing findings indicatingthat PFC lesions may impair selection and organizationprocesses that facilitate LTM encoding. Neuroimagingstudies of cognitive control have suggested that VLPFC isrecruited during tasks that require the selection of specificitem information, whereas DLPFC is additionally recruitedduring tasks that require organizational processing. As wewill describe below, these findings may provide criticalinsights into the ways in which lateral prefrontal regionscontribute to successful LTM formation.

Subsequent Memory Effects and the PFC

Event-related fMRI studies have investigated LTMencoding by identifying subsequent memory or dif-ference due to memory effects (see Paller and Wagner2002 for review). In these studies, activation duringencoding of a particular item or set of items is analyzedas a function of later memory success or failure. Regionswhere activation is increased during encoding of itemsthat are subsequently remembered (relative to activationduring encoding of items that are subsequently forgot-ten) are thought to play a role in promoting successfulLTM formation.

As shown in Figure 2 and Table 1, studies using the subsequent memory paradigm have demonstrated

significant prefrontal involvement in LTM encoding.Inspection of the spatial distribution of activation peaks(or local maxima) from these studies also suggeststhat the degree of involvement seems to differ betweendifferent PFC subregions. The data provide overwhelm-ing support for the notion that the VLPFC contributesto LTM formation. Out of 150 local maxima associatedwith subsequent memory within the PFC, 132 fall withinthe VLPFC. Furthermore, all but two studies that


Fig. 2. Prefrontal activations reported in 37 fMRI studiesof long-term memory (LTM) formation. Each green dotrepresents an activation peak in an analysis that reportedincreased activation during encoding of items that weresubsequently remembered, as compared with items thatwere subsequently forgotten (i.e., a subsequent memoryeffect). Each red dot represents an activation peak in ananalysis that reported increased activation during encod-ing of items that were subsequently forgotten, as com-pared with items that were subsequently remembered(i.e., a subsequent forgetting effect). Of the 150 localmaxima associated with subsequent memory, only 18 fallwithin the dorsolateral prefrontal cortex (DLPFC) (BA 46and 9) compared with 132 in the ventrolateral prefrontalcortex (VLPFC) (BA 6, 44, 45, and 47). Furthermore, 10 ofthe 11 local maxima associated with subsequent forget-ting fall within the DLPFC. L = left; R = right.



Table 1. Prefrontal Activation in Studies of LTM Encoding

A. Remember >> Forgotten Contrasts

Activated Prefrontal

Study Study Description Regions

Baker and others 2001 Semantic (abstract) vs. structural (uppercase) VLPFCBlumenfeld and Ranganath 2006 Organizational processing VLPFC, DLPFCBrassen and others 2006 Item memory and free-recall VLPFC, DLPFCBuckner and others 2001 Encoding during retrieval VLPFCChee and others 2002 Word frequency VLPFCClark and Wagner 2003 Novel word learning VLPFCDavachi and others 2001 Maintenance vs. elaborative rehearsal VLPFCDolcos and others 2004 Arousal, valence, and emotion VLPFCErk and others 2003 Emotional processing VLPFCFletcher and others 2003 Levels of processing VLPFCGaroff and others 2005 Specific vs. general encoding VLPFC, DLPFCHenson and others 1999 Recollection and familiarity VLPFCJackson and Schacter 2004 Associative memory VLPFCJohnson and others 2004 Refreshing DLPFCKensinger and Schacter 2005 Emotion and reality monitoring DLPFC, VLPFCKirchhoff and others 2000 Novelty and content dependency VLPFCMacrae and others 2004 Self-referential processing VLPFCMaril and others 2003 Feeling of knowing VLPFCMorcom and others 2003 Effect of aging VLPFCOtten and Rugg 2001a Content dependency VLPFCOtten and others 2001 Levels of processing VLPFCOtten and others 2002 Item vs. task processing VLPFCPrince and others 2005 Associative memory VLPFCRanganath and others 2004 Recollection and familiarity VLPFCRaye and others 2002 Refreshing DLPFCReber and others 2002 Encoding effort VLPFCReynolds and others 2004 Item vs. task processing VLPFCSergerie and others 2005 Emotional processing of faces VLPFC, DLPFCSommer and others 2005 Object-location associations VLPFC, DLPFCSperling and others 2003 Face-name associations VLPFCStaresina and Davachi 2006 Associative memory and free recall VLPFC, DLPFCSummerfield and others 2006 Face-house associative binding VLPFC, DLPFCUncapher and Rugg 2005 Memory durability VLPFCWagner and others 1998 Semantic processing VLPFCWeis and others 2004 Temporal lobe & cerebellum during encoding VLPFCde Zubicaray and others 2005 Word frequency and strength effects VLPFC

B. Forgotten >> Remember Contrasts

PrefrontalStudy Study Description Regions

Clark and Wagner 2003 Novel word learning DLPFCDaselaar and others 2004 Forgetting VLPFC, DLPFCKensinger and Schacter 2005 Emotion and reality monitoring DLPFCOtten and Rugg 2001b Content dependency DLPFCWagner and others 1998 Semantic processing DLPFC

A, Studies that revealed local maxima in prefrontal regions where activity during encoding was greater for items that weresubsequently remembered than for items that were subsequently forgotten. B, Studies that revealed local maxima in pre-frontal regions where activity during encoding was greater for items that were subsequently forgotten than for items thatwere subsequently remembered. LTM = long-term memory; VLPFC = ventrolateral prefrontal cortex; DLPFC = dorsolateralprefrontal cortex.


reported subsequent memory effects reported local max-ima within the VLPFC. Given that the ability to selectrelevant item information is essential for many forms ofgoal-directed cognitive processing, including memoryencoding, it makes sense that the VLPFC activity shouldbe strongly linked to memory encoding in a wide varietyof behavioral contexts. These findings will be discussedin more detail below.

The imaging literature seems to tell a different storyabout the DLPFC. Out of 150 local maxima throughoutthe PFC, only 18 fall within the DLPFC. Furthermore,some studies have reported that DLPFC activation wasincreased during encoding of items that were subse-quently forgotten, as compared with those that were sub-sequently remembered (i.e., a subsequent forgettingeffect). These findings do not seem to support the notionthat the DLPFC contributes to LTM encoding. However,a close analysis of the studies summarized in Table 1reveals that the majority of imaging studies have usedencoding and retrieval tasks that deemphasize organiza-tional processing. As described below, DLPFC activa-tion should be associated with successful LTM encodingspecifically under encoding and/or retrieval conditionsthat emphasize organization.

VLPFC and Selection during LTM Encoding

As shown in Figure 2, there is overwhelming support forthe notion that the VLPFC supports LTM encoding.Behavioral research suggests that engaging selectionprocesses at the time of encoding can strongly impact thestrength or distinctiveness of a memory for the relevantitem (Craik and Lockhart 1972; Hunt and Einstein 1981).For instance, rehearsing an item (Naveh-Benjamin andJonides 1984; Greene 1987; Davachi and others 2001;Raye and others 2002; Ranganath and others 2005) orattending to its high-level or semantic features (Craik andLockhart 1972) can improve its subsequent memorability.Given that most fMRI studies emphasize encoding andretrieval of individual items, and that selection processesfigure so prominently in these contexts, it makes sensethat the VLPFC has been so consistently implicated insuccessful memory formation.

As noted in the VLPFC and Selection Processessection above, anterior regions of the VLPFC (BA47/12, 45; see Fig. 1) have been implicated in tasks thatrequire high levels of control and/or selection of seman-tic information (i.e., deep processing; cf. Craik andLockhart 1972), whereas more posterior regions of theVLPFC (BA 44,6) have been implicated in tasks thatrequire less control during selection and/or selection ofphonological or orthographic information (i.e., shal-low processing). Interestingly, some LTM encodingstudies have found corresponding patterns of subsequentmemory effects in the VLPFC. For instance, some stud-ies have found that more anterior regions of the VLPFC(BA 47/12, 45) showed subsequent memory effects foritems studied under deep encoding conditions (e.g.,Wagner and others 1998; Baker and others 2001; Ottenand others 2001; Fletcher and others 2003; Ranganathand others 2004), whereas more posterior regions of the

VLPFC (BA 44, 6) showed equivalent subsequent mem-ory effects following both deep and shallow encoding(Baker and others 2001; Fletcher and others 2003).However, the literature is not consistent in this regard, asseveral studies have failed to find such a dissociation inthe PFC (Baker and others 2001; Davachi and others2001; Otten and others 2001; Otten and Rugg 2001a; butsee Fletcher and others 2003). Thus, more research isneeded to further characterize how VLPFC subregionscontribute to LTM encoding.

Until now, we have highlighted how selection processes,putatively implemented by the VLPFC, may support suc-cessful LTM formation in a very wide range of behav-ioral contexts. However, there are some situations inwhich selection processing may be deleterious to mem-ory formation. For example, if selection processes directattention away from items that are to be encoded,VLPFC activation can be negatively correlated with sub-sequent memory performance (Otten and others 2002;Reynolds and others 2004). In one such study, partici-pants were scanned while performing several blocks ofencoding trials, and subsequent memory was tested out-side of the scanner (Otten and others 2002). Consistentwith many previous studies, VLPFC activity wasincreased during encoding of items that were later remem-bered, as compared to those that were later forgotten.However, in this report, a different analysis method wasused to examine sustained activity across each block oftrials that was not specifically related to item processing.Interestingly, this sustained activity in the VLPFC acrosseach block of encoding trials was negatively correlatedwith subsequent memory for the items in each block.This suggests that when processing is directed awayfrom specific items, VLPFC activity will be negativelycorrelated with subsequent memory performance.

DLPFC and Organization in LTM Encoding

As mentioned above, organizational processing duringencoding is thought to be an important determinant oflater memory (Bower 1970; Sternberg and Tulving 1977;Hunt and Einstein 1981), and patients with lesions thatinclude the DLPFC are thought to have deficits in sponta-neous usage of organizational strategies. Given that imag-ing studies have implicated the DLPFC in controlprocesses that support organization in WM, it is surpris-ing that few studies have reported DLPFC subsequentmemory effects and that some studies have reported sub-sequent forgetting effects in the DLPFC. This pattern offindings is consistent with at least two possibilities: 1) theDLPFC implements processes that do not typically pro-mote successful LTM formation or 2) most previous stud-ies were insensitive to detect the way in which the DLPFCcontributes to LTM formation. Relevant to the latter pos-sibility, most LTM encoding studies have used encodingtasks that orient attention toward specific attributes of sin-gle items and away from relationships between items. Inthese studies, engaging in organizational processing dur-ing encoding was irrelevant or possibly deleterious to latermemory performance (i.e., because allocating resourcestoward processing the relationships among items might




take attentional resources away from processing the dis-tinct features of the items themselves; Bower 1970; Begg1979; McGee 1980). Accordingly, it is possible that theencoding conditions in many previous imaging studieswere not conducive to revealing subsequent memoryeffects in the DLPFC. The retrieval tests used in subse-quent memory paradigms might also be a relevant factor.Most imaging studies of encoding assess LTM with testsof item recognition memory (see Box 1). However, organ-ization facilitates memory by enhancing associationsamong items (Bower 1970; Begg 1979; McGee 1980).Thus, averaging encoding activation as a function of sub-sequent item recognition performance might mask therole of the DLPFC in successful LTM encoding.

If the DLPFC contributes to LTM encoding through itsrole in organizational processing, the ability to detect thiscontribution may depend on the kinds of encoding andretrieval tasks that are used. Indeed, in studies that usedencoding tasks that encouraged formation of inter-itemassociations (Sommer and others 2005; Summerfield andMangels 2005; Blumenfeld and Ranganath 2006), orstudies that used retrieval tests that are sensitive to mem-ory for associations among items (Brassen and others

2006; Staresina and Davachi 2006), DLPFC activity dur-ing encoding was positively correlated with subsequentmemory performance.

In one such study (Blumenfeld and Ranganath 2006),we found evidence that DLPFC activation is related tosuccessful LTM encoding specifically under conditionsthat emphasize organization. In this study, participantswere scanned while they performed two WM tasks: Onrehearse trials, participants were presented with a set ofthree words and were required to maintain the set acrossa 12-second delay period, in anticipation of a questionprobing memory for the identity and serial position of theitems (see Fig. 3). On reorder trials, participants wererequired to rearrange a set of three words based on theweight of the object that each word referred to. Theymaintained this information across a 12-second delayperiod (see Fig. 3) in anticipation of a question probingmemory for serial order of the items in the rearranged set.Although both rehearse and reorder trials required main-tenance of the three-item set, reorder trials additionallyrequired participants to compare the items in the set andtransform their serial order. Consequently, reorder trialsrequired substantial organizational processing on each

Fig. 3. Dorsolateral prefrontal cortex (DLPFC) activity during working memory (WM) organization predicts successfullong-term memory (LTM) formation (Blumenfeld and Ranganath 2006). (a) Schematic depiction of the two tasks per-formed during fMRI scanning. (b) Behavioral results, showing that participants recalled significantly more triplets fromeach reorder trial (yellow) than would be expected based on the overall hit rate. This finding suggests that, on reordertrials, memory performance was supported by associations among the items in the memory set. (c) fMRI data showingthat DLPFC (top) activation was increased during the delay period of reorder trials for which two to three items weresubsequently remembered (solid yellow), relative to trials in which zero to one items were remembered (dashed yellow).No such effect is seen on rehearse trials (gray lines). At bottom, activation in a region of the posterior ventrolateral pre-frontal cortex (pVLPFC) is plotted, showing that delay period activation in this region during both rehearse and reordertrials was predictive of subsequent memory performance.



trial, in addition to WM maintenance. Analyses of resultsfrom a postscan recognition memory test showed thatthere were significantly more reorder trials in which allthree items were recollected than would be expectedbased on the overall item hit-rates alone (see Fig. 3). Incontrast, the proportion of rehearse trials on which allthree items were subsequently recollected was no differ-ent than would be expected by the item hit-rates. Thesefindings suggest that active organization of the memoryset on reorder trials resulted in successful encoding of theassociations among the items in each set.

Consistent with the notion that the DLPFC is involvedin organization in WM, analyses of fMRI data revealedthat DLPFC activation was increased during reorder trials,as compared with rehearse trials (Fig. 3). Furthermore,DLPFC activation during reorder, but not rehearse, trialswas positively correlated with subsequent memory per-formance. Specifically, DLPFC activation was increasedduring the delay period of reorder trials for which two tothree items were later recollected, relative to trials forwhich one or no items were later recollected. Critically,no such relationship was evident during rehearse trials. Incontrast, activation in a posterior region of the left VLPFC(BA 44, 6) was correlated with subsequent memory per-formance on both rehearse and reorder trials. Thus, resultsfrom this study suggest that DLPFC activation promotessuccessful LTM formation through its role in organiza-tional processing during encoding.

A study by Summerfield and colleagues (2006) pro-vides further evidence for the role of the DLPFC in LTMformation. In this study, participants learned associa-tions between faces and houses by using a mnemonic(he lives in this house). Results showed that DLPFCand VLPFC activation was enhanced during encoding ofpairs that were subsequently remembered. In addition,posterior regions thought to be involved in perception offaces (the fusiform face area; Puce and others 1995;Kanwisher and others 1997) and houses (the parahip-pocampal place area; Aguirre and others 1998; Epsteinand Kanwisher 1998) showed subsequent memoryeffects. Functional connectivity analyses demonstratedthat the correlation between the DLPFC, FFA, and thePPA was increased on trials that led to successful mem-ory for the face-house pairing. This suggests that theDLPFC may promote memory for associations throughincreased connectivity with regions that represent theitems that are to be associated.

Results from another recent study demonstrated the spe-cific nature of DLPFC contributions to memory encodingby comparing the relationship between activation and sub-sequent performance on free recall and item recognitionmemory tests (Staresina and Davachi 2006). As describedabove, recognition tests are often insensitive to detectingthe presence of inter-item associations in LTM, whereasrecall performance is significantly influenced by organiza-tion during encoding (see Box 1). Critically, DLPFC acti-vation was specifically enhanced during encoding of itemsthat were recalled compared to those that were not.DLPFC activation was not correlated with subsequent item

recognition performance. In contrast, encoding time acti-vation in the VLPFC was positively correlated with subse-quent memory performance on both the recall and therecognition tests. These results are consistent with the ideathat DLPFC activation will contribute to subsequent mem-ory performance specifically under retrieval conditionsthat are sensitive to memory for associations among items.

Before concluding, it is important to address potentialchallenges to our proposal that the DLPFC contributes toLTM formation via its role in organization. A few studieshave reported increased DLPFC during processing ofitems that were subsequently forgotten, relative to itemsthat were subsequently remembered. Importantly, manyof the studies that reported forgetting effects examinedencoding of single items and used encoding and retrievaltasks that directed processing toward specific features ofthe individual items (Otten and Rugg 2001b; Wagner andDavachi 2001; Clark and Wagner 2003; Daselaar and oth-ers 2004; Kensinger and Schacter 2005). As mentionedabove, these conditions deemphasize organizational pro-cessing, which according to our proposal, will reduce thecontribution of the DLPFC in LTM formation. Moreover,to the extent that organizational processing competes withitem encoding (Begg 1979; McGee 1980; but see Hockleyand Cristi 1996), DLPFC forgetting effects may reflecttask-irrelevant processing that interferes with successfulmemory formation. Indeed, during item-specific encodingconditions, organizational processing (and hence DLPFCactivity) may need to be inhibited to promote item mem-ory (Daselaar and others 2004). Future research shouldtest this intriguing yet speculative proposal.

Conclusion

In conclusion, we have reviewed evidence suggestingthat different regions of the PFC implement cognitivecontrol processes that support successful LTM encoding.Neuropsychological research has suggested that the PFCsupports LTM encoding by supporting controlled selec-tion of goal-relevant item information as well as organi-zation among items. Neuroimaging studies havesuggested that these cognitive control processes maydepend upon distinct regions of the PFC. For instance,studies investigating resolution of proactive interferenceand semantic categorization suggest that the VLPFCmay be disproportionately involved in controlled selec-tion of item information. Studies investigating manipu-lation, transformation, or chunking of multiple items inWM provide evidence for the notion that the DLPFCmay be disproportionately involved in organizationalprocessing. Recent imaging studies suggest that theroles of these prefrontal regions in cognitive control aredirectly linked to their roles in LTM encoding. Morespecifically, results from functional imaging studiesinvestigating LTM suggest that the VLPFC may supportLTM formation through its role in selecting relevantitem information, whereas the DLPFC may supportLTM by building associations among items that areactive in memory.


References

Aguirre GK, Zarahn E, DEsposito M. 1998. Neural components oftopographical representation. Proc Natl Acad Sci U S A 95:83946.

Alexander MP, Stuss DT, Fansabedian N. 2003. California VerbalLearning Test: performance by patients with focal frontal and non-frontal lesions. Brain 126:1493503.

Aron AR, Robbins TW, Poldrack RA. 2004. Inhibition and the rightinferior frontal cortex. Trends Cogn Sci 8:1707.

Badre D, Poldrack RA, Pare-Blagoev EJ, Insler RZ, Wagner AD. 2005.Dissociable controlled retrieval and generalized selection mecha-nisms in ventrolateral prefrontal cortex. Neuron 47:90718.

Badre D, Wagner AD. 2005. Frontal lobe mechanisms that resolveproactive interference. Cereb Cortex 15:200312.

Baker JT, Sanders AL, Maccotta L, Buckner RL. 2001. Neural corre-lates of verbal memory encoding during semantic and structuralprocessing tasks. Neuroreport 12:12516.

Barde LH, Thompson-Schill SL. 2002. Models of functional organi-zation of the lateral prefrontal cortex in verbal working memory:evidence in favor of the process model. J Cogn Neurosci 14:105463.

Barnes JM, Underwood BJ. 1959. Fate of first list associations intransfer theory. J Exp Psychol 58:97105.

Begg I. 1979. Trace loss and the recognition failure of unrecalledwords. Mem Cognit 7:11323.

Blumenfeld RS, Ranganath C. 2006. Dorsolateral prefrontal cortexpromotes long-term memory formation through its role in workingmemory organization. J Neurosci 26:91625.

Bokde AL, Tagamets MA, Friedman RB, Horwitz B. 2001. Functionalinteractions of the inferior frontal cortex during the processing ofwords and word-like stimuli. Neuron 30:60917.

Bor D, Cumming N, Scott CE, Owen AM. 2004. Prefrontal corticalinvolvement in verbal encoding strategies. Eur J Neurosci19:336570.

Bor D, Duncan J, Wiseman RJ, Owen AM. 2003. Encoding strategiesdissociate prefrontal activity from working memory demand.Neuron 37:3617.

Bower GH. 1970. Imagery as a relational organizer in associativelearning. J Verbal Learn Verbal Behav 9:52933.

Brassen S, Weber-Fahr W, Sommer T, Lehmbeck JT, Braus DF. 2006.Hippocampal-prefrontal encoding activation predicts whetherwords can be successfully recalled or only recognized. BehavBrain Res 171:2718.

Buckner RL. 2003. Functional-anatomic correlates of controlprocesses in memory. J Neurosci 23:39994004.

Buckner RL, Wheeler ME, Sheridan MA. 2001. Encoding processesduring retrieval tasks. J Cogn Neurosci 13:40615.

Bunge SA, Ochsner KN, Desmond JE, Glover GH, Gabrieli JD. 2001.Prefrontal regions involved in keeping information in and out ofmind. Brain 124:207486.

Bunge SA, Wendelken C, Badre D, Wagner AD. 2005. Analogical rea-soning and prefrontal cortex: evidence for separable retrieval andintegration mechanisms. Cereb Cortex 15:23949.

Cannon TD, Glahn DC, Kim J, Van Erp TG, Karlsgodt K, Cohen MS,and others. 2005. Dorsolateral prefrontal cortex activity duringmaintenance and manipulation of information in working memoryin patients with schizophrenia. Arch Gen Psychiatry 62:107180.

Chee MW, Westphal C, Goh J, Graham S, Song AW. 2003. Word fre-quency and subsequent memory effects studied using event-relatedfMRI. Neuroimage 20:1042-51.

Clark D, Wagner AD. 2003. Assembling and encoding word represen-tations: fMRI subsequent memory effects implicate a role forphonological control. Neuropsychologia 41:30417.

Craik FIM, Lockhart RS. 1972. Levels of processing: a framework formemory research. J Verbal Learn Verbal Behav 12:599607.

Crone EA, Wendelken C, Donohue S, van Leijenhorst L, Bunge SA.2006. Neurocognitive development of the ability to manipulateinformation in working memory. Proc Natl Acad Sci U S A103:931520.

DEsposito M, Postle BR, Ballard D, Lease J. 1999. Maintenance ver-sus manipulation of information held in working memory: anevent-related fMRI study. Brain Cogn 41:6686.

DEsposito M, Postle BR, Jonides J, Smith EE. 1999. The neural sub-strate and temporal dynamics of interference effects in workingmemory as revealed by event-related functional MRI. Proc NatlAcad Sci U S A 96:751419.

Daselaar SM, Prince SE, Cabeza R. 2004. When less means more:deactivations during encoding that predict subsequent memory.Neuroimage 23:9217.

Davachi L, Maril A, Wagner AD. 2001. When keeping in mind supportslater bringing to mind: neural markers of phonological rehearsal pre-dict subsequent remembering. J Cogn Neurosci 13:105970.

de Zubicaray GI, McMahon KL, Eastburn MM, Finnigan S,Humphreys MS. 2005. fMRI evidence of word frequency andstrength effects during episodic memory encoding. Brain ResCogn Brain Res 22:43950.

della Rocchetta AI. 1986. Classification and recall of pictures afterunilateral frontal or temporal lobectomy. Cortex 22:189211.

della Rocchetta AI, Milner B. 1993. Strategic search and retrievalinhibitionthe role of the frontal lobes. Neuropsychologia 31:50324.

Dimitrov M, Granetz J, Peterson M, Hollnagel C, Alexander G,Grafman J. 1999. Associative learning impairments in patientswith frontal lobe damage. Brain Cogn 41:21330.

Dobbins IG, Foley H, Schacter DL, Wagner AD. 2002. Executive con-trol during episodic retrieval: multiple prefrontal processes sub-serve source memory. Neuron 35:98996.

Dobbins IG, Wagner AD. 2005. Domain-general and domain-sensitiveprefrontal mechanisms for recollecting events and detecting nov-elty. Cereb Cortex 15:176878.

Dolcos F, LaBar KS, Cabeza R. 2004. Dissociable effects of arousal andvalence on prefrontal activity indexing emotional evaluation andsubsequent memory: an event-related fMRI study. Neuroimage23:6474.

Ebbinghaus H. 1885/1964. Memory: a contribution to experimentalpsychology. New York: Dover.

Epstein R, Kanwisher N. 1998. A cortical representation of the localvisual environment. Nature 392:598601.

Erk S, Kiefer M, Grothe J, Wunderlich AP, Spitzer M. 2003. Emotionalcontext modulates subsequent memory effect. Neuroimage 18:43947.

Eslinger PJ, Grattan LM. 1994. Altered serial position learning afterfrontal lobe lesion. Neuropsychologia 32:72939.

Fletcher PC, Henson RN. 2001. Frontal lobes and human memory:insights from functional neuroimaging. Brain 124:84981.

Fletcher PC, Stephenson CM, Carpenter TA, Donovan T, BullmorelET. 2003. Regional brain activations predicting subsequent mem-ory success: an event-related fMRI study of the influence of encod-ing tasks. Cortex 39:100926.

Garoff RJ, Slotnick SD, Schacter DL. 2005. The neural origins of spe-cific and general memory: the role of the fusiform cortex.Neuropsychologia 43:84759.

Gershberg FB, Shimamura AP. 1995. Impaired use of organizationalstrategies in free-recall following frontal lope damage.Neuropsychologia 33:130533.

Glahn DC, Kim J, Cohen MS, Poutanen VP, Therman S, Bava S, and oth-ers. 2002. Maintenance and manipulation in spatial working memory:dissociations in the prefrontal cortex. Neuroimage 17:20113.

Gold BT, Balota DA, Kirchhoff BA, Buckner RL. 2005. Common anddissociable activation patterns associated with controlled semanticand phonological processing: evidence from fMRI adaptation.Cereb Cortex 15:143850.

Gold BT, Buckner RL. 2002. Common prefrontal regions coactivatewith dissociable posterior regions during controlled semantic andphonological tasks. Neuron 35:80312.

Greene RL. 1987. Effects of maintenance rehearsal on human-memory.Psychol Bull 102:40313.

Henson RNA, Rugg MD, Shallice T, Josephs O, Dolan RJ. 1999.Recollection and familiarity in recognition memory: an event-relatedfunctional magnetic resonance imaging study. J Neurosci 19:396272.

Hirst W, Volpe BT. 1988. Memory strategies with brain damage. BrainCogn 8:379408.

Hockley WE, Cristi C. 1996. Tests of encoding tradeoffs between itemand associative information. Mem Cognit 24:20216.




Hunt RR, Einstein GO. 1981. Relational and item-specific informationin memory. J Verbal Learn Verbal Behav 20:497514.

Jackson O 3rd, Schacter DL. 2004. Encoding activity in anteriormedial temporal lobe supports subsequent associative recognition.Neuroimage 21:45662.

Janowsky JS, Shimamura AP, Squire LR. 1989a. Memory andmetamemory: comparisons between patients with frontal lobelesions and amnesic patients. Psychobiology 17:311.

Janowsky JS, Shimamura AP, Squire LR. 1989b. Source memoryimpairment in patients with frontal lobe lesions. Neuropsychologia27:104356.

Jetter W, Poser U, Freeman RB Jr, Markowitsch HJ. 1986. A verballong term memory deficit in frontal lobe damaged patients. Cortex22:22942.

Johnson MK, Mitchell KJ, Raye CL, Greene EJ. 2004. An age-relateddeficit in prefrontal cortical function associated with refreshinginformation. Psychol Sci 15:12732.

Jonides J, Nee DE. 2006. Brain mechanisms of proactive interferencein working memory. Neuroscience 139:18193.

Jonides J, Smith EE, Marshuetz C, Koeppe RA, Reuter-Lorenz PA.1998. Inhibition in verbal working memory revealed by brain acti-vation. Proc Natl Acad Sci U S A 95:841013.

Kanwisher N, McDermott J, Chun MM. 1997. The fusiform face area:a module in human extrastriate cortex specialized for face percep-tion. J Neurosci 17:430211.

Kensinger EA, Schacter DL. 2005. Emotional content and reality-monitoring ability: fMRI evidence for the influences of encodingprocesses. Neuropsychologia 43:142943.

Kesner RP, Hopkins RO, Fineman B. 1994. Item and order dissociationin humans with prefrontal cortex damage. Neuropsychologia32:88191.

Kirchhoff BA, Wagner AD, Maril A, Stern CE. 2000. Prefrontal-temporalcircuitry for episodic encoding and subsequent memory. J Neurosci20:617380.

Luria AR. 1966. Higher cortical functions in man. New York: BasicBooks.

Luria AR. 1973. The working brain. New York: Basic Books.Macrae CN, Moran JM, Heatherton TF, Banfield JF, Kelley WM.

2004. Medial prefrontal activity predicts memory for self. CerebCortex 14:64754.

Maril A, Simons JS, Mitchell JP, Schwartz BL, Schacter DL. 2003.Feeling-of-knowing in episodic memory: an event-related fMRIstudy. Neuroimage 18:82736.

McAndrews MP, Milner B. 1991. The frontal cortex and memory fortemporal order. Neuropsychologia 29:84959.

McDermott KB, Petersen SE, Watson JM, Ojemann JG. 2003. A pro-cedure for identifying regions preferentially activated by attentionto semantic and phonological relations using functional magneticresonance imaging. Neuropsychologia 41:293303.

McGee R. 1980. Imagery and recognition memory: the effects of rela-tional organization. Mem Cognit 8:3949.

Mecklinger A, Weber K, Gunter TC, Engle RW. 2003. Dissociablebrain mechanisms for inhibitory control: effects of interferencecontent and working memory capacity. Brain Res Cogn Brain Res18:2638.

Miller EK, Cohen JD. 2001. An integrative theory of prefrontal cortexfunction. Annu Rev Neurosci 24:167202.

Miller GA, Galanter E, Pribram KH. 1960. Plans and the structure ofbehavior. New York: Holt, Rinehart and Winston.

Milner B. 1962. Some cognitive effects of frontal lobe lesions in man.Philos Trans R Soc Lond B 296:21126.

Milner B, Petrides M, Smith ML. 1985. Frontal lobes and the tempo-ral organization of memory. Hum Neurobiol 4:13742.

Milner B, Corsi P, Leonard G. 1991. Frontal-lobe contribution torecency judgements. Neuropsychologia 29:60118.

Mohr HM, Goebel R, Linden DE. 2006. Content- and task-specificdissociations of frontal activity during maintenance and manipula-tion in visual working memory. J Neurosci 26:446571.

Monsell S. 1978. Recency, immediate recognition, and reaction time.Cogn Psychol 10:465501.

Morcom AM, Good CD, Frackowiak RS, Rugg MD. 2003. Age effectson the neural correlates of successful memory encoding. Brain126:21329.

Moscovitch M, Melo B. 1997. Strategic retrieval and the frontal lobes:evidence from confabulation and amnesia. Neuropsychologia35:101734.

Moscovitch M, Winocur G. 1995. Frontal lobes, memory, and aging.Ann N Y Acad Sci 769:11950.

Naveh-Benjamin M, Jonides J. 1984. Maintenance rehearsal: a two-component analysis. J Exp Psychol Learn Mem Cogn 10:36985.

Nelson JK, Reuter-Lorenz PA, Sylvester CY, Jonides J, Smith EE.2003. Dissociable neural mechanisms underlying response-basedand familiarity-based conflict in working memory. Proc Natl AcadSci U S A 100:111715.

Otten LJ, Henson RN, Rugg MD. 2001. Depth of processing effects on neural correlates of memory encoding: relationship between find-ings from across- and within-task comparisons. Brain 124:399412.

Otten LJ, Henson RN, Rugg MD. 2002. State-related and item-relatedneural correlates of successful memory encoding. Nat Neurosci5:133944.

Otten LJ, Rugg MD. 2001a. Task-dependency of the neural correlates ofepisodic encoding as measured by fMRI. Cereb Cortex 11:115060.

Otten LJ, Rugg MD. 2001b. When more means less: neural activityrelated to unsuccessful memory encoding. Curr Biol 11:152830.

Paller KA, Wagner AD. 2002. Observing the transformation of experi-ence into memory. Trends Cogn Sci 6:93102.

Perret E. 1974. The left frontal lobe and the suppression of habitualresponses in verbal categorical behavior. Neuropsychologia12:32330.

Petrides M. 2000a. Dissociable roles of mid-dorsolateral prefrontaland anterior inferotemporal cortex in visual working memory. JNeurosci 20:7496503.

Petrides M. 2000b. The role of the mid-dorsolateral prefrontal cortexin working memory. Exp Brain Res 133:4454.

Petrides M, Milner B. 1982. Deficits on subject-ordered tasks after frontal-and temporal-lobe lesions in man. Neuropsychologia 20:24962.

Poldrack RA, Wagner AD, Prull MW, Desmond JE, Glover GH, GabrieliJD. 1999. Functional specialization for semantic and phonologicalprocessing in the left inferior prefrontal cortex. Neuroimage 10:1535.

Postle BR, Berger JS, DEsposito M. 1999. Functional neuroanatomi-cal double dissociation of mnemonic and executive controlprocesses contributing to working memory performance. Proc NatlAcad Sci U S A 96:1295964.

Postle BR, Brush LN. 2004. The neural bases of the effects of item-nonspecific proactive interference in working memory. CognAffect Behav Neurosci 4:37992.

Postle BR, Brush LN, Nick AM. 2004. Prefrontal cortex and the medi-ation of proactive interference in working memory. Cogn AffectBehav Neurosci 4:6008.

Postle BR, DEsposito M. 1999. What-Then-Where in visual workingmemory: an event-related fMRI study. J Cogn Neurosci 11:58597.

Prince SE, Daselaar SM, Cabeza R. 2005. Neural correlates of rela-tional memory: successful encoding and retrieval of semantic andperceptual associations. J Neurosci 25:120310.

Puce A, Allison T, Gore JC, McCarthy G. 1995. Face-sensitive regionsin human extrastriate cortex studied by functional MRI. JNeurophysiol 74:11929.

Ranganath C, Cohen MX, Brozinsky CJ. 2005. Working memorymaintenance contributes to long-term memory formation: neuraland behavioral evidence. J Cogn Neurosci 17:9941010.

Ranganath C, Knight RT. 2003. Prefrontal cortex and episodic mem-ory: integrating findings from neuropsychology and event-relatedfunctional neuroimaging. In: Parker A, Wilding EL, Bussey TJ,editors. Memory encoding and retrieval: a cognitive neuroscienceperspective. New York: Psychology Press. p 8399.

Ranganath C, Yonelinas AP, Cohen MX, Dy CJ, Tom SM, DEspositoM. 2004. Dissociable correlates of recollection and familiaritywithin the medial temporal lobes. Neuropsychologia 42:213.

Raye CL, Johnson MK, Mitchell KJ, Reeder JA, Greene EJ. 2002.Neuroimaging a single thought: dorsolateral PFC activity associated



with refreshing just-activated information. Neuroimage 15:44753.

Reber PJ, Wong EC, Buxton RB. 2002. Encoding activity in the medialtemporal lobe examined with anatomically constrained fMRI analy-sis. Hippocampus 12:36376.

Reynolds JR, Donaldson DI, Wagner AD, Braver TS. 2004. Item- andtask-level processes in the left inferior prefrontal cortex: positiveand negative correlates of encoding. Neuroimage 21:147283.

Roskies AL, Fiez JA, Balota DA, Raichle ME, Petersen SE. 2001.Task-dependent modulation of regions in the left inferior frontalcortex during semantic processing. J Cogn Neurosci 13:82943.

Sergerie K, Lepage M, Armony JL. 2005. A face to remember: emo-tional expression modulates prefrontal activity during memory for-mation. Neuroimage 24:5805.

Shimamura AP. 1995. Memory and the prefrontal cortex. Ann N YAcad Sci 769:1519.

Shimamura AP, Janowsky JS, Squire LR. 1990. Memory for the tempo-ral order of events in patients with frontal lobe lesions and amnesicpatients. Neuropsychologia 28:80313.

Shimamura AP, Jurica PJ, Mangels JA, Gershberg FB, Knight RT.1995. Susceptibility to memory interference effects followingfrontal lobe damage: findings from tests of paired-associate learn-ing. J Cogn Neurosci 7:14452.

Sommer T, Rose M, Weiller C, Bchel C. 2005. Contributions ofoccipital, parietal and parahippocampal cortex to encoding ofobject-location associations. Neuropsychologia 43:73243.

Sperling R, Chua E, Cocchiarella A, Rand-Giovannetti E, Poldrack R,Schacter DL, and others. 2003. Putting names to faces: successfulencoding of associative memories activates the anterior hippocam-pal formation. Neuroimage 20:140010.

Stanhope N, Guinan E, Kopelman MD. 1998. Frequency judgementsof abstract designs by patients with diencephalic, temporal lobe orfrontal lobe lesions. Neuropsychologia 36:138796.

Staresina BP, Davachi L. 2006. Differential encoding mechanisms forsubsequent associative recognition and free recall. J Neurosci26:916272.

Stern CE, Owen AM, Tracey I, Look RB, Rosen BR, Petrides M. 2000.Activity in ventrolateral and mid-dorsolateral prefrontal cortex dur-ing nonspatial visual working memory processing: evidence fromfunctional magnetic resonance imaging. Neuroimage 11:3929.

Sternberg RJ, Tulving E. 1977. Measurement of subjective organiza-tion in free-recall. Psychol Bull 84:53956.

Stuss D, Benson DF, eds. 1986. The frontal lobes. New York: Raven.

Stuss DT, Alexander MP, Palumbo CL, Buckle L, Sayer L, Pogue J. 1994.Organizational strategies of patients with unilateral or bilateral frontallobe injury in word list learning tasks. Neuropsychology 8:35573.

Summerfield C, Greene M, Wager T, Egner T, Hirsch J, Mangels J.2006. Neocortical connectivity during episodic memory formation.PLoS Biol 4:e128.

Summerfield C, Mangels JA. 2005. Functional coupling betweenfrontal and parietal lobes during recognition memory. Neuroreport16:11722.

Swick D, Knight RT. 1996. Is prefrontal cortex involved in cued recall?A neuropsychological test of PET findings. Neuropsychologia34:101928.

Thompson-Schill SL, Bedny M, Goldberg RF. 2005. The frontal lobesand the regulation of mental activity. Curr Opin Neurobiol 15:21924.

Thompson-Schill SL, DEsposito M, Aguirre GK, Farah MJ. 1997.Role of left inferior prefrontal cortex in retrieval of semanticknowledge: a reevaluation. Proc Natl Acad Sci U S A 94:147927.

Uncapher MR, Rugg MD. 2005. Encoding and the durability ofepisodic memory: a functional magnetic resonance imaging study.J Neurosci 25:72607.

Veltman DJ, Rombouts SA, Dolan RJ. 2003. Maintenance versusmanipulation in verbal working memory revisited: an fMRI study.Neuroimage 18:24756.

Vilkki J, Servo A, Surma-aho O. 1998. Word list learning and predictionof recall after frontal lobe lesions. Neuropsychology 12:26877.

Wagner AD, Davachi L. 2001. Cognitive neuroscience: forgetting ofthings past. Curr Biol 11:R9647.

Wagner AD, Maril A, Bjork RA, Schacter DL. 2001. Prefrontal con-tributions to executive control: fMRI evidence for functional dis-tinctions within lateral prefrontal cortex. Neuroimage 14:133747.

Wagner AD, Pare-Blagoev EJ, Clark J, Poldrack RA. 2001.Recovering meaning: left prefrontal cortex guides controlledsemantic retrieval. Neuron 31:32938.

Wagner AD, Schacter DL, Rotte M, Koutstaal W, Maril A, Dale AM,and others. 1998. Building memories: remembering and forgettingof verbal experiences as predicted by brain activity. Science 281:118891.

Weis S, Klaver P, Reul J, Elger CE, Fernandez G. 2004. Temporal andcerebellar brain regions that support both declarative memory for-mation and retrieval. Cereb Cortex 14:25667.

Wheeler MA, Stuss DT, Tulving E. 1995. Frontal lobe damage producesepisodic memory impairment. J Int Neuropsychol Soc 1:52536.


/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (U.S. Web Coated \050SWOP\051 v2) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown

/SyntheticBoldness 1.000000 /Description >>> setdistillerparams> setpagedevice

neurology_prefrontal cortex and long-term memory encoding an integrative review of findings from...

Documents