What does neuroimaging tell us about the role of

seminars in

THE NEUROSCIENCES, Vol 8, 1996: pp 47–55

What does neuroimaging tell us about the role of prefrontal cortex in memory retrieval? Randy L. Buckner and Steven E. Petersen

certain kinds of long-term memory retrieval, the discussion will focus on this area of memory research. For reviews of neuroimaging work on memory storage and working memory see refs 12 and 13. The discussion is divided into two parts. The first section deals with conceptual considerations and how these relate to neuroimaging data. The remaining sections deal with recent neuroimaging findings and how these inform us about the role of the prefrontal cortex in memory retrieval.

The past five years have seen an outpouring of neuroimaging studies of memory — using both positron emission tomography (PET) and functional magnetic resonance imaging (fMRI). These studies have convincingly demonstrated that neuroimaging can be used to study the functional anatomy of normal human memory and that neuroimaging can precisely localize memory related brain activations within small areas of cortex. To illustrate the application of neuroimaging in the study of memory, this review shows how several laboratories have produced data that converge on the notion that specific areas in the prefrontal cortex are active during long-term memory retrieval. Moreover, the data further suggest that distinct prefrontal brain areas might make differential contributions to different kinds of long-term memory retrieval.

Conceptual considerations Important to a discussion of prefrontal contributions to memory functions are two ideas that have become well accepted in recent years, one in relation to the organization of the prefrontal cortex, and the other in relation to cognitive memory function. With regard to prefrontal organization, it has become clear that multiple subregions exist within the larger area we label the prefrontal cortex. This has been demonstrated in the rhesus monkey by showing that localized prefrontal areas have distinct cytoarchitectonic structure, labeling patterns, physiological properties, and receive projections from other areas of cortex in a specific fashion.1,14-16 The multiple areas within monkey prefrontal cortex are likely to have distinct processing roles — a property that almost certainly extends to humans. Already, human prefrontal cortex has been tentatively subdivided based on anatomy,17 and lesions to different regions of the prefrontal cortex are associated with distinct behavioral outcomes.1 Thus, it is essential that the human prefrontal cortex be treated as a region containing multiple areas. Our challenge, then, is to use neuroimaging to aid in identifying these individual areas, in determining their processing roles, and in understanding how their individual roles might collectively contribute to memory task performance. Before attempting to use neuroimaging data in this manner, it is important to demonstrate the limits of

Key words: neuroimaging / PET / fMRI / memory / prefrontal cortex / semantic memory / episodic memory / declarative memory ©1996 Academic Press Ltd

ALTHOUGH THE PREFRONTAL CORTEX has long been accepted as an important brain region involved in short-term (working) memory,1,2 a flurry of recent neuroimaging studies (both PET and fMRI) have generated excitement about the possibility that the prefrontal cortex participates in a variety of memory processes — including long-term memory. The idea that areas in the frontal cortex contribute to human long-term memory performance has been suggested previously, based on neuropsychological studies.3,4 Neuroimaging studies, reporting prefrontal cortex activation during various long-term memory tasks, have focused interest on this possibility.5-11 In the following discussion, we will describe some of these findings and their implications. Since the most novel aspect of these studies has been the demonstration that specific prefrontal cortex regions participate in From the Departments of Neurology and Neurological Surgery and Radiology, Washington University School of Medicine, St Louis, MO 63110, USA ©1996 Academic Press Ltd 1044-5765/96/010047 + 09 $12.00/0

47

R. L. Buckner and S. E. Petersen neuroimaging to localize and resolve memory-related brain activations within subregions of the prefrontal cortex. Localization refers to the accuracy with which an activation can be placed in a specified space; resolution refers to the distance by which two regions of activation must be separated before they can be identified as separate responses. Although many theoretical models and imaging data suggest the localization limits of neuroimaging data, we feel it is also important to demonstrate the empirical boundaries derived from data relevant to that presented in this review. For this demonstration,

we turn to an instance in which nearly the same memory conditions were studied across three independent experiments.5 This study allowed us to observe independent subject-groups manifesting the same physiologic activation (Figure 1A). As can be seen, the localization of the responses across groups was remarkably consistent suggesting that current neuroimaging methods yield data that can be reliably localized to within 10 mm range across experiments. Since it is equally essential to establish that neuroimaging can detect activations originating from

Figure 1. Schematic diagrams show between subject-group and between task data in order to illustrate the ability of PET neuroimaging to reliably localize and resolve prefrontal brain activations. (A) Two activations, one in the left hemisphere (squares) and one in the right hemisphere (circles), are shown across three independent subject groups.5 Each symbol represents data from a single subject group. Across groups, the activations are reliably localized to small regions of prefrontal cortex. (B) Two functionally distinct areas of left prefrontal activation are shown that were reliably dissociated across tasks using PET.19 This level of resolution probably represents an upper boundary of the ability of functional neuroimaging to resolve activations originating from distinct prefrontal areas.

48

Memory retrieval and prefrontal cortex Table 1. Different kinds of memory retrieval discussed in this review Memory type

Other name

Examples

Semantic

Fact memory

Episodic

Event memory

Retrieval of foods that fit the category vegetable. Retrieval of the capital of France. Retrieval of words that begin with ‘L’. Recollection of what article you read yesterday. Recollection of the vegetable you ate for dinner. Recollection of words presented on a study list.

Both semantic and episodic memories are types of conscious (or declarative52) memories. They differ with regard to the type of information being retrieved. Retrieval of semantic memory involves accessing facts or pieces of information independently of where or when the information was learned. Episodic retrieval involves recollecting information that was learned and is associated with a specific point in time and context. Some researchers believe the two forms of memory are quite distinct21 while others believe they form a continuum with episodic memories relying on richer contextual information53.

conscious (or explicit) forms of memory. Learning to skillfully ride a bike takes months of practice and does not continuously require conscious retrieval of how to balance and move. On the other hand, learning to recognize a new face can take place in a single introduction and recognition of the face often brings to mind many aspects of the meeting such as the person’s name. A second distinction, which is important to this review (see Table 1), is between the recollection of general information (semantic or fact memory) and information associated with a specific event (episodic or event memory).21 Both semantic and episodic memories are considered examples of conscious memory because subjects are aware of the information being retrieved. Reciting a list of words beginning with the letter ‘L’ is an example of semantic memory because learned information is being retrieved but the context in which that information was learned is not accessed. Recollecting what you had for breakfast yesterday morning is an example of episodic memory because the specific source and the context associated with the learned information is required. Importantly, several of these distinctions at the behavioral level have gained support at the biological level. Patients with amnesia can show selective impairment in conscious memory (both semantic and episodic) in comparison with preserved skill learning.23 These findings make clear that it is necessary to consider that different kinds of memory might be mediated by distinct brain pathways. In this regard, functional neuroimaging data from memory studies should be considered in terms of the kind of memory being addressed as well as the specific processes (e.g. encoding, storage, retrieval) involved.12 As will be seen, these behavioral distinctions have predicted some of the biological distinctions obtained from the neuroimaging data.

distinct areas within the prefrontal cortex, evidence about the capability of neuroimaging to resolve separate activations within the frontal cortex is also important. The first evidence that neuroimaging can resolve multiple prefrontal areas comes from the observation that different experiments have reported activations that span nearly the entire volume of the prefrontal cortex — a region that is known to contain multiple areas. However, the variability across experiments represents an unknown combination of methodological and between-group factors that might cause variability unrelated to the participation of distinct anatomical areas.18 To determine a more appropriate (and conservative) estimate of the ability of neuroimaging to distinguish among multiple prefrontal areas, it is necessary to systematically observe dissociations across prefrontal areas. This has been done in several instances and suggests a current upper limit to the resolution of neuroimaging at about 16 mm19,20 (Figure 1B). In humans this corresponds to the difference between the approximate locations of Brodmann’s area 10 versus area 45, or between areas 44 and 47. Thus, there is every reason to believe that neuroimaging is capable of detecting distinct functional roles for multiple areas within the prefrontal cortex. The second concept, related to cognitive aspects of memory function, is that multiple forms of memory exist.21 A complete description of current classification schemes, which are continuously being modified as more memory phenomena are elucidated, is well beyond the scope of this review (see ref 22). Nonetheless, a few brief examples are worth illustrating. At the behavioral level, multiple forms of memory have been demonstrated by showing that acquisition and retention during different learning paradigms have distinct characteristics. For example, a distinction has been made between skill learning and 49

R. L. Buckner and S. E. Petersen

Recent neuroimaging findings

As an initial step in this process, composite activations from a number of long-term memory tasks examining semantic and episodic memory retrieval are presented. The studies have also activated many areas outside of the prefrontal cortex indicating that multiple distributed areas are used to accomplish these tasks. However, for the purpose of this discussion, we focus only on the prefrontal activations (Figures 2 and 3). In addition, many other kinds of long-term memory retrieval have also been profitably studied using neuroimaging, sometimes revealing prefrontal activations (e.g. procedural memory, classical conditioning). Each of these kinds of memory could be discussed at length and are not included simply because of space limitations.

Numerous studies have observed prefrontal cortex activations during long-term memory tasks. Those studying semantic memory retrieval and episodic memory retrieval have been particularly reliable in producing robust prefrontal activations (semantic memory retrieval, refs 5,8,24-29; episodic memory retrieval, refs 5,9-11,30-34). These studies demonstrate convincingly that areas within the prefrontal cortex are active during tasks requiring retrieval from long-term memory. The current focus of the field has moved toward trying to understand which prefrontal brain areas are activated selectively during the different kinds of memory and what the processing roles of these areas might be.

Figure 2. Activations are shown for a number of neuroimaging studies that have examined subjects performing verbal semantic memory retrieval tasks. The activations are plotted on a horizontal brain section51 representing Z = 12 with all prefrontal activations falling 10 mm above or below the Z = 12 plane included. The lateral view of the brain shows the approximate position of the horizontal section. The tasks include overtly retrieving verbs (circles 1 and 8);25,27 silently retrieving words beginning with ‘F’ or ‘S’ (circle 2);24 silently retrieving verbs (circle 3);29 retrieval of whether words represent living objects (circle 4);7 and overtly retrieving words from three-letter word beginnings (circles 5,6 and 7).5 A region of common activation is reliably observed in the leftinferior prefrontal cortex.

50

Memory retrieval and prefrontal cortex

Left-inferior prefrontal involvement in semantic memory retrieval

cortex contributes to the access and/or representation of verbal information. The participation of the left inferior prefrontal cortex can occur within the context of word retrieval or with the goal of saying the word aloud. Importantly, the role of this region seems to be at a higher level than simply controlling speech output as simple speech tasks have shown little or no activation here.12,27 The exact nature of the information operated on by this region is still unresolved but there is some suggestion that the information can be coded phonologically.35,36 Future experiments may find that there are multiple areas within this left-inferior prefrontal region which contain further functional subdivisions. Perhaps these areas will be found to represent different aspects of verbal information, or maybe even extend beyond verbal information.

A number of studies involving the semantic retrieval of words, or information about words, have activated a cluster of areas within a region of the left-inferior prefrontal cortex (Figure 2). It seems reasonable to speculate that areas within this region might be used to access and maintain a representation of words during their retrieval — a demand common to all of the tasks. This idea is consistent with notions that have evolved from the study of speech production tasks. These studies have suggested that left prefrontal areas are used during more elaborate forms of production when words must be generated in a non-automatic27 or internally guided fashion.24 Perhaps all of these areas express a similar concept: left-inferior prefrontal

Figure 3. Activations are shown for a number of neuroimaging studies that have examined subjects performing episodic memory retrieval tasks. The activations are plotted in the same way as Figure 2. The tasks include recall of words using three-letter word beginnings as cues (circles 1,2,3 and 10);5,9,50 recognition of faces by young subjects (circle 4);46 recognition of faces by old subjects (circle 5);32 recognition of auditory sentences (circle 6);11 paired-associate recall of pictures (circle 7);31 paired-associate recall of auditory words (circle 8);31 and yes/no recognition of words (circle 9).30 A region of common activation is reliably observed in right-anterior prefrontal cortex. This area(s) is often activated in addition to other prefrontal areas which are less consistent across tasks and might be related to the specific type of information being retrieved.

51

R. L. Buckner and S. E. Petersen Table 2. Studies demonstrating activation of the left-inferior prefrontal cortex during tasks involving verbal semantic memory retrieval Circle number

X

Y

Z

Target task

Reference task

1

–37 –34 –43 –40 –38 –37 –40 –40 –43

23 23 29 14 28 20 18 20 28

21 9 20 16 16 9 15 17 13

Verb generation

Noun repetition

Generate words beginning with ‘F’ Think of verbs Semantic decision Stem completion Stem completion Stem completion Verb generation

Word repetition Rest Orthographic decision Fixation Fixation Fixation Noun repetition

2 3 4 5 6 7 8

Listed are Talairach and Tournoux (1988)51 locations for each of the left-inferior prefrontal activations that are plotted in Figure 2. The studies are listed in relation to the number of the plotted circle in Figure 2. The target verbal semantic retrieval task that was studied and the reference (control)task used are listed in the right-most columns.

verbal material into episodic memory have also activated this area.7,10 This too is not surprising if one considers what the consequence of maintaining a word item might be. Representing words internally might be a partial source of memory traces leading to episodic long-term storage. That is, information represented within a pathway utilizing the left-inferior prefrontal cortex may be easily converted into a more enduring form.7,12 This idea is quite consistent with cognitive findings that have shown words generated by subjects (generation effect)40 or semantically elaborated on by subjects (levels of processing effect)41 are more likely to be remembered then those that are simply read aloud. In this regard, the process of retrieving word information from semantic memory (which activates the left-inferior prefrontal cortex) might simultaneously be storing knowledge of that retrieval event (or aiding in the storing of that knowledge) into episodic long-term memory.12,42 An interesting future question currently left unanswered is how these areas might interact with other brain areas thought to be important for memory storage, such as areas in the medial temporal lobe44-45 and other prefrontal areas.

Already there is some evidence that areas in the more anterior region of the left prefrontal cortex, at or near Brodmann’s areas 46, 47 and/or 10, are activated during tasks requiring access to word meaning.19 This possibility is supported by studies which activate these anterior areas during tasks in which word meaning is accessed as a component of the information being retrieved.7,25,26 Gabrieli and colleagues have further shown that activation within a subset of these areas are sensitive to repeated retrieval of word meaning — which is often referred to as conceptual priming. Using fMRI, they found that left prefrontal areas active during retrieval of word meaning are less active the second time the same information is retrieved.28 Given these findings, it might be the case that these anterior areas are recruited to maintain or control access to a second, higher-level of semantic information — which, in some instances, involves word meaning.

Left-inferior prefrontal cortex is involved in other kinds of memory Given the tentative ideas about the role the leftinferior prefrontal cortex might be playing in semantic retrieval, it is not surprising that other kinds of memory tasks and memory processes activate this region. Some studies of verbal working memory have activated the left-inferior prefrontal cortex.12,37,38 Verbal working memory requires the maintenance and manipulation of words internally. This demand is inherent to the elaborate speech production and semantic retrieval tasks that have already been discussed.12,27,39 Studies of tasks known to encourage storage of

Episodic memory retrieval Almost all of the episodic retrieval tasks published to date have activated the right-anterior prefrontal cortex (Figure 3). This is in strong contrast to the more posterior left prefrontal activations that are discussed above in relation to semantic memory retrieval. Right prefrontal activation during episodic retrieval was initially observed in a series of studies using purely verbal materials5,9 but has been shown to generalize to the retrieval of faces,32,46 pictures,31 and another 52

Memory retrieval and prefrontal cortex Table 3. Studies demonstrating right-anterior prefrontal cortex activation during tasks involving episodic memory retrieval Circle number

X

Y

Z

Target task

Reference task

1 2 3 4 5 6 7 8 9 10

28 31 31 34 30 32 25 27 30 30

45 50 50 54 52 44 47 47 44 46

4 9 11 4 8 12 16 18 10 8

Stem-cued recall Stem-cued recall Stem-cued recall Face recognition (Young subjects) Face recognition (Old subjects) Auditory recognition (all old sentences) Paired associate recall of pictures Paired associate recall of words Word recognition Stem-cued recall (Low recall version)

Stem completion Stem completion Stem completion Face matching Face matching Auditory recognition (all new sentences) Word repetition Word repetition Word reading Stem completion

Listed are Talairach and Tournoux (1988)51 locations for each of the right-anterior prefrontal activations that are plotted in Figure 3. The studies are listed in relation to the number of the plotted circle in Figure 3. The target episodic retrieval task that was studied and the reference (control) task used are listed in the right-most columns. The consistency of the results is remarkable but it should be noted that at least two studies of episodic retrieval have not detected right-anterior prefrontal activation in the same area as the studies listed above, although both of these studies did detect right prefrontal activation in different areas10,54.

verbal task using auditory sentences.11 Various delay intervals between study and test have been used ranging from a few minutes5 to one-week30 and the format of retrieval task has been yes/no recognition,11,30 forced choice recognition,32,46 perceptually cued recall,5 and paired associate recall.31 Old and young subjects32 and men and women (Buckner, unpublished observations) have independently demonstrated activation within this focal area during episodic retrieval. The generality of this finding is remarkable and suggests that a process common to many episodic memory retrieval tasks is being facilitated by this rightanterior prefrontal region. The exact processing role this area plays in episodic retrieval is still unclear, but attempts to elucidate the role should pay careful attention to the localization of the activation.47 Current data suggest that the common activation is not across a distributed set of areas within the prefrontal cortex. Rather, all of these activations appear to be focused in a localized region in the anterior cortex. The finding of right prefrontal activation across a wide range of episodic retrieval tasks, in the presence of the left prefrontal activations during episodic storage mentioned earlier, stimulated a model of hemispheric asymmetry in relation to prefrontal memory function.42 Tulving and colleagues suggested that the right prefrontal cortex is active more during episodic retrieval than is the left prefrontal cortex. This model helped greatly in drawing attention to the commonalities observed across the many laboratories studying these phenomena. However, the model is only minimally constrained in terms of prefrontal anatomy. As discussed above, prefrontal cortex is

made up of multiple subregions. Given the reliability and resolution of our current neuroimaging techniques (Figure 1), we are already beyond classifying activation of prefrontal brain areas only in terms of which hemisphere they fall in. Moreover, posterior prefrontal areas, sometimes bilaterally, have been activated during episodic memory retrieval in addition to the right-anterior prefrontal area that has often been observed. This argues against a strict laterality hypothesis. Thus, although Tulving’s model was useful as an initial heuristic, recent data have suggested an even more specific set of hypotheses about which prefrontal areas are activated during episodic memory retrieval.5,12 The observation of additional, and sometimes bilateral, prefrontal brain activations during episodic retrieval generates another set of interesting questions. The pattern of the posterior activations may be related to the kind of episodic information being retrieved. Purely verbal retrieval has activated posterior left frontal areas in addition to the right-anterior prefrontal area5 while face (pictorial) retrieval has exclusively activated right prefrontal areas.32,46 Interestingly, a task requiring the paired associate recall of picture names (a task that demands both pictorial and verbal retrieval) has activated the posterior prefrontal cortex within the frontal operculum bilaterally.31 It will be interesting to see what emerges as more data are collected, but the notion that participation of certain prefrontal areas is dependent upon the kind of information being retrieved is consistent with, and predicted by, a great deal of non-human primate work.2,48 An intriguing possibility is that the episodic memory tasks rely on multiple, distinct prefrontal areas to 53

R. L. Buckner and S. E. Petersen work was supported by NIH grants NS06833, EY08775, HL13851, AG08377, NS32979, and grants from the Charles A. Dana Foundation and the McDonnell Center for Higher Brain Function. Reprint requests should be sent to Randy L. Buckner, Washington University School of Medicine, Department of Neurology and Neurological Surgery, Campus Box 8111, St. Louis, MO 63110, USA.

access the specific kind of information being retrieved. In addition to these areas, the common activation in right-anterior prefrontal cortex that has been reliably identified across laboratories may be more uniformly recruited to guide processes that generalize across episodic memory retrieval tasks.12 Whether this common area plays a role only in episodic memory retrieval remains an open question. Some have speculated that this area may be recruited to guide processes related to searching for information in memory and may be specific to episodic retrieval.49,50 Others have suggested its role is more general and involves relating information to a specific context — a prominent component of many episodic retrieval tasks but potentially relied on by other kinds of tasks as well.47 More data are needed to further resolve this issue, but it is noteworthy that in a period of a few years the field has already progressed from simply identifying prefrontal areas active during episodic retrieval to trying to determine their specific processing contributions.

References 1. Fuster JM (1989) The Prefrontal Cortex. Raven Press, New York 2. Goldman-Rakic PS (1987) Circuitry of primate prefrontal cortex and regulation of behavior by representational memory, in The Handbook of Physiology. Section 1: The Nervous System, Volume V. Higher Functions of The Brain. Part 1 (Plum F, Mountcastle V, eds) pp 373-417. American Physiological Society, Bethesda, MD, USA 3. Moscovitch M (1982) Multiple dissociations of function in amnesia, in Human Memory and Amnesia (Cermak L, ed) pp 337-370. Erlbaum, Hillsdale, NJ, USA 4. Shimamura AP, Janowsky JS, Squire LR (1991) What is the role of frontal lobe damage in memory disorders? in Frontal Lobe Function and Dysfunction (Levin H, Eisenberg HM, Benton AL, eds) pp 173-198. Oxford, New York, USA 5. Buckner RL, Petersen SE, Ojemann JG, Miezin FM, Squire LR, Raichle ME (1995) Functional anatomical studies of explicit and implicit memory retrieval tasks. J Neurosci 15:12-29 6. Grasby PM, Frith CD, Friston KJ, Bench C, Frackowiak RSJ, Dolan RJ (1993) Functional mapping of brain areas implicated in auditory-verbal memory function. Brain 116:1-20 7. Kapur S, Craik FIM, Tulving E, Wilson AA, Houle S, Brown GM (1994) Neuroanatomical correlates of encoding in episodic memory: levels of processing effect. Proc Natl Acad Sci USA 91:2008-2011 8. McCarthy G, Blasmire AM, Rothman DL, Gruetter R, Shulman RG (1993) Proc Natl Acad Sci USA 90:4952-4956 9. Squire LR, Ojemann JG, Miezin FM, Petersen SE, Videen TO, Raichle ME (1992) Activation of the hippocampus in normal humans: a functional anatomical study of memory. Proc Natl Acad Sci USA 89:1837-1841 10. Shallice T, Fletcher P, Frith CD, Grasby P, Frackowiak RSJ, Dolan RJ (1994) Brain regions associated with acquisition and retrieval of verbal episodic memory. Nature 368:633-635 11. Tulving E, Kapur S, Markowitsch HJ, Craik FIM, Habib R and Houle S (1994) Neuroanatomical correlates of retrieval in episodic memory: auditory sentence recognition. Proc Natl Acad Sci USA 91:2012-2015 12. Buckner RL, Tulving E (1995) Neuroimaging studies of memory: Theory and recent PET results, in Handbook of Neuropsychology, Vol 10 (Boller F, Grafman J, eds) pp 439-466. Elsevier, Amsterdam 13. Smith EE, Jonides J, Koeppe RA, Awh E, Schumacher EH, Minoshima S (1995) Spatial versus object working memory: PET investigations. J Cog Neurosci 7:337-356 14. Barbas H, Pandya DN (1989) Architecture and intrinsic connections of the prefrontal cortex in the rhesus monkey. J Comp Neurol 286:353-375 15. Preuss TM, Goldman-Rakic PS (1991) Myelo- and cytoarchitecture of the granular frontal cortex and surrounding regions in the strepsirhine Galago and the anthropoid primate Macaca. J Comp Neurol 310:429-474 16. Carmichael ST, Price JL (1994) Architectonic subdivision of the orbital and medial prefrontal cortex in the macaque monkey. J Comp Neurol 346:366-402

Conclusions Neuroimaging techniques have progressed to the point where they can reliably localize activations in the prefrontal cortex and can dissociate activations originating from functionally distinct areas. These technical accomplishments provide a foundation for using neuroimaging to aid in characterizing the prefrontal cortex’s role in memory function. Progress has already been made in determining prefrontal areas that may play a role in long-term memory retrieval. An area in the left-inferior prefrontal cortex has been observed to be active across a wide range of tasks requiring subjects to retrieve words or information about words from semantic memory. Tasks requiring subjects to retrieve information from a specific study episode (episodic memory) have reliably activated the right-anterior prefrontal cortex in addition to more posterior prefrontal areas which vary across tasks. The functional identification and preliminary characterization of these two areas (left-inferior prefrontal cortex and right-anterior prefrontal cortex) in healthy normal subjects provides a clear demonstration of the utility of functional neuroimaging in the study of memory function.

Acknowledgements We thank Jonathan Demb who thoughtfully read and commented on an early version of this manuscript. This

54

Memory retrieval and prefrontal cortex 36. Fiez JA, Tallal P, Raichle ME, Miezin FM, Katz WF, Dobmeyer S, Petersen SE (1995) PET studies of auditory and phonological processing: Effects of stimulus type and task condition. J Cog Neurosci 7:357-375 37. Awh E, Jonides J, Smith EE, Schumacher EH, Koeppe R, Katz S (1996) Dissociation of storage and rehearsal in verbal working memory: Evidence from PET. In press 38. Petrides M, Alvisatos B, Meyer E, Evans AC (1993) Functional activation of the human frontal cortex during the performance of verbal working memory tasks. Proc Natl Acad Sci USA 90:878-882 39. Goldman-Rakic PS (1994) The issue of memory in the study of prefrontal function, in Motor and Cognitive Functions of the Prefrontal Cortex (Thierry A-M, Glowinski J, Goldman-Rakic PS, Christen Y, eds) pp 112-121. Springer, New York 40. Slamecka NJ, Graf P (1978) The generation effect: Delineation of a phenomenon. J Exp Psychol: Human Learning and Memory 4:592-604 41. Craik FIM, Tulving E (1975) Depth of processing and retention of words in episodic memory. J Exp Psychol: General 104:268-294 42. Tulving E, Kapur S, Craik FIM, Habib R, Houle S (1994) Hemispheric encoding/retrieval asymmetry in episodic memory: Positron emission tomography findings. Proc Natl Acad Sci USA 91:2016-2020 43. Cohen NJ, Eichenbaum H (1993) Memory, Amnesia, and the Hippocampal System. MIT Press, Cambridge, MA 44. Squire LR (1992) Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychol Rev 99:195-231 45. Murray EA (1996) Multiple functional subdivisions contributing to memory in the medial temporal lobe in monkeys. Semin Neurosci, in press 46. Haxby JV, Horwitz B, Maisog JM, Ungerleider LG, Mishkin M, Schapiro MB, Rapoport SI, Grady CL (1993) Frontal and temporal participation in long-term recognition memory for faces: A PET-rCBF activation study. J Cereb Blood Flow Metab Suppl 1 13:S499 47. MacLoed AK, Buckner RL, Petersen SE, Raichle ME (1995) Activation of anterior right prefrontal cortex during semantic monitoring. Soc Neurosci Abstr 21:1441 48. Wilson FAW, O Scalaidhe SP, Goldman-Rakic PS (1993) Dissociation of object and spatial processing domains in primate prefrontal cortex. Science 260:1955-1958 49. Nyberg L, Tulving E, Habib R, Nilsson LG, Kapur S, Houle S, Cabeza R, McIntosh AR (1996) Functional brain maps of retrieval mode and recovery of episodic information. NeuroReport in press 50. Schacter DL, Alpert NM, Savage CR, Rauch SL, Albert MS (1996) Conscious recollection and the human hippocampal formation: Evidence from positron emission tomography. Proc Natl Acad Sci USA in press 51. Talairach J, Tournoux P (1988) Co-Planar Sterotaxic Atlas of the Human Brain. Thieme Medical Publishers, Inc., New York 52. Squire LR (1987) Memory and Brain. Oxford University Press, New York 53. Baddeley A (1992) Working memory: The interface between memory and cognition. J Cog Neurosci 4:281-288 54. Petrides M, Alvisatos B, Evans AC (1995) Functional activation of the human ventrolateral frontal cortex during mnemonic retrieval of verbal information. Proc Natl Acad Sci USA 92:5803-5807

17. Petrides M, Pandya DN (1994) Comparative architectonic analysis of the human and the macaque frontal cortex, in Handbook of Neuropsychology, Vol 9, (Boller F, Grafman J, eds) pp 17-58. Elsevier, Amsterdam 18. Hunton DL, Miezin FM, Buckner RL, Raichle ME, Petersen SE (1994) An assessment of replicability and anatomical variability in functional neuroimaging studies. Soc Neurosci Abstr 20:354 19. Buckner RL, Raichle ME, Petersen SE (1995) Dissociation of human prefrontal cortex areas across different speech production tasks and gender groups. J Neurophy 74:2163-2173 20. Petrides M, Alivisatos B, Evans AC, Meyer E (1993) Dissociation of human mid-dorsolateral from posterior dorsolateral frontal cortex in memory processing. Proc Natl Acad Sci USA 90:873-877 21. Tulving E (1983) Elements of Episodic Memory. Oxford University Press, New York 22. Schacter DL, Tulving E (1994) Memory Systems 1994 (Schacter DL, Tulving E, eds). MIT Press, Cambridge, MA 23. Corkin S (1968) Acquisition of a motor skill after bilateral medial temporal lobe excision. Neurophyschologia 6:255-265 24. Frith CD, Friston K, Liddle PF, Frackowiak RSJ (1991) Willed action and the prefrontal cortex in man: A study with PET. Proc R Soc Lond B 244:241-246 25. Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME (1988) Positron emission tomographic studies of the cortical anatomy of single-word processing. Nature 331:585-589 26. Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME (1989) Positron emission tomographic studies of the processing of single words. J Cog Neurosci 1:153-170 27. Raichle ME, Fiez JA, Videen TO, MacLeod AK, Pardo JV, Fox PT, Petersen SE (1994) Practice-related changes in human brain functional anatomy during non-motor learning. Cereb Ctx 4:8-26 28. Demb JB, Desmond JE, Wagner A, Vaidya CJ, Glover G, Gabrieli JDE (1995) Semantic encoding in the left inferior prefrontal cortex: A functional MRI study of task difficulty and process specificity. J Neurosci 15:5870-5878 29. Wise R, Chollet F, Hadar U, Friston K, Hoffner E, Frackowiak R (1991) Distribution of cortical neural networks involved in word comprehension and word retrieval. Brain 114:1803-1817 30. Andreasen NC, O’Leary DS, Arndt S, Cizadlo T, Hurtig R, Rezai K, Watkins GL, Boles L, Ponto B, Hichwa R (1995) Short-term and long-term verbal memory: A positron emission tomography study. Proc Natl Acad Sci USA 92:5111-5115 31. Buckner RL, Raichle ME, Miezin FM, Petersen SE (1995) PET studies of the recall of pictures and words from memory. Soc Neurosci Abstr 21:1441 32. Grady CL, McIntosh AR, Horwitz B, Maisog JM, Ungerleider LG, Mentis MJ, Pietrini P, Schapiro MB, Haxby JV (1995) Agerelated reductions in human recognition memory due to impaired encoding. Science 269:218-221 33. Roland PE, Guly´as B (1995) Visual memory, visual imagery, and visual recognition of large field patterns by the human brain: Functional anatomy by positron emission tomography. Cereb Ctx 5:79-93 34. Stern CE, Corkin S, Guimaraes AR, Sugiura R, Carr CA, Baker JR, Jennings PJ, Gonz´ales, Rozen BR (1994) A functional MRI study of long-term explicit memory in humans. Soc Neurosci Abstr 20:1290 35. D´emonet JF, Chollet F, Ramsay S, Cardebat D, Nespoulous JL, Wise R, Rascol A, Frackowiak R (1992) The anatomy of phonological and semantic processing in normal subjects. Brain 115:1753-1768

55