ryan_cohen_03 - Cognitive Afi‘ective& Behavioral...

Info iconThis preview shows pages 1–18. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 16
Background image of page 17

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 18
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Cognitive, Afi‘ective, & Behavioral Neuroscience 2003, 3 (3), 168-185 Evaluating the neuropsychological dissociation evidence for multiple memory systems JENNIFER D. RYAN R0 tman Research Institute, Toronto, Ontario, Canada and NEAL J. COHEN University of Illinois at Urbana—Champaign, Urbana, Illinois This article presents a critical evaluation of the logic and nature of the neuropsychological dissociation evidence that has provided one of the essential lines of support for claims of multiple memory systems—- specifically, suggesting that amnesia sele ctivelycompromises, and an intact hippocampal system selec- tively supports, a particular form of memory. An analysis of the existing neuropsychological dissocia- tion evidence is offered in which different classes of evidence—different dissociation approaches—are identified and characterized. The logic of these neuropsychological dissociation approaches is evalu- ated critically in terms of their ability to distinguish among alternative theoretical views. We conclude that although they support a multiple memory systems account, the findings from these types of neuro- psychological dissociation, taken individually and without support from other converging lines of cog- nitive neuroscience evidence, cannot definitively rule out alternative formulations. A more powerful neuropsychological dissociation approach is then outlined, involving dissociationwithin condition that, by more effectively limiting the critical domains of difference between the dissociated performances, can successfullyrule out alternative accounts. Its application in Ryan, Althoff, Whitlow, and Cohen (2000) is described, providing strong support for the power of the dissociation Within condition approach. One of the major developments in memory research in the last several decades has been the advancement of the claim that there are multiple memory systems in the brain and the research it has engendered whose aim has been to identify and characterize (some of ) these systems (e.g., Cohen, 1984; Cohen & Eichenbaum, 1993; Cohen, Poldrack, & Eichenbaum, 1997; Cohen & Squire, 1980; Eichenbaum & Cohen, 2001; Graf & Schacter, 1985 ; Nadel, 1994; Schacter & Tulving, 1982, 1994; Squire, 1987, 1992; Tulving, 1972, 1985; Weiskrantz, 1987). Historically, the key data in arguing for and testing ideas about multiple memory systems has come in the form of neuropsychological dissociation evidence—the ability of patients with amnesia to reliably exhibit instances of fully preserved learning and memory in the face of otherwise profound and pervasive memory deficits. This dissocia— tion between classes of impaired and classes of spared memory abilities in patients with hippocampal system damage has been central in suggesting that amnesia se— lectively compromises, and an intact hippocampal sys— tem selectively supports, a particular aspect or form of memory. Correspondence concerning this article should be addressed to J. D. Ryan, Rotman Research Institute, Baycrest Centre for Geriatric Care, 3560 Bathurst St., Toronto, ON, M6A 2E1 Canada (e-mail: jryan@ rotman-baycrest.on.ca). However, not all researchers have interpreted the neuro— psychologicaldissociation evidence as requiring a claim of multiple memory systems. Accordingly, rather than re— lying exclusively on the neuropsychological dissociation evidence, we and others have derived critical support for the claim of multiple memory systems from several con- verging fines of evidence (e.g., Cohen & Eichenbaum, 1993; Eichenbaum & Cohen, 2001; Gabrieli, 1999; Squire, 1987, 1992; Tulving, 1999). These converging lines in— clude performance dissociations in animal models of am- nesia, findings of hippocampal system activation during the performance of some, but not other, memory tasks in functional neuroimaging studies of normal control sub— j ects, and recordings of activity of single cells in the hip— pocampus in freely behaving animals for some, but not other, memory challenges. Here, as in other areas of cog— nitive neuroscience, each of the different lines permits the addressing of somewhat different questions about the na— ture of memory and its organization in the brain; taken a1- together, they permit the investigation of memory in both humans and animals, using methods that have very differ— ent assumptions. They provide multiple sources of data and multiple sources of constraint on theories about the organization of memory. To the extent that the answers forth- coming from these different lines all converge on a partic— ular View of memory organization—that is, to the extent that a single view can encompass the data across different measures, methods, species, and assumptions—that View Copyright 2003 Psychonomic Society, Inc. 168 NEUROPS YCHOLO GICAL DIS S OCIATION EVIDENCE 169 can be endorsed with more confidence. It is the conver- gence of these many lines of evidence that has been most persuasive that there are multiple memory systems in the brain. Despite the power of the approach involving converging methods, the fact that much of the debate about the mul- tiple memory systems claim has centered on neuro— psychological dissociation evidence leads us, in this arti— cle, to consider that evidence in isolation, without support from the other converging lines. The question that is cen- tral to this article is why the neuropsychologicaldissocia— tion evidence alone has been less than fully persuasive. We do not attempt to summarize here the full range of data from neuropsychological studies of amnesia; such a re— view is beyond the scope of this article and can be found elsewhere (Cohen & Eichenbaum, 1993; Eichenbaum & Cohen, 2001). Rather, in this article, we consider the logic of work using neuropsychological dissociation evidence, raising some critical questions about why such evidence may be unable to unambiguously support multiple mem— ory systems theories. We point out weaknesses that make it difficult for most existing dissociation findings in the literature, taken in isolation, to definitively rule out alter— native formulations of the organization of memory. We then describe a more powerful approach to obtaining neuropsychological dissociation evidence that avoids these pitfalls, and we go on to illustrate its application. CLASSES OF NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE Consideration of the various neuropsychological find— ings of dissociation in the literature shows important dif— ferences among them with regard to the logic or nature of the dissociation—that is, what exactly is being disso— ciated. These differences are consequential, we will show, with regard to the power of the dissociation to dis- tinguish multiple memory systems accounts from uni- tary system accounts and to distinguish among different multiple memory systems accounts. On our analysis, we can identify two different classes of neuropsychological dissociation evidence—two neuropsychological dissoci— ation appro aches—represented in the existing literature. We will call these dissociation across tasks and dissoci- ation across instruction conditions. In the sections that follow, we will characterize the two classes in turn, lay- ing out their underlying logic and illustrating the logic with representative examples of their application in the literature. We then will evaluate them critically, examin- ing whether the evidence derived from these representa- tive examples, taken individually, can unambiguously support multiple memory systems theories over alterna- tive formulations of the organization of memory. We will show how each class of neuropsychological dissociation evidence has been used in particular experiments to argue for and against certain theories of memory and amnesia and will point out some critical weaknesses that limit their ability to support a multiple memory systems ac- count exclusively Dissociation Across Tasks The first of the two classes of neuropsychological dis— sociation evidence used extensively in the literature to support the multiple memory systems claim involves dissociation across tasks. Here, various classes of mem— ory tasks are identified on which amnesic patients are impaired, as contrasted with other classes of memory tasks on which amnesic patients are spared. The logic here is as follows: (1) There are multiple memory sys— tems; (2) one of the memory systems depends critically on the hippocampal system (and/or other) structures that are damaged in amnesia, and this memory system is selec— tively compromised in amnesia; (3) the different memory systems support different classes of performance; and (4) all those—and only those—task performances de- pendent on the hippocampal—mediated memory system should be impaired in amnesia. This approach conforms with the long tradition of work in neuropsychology, in which different tasks are used to tap into and illuminate the functioning of differ— ent cognitive systems. Clinically, performance deficits on one or another neuropsychological test, when occur— ring selectively—that is, in the absence of deficit on other neuropsychological tests———is interpreted as re— flecting an isolated, specific impairment to language, at- tention, executive, or memory systems. Thus, amnesic patients are identified clinically by exhibiting impaired performance on various standardized tests of long-term memory (e. g., on the [3 O-min] delayed recall tests of word pairs, paragraphs, and line drawings on the Wechsler Memory Scale) in the absence of, or disproportionate to, any impairment on IQ tests and tests of perceptual, at— tentional, and linguistic performance. Such a pattern of performance of impairment and sparing across tasks iso- lates the deficit to the domain of memory. That the def— icit can be further specified to the domain of long-term memory is illustrated by the many findings of dissociation in amnesic patients between their performance on tasks tapping long-term memory versus tasks tapping working memory; amnesia is accompanied by an impairment on the former in the face of sparing on the latter. Much evidence from this approach has been offered in support of the multiple memory systems claim, in which the pattern of impaired versus spared memory performances across tasks in amnesia has seemed to iso— late it to a particular form of long-term memory. The now classic dissociation evidence of this type involves spar— ing of performance on a variety of skill-learning tasks despite profound amnesia. For example, H.M. showed impressive learning across trials in mirror tracing (trac- ing the outline of a star when all visual input about the star and the subject’s hand is mirror reversed), and many amnesic patients have shown successful incremental learning across trials and across sessions on manual— l7O RYAN AND COHEN tracking tasks such as rotary pursuit (maintaining con— tact of a hand—held stylus with a small disk on a rotating platter; Brooks & Baddeley, 1976; Cohen, 1981 ; Milner, 1962; Milner, Corkin, & Teuber, 1968). Those preserved learning and memory performances, and others, stand in sharp contrast to the thoroughly documented impairment in amnesia on recall or recognition memory tests for words, sentences, faces, scenes, routes, personal and public events, and so forth. However, as will be outlined later in detail, spared and impaired performance here is documented across tasks that differ on a number of di— mensions, including stimulus materials, task instruc— tions, and response requirements. These differences cre— ate divergence in the type or mode of processing that may be engaged in performing the task. Therefore, such examples of spared and impaired performance may be due to the use either of different memory systems or of different modes of processing engaged by the two tasks. Different modes of processing may arise from a single, unitary system of memory, thus providing a plausible al— ternative to the multiple memory systems account. Dissociation Across Instruction Conditions The second class of neuropsychological dissociation evidence used to support the multiple memory systems claim involves dissociation across instruction condi- tions. Here, in each experiment, after subjects are pre- sented with a single kind of learning experience with a single set of materials, performance is assessed under two different types of test instructions that place differ- ent demands on memOry. The logic here is as follows: (1) There are multiple memory systems; (2) one of the memory systems depends critically on the hippocampal system (and/or other) structures that are damaged in am- nesia, and this memory system is selectively compro- mised in amnesia; (3) the different memory systems sup- port different classes of performances; (4) but many tasks can be supported by more than one of the memory systems, operating separately or in combination; (5) we can constrain how subjects perform a task and, hence, which memory system is engaged, through the use of certain instructions; and (6) with instructions that en- gage the hippocampal-mediated memory system, per- formance should be impaired in amnesia, but with in— structions that engage other memory systems that are capable of supporting successful performance on a given task, performance should be intact in amnesia. Two well-cited studies that we take as exemplars of this strategy are those of Cohen and Squire (1980) and Graf, Squire, and Mandler (1984). In Cohen and Squire, subjects read a series of novel and repeated word triplets presented in mirror-reversed text. Memory was assessed under two different instruction conditions. Under one set of instructions, subjects were presented with more mirror- reversed text and were to read the word triplets as quickly and accurately as possible. Memory was assessed in terms of increased speed of reading. Under the other set of in- structions, individual words were presented in standard orientation, and memory was assessed with recognition judgments as to which ones had and had not been previ- ously read during the experiment. Arnnesic patients exhib- ited fully intact performance when they were given speeded reading instructions but showed profoundly im- paired performance when given recognition memory in— structions. In Graf et al. (1984), subjects were given a word list to study and then were presented at test with word stems, consisting of the first several letters of some studied and some nonstudied words. Memory was tested under two different instruction conditions, differing with regard to whether or not they required the subjects to gain explicit access to (or conscious awareness of) a particular study experience. One set of instructions involved cued recall (complete the stem with a word from the study list); the other set of instructions asked for stem completion with- out reference to the study list (complete the stem with the first word that comes to mind). Under stem completion instructions, amnesic patients showed the same bias as did normal control subjects to complete the stem with a word that had been on the study list, exhibiting intact repetition priming. But under cued recall instructions, amnesic patients were impaired, as compared with con— trol subjects, recalling fewer words from the study list than did the control subjects. Both of these dissociation across instruction condi— tions experiments, and the many others like them in the literature, have much in common with the dissociation across tasks experiments discussed earlier. Specifically, two test conditions involving different instructions cer— tainly could be described as two different tasks. How- ever, they differ in that the dissociation across instruc- tion conditions experiments use a single learning experience. In this way, whatever memory processes and representations were entailed in the initial encoding or learning of information can be held constant—equiva- lent for the different instruction conditions. This addi— tional level of experimental control is what distinguishes this class of neuropsychological dissociation studies from the first class. However, differences in instructions across task conditions would seem to invoke different modes of processing, along with the intended differences in memory requirements. Thus, again, as in the dissoci- ation across tasks condition, it is possible that spared versus impaired performances may be due to different memory systems that are engaged by the different tasks or to different modes of processing using a single mem— ory system engaged by the different tasks. EVALUATING THE CLASSES OF NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE We turn now to a more detailed description of the two classes of dissociation evidence identified above, illus- NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE trating how each has been used to support various claims about the organization of normal memory and the nature of amnesia. It must be noted that the various competing theories all derive support from the same neuropsycho- logical dissociation data. Our analysis here will show that the evidence from studies using the dissociation across tasks and the dissociation across instruction con- ditions approaches, each taken in isolation, is suscepti— ble to multiple interpretations, including both multiple memory systems accounts and accounts claiming multi— ple modes of processing. The analysis will also show that this reflects a shortcoming in the ability of these ap— proaches to successfully limit the number of dimensions on which the given example of dissociated (impaired vs. spared) performances differ, preventing identification of the truly critical dimension of difference. This shortcoming will be illustrated by considering a few well—known experimental findings in the empirical literature. Here, we identify seven dimensions on which the dissociated performances may differ that are plausi— ble candidates for accounting for an observed dissocia— tion in any given case (see Tables 1—5): type of stimuli, methodological and instructional conditions, obtained measures, study—test match (i.e., the extent to which study time and test time tasks are matched in their con- ditions and requirements), processing requirements, memory representation requirements, and time of test administration (i.e., whether or not the dissociated per- formances are tested at the same time). To offer one il— lustrative example, the tests of motor skill learning on which amnesic patients are spared differ from the tests of recall and recognition memory on which amnesic pa- tients are imp aired not only with regard to their memory representation requirements, but also in each of the fol— lowing ways: Testing of these different classes of per- formances occurs at different times with different types of stimuli, different instructions and processing require— ments, and different types of measures (see Table 1 and the discussion below). The obvious dilemma concerns how to show that it is the differences in their memory re- quirements, and not one or more of these other differ- 171 ences, that should be considered critical for understand— ing this particular performance dissociation. Several experiments are considered here with regard to these dimensions. Both of the two neuropsychological dissociation approaches will be seen to be limited in their ability to minimize the dimensions of difference between the dissociated performances. The greater experimental control of the dissociation across task instructions ap— proach over the dissociation across tasks approach will be seen to result in some reduction of the number of po— tentially relevant dimensions of difference, but it still leaves open too many alternative accounts. Accordingly, a more powerful class of dissociation evidence, with a greater potential for limiting the dimensions of difference between the dissociated performances to their memory representation requirements, needs to be developed. Before undertaking this analysis, however, we must first introduce several theoretical accounts that will fea- ture prominently in the discussion to follow. We cannot cover all of the various theories of memory in all of their various forms here (see Cohen & Eichenbaum, 1993, and Cohen et al., 1999, for more extended treatments of the range of competing theories). Instead, we will limit our discussion here to three major theories. Although this does not cover the full theoretical landscape in the field of memory, these are the theories that have received the most discussion in the literature, have each claimed support from existing neuropsychological dissociation evidence, and fully serve our present purposes of per— mitting a critical analysis of different approaches to neuropsychological dissociation evidence. Theoretical Accounts For many authors, the neuropsychological dissocia- tion evidence has been interpreted as supporting a mul- tiple memory systems view, in which there are function— ally and anatomically distinct memory systems in the brain. Hippocampal amnesia selectively compromises, and an intact hippocampal system selectively supports, one of those memory systems mediating a particular form of memory. Two multiple memory systems ac— Table 1 Dissociated Performances Using the Dissociation Across Tasks Approach in Early Studies (e.g., Milner, 1962; Milner, Corkin, & Teuber, 1968) Dimension Mirror—Tracing Task Recognition Memory Task Type of stimuli outlined star faces Obtained measures no. of errors per trial % correct recognition Study—test match yes no Methods/instructions mirror tracing face recognition Processing requirements visual form analysis perceptual—motor integration Memory requirements acquisition and expression of perceptual—motor skill viSual form analysis face processing conceptual/attributional processing memory for faces conscious recollection memory for relations representational flexibility Note—Dimensions of difference between dissociated performances appear in bold. 172 RYAN AND COHEN counts that have received particular currency in the lit- erature will be considered here, together with the com- peting view that there is only one form of memory and that the performance dissociations reflect different forms of processing. Declarative versus procedural memory. One promi— nent account of the findings of dissociation between classes of impaired versus spared memory performances in amnesia distinguishes between declarative and proce— dural memory, proposing that amnesia is a selective def— icit of, and the hippocampal system is selectively in— volved in, declarative long—term memory (Cohen, 1981, 1984; Cohen & Eichenbaum, 1993; Cohen & Squire, 1980; Eichenbaum, 1997, 1999; Eichenbaum & Cohen, 2001; Squire, 1987, 1992). Declarative memory sup— ports the encoding, retention, and retrieval of memory for facts and events, whereas procedural memory sup— ports the acquisition and expression of skilled perfor— mance. We will focus here specifically on the particular elaboration and extension of the declarative—procedural distinction that we have offered, based on multiple con- verging lines of cognitive neuroscience evidence, which holds that the hippocampal system supports long-term memory for all manner of (even arbitrary or accidental) relations among the constituent elements of a scene or an event (see, e.g., Cohen & Eichenbaum, 1993; Cohen et al., 1997; Cohen & Ryan, 2003; Eichenbaum, 1997, 1999; Eichenbaum & Cohen, 2001). Being able to remember which names go with which faces, the phone numbers of one ’3 various friends or col- leagues, the name of the player who scored the winning touchdown in the most recent Super Bowl, or who one ran into at the movie theater two weekends ago is a major challenge for anyone’s memory, because the associations between names and faces, or between people and phone numbers, or among the actors and actions that enter into unstaged events are only arbitrary or accidental and can- not be derived fully from other information. This challenge is handled, on our account, by hippocampal-dependent declarative memory through its mediation of a fundamen— tally relational form of memory. It supports representa— tions of the outcomes of processing of all the various net- works or modules that are engaged in comprehending the event or scene—all the items that have been coactivated during processing of any given learning episode. This in- formation is then bound together into long-term memories, thereby capturing the relations among the coactivated items. Tasks that depend critically on such relational memories, such as recall or recognition of the words, ob- jects, or faces that appeared on a list in a specific exper— iment and the recollection of specific events, are im— paired in amnesia. The learning of new domains of semantic knowledge— that is, of integrated knowledge structures—also de— mands memory for relations and is likewise impaired in full-blown cases of amnesia. It is for this reason that we have long included both episodic and semantic memory in our description of declarative memory (e.g., Cohen, 1981, 1984; Cohen & Eichenbaum, 1993; Eichenbaum & Cohen, 2001; cf. Schacter & Tulving, 1994; Tulving, 1985). ‘ One other characteristic of the representations sup— ported by declarative memory is what we call represen— rational flexibility (Cohen, 1984; Cohen & Eichenbaum, 1993). Declarative memories are not inextricably tied to the original learning context or the original modality of input; rather, they can be accessed and used in novel con— texts, under testing situations that are very different from the circumstances of initial learning. Thus, one can re— member whom one ran into at the movie theater two week- ends ago, regardless of whether one is asked about it ver— bally or is shown a photo line—up, and regardless of whether one is asked about it at home, while driving in a car, or when back at the movie theater, and one can recall con- ceptual information about the story, perceptual informa— tion about particular scenes from the movie, and all sorts of information about the actors in the movie, regardless of whether one is responding to an oral query, to a writ— ten questionnaire, or to photos. Procedural memory, by contrast, supports nonrela— tional forms of memory. It is mediated by the ongoing tuning and modification of particular processing net— works, making the operations of those processors faster and more efficient for recently experienced items when they are encountered again soon thereafter. Such effects of previous exposure to (i.e., prior processing of) items, in the absence of any requirement to remember such re- lational information as which scene or events those items appeared in or which other items co-occurred with them, are independent of the hippocampal system. Accord— ingly, tests that take as their measure of memory in— creased speed or efficacy of processing of individual stimulus items or a bias toward specific items as a result of recent previous exposure can be fully supported by procedural memory and can elicit intact performance in amnesia. Note that procedural memory is inextricably tied to the changes in specific processing networks and can be expressed only when those networks are again en— gaged; that is the sense in which such memories are said to be inflexible. Explicit versus implicit memory. The other dominant multiple memory systems account of the neuropsycho— logical dissociation evidence emphasizes the critical role of conscious recollection of prior study episodes. Based originally on Graf and Schacter (1985) and Schacter (1987), this View addresses the neuropsychological dis- sociation evidence by holding that amnesic patients are selectively impaired and that the hippocampal system is selectively involved, in explicit memory—that is, in gaining conscious access to memory for, or explicitly re— membering, previous learning experiences. Tasks such as recall or recognition of words, objects, or faces from a previously studied list are said to tap the ability to gain conscious access to the memory of the study episode (re— quiring explicit knowledge of having studied that partic- ular list in that particular lab oratory experiment) in order NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE 173 to then determine which items were and were not part of that study episode. The system(s) damaged in amnesia selectively supports the ability to consciously recollect and explicitly remember. A deficit in explicit memory, then, causes amnesic patients to be impaired on such tasks. By contrast, tasks that take any change in pro— cessing performance induced by prior exposure as a manifestation of memory do not require explicit remem- bering of the experience. These tasks are said to invoke implicit memory, independently of the operation of the hippocampal system, and are performed normally by amnesic patients. This View seems similar to the declarative—procedural memory account that we have offered and, in fact, is often confused with it. However, note that on the explicit— implicit memory view, the critical element determining impaired versus spared memory performance is how mem— ory is assessed—that is, whether it is assessed explicitly and, thus, requires conscious recollection versus being assessed implicitly, without needing conscious recollec- tion of any particular study episode. This dependence on the means of testing was emphasized by Richardson— Klavehn and Bjork (1988), who, rather than distinguish- in g between explicit and implicit memory, favored draw— ing a distinction between direct and indirect tests of memory, respectively. On the declarative—procedural memory view, by contrast, the critical issue is whether or not performance requires memory representations of the relations among the elements of some experience or scene. The explicit—implicit account does not make such representational claims. The reason that these accounts have often been con— fused with one another is that any explicit (direct) mem— ory test necessarily involves declarative memory for re— lations. That is, explicit (direct) memory tests require, by definition, the ability to gain conscious access to the prior learning episode associated with the test item, thereby requiring memory for some relation between the to—be—tested item and the prior learning experience in which it occurred. Thus, any deficit on an explicit (di- rect) memory test, such as that seen in amnesia, could reflect a deficit of explicit memory, of relational mem— ory, or of both. To illustrate with One of our earlier real- world examples, a failure to recall now who it was that one bumped into at the movie theater two weekends ago could reflect (on the declarative—procedural account) a failure of memory for the set of relations between that night, that movie theater, that movie, and that set of peo— ple who converged at that time, it could reflect (on the explicit—implicit account) an inability to gain conscious access to the perfectly intact memory of the episode of that particular night at the movies, or it could reflect both types of failure. Accordingly, the same evidence can be—and has been—taken as support for these two com— peting multiple memory systems theories. Processing views. The alternative framework for un- derstanding neuropsychological dissociation evidence endorses processing—based views (see, e.g., Roediger, 1990, 2000), in which only a single memory system is postulated. On such views, the different classes of mem— ory test performances that are impaired versus spared in amnesia depend on different processing modes, rather than on different forms of memory—that is, it is a par- ticular mode of processing, rather than a particular form of memory, that is compromised in amnesia. Here, the important difference between, for example, skill learn— ing and recognition memory is said to concern the dif- ferent processing demands imposed by these tasks. More particularly, on Roediger’s (1990, 2000) version of this account, the tasks differ in the match between study time and test time processes and, thus, in transfer appropriate processing. Consider the typical skill—learning task (e.g., mirror tracing or manual tracking), in which the subject per— forms the same operations over and over throughout the experiment and memory is assessed in terms of the in- crease in performance across trials. There, the operations being performed at “test” are the same as those being performed at “study;” the distinction between “study” and “test” is only a virtual one, with complete transfer of the processing requirements. Subjects need only express their increasingly skilled performance on each successive trial. By contrast, a recognition memory task involves separate study and test phases, with different processing requirements during the two phases. During study, when subjects first encounter the to—be—remembered stimuli, they must engage in processes that will permit them to identify and comprehend the meaning of each item on the list and then in whatever rehearsal and maintenance strategies or other encoding strategies will help them to remember the items as well as they can. At test, they must invoke retrieval processes, using various search strategies to examine their memories—including the conceptual processing that is associated with gaining conscious or explicit access to memory for the previous study episode~and making attributional judgments as to whether or not a given item that seems familiar is part of that previous study episode. To return to our real—world example, when asked to re- member who it was that one bumped into at the movies two weekends ago, one must engage a particular mode of processing that involves an effortful attempt to gain ac— cess to memory for and consciously recollect a specific episode and to make attributional judgments about the source of any information that is retrieved (Was that when I saw the movie Gladiator or Memento or the Bourne Identity? Was it young Ricky whom we ran into at the movies, or was it at the restaurant beforehand? Was it that occasion, two weekends ago, or any one of the many other movie going occasions in our life .7). This sort of processing is very different from what one did while at the movies, when one was engaged in taking in the movie, visiting with friends, eating popcorn, and so forth. Clearly, there is a large mismatch here between the mode of processing at study time and the mode of pro— cessing at test time. 174 RYAN AND COHEN Tasks that differ more in terms of study—test match are said to require more conceptual or attributional process- ing at test time in order to bridge the difference in pro- cessing between study time and test time. On processing views, it is this conceptual or attributional processing that is selectively compromised in amnesia, interfering with the ability to deal with mismatch between study time and test time processes and preventing the subjec- tive component of remembering. It is important to note that multiple memory systems Views do not deny the critical role of different memory processes. Where they differ from processing theories, in our view, is in the claim that there are different forms of memory—different types of memory representations supported by the multiple systems. Thus, on a multiple memory systems view, the deficits in new learning in amnesia result from the failure to form and/or maintain and/or retrieve and use a particular type of memory rep— resentation (e. g., relational memories, in our declarative— procedural account). Note here the specification of am- nesia in terms of particular processes operating on a par- ticular class of memory representations. However, on a processing view, the deficits are said to arise from a failure specific only to a particular mode or type of processing (e. g., conceptual, attributional, or conscious processing), without regard to which representations are to be oper- ated upon. (For other views of the issues associated with distinguishing between multiple memory systems and processing accounts, see Foster & Jelic, 1999; Roediger, 2000; Sherry & Schacter, 1987; Tulving, 1999). Having briefly sketched these three alternative theo- retical perspectives, we will use them, in the following analysis of neuropsychological dissociation evidence, to examine the strengths and weaknesses of the two neuro— psychological dissociation approaches we identified above. Dissociation Across Tasks The first of the two classes of neuropsychological dis- sociation evidence used extensively in the literature to support the multiple memory systems claim is what we have called the dissociation across tasks approach. We will examine its ability to isolate the critical dimensions of difference between the dissociated performances and, thereby, to determine what it is capable of telling us about the nature of amnesia and the organization of nor- mal memory. The classic evidence using this approach comes from the findings of spared motor skill learning despite impaired recall or recognition memory in amne— sia, as we have seen. Two specific examples will be con- sidered here. In the mirror—tracing task used by Milner (Milner, 1962; Milner et al., 1968), a drawing of a star is presented, which can be seen only in mirror—reversed form. The subject is to trace the outline of the star while viewing it and his or her hand through a mirror, trying to stay be— tween a pair of lines representing the outer edge of the star. This is done over and over again across a set of iden— tical trials. The measure of memory is the reduction across trials in the number of errors (i.e., in the number of times the pencil falls outside the lines while tracing). This task is the same for the “study” exposures as for the “test” exposures. As is the nature of skill—learnin g tasks, the same task is presented on every trial of the experi— ment, with the same stimulus, and performance is as- sessed throughout The processing requirements of this task include visual form analysis and perceptual—motor integration, to appreciate the shape and dimensions of the star and to use that information to guide the hand along the outline of the star with sufficient precision; those processing requirements are the same for “study” as for “test.” The memory requirements of the task in- volve the acquisition and expression of perceptual— motor skill, in order to show improvement in perfor— mance across trials.1 The descriptions of this task with regard to each of the seven dimensions are summarized in Table 1. Consideration of other examples of motor skill learning in amnesia leads to a nearly identical analysis. For ex— ample, the rotary pursuit task on which amnesic patients have successfully shown incremental learning (e. g., Brooks & Baddeley, 1976; Cohen, 1981; Corkin, 1968) involves the repeated presentation, on each trial, of a small target circle on a platter rotating at a certain speed. The subject is to maintain contact of a hand-held stylus with the target circle, over and over again across a set of identical trials. The measure of memory, as above, is the reduction across trials in the number of errors—in this case, the number of times the stylus loses contact with the target circle. (An additional measure is the correspond— ing increase in the percentage of time the stylus stays on target.) For this task, too, there is a complete match be- tween “study” and “test,” since the same task and stimu— lus are presented on every trial of the experiment, and performance is assessed throughout The processing re- quirements of this task include visual form analysis, analysis of target speed and trajectory, and perceptual— motor integration, in order to apprehend the “path” that the target circle is taking in time and space and to use that information to track the target with the stylus with sufficient precision. The memory requirements of the task involve the acquisition and expression of perceptual— motor skill, involved in getting better able to see and an— ticipate the moving position of the target circle in order to show improvement across trials in tracking perfor— mance. By contrast, consider the tests of recall or recognition memory on which amnesic patients are so impaired. As an example, let us take the recognition memory for faces, on which H.M.’s impaired memory performance was contrasted to his impressive learning of mirror trac— ing in Milner et al. (1968). In this task, a set of previ— ously unfamiliar faces is presented, and the subject is to make either a gender decision or an age decision about each face. This constitutes the “study” phase of the ex— periment, although no mention is made to the subject of NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE 175 a subsequent memory test (i.e., incidental learning in— structions are used). Then there is a subsequent test phase, in which there is another presentation of faces, some of which were previously presented in the study phase (old) and some of which are presented for the first time at test (new), and the subject is to make a yes/no recognition decision for each face. The measure of mem— ory is the degree of success in correctly identifying which of the faces had been presented previously. The task is thus quite different for the study phase than for the test phase, with different processing and memory require- ments for study than for test. At study, the processing and memory requirements include the following: visual analysis of faces, involving the extraction of information necessary to determine age and gender; memory re— trieval of preexperimental knowledge about the features of faces that are diagnostic of male versus female and old versus young, such as hair length, smoothness of skin, and so forth; and incidental encoding of the study time faces. The processing and memory requirements at test include visual analysis of faces, involving the ex- traction of information necessary to determine previous exposure, and memory retrieval of previously presented faces. There are some additional requirements that are more theory bound. Proponents of the explicit—implicit memory distinction would emphasize the requirement in this task (but not in the skill-learning tasks) for conscious recollec- tion of the study episode, to permit the subject to judge which faces were encountered during that episode. Adher- ents of the declarative—procedural memory distinction would highlight the requirement in this task (but not in the skill—learnin g tasks) for memory for relations among the various faces presented during study and the other cues that were part of the study event, so as to permit the subject to judge which faces were associated with that event. They would also emphasize the requirement for representational flexibility, permitting the subject to take memory for information presented in the context of mak— ing gender or age judgments and to use that memory in the context of making judgments of previous occurrence. Supporters of a processing View would point to the re— quirement in this task (but not in the skill—learnin g tasks) for conceptual and attributional processing, to permit the subject to correctly attribute his or her familiarity for, or knowledge of, the old test faces to their having been Viewed during the study phase. And they would appeal to transfer—appropriate processing, emphasizing the diffi— culty in transferring the fruits of processing done in the study phase to the processing demands of the test phase. The descriptions for each of the seven dimensions for the face recognition task are summarized in Table 1, pre- sented in comparison with those for the mirror—tracing task. It is clear that the two tasks differ not only in their memory representation demands, but in all the other di- mensions as well. Thus, the dissociation of performance seen in amnesia here is between two different tasks, tested at different times with different stimulus materi- als, different methods and instructions, and different types of measures, and differing in their processing and memory requirements, as well as in the match between study time and test time requirements. On the basis of these differences across dimensions, one may argue, for example, that amnesic patients are impaired at process- ing the fine differences among various face stimuli, whereas perceptual—motor integration is intact. That ac— count then would highlight differences in processing re— quirements across the two tasks. These two modes of processing may arise from a single memory system, or the processes may be unique to separate systems of mem- ory. Therefore, a multiple memory systems explanation is not necessarily required to explain the observed dis- sociation. Accordingly, although multiple memory sys— tems claims provide a good account of this performance dissociation by attributing the dissociation to the differ— ence in memory representation requirements, alternative explanations that emphasize the other dimensions of dif- ference can be supported just as well. For that reason, these data do not provide much leverage in permitting us to distinguish among the major alternative accounts con- sidered above, each of which draws support from these findings. Table 2 Dissociated Performances Using the Dissociation Across Instruction Conditions Approach ‘ in Cohen and Squire (1980) - Dimension Mirror-Reading Instructions Recognition Memory Instructions Type of stimuli mirror-reversed word triplets standard orientation words Obtained measures reading time per triplet % correct recognition Study—test match yes no Methods/instructions mirror reading word recognition Processing requirements visual pattern analysis word form analysis speech production Memory requirements perceptual skill word/letter knowledge retrieval acquisition and expression of visual pattern analysis word form analysis conceptual/attributional processing memory for words conscious recollection memory for relations representational flexibility Note—Dimensions of diflerence between dissociated performances appear in bold. 176 RYAN AND COHEN Dissociation Across Instruction Conditions The second class of neuropsychological dissociation evidence used to support multiple memory systems claims comes from what we have called the dissociation across instruction conditions approach. In order to pro— vide a more focused test of the possibility of multiple memory systems, Cohen and Squire (1980) used a single kind of learning experience, with a single set of stimulus materials, and then tested with different instructions that placed different demands on memory. The performance dissociation that emerged in amnesia (normal acquisi- tion and expression of skill in mirror reading but im— paired recognition memory for the same words) was very influential in pressing the multiple memory systems claim. But how well does this paradigm really fare with regard to reducing the dimensions on which the dissoci— ated performances differ, with respect to our seven di— mensions? As will be outlined in the following discus— sion and summarized in Table 2, despite the power of Cohen and Squire’s dissociation, it will fail, in the analysis here, to isolate memory differences as the soledimension of difference. In our second example of dissociation across instructions conditions, Graf et al. (1984) presented a more controlled study that was an improvement over Cohen and Squire’s design, but this study, in the end, will also fail to isolate the memory dimension as the one to which dissociations in amnesic performance could be at— tributed. Thus, despite increased methodological con— trols from Cohen and Squire’s to Graf et al.’s study, the critical debate between a multiple memory systems ac— count and a unitary system view remains unresolved. In Cohen and Squire’s (1980) paradigm, the subject reads a series of word triplets presented in mirror—reversed text, some presented only once during the course of the experiment (novel) and others presented in each succes— sive block of the experiment (repeated). The task is to read each triplet aloud as quickly and accurately as pos- sible; a given triplet stays on the screen until the subject reads all three words correctly. This constitutes the “study” phase for both to—be—tested performances, al— though the subjects only instructions are to read the words, and no mention is made of a subsequent memory task. Memory is then assessed in two different ways, under two test instruction conditions. Under mirror—reading instructions, the subject is once again instructed to read the mirror—presented word triplets as quickly and accurately as possible. Thus, the task and the performance measures remain equivalent across the study and the test conditions. The processing requirements include visual pattern analysis, word form analysis, and speech production, in order to correctly discriminate and interpret the mirror-reversed letters (especially, in dis- tinguishing bs from ds, ps from 613, etc.) so as to correctly identify and read the words. The memory requirements involve memory retrieval of preexperimental knowledge about visual features of letters and words, to help guide analysis of letter and word forms, and the acquisition and expression of perceptual skill, involved in learning the mapping between mirror-reversed and standard orienta— tion text, in order to show improvement across trials in reading performance. Under recognition memory instructions, the subject is to make a yes/no recognition decision for each item, old and new, presented in standard orientation. As with the face recognition task described above, the measure of memory is the ability to correctly identify which of the words have been presented previously. Different types of processing and memory representations are required for study than for test in recognition memory, unlike with mirror reading. At study, the processing and memory re— quirements are those of mirror reading. At test, the pro— cessing and memory requirements include visual pattern and word form analysis of the standard orientation words and memory for the recently read words, extracting and using whatever information may be helpful in determin— ing previous exposure. In addition, the major competing theories under con- sideration here hold that recognition memory has other requirements, including (depending on which theory one consults; see above) conscious recollection of the study episode, memory for relations among the various items presented during study and the other cues that were part of the study event, representational flexibility, and con- ceptual and attributional processing, in order to correctly identify old items as having been encountered during this particular experiment. When one compares across the two columns in Table 2, it is clear that the dissociated conditions differ on di- mensions other than memory representation demands, as imposed by the two distinct test instruction conditions. The dissociated performances are assessed at different times with different stimulus formats (single words in standard orientation vs. word triplets in mirror—reversed form), using different instructions and obtaining differ— ent types of measures. As well, the two different test time instructions impose differences in processing and mem- ory requirements—for example, in the need for concep— tual or attributional processing in one but not the other task and in the match between study time and test time requirements. Therefore, these differences in processing requirements do not necessarily require invoking a mul- tiple memory systems account to explain the observed dissociation in performance. That is, multiple processes may arise from a unitary memory system, with spared and impaired performance being observed on those processes, rather than on any memory system itself. Accordingly, although this paradigm does reduce the number of di— mensions on which the dissociated performances differ, as compared with the paradigms using the dissociation across tasks approach, there are many differences that re- main. As a result, various alternative explanations of the dissociation that emphasize those other differences can be offered, keeping alive all of the major competing ac- counts under consideration here. Our second example is Graf et al. (1984), in which they found normal performance in amnesia under stem com- NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE 177 Table 3 Dissociated Performances Using the Dissociation Across Instruction Conditions Approach in Graf, Squire, and Mandler (1984) Dimension Stem Completion Instructions Cued Recall Instructions Type of stimuli word stems word stems Obtained measures no. of study list words generated no. of study list words generated Study—test match no no Methods/instructions word stem completion: cued recall: (fill in the stem with the first word that comes to mind) Processing requirements visual pattern analysis letter and word form analysis (word) speech production Memory requirements retrieval of word knowledge memory for study list items (fill in the stem with a word from the study list) visual pattern analysis word form analysis (word) speech production conceptuallattributional processing memory for study list items conscious recollection memory for relations representational flexibility Note—Dimensions of difference between dissociated performances appear in bold. pletion instructions but impaired performance for the same items under cued recall instructions. Let us consider this paradigm with regard to our seven dimensions. The comparison between conditions is summarized in Table 3. Here, as in Cohen and Squire’s (1980) paradigm, there is a single learning experience with a single set of stim- ulus materials and study instructions. The subject is pre— sented with a series of words and is to make a judgment on each one; the decision involves either how many vow- els the word contains or how “likeable” it is. This con— stitutes the “study” phase for both to-be-tested perfor— mances, although the subject’s only instructions concern the judgments that are to be made, and no mention is made of a subsequent memory task. Memory is then as— sessed in two different ways, under two test instruction conditions, but with the same set of test items. The test items are word stems, constituting the initial letters of words, half of which had been presented on the study time list for that subject and half of which had not been presented previously to that subject, and Vice versa for another subject. For example, one subject’s study list will include the word motel, but not the word cycle, and the opposite will be the case for the other subject; but at test, both subjects will see the stems mot_ and cyc__. Under stem completion instructions, the subject is to complete the stem with the first word that comes to mind. Each stem can be completed to form several dif— ferent words. The measure of memory is the increased likelihood of completing a stem to form one of the re— cently studied words for stems of words that had been on the study list—that is, in the example above, a higher probability of completing mot_ to motel than of cyc_ to cycle for the first subject, and vice versa for the second subject. The processing requirements include visual pat— tern analysis, letter and word form analysis, and (word) speech production, in order to generate legal word com— pletions. The memory requirements involve memory re— trieval of (preexp erimental) word knowledge, with which to compare the presented word stems, as well as some aspect of memory for (or increased fluency, activation, or accessibility of ) recently presented words from the study list that would make those items more likely to be generated at test. These requirements differ in various ways from those at study, which include letter analysis and processing of word form and word meaning, so as to enable decisions about the number of vowels in and the likeability of any given study list word. They also in— clude incidental encoding of the words. Under cued recall instructions, the subject is to com— plete each word stem with a word from the study list. That is, the stems are to be treated as cues for recalling partic- ular words from the list. The measure of memory is the number of study list words generated in response to the recall cues. The processing requirements include visual pattern and word form analysis and (word) speech pro— duction, in order to compare the presented word stems with words retrieved from memory so as to correctly gen— erate words that were on the study list and that match the stem cue. The memory requirements include memory re- trieval of recently presented words. In addition, the major competing theories under consideration here would also include conscious recollection of the study episode, memory for relations among the various items presented during study and the other cues that were part of the study event, representational flexibility, and conceptual and at- tributional processing, so that words from the specific list of words presented in the study phase of the experiment can be correctly elicited by the stem cues. Just as in the other condition, the test time require- ments differ in various ways from those at study, which, as was described above, include letter analysis and pro— cessing of word form and word meaning, so as to enable decisions about the number of vowels in and the like- ability of any given study list word. They also include in- cidental encoding of the words. An inspection of Table 3 and a comparison with the earlier tables makes clear that this paradigm does reduce the number of dimensions on which the dissociated per- 17 8 RYAN AND COHEN formances differ, moving us in the right direction. Graf et al.’s (1984) study improves over Cohen and Squire’s (1980) study in that the type of stimuli and the measures obtained are now equivalent across the instruction con— ditions. But just as clearly, the dissociated conditions dif- fer not only in their memory demands, but on other di— mensions as well. These differences are imposed by the distinct test instructions, which necessarily entail differ— ent processing and memory requirements, as is illus— trated in Table 3. Thus, despite the methodological im- provements over Cohen and Squire’s study, Graf et al.’s work did not isolate the memory dimension, which is critical to distinguishing a multiple memory systems ac- count from a unitary system (or processing) view. Ac— cordingly, various alternative explanations of the disso— ciation that emphasize those other differences can be offered. For example, for both this experiment and Cohen and Squire (1980), it is possible to argue that it is the differ- ence in processing demands at test time elicited by the different test instruction conditions that is the critical el- ement for understanding the performance dissociations. Thus, processing-based views, in which only a single memory system is postulated, can be maintained as an alternative to multiple memory systems accounts of these particular findings by arguing that amnesic pa- tients have a deficit in conceptual, attributional, or con— scious processing. It is worth noting here that, with the dissociation across instruction conditions approach, it is especially difficult to distinguish between the process— ing view and the particular variant of a multiple memory systems view embodied in the explicit—implicit memory distinction. The explicit—implicit memory distinction is the most process oriented of the multiple memory sys— tems accounts. Its crucial assertion is that the memory systems differ fundamentally in the extent to which they support conscious recollection or conscious awareness of some prior learning episode and that tasks that require the ability to consciously recollect are impaired in am- nesia. Tasks that would test this View with the dissocia- tion across instruction conditions approach necessarily involve a comparison between pairs of conditions that require versus do not require the particular mode of pro— cessing used in remembering consciously, which in turn necessarily leaves both memory requirements and pro— cessing requirements as dimensions of difference be— tween the dissociated performances. Moreover, the limitations of evidence from this disso— ciation approach are apparent even if we choose to in- terpret the evidence wholly within a multiple memory systems framework. Our analysis makes clear that one cannot use the examples of neuropsychological dissocia- tion evidence considered above to adjudicate between the declarative—procedural memory account and the explicit— implicit memory account. As long as the different in— structional sets used to dissociate performance empha- size explicit (direct) versus implicit (indirect) testing of memory, as is the case in the examples here, the con— scious component of explicit memory is confounded with the relational component of declarative memory, as has been described above. Limitations and Solutions The foregoing considerations illustrate weaknesses in the existing neuropsychological dissociation approaches that prevent current findings in the literature, taken in isolation, from definitively ruling out alternative formu— lations of the organization of memory. However, identi— fying limitations in the existing approaches to neuro— psychological dissociation evidence should not be confused with an indictment against the multiple mem— ory systems view. That is, the issues raised here cast doubt on the power of existing empirical approaches to adequately test the theoretical claims or, more accu— rately, to adjudicate among competing theories; the lim- itations considered here are not about the substance of the theoretical claims. This leaves the question, How can we garner the kind of evidence needed to decide among the theories? Two solutions are available: to not consider the neuropsychological dissociation evidence in isola- tion but, rather, to always bring to bear evidence from other converging lines or to develop a more powerful neuropsychological dissociation approach that avoids the critical weaknesses identified with the existing approaches discussed above. These will be discussed in turn. The first of these two suggestions is the converging methods approach that has been widely adopted in cog- nitive neuroscience. As was indicated earlier, claims about multiple memory systems in general and about our multiple memory systems theory in particularhave, from the beginning,taken evidence from and endeavored to pro- vide an account not only of human neuropsychological dissociations, but also of human functional imaging data and findings from both lesion and electrophysiological recording studies of animals. We have explicated the rele— vant findings in detail elsewhere (Cohen & Eichenbaum, 1993; Eichenbaum & Cohen, 2001) and will not recount them here. But the fact of the matter is that there are data from other converging cognitive neuroscience lines of evidence that existing processing views just do not explain and that, taken together with the neuropsychological dis— sociation evidence, provide a way to choose among the potential theories discussed above. Consider, for example, the various neuroimaging findings of hippocampal activation during incidental en- coding of words, faces, or pictures and the correlation between that encoding time activity and success in sub- sequent memory tasks (Brewer, Zhao, Desmond, Glover, & Gabrieli, 1998; Kelley et al., 1998; Kirchoff, Wagner, Maril, & Stern, 2000; Martin, Wiggs, & Weisberg, 1997; Stern et al., 1996; Wagner et al., 1998). In these studies, there was no requirement for conceptual or attributional processing and no need to consciously recollect some previous learning experience, because brain activity was NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE 179 assessed not at test time but at encoding time. One might argue that conscious recollection is part of what subjects do in the course of encoding, but some of the successful demonstrations of hippocampal activation at encoding have involved studies in which nonfamous faces or un- familiar scenes were used as the stimuli, for which sub- jects had no previous experience to recollect (see the re- view in Cohen et al., 1999). Or consider the well-established findings that single neurons in the hippocampi of freely behaving rats fire preferentially to particular conjunctions of cuesthat is, the relations among various objects, locations, and task- relevant behaviors while the animals navigate through their environment (see the review by Eichenbaum, 2000). During the course of whatever processing the animal is doing, different hippocampal neurons take on different firing preferences, each firing preferentially to one or another specific combination of and relations among the constituent elements of the environment. This apparent mapping of the spatial and other relations in the envi- ronment by the set of hippocampal neurons is well han— dled by a theory that attributes a special role to the hip— pocampus in the formation and use of (declarative) memory for relations but has no obvious explanation in processing accounts that posit conceptual, attributional, or, especially, conscious processing as the role of the hip- pocampus. Nor does it find an explanation in the multiple memory systems account that claims a selective role of the hippocampus in explicit memory; it is difficult to see how the tasks that elicit such hippocampal neuron activ- ity in these animals could be said to require conscious recollection of previous episodes. This criticism is true of the animal studies in general, in which one is hard pressed to show that the memory tasks on which animals with hippocampal lesions fail are exclusively those for which normal animals use conscious recollection. Accordingly, it seems that the limitations of the neuro- psychological dissociation evidence can be overcome by making use of these other lines of evidence to success- fully constrain the theories we are considering. Here, as elsewhere, science is well served by bringing to bear as much evidence as possible. Rarely is there the single crit- ical experiment capable of resolving all debates in a given field; rather, multiple findings are used to con— strain the problem space of possible solutions. But a rea— sonable goal would be to expect each line of evidence to be strong enough to be taken in isolation and, of partic- ular relevance here, to seek neuropsychological dissoci- ation evidence that does not suffer from the limitations discussed above. This brings us to our second suggestion—namely, to develop a more powerful neuropsychological dissocia— tion approach that avoids the critical weaknesses that have prevented us from successfully ruling out alterna- tive accounts, without appealing to other converging lines of evidence. One such dissociation approach is out— lined and illustrated in the next section. Regardless of whether one is considering the particular theories under discussion here or other theories and whether the domain is memory or some other cognitive ability, an approach to neuropsychological dissociation evidence that brings us closer to being able to isolate a single dimension of difference would be an important advance. TOWARD A MORE POWERFUL CLASS OF NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE We have argued that the limitations of existing classes of neuropsychological dissociation evidence arise from the failure to successfully minimize the dimensions of difference between the dissociated performances. The ideal approach would permit isolation of memory re— quirements as the single critical dimension of difference between conditions showing performance dissociations in amnesia. We now describe a dissociation within con- dition approach, suggesting that it has the greatest po— tential for targeting differences in memory representa— tion requirements to the exclusion of other differences and, thereby, for providing more definitive tests of mul- tiple memory systems claims. Dissociation Within Condition This strategy involves testing with a single set of ma- terials on a single task with a single set of instructions and deriving measures sensitive to (or revealing of) dif- ferent forms of memory simultaneously, thereby holding constant everything except the memory requirements. The logic here is as follows: (1) There are multiple mem- ory systems; (2) one of the memory systems depends critically on the hippocampal system (and/ or other) structures that are damaged in amnesia, and this memory system is selectively compromised in amnesia; (3) the different memory systems are always processing in their own modes, at the same time, for the same materials; (4) we can devise measures sensitive selectively to the oper- ation of each of the separate systems; (5) normal sub— jects should show evidence of the operation of the sepa— rate systems, assessed simultaneously; and (6) amnesics should show evidence of intact operation only of those memory systems that work independently of the hippo— campal system. This approach has been applied to the issue of preserved versus impaired memory performances in amnesia in a recent series of studies in which subjects were presented with a set of scenes under a single instructional set and their eye movements were monitored while they viewed the scenes (Ryan, Althoff, Whitlow, & Cohen, 2000). We found that different aspects of eye movement behavior, collected simultaneously, revealed different aspects of memory for subjects’ prior viewing history with the scenes. One set of eye movement measures was sensitive to previous exposure to (i.e., repetition of) scenes, pro— vidin g an index of memory for items (here, whole scenes), whereas another set of eye movement measures was sen— sitive to manipulations of’the relations among elements 180 RYAN AND COHEN of those previously viewed scenes, providing an index of memory for relations. We found that amnesic patients showed the effects of repetition just as did normal con— trol subjects but failed to show the normal effects of ma— nipulation of relations. Given that there was a single class of materials, a single set of instructions, and a sin— gle class of (eye movement) measures collected at the same time, this evidence of dissociation within condition makes a particularly compelling case that amnesia is a deficit in, and the hippocampal system is specialized for, a particular aspect or form of memory. First, we will describe the paradigm and results in a lit— tle more detail. We then will explicate the way in which it constitutes an application of the dissociation within con— dition approach and show how it manages to avoid the limitations of the previous dissociation approaches. In this work, the subject sees a series of images of real—world scenes, some presented only once during the experiment (novel scenes), some presented in the same form once in each of the three blocks of the experiment (repeated scenes), and some presented once each in the first two blocks and then presented in manipulated form in the final, critical block (manipulated scenes). Manip- ulated scenes involve a modification of the relations among the scene elements, including deletions, addi— tions, or left—right (or right—left) shifts of one of the ob— jects from the original form of the scene. Any given scene is viewed by the subject in only one form (novel, repeated, or manipulated) but is viewed equally often in the different forms across subjects. Thus, the physically identical scenes that are viewed as manipulated by one subject are viewed as unmanipulated (novel or repeated) by other subjects. The subject’s task is to answer a question about the re- lations among the depicted elements for each scene dis- played on the monitor (e. g., are there kittens behind the boy ?). The instructions make no mention of a subse- quent memory test or of the possibility of manipulations of the scenes. Eye movements are monitored throughout viewing. Note that there is no eye movement response required of the subject; it is incidental to the subject’s task. There is no distinction to be drawn between study and test phases, because the same task, the same type of stimulus materials, and the same instructions are used throughout the experiment. The measure of memory is providedby differences in eye movement patterns elicited to physically identical scenes as a function of differences in their viewing histories. From the eye movements recorded continuously, we derive two classes of measures, obtained simultaneously. One set of measures compares eye movements elicited to previously viewed scenes versus the very same scenes when they are novel. Differences in viewing scenes when they are novel versus repeated constitute a repetition ef- fect, reflecting item memory for (previous exposure to or repetition of) the scenes. The other set of measures compares eye movements elicited to manipulated scenes versus the (very same) scenes when they are repeated in the same unmanipulated form throughout the experi- ment. Differences in viewing scenes when they are ma— nipulated versus repeated unmanipulated and, particu— larly, increased viewing directed to the very region of the manipulation in manipulated scenes (e. g., greater view— ing directed to the now empty region of a scene where a kitten had been in previous viewings of that scene, as compared with viewing of that same region when it had always been empty) constitute a relational manipulation eflect, reflecting memory for relations among the con- stituent elements of scenes. Scenes with these different types of viewing histories (novel, repeated unmanipu— lated, and manipulated) are intermixed. The findings were that normal control subjects showed both types of eye movement effects. They showed mem- ory for items: There was significantly greater sampling of the scene by the eyes (higher number of fixations within the fixed viewing period) and significantly more regions of the scene sampled by the eyes when the scene was novel than when it was repeated. And they showed memory for relations: There was significantly greater viewing directed to the very region of manipulation (higher proportion of total fixations and higher propor- tion of total viewing time) when the scene was manipu— lated than when it was repeated. The latter effect oc- curred whether or not the subjects were consciously aware that a manipulation had occurred; that is, it oc— curred even for scenes that the subjects could not tell had been manipulated. But amnesic patients showed only the repetition effect, which was of the same magnitude as Table 4 Dissociated Performances Using the Dissociation Within Condition Approach in Ryan, Althoff, Whitlow, and Cohen (2000) Dimension Measure 1: Repetition Effect Type of stimuli images of real—world scenes Obtained measures patterns of eye movements Study—test match yes Methods/instructions answer yes/no orienting question: (Are there kittens behind the boy ?) visual obj ect and scene perception Processing requirements processing of spatial relations Memory requirements memory for items (scenes) knowledge retrieval of objects and scenes Measure 2: Manipulation Effect images of real-world scenes patterns of eye movements yes - answer yes/no orienting question: (Are there kittens behind the boy?) visual object and scene perception processing of spatial relations knowledge retrieval of objects and scenes memory for relations Note——Dimensions of difference between dissociated performances appear in bold. NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE that seen in the normal control subjects; none of the am— nesic patients showed a relational manipulation effect. The logic of the dissociation within condition ap- proach and how it is implemented in Ryan et a1. (2000) are illustrated in Table 4, in which that work is consid- ered with regard to our seven dimensions of difference. Looking at the two columns in Table 4, we see not a comparison of two different tasks or of two test instruc— tion conditions but of two simultaneously obtained mea— sures in a single condition of a single task with a single set of stimulus materials under a single set of test in- structions—thus, a dissociation within condition. The processing requirements for both of the dissociated per— formances include Visual object and scene perception and processing of spatial relations, in order to correctly answer the orienting questions. The memory require— ments for both of the dissociated performances include memory retrieval of preexperimental knowledge about visual objects and scenes, to help in answering the ori- enting questions. The only difference emerges in the memory representation requirements for showing the repetition effect, which requires memory for the previ- ous occurrence of scenes, versus those for showing the relational manipulation effect, which requires memory for relations among the constituent elements of scenes. The findings of dissociation—with impairment in am- nesia on measures of memory for relations but sparing on simultaneously collected measures of memory for repetition of items—when subjects are performing within a single processing mode illustrate the power of the dis- sociation within condition approach. Pointing to the spe— cial status of memory for relations, they provide strong support for the multiple memory systems View. Moreover, as we will see next, the alternative formulations under consideration here do not provide a means for accounting for these data. The dissociationreported in Ryan et a1. (2000) emerges between two simultaneously obtained measures in a sin— gle condition of a single task with a single set of stimu— lus materials under a single set of test instructions. For each stimulus in the list, subjects make a decision about the relations present in the scene, with all information required for that decision in full view. Thus, subjects are engaged in just a single mode of processing throughout the experiment, and the dissociation emerges as a func- tion of the previous viewing history with these stimuli. There are no processing mode differences here to be con- founded with the differences in memory requirements that could be used to explain how amnesic patients show a normal repetition effect but no relational manipulation effect. One could conceivably take a narrower View of pro- cessing differences and argue that the different stimulus types (novel, repeated unmanipulated, and manipulated) invoke different processing modes on a trial-by-trial basis. But this is not persuasive for three reasons. First, note that the data showing the two eye movement effects 181 are for physically identical stimuli. For example, for the relational manipulation effect, the finding of increased viewing being directed to the (now empty) region of a manipulated scene where there had previously been a kitten involves a comparison with the amount of viewing of the same scene when there had never been a kitten in that (always empty) location. Hence, any difference in viewing of the empty region is elicited not by the stimu— lus itself but by the comparison of the stimulus with a memory representation of its previous occurrence. Given that every trial can be conceived of as involving a comparison of the stimulus with memory representa- tions of the previously presented scenes, there is no ob— vious processing mode distinction to be drawn among trials. But it is essential that the subject have memory for the relations among the constituent elements of the ear— lier presented scenes in order for the eyes to be attracted to the now empty region. Second, note that the relational manipulation effect occurs whether or not the subjects are aware that there has been a manipulation. That is, the effect is seen just as well for scenes that the subjects claim have not been manipulated. Thus, one cannot argue that the subjects identify the manipulated stimuli as a special class of stimuli and then engage in a different mode of process— ing that manifests itself in a different pattern of eye movements. Third, with regard to the failure of amnesic patients to show this effect, one cannot argue that the particular kind of eye movement pattern elicited by manipulated scenes reflects a certain mode of relational processing that they simply cannot perform. In follow—up studies (Ryan & Cohen, in press), we have been able to show that amnesic patients exhibit the normal effect of increased viewing being directed at the critical region of change in manip— ulated scenes in a short—delay variant of the original par- adigm; they show the deficit only when there are enough intervening items and a long enough delay to tax long— term memory. And it is only the relational manipulation effect that they fail to show at long delays; amnesic pa- tients show an intact repetition effect even at long delays. Taken together, these results show that there is a deficit only in the measure that is sensitive to the comparison of the current stimulus with long-term memory representa- tions of the relations among the elements of previously presented scenes. This set of findings is not accommo- dated by processing accounts. Rather, it is the memory representationrequirements—Specifically, long—term re— lational (declarative) memory—that produce the disso- ciation. Ryan et al.’s (2000) findings also succeed in disam- biguating among competing multiple memory systems accounts. The task and instructions used here assessed memory implicitly (indirectly). The subjects answered a question about the relations among the depicted elements for each scene on the screen without any reference to a memory task; there were no instructions either to study 182 RYAN AND COHEN the scenes for a later test or to refer back to any previous occasions. The eye movements used to assess memory, monitored throughout viewing, were incidental to the subjects’ actual task. Furthermore, the relational manipu- lation effect, which was found to be selectively impaired in amnesia, did not depend in any way on explicit remem— bering or conscious recollection; it occurred (in normal control subjects) whether or not the subjects were con— sciously aware of the manipulations. This is critical, pro— vidin g the means to pit the declarative—procedural mem— ory distinction against the explicit—implicit memory distinction. Because memory is assessed implicitly (in- directly), the explicit—implicit account, holding that the hippocampal system specifically mediates explicit mem— ory, must predict that the relational manipulation effect would be preserved along with the repetition effect; that is, they should share the same fate. However, the declarative— procedural memory distinction, holding that the hippo— campal system specifically mediates memory for rela— tions, predicts that amnesic patients would be impaired selectively. The findings that amnesic patients fail se— lectively to show the relational manipulation effect while, nonetheless, exhibiting the repetition effect normally, despite both effects being assessed implicitly (indi- rectly), provide strong support for the view that the hippocampal system is tied to memory for relations and not to explicit memory. It should be noted that additional support for this con- clusion comes from Chun and Phelps (1999), who re— ported the failure of amnesic patients to show contextual cuing, a form of implicit perceptual learning of the re— peated configurations of targets and background con— texts. In this visual search task, subjects were to identify a rotated T among a set of distractor Ls presented at var— ious locations and orientations. In a subset of the trials, the locations of the target and the distractors remained constant, thus providing a repeated spatial context or spatial relations that could be used to guide search per- formance. In normal subjects, such trials elicited a de— crease in search time, as compared with novel trials, even though the task assessed memory implicitly (indirectly) and the subjects were unable to distinguish explicitly the trials with repeated contexts from novel trials; amnesic patients failed to show this relational effect (but see Manns & Squire, 2001). Contrasting the Different Classes of Dissociation Evidence Why is it that whereas various competing theories can all provide an account of earlier neuropsychological dis- sociation evidence, they cannot provide an account of this class of evidence? Let us examine these accounts with respect to the different neuropsychological dissoci— ation approaches. Processing views can be maintained in findings of dissociation across tasks, because the two tasks inevitably end up engendering quite different modes of processing. As we have seen, in the dissocia— tions observed between motor skill learning and recog— nition memory, subjects are confronted with fundamen- tally different cognitive challenges at two different test occasions—they are instructed to perform different op— erations with different classes of stimuli—and their per— formance is assessed in fundamentally different ways (see Table 1). These differences can elicit differences in mode of processing that are as compelling as the differ— ences in memory requirements. Likewise, processing views can be maintained in findings of dissociation across instruction conditions, because the different in- structions given at different times necessarily invoke dif- ferent modes of processing, even for the same test mate- rials (see Table 3). For example, as we saw in stem completion versus cued recall (Graf et a1., 1984), the in- structions at test time purposely directed subjects to per— form different processing tasks that differed with regard to whether or not they encouraged conscious recollec- tion. Thus, the differences in memory requirements are confounded with difference in processing requirements. These processing requirements may all arise from a sin- gle system of memory. By comparing across conditions that differ with respect to processing requirements, the argument remains that the dissociation in performance may arise either from a single or from multiple systems of memory. That is, the dissociation may occur across processes that are each part of a single system, or the dis— sociation may persist in types of processing that are unique to separate memory systems. But the dissociation within condition approach avoids different tasks and different instruction conditions. More generally, its power comes from its ability to reduce the dimensions of difference among the dissociated perfor— mances. As can be seen in Table 5, the dissociation within Table 5 Dimensions of Difference Between Dissociated Performances That Remain in Each Dissociation Approach Dissociation Dissociation Across Dissociation Dimension of Difference Across Tasks Instruction Conditions Within Condition Time of administration ‘I ‘1 Type of stimuli V Obtained measures ‘/ Study—test match \/ Methodology and instructions \I *1 Processing requirements \I ‘1 Memory requirements \I \I ‘l NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE condition approach has the potential to isolate a sin- gle critical dimension of difference, permitting us to en— tertain targeting memory representation requirements selectively. Precedents and Other Examples of the Dissociation Within Condition Approach Ryan et al.’s (2000) work, discussed above, was ex— pressly designed as an implementation of the dissocia— tion within condition approach, although the explication of the logic of the approach and the evaluation of its po- tential for addressing critical claims about multiple memory systems are done for the first time in the pres— ent paper. There are, however, two precedents for the ap- proach within the amnesia literature on preserved versus impaired memory of which we are aware. The first is the mirror-reading study of Cohen and Squire (1980), dis- cussed earlier. In addition to the dissociation it offered between intact perceptual (visual pattern analysis) skill learning and impaired recognition memory—a dissocia- tion across instruction conditions—the study also included a within—conditioncomp arison. Within the mirror—reading component of the study, some of the trials involved novel word triplets, seen only once during the study, and some were repeated word triplets, seen many times throughout the study; these trial types were intermixed. Enhance- ment of reading speed across novel triplets provided a measure of generalized skill learning; amnesic patients proved to be fully intact on this measure. An additional increment in speed for the repeated triplets, as compared with novel triplets, was observed. But the results were mixed. Amnesic patients showed an advantage for re- peated triplets over novel triplets, but less than that shown by normal control subjects. As was indicated in Cohen (1984), this is likely due to that element of per- formance being supported by a combination of two dis- tinct phenomena: repetition priming, producing en- hanced speed of reading of each of the often repeated words, and consciously aware (i.e., explicit) declarative memory for relations among the words within each triplet, producing the ability to generate the second and third words of a triplet upon reading the initial word of the triplet without having to complete the process of reading. To the extent that repetition priming is intact in amnesia, one would expect to see enhancedperformance for repeated triplets over novel triplets; but to the extent that memory for relations among the words within triplets is impaired in amnesia, one would expect failure to show the full advantage for repeated triplets that normal subjects show. Accordingly, although on our present analysis the Cohen and Squire study included a within-condition comparison, the results did not produce the full dissoci- ation of normal performance on one measure and failure on the other, simultaneously obtained measure that would signal the separate, simultaneous operation of hippocamp a1 and nonhippocampal systems, respectively. The second study is the one by Chun and Phelps (1999), discussed at the end of the preceding section. 183 The design of this perceptual—learning experiment is much like that of Cohen and Squire (1980), with novel trials and often repeated trials intermixed throughout the study. Enhancement of search speed across novel trials provided a measure of generalized skill learning, whereas any additional increment in performance for the repeated trials, as compared with the novel trials, provided the measure of memory for relations (here, the spatial con- textual relations between targets and distractors). As we have seen, amnesic patients failed to show any improve- ment in performance for the repeated trials; they also showed generalized skill learning, manifested as in- creasingly faster search across novel trials. However, performance on the novel trials was not specifically compared between amnesic patients and control sub- jects, so it is not clear whether skill learning was fully preserved in amnesia here. Thus, the results did not quite produce the full dissociation of normal performance on one measure and failure on the other, simultaneously ob- tained measure that would demonstrate the separate, si— multaneous operation of hippocampal and nonhip- pocampal systems, respectively. Nonetheless, this and the previous study have the within—condition comparison that is fully realized in Ryan et a1. (2000), and they pro— vide data strongly supportive of the same conclusion. In work on long-term versus short-term or working memory, there is one well—known study of amnesic pa— tients that ostensibly makes use of a dissociation within condition strategy. Baddeley and Warrington (1970) looked at performance of amnesic patients on immediate free recall. In normal subjects, performance is sensitive to the serial position of items in the study list: The initial items in the study list are relatively well recalled, as are the terminal items in the list, as compared with those items that appeared in the middle portion of the study list. The advantage for the early items is called the pri— macy effect, and the advantage for the final items is called the recency effect. Baddeley and Warrington found that amnesic patients showed the recency effect but had a reduced primacy effect. Attributing the recency effect to short—term or working memory and the primacy effect to long—term memory, these authors took these data as supporting a fundamental distinction between short-term and long-term memory stores. It is not clear that the primacy effect and the recency effect can be mapped directly and completely onto long-term and short—term memory; nor is it clear that subjects are using the same strategies and processes in recalling items that were just presented and were just being rehearsed and still have sensory representations available, as compared with recalling items from 'the very beginning of the list. Nonetheless, the logic being used here—that there are multiple systems making separate contributions to per- formance on the single task being performed—is very much in the dissociation within condition spirit. There is also a parallel example of the dissociation within condition approach from a different domain of neuropsychology that provides compelling evidence 184 RYAN AND COHEN concerning the question Of whether there are function- ally distinct systems for recognition of faces versus other (nonface) visual objects. Investigators have addressed this nearly exclusivelyby using the standard dissociation across tasks approach. There are many reports of face agnosia—a deficit in identifying people by their faces— despite intact recognition of nonface visual objects, and there are some recent demonstrations of (nonface) visual Object agnosia with no deficit of face recognition. But this issue is complicated by the fact that faces and non- face stimuli differ in many regards, including the physi— cal nature Of the stimuli, the amount of expertise with the stimuli, the degree to which items need to be specif— ically individuated to support performance, and so forth. Yet, in the course Of a set of 19 experiments with a pa— tient with (nonface) visual object agnosia, Moscovitch, Winocur, and Behrmann (1997) offered two experiments that, by using a dissociation within condition sensibility, provide a different angle on the issue. Presented with a set of paintings (by such artists as Arcimbaldo, Biblico, and Terra) of faces whose compo- nent features were fruits, vegetables, and/or other ob— jects, normal subjects were simultaneously aware of both the faces and the objects, but patient C.K. was not; he had no problem seeing the faces but was rarely able to detect that the faces were made up of Obj ects or were pe- culiar in any way. In the other experiment of interest here, subjects viewed the painting “The Faces in the For- est,” showing a scene involving trees, rocks, streams, and other natural objects. But the scene was composed in such a way that, by careful configuration of these natural objects, it was possible to see a set of faces of various sizes in various places in the forest. Normal subjects found it difficult to see the faces, and it took them a long time to discover them among the trees, rocks, and so forth of which they were composed; they had no trouble at all identifying the nonface objects. Patient C.K., by con— trast, had little problem perceiving the faces, discovering them more rapidly than did the normal subjects, but had great difficulty in appreciating the nonface objects. What makes this work particularly compelling is its ability to give a sense of face and object processing Op- eratin g simultaneously under normal circumstances, but not in this patient with brain insult, thus realizing one of the critical premises and strengths of the dissociation within condition approach. In doing so, these findings provide particularly compelling evidence for a dissocia— tion between face processing and (nonface) visual object processing, however those systems are conceived. Extensions of the Dissociation Within Condition Approach Dissociation evidence is used throughout cognitive neuroscience; not just within neuropsychology. Accord— ingly, it is not the only method for which our analysis of different classes of dissociation evidence is relevant. The comparisons across conditions or across trials that are an essential part of functional imaging and psychophysio- logical studies raise the very same issues as those consid— ered in this article for neuropsychological dissociations. Indeed, the logic of the dissociation within condition ap- proach seems ideal for functional imaging and psy- chophysiological work. The approach is based on the idea that, for the domain of memory, various systems are always processing in their own modes, at the same time, and for the same materials and that the ideal research strategy is to derive measures sensitive selectively to the operation of each of the separate systems. Any technique that permits monitoring of the participation of multiple brain regions/systems from trial to trial throughout the experiment can make use of the dissociation within con- dition approach. Thus, for example, fMRI or ERP evi- dence of changes in the extent to which different systems are actively engaged, while sampling them continuously and simultaneously during a single condition Of a single task with a single class of stimuli differing only on the dimension being tested (specifically, previous viewing history), can provide dissociation within condition evi- dence of the separate, simultaneous operation of differ- ent memory systems. Taken altogether, if researchers use converging lines of evidence and are armed with well-reasoned ideas and well-motivated experiments, the dissociation within con- dition approach outlined here is capable of providing a particularly powerful tool for testing claims about the or— ganizationof memory (and other cognitive capacities) in the brain. Applied to the issue of preserved versus im— paired memory performances in amnesia, this approach permits disambiguation among competing theories, pro- viding strong support for the multiple memory systems view that distinguishes between declarative (relational) and procedural memory. Its application to other neuro- psychological data and to other cognitive neuroscience data holds great promise. REFERENCES BADDELEY,A. D., & WARRINGTON, E K. (1970). Amnesia and the dis- tinction between long-term and short—term memory. Joumal of Ver- balLearning & VerbalBehavior, 9, 176-189. BREWER,J. B., ZHAO, Z, DESMOND, J. E., GLOVER, G. 1-1., & GABRIEL], J. D. E (1998). Making memories: Brain activity that predicts hOW well visual experience will be remembered. Science, 281, 1185-1187. BROOKS, D. N., & BADDELEY,A. D. (1976). What can amnesic patients learn? Neuropsychologia, 14, 111—122. CHUN, M. M., & PHELPS, E A (1999.) Memory deficits for implicit contextual information in amnesic subjects with hippocampal dam- age. Nature Neuroscience, 2, 844-847. COHEN, N. J. (198 l). Neuropsychological evidence for a distinction be- tween procedural and declarative knowledge in human memory and amnesia. Unpublished doctoral dissertation, University of Califor- nia, San Diego. . COHEN, N. J. (1984). Preserved learning capacity in amnesia: Evidence for multiple memory systems. In L. R. Squire & N. Butters (Eds), Neuropsychology of memory (pp. 83-103). New York: Guilford. COHEN, N. J., & EICHENBAUM, H (1993). Memory, amnesia, and the hippocampalsystem. Cambridge, MA: MIT Press. COHEN, N. J., POLDRACK, R. A, & EICHENBAUM, H (1997). Memory for items and memory for relations in a procedural/declarative mem- ory framework. Memory, 5, 131-178. NEUROPSYCHOLOGICAL DISSOCIATION EVIDENCE 185 COHEN, N. J., & RYAN, J. D. (2003). Hippocampal binding theory. Man- uscript in preparation. COHEN, N. J., RYAN, J. [D.], HUNT, C., ROMINE, L., WSZALEK, T., s; NASH, C (1999). Hippocampal system and declarative (relational) memory: Summarizing the data from functional neuroimaging stud- ies. Hippocampus, 9, 83-98. COHEN, N. [l], & SQUIRE, L. R (1980). Preserved learning and reten- tion of pattern-analyzing skill in amnesia: Dissociation of knowing how and knowing that. Science, 210, 207-210. CORKIN, S. (1968). Acquisition of motor skill after bilateral medial temporal-lobe excision. Neuropsychologia, 6, 255-265. EICHENBAUM, H. (1997). Declarative memory: Insights from cognitive neurobiologyAnnual Review of Psychology, 48, 547—572. EICHENBAUM, H (1999). Conscious awareness, memory and the hip- pocampus. Nature Neuroscience, 2, 775-7 76. EICHENBAUM, H. (2000). Hippocampus: Mapping or memory? Current Biology, 10, R785-R787. EICHENBAUM, H, & COHEN, N. J. (2001). From conditioning to con- scious recollection: Memory systems of the brain. Oxford: Oxford University Press. FOSTER, J. K, & JELIC, M. (1999).Memory: Systems, process or function ? Oxford: Oxford University Press. GABRIELI, J. D. E. (1999). The architecture of human memory. In J. K. Foster & M. Jelic (Eds.), Memory: systems, process or function? (pp. 205-231). Oxford: Oxford University Press. GRAF, P., & SCHACTER, D. L. (1985). Implicit and explicit memory for new associations in normal and amnesic patients. Journal of Experi- mental Psychology: Learning, Memory, & Cognition, 11, 501-518. GRAF, P., SQUIRE, L. R, & MANDLER, G. (1984). The information that amnesic patients do notforget. Journal of Experimental Psychology: Learning, Memory, & Cognition, 10, 164-178. KELLY, W. M. MIEer, F. M., MCDERMOTT, K. B., BUCKNER, R L., RAICHLE, M. E, COHEN, N. J., OLLINGER,J. M., AKBUDAK, E, CON— TURO, T. E, SNYDER, A Z, & PETERSEN, S. E (1998). Hemispheric specialization in human dorsal frontal cortex and medial temporal lobe for verbal and nonverbal memory encoding. Neuron, 20, 927—936. KIRCHOFF, B. A, WAGNER, A D., MARIL, A, & STERN, C. E (2000). Prefrontal—temporal circuitry for episodic encoding and subsequent memory. Journal of Neuroscience, 20, 6173—6180. MANNS, J. R, & SQUIRE, L. R (2001). Perceptual learning, awareness, and the hippocampus. Hippocampus, 11, 776-782. MARTIN, A, WIGGS, C., & WEISBERG, J. (1997). Modulation of human medial temporal lobe activity by form, meaning and experience. Hip- pocampus, 7, 587-593. MILNER, B. (1962). Les troubles de la mémoire accompagnant des le- sions hippocampiques bilatérales. In P. Passouant (Ed.), Physiologic de l’hippocampe (pp. 257-262) Paris: Centre de la Recherche Sci- entifique. MILNER, B., CORKIN,S., & TEUBER,H L. (1968). Further analysis of the hippocampal amnesic syndrome: 14-year follow-up study of HM. Neuropsychologia, 6, 215-234. Moscovrrcn, M., WINOCUR, G., & BEHRMANN, M. (1997). What is spe- cial about face recognition? Nineteen experiments on a person with visual object agnosia and dyslexia butnormal face recognition. Jour- nal of Cognitive Neuroscience, 9, 555-604. NADEL, L. (1994). Multiple memory systems: What and why, an up- date. In D. L. Schacter & E. Tulving (Eds.), Memory Systems 1994 (pp. 39—63). Cambridge, MA: MIT Press. RICHARDsoN-KLAVEHN,A, & BJORK, R A (1988).Measures of memory. Annual Review of Psychology, 39, 475—543. ROEDIGER, H L., 111 (1990). Implicit memory: Retention without re- membering.American Psychologist, 45, 1043-1056. ROEDIGER, H. L., III (2000). Why retrieval is the key process in under— standing human memory. In E. Tulving (Ed.), Memory, conscious— ness, and the brain: The Tallinn Conference (pp. 52-75). Philadel- phia: Psychology Press. RYAN, J. D., ALTHOFF, R R, WHITLow, S., & COHEN, N. J. (2000). Am- nesia is a deficit in declarative (relational) memory. Psychological Science, 11, 454-461. RYAN, J. D., & COHEN, N. J. (in press). Processing and short-term re— tention of relational information in amnesia. Neuropsychologia. SCHACTER, D. L. (1987). Implicit memory: History and current status. Journal of Experimental Psychology: Learning, Memory, & Cogni- tion, 13, 501-508. SCHACTER, D. L., & TULVING, E (1982). Memory, amnesia and the episodic/semantic distinction. In R. L. Isaacson & N. E. Spear (Eds.), Expression of knowledge (pp. 33-61). New York: Plenum. SCHACTER, D. L., & TULVING, E (1994). What are the memory systems of 1994? In D. L. Schacter & E. Tulving, (Eds.), Memory systems 1994 (pp. 1-38). Cambridge, MA: MIT Press. SHERRY, D. E, & SCHAcrER, D. L. (1987). The evolution of multiple memory systems. PsychologicalReview, 94, 439-454. SQUIRE, L. R. (1987). Memory and brain. New York: Oxford University Press. SQUIRE, L. R (1992). Memory and the hippocampus: A synthesis of findings with rats, monkeys, and humans. Psychological Review, 99, 195 -23 l. STERN, C E, CORKIN, S., GONZALEZ, R G., GUIMARAEs,A R, BAKER, J. R, JENNINGS, P. J., CARR, C. A, SUGIURA, R M., VEDANTHAM, V., & ROSEN, B. R (1996). The hippocampal formation participates in novel picture encoding: Evidence from functional magnetic reso- nance imaging. Proceedings of the National Academy of Sciences, 93, 8660-8665. TULVING, E (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization of memory (pp. 381-403). New York: Academic Press. TULVING, E (1985). How many memory systems are there? American Psychologist, 40, 385-398. TULVING, E (1999). Study of memory: Processes and systems. In J. K. Foster & M. Jelic (Eds.), Memory: Systems, process or function? (pp. 1 1—3 0). Oxford: Oxford University Press. WAGNER, A D., SCHACTER, D. L, ROTTE, M., KOUTsrAAL,W., MARIL,A, DALE, A. M., ROSEN, B. R, & BUCKNER, R L. (1998). Building memories: Remembering and forgetting of verbal experiences as predicted by brain activity. Science, 281, 1 188-1 191. WEISKRANTZ, L. (1987). Neuroanatomy of memory and amnesia: A cam for multiple memory systems. Human Neurobiology, 6, 93-105. NOTE 1. Regarding the description, above, of the processing and memory requirements of the task, it is neither possible nor necessary to offer a full analysis or complete listing here. Clearly, to fully understand and model performance on this task or the others to be considered in this paper, a more complete account would be in order. However, for the pur- poses here of considering the ability of each of the three classes of dis— sociation evidence to isolate a single critical dimension on which the dissociated performances differ, only a general description of the major requirements will prove necessary. (Manuscript received August 14, 2001; revision accepted for publication April 1, 2003.) ...
View Full Document

This note was uploaded on 12/01/2011 for the course CLP 7934 taught by Professor Staff during the Fall '08 term at University of Florida.

Page1 / 18

ryan_cohen_03 - Cognitive Afi‘ective& Behavioral...

This preview shows document pages 1 - 18. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online