Memory is a biological adaptation that is crucial to numerous aspects of everyday existence. Although people are often able to remember past experiences with a reasonable degree of accuracy, under certain circumstances, memory can be subject to substantial distortion.1 Issues concerning accuracy and distortion in human memory have recently achieved widespread scientific and social prominence in debates concerning the veracity of recovered memories of childhood abuse.2,3
The recovered memories debate is complex and frequently vexing. Numerous biological, psychological, social, and even cultural factors are involved, and they no doubt interact in complex ways. We do not attempt to resolve this heated debate. Instead, we seek to illuminate a key issue in the recovered memories controversy - the nature of false memories - by considering evidence and ideas from recent research in cognitive neuroscience. Whether one believes that false memories of abuse are constructed rarely or frequently, it seems clear that understanding more about the nature of false memories is essential to clarifying and perhaps resolving the recovered memories debate. Both cognitive psychology and psychiatry have provided useful insights into various aspects of false memory,1 and we believe that a cognitive neuroscience approach, too, can offer important findings and ideas.
Cognitive neuroscience is an interdisciplinary enterprise whose main purpose is to understand the relation between the brain and various forms of cognition - memory, perception, language, action, problem solving, and so forth.4 Cognitive neuroscience approaches to memory include studies of patients with memory disorders that result from discrete brain lesions, research with neuroimaging techniques such as positron emission tomography and functional magnetic resonance imaging, and computational modeling of memory performance.
In this article, we consider three different yet related phenomena that, we believe, provide important insights concerning possible neural bases of false memories: source amnesia, confabulation, and false recognition. We focus on evidence from patients with lesions to specific brain structures, but we also touch on relevant observations from neuroimaging studies and behavioral studies of people with intact memory functions. In addition to reviewing pertinent literature, we also present a brief case report of false memory in a patient with a focal brain lesion. It should be noted that although we focus on patients with brain lesions, we do not wish to imply that false memories of childhood abuse or other experiences are indicative of, or caused by, brain damage. Rather, we wish to highlight that by carefully examining false memories in patients with brain lesions, it should be possible to better understand the brain mechanisms that underlie true and false memories in neurologically intact individuals.
Source amnesia refers to situations in which people remember a fact or occurrence from the past but fail to remember the source of their knowledge - who told them the fact, whether an event actually occurred or was merely imagined, and so forth. Schacter, Harbluk, and McLachlan5 documented source amnesia in studies of patients with severe deficits in memory for recent experiences that were produced by various kinds of brain lesions. Schacter et al. found that these amnesic patients could sometimes learn new "fictitious facts" that were imparted to them by one of two experimenters (e.g., "Bob Hope's father was a fireman."). However, when amnesic patients recalled one of these newly learned "facts" correctly, they rarely were able to remember the correct source. Indeed, patients frequently failed to recollect that either experimenter had told them a fact, and instead claimed that they were guessing or that they had acquired the fact from an extraexperimental source, such as television or radio.
Source amnesia is relevant to the understanding of false memory because subsequent research has revealed that when people fail to recollect the source of their knowledge, they are subject to various kinds of memory illusion and distortions, such as confusing real and imagined events,6 making incorrect attributions about why something seems familiar,7 or failing to remember whether something actually happened or was only suggested.1,28 Similar observations have been made in developmental studies that have shown that source amnesia plays a major role in some false memories exhibited by young children.9
Neuropsychological research has provided evidence that susceptibility to source amnesia is associated with damage in frontal regions of the brain. Lesions that are restricted to the frontal lobes do not ordinarily produce a severe amnesic syndrome of the kind that is observed after damage to the medial temporal lobes or thencephalic brain structures, but they can impair recollection for temporal order and related aspects of memory.10,11 A linkage between frontal damage and source amnesia was initially reported by Schacter et al.,5 who found that those amnesic patients who performed poorly on tests sensitive to frontal lobe dysfunction exhibited more source amnesia than did patients who performed well on frontal-sensitive tests. Shimamura and Squire12 later reported a similar finding. A subsequent study demonstrated that patients with focal frontal lobe lesions, who are not globally amnesic, nevertheless exhibit extremely high levels of source amnesia.13 Additional evidence on this point has been provided by studies that have documented source memory deficits in elderly adults, particularly those with neuropsychological signs of frontal impairment.14
Taken together, the foregoing studies point to a clear linkage among source amnesia, frontal lobe functions, and false memory. Future research will be needed to clarify precisely which regions of frontal cortex are most strongly implicated in source amnesia and to clarify the role played by source amnesia and frontal functions in the false memories exhibited by normal, non-brain-damaged people in the laboratory and in everyday life.
Confabulation is a symptom that occurs in a variety of neurological and psychiatric syndromes. The major feature of confabulation is an inaccurate and sometimes bizarre narrative account of a present or past event. Although the term "confabulation" has sometimes been applied rather broadly to almost any kind of false memory, we prefer to use the term strictly in relation to the inaccurate stories or narratives that are produced by clinical populations. Moscovitch15 characterizes confabulation as "honest lying," in the sense that patients believe what they are saying even though it is demonstrably false. In cases of brain damage, confabulation is often observed in connection with lesions to ventromedial aspects of the frontal lobes.15,16 While most investigators agree that frontal lobe damage is strongly associated with confabulation, there is also reason to believe that additional damage in the basal forebrain may be necessary for extensive confabulation to occur.17
Confabulation seems to be observed most readily when patients are asked about personal experiences (episodic memory); it is observed relatively rarely when patients are asked about general knowledge (semantic memory).18 Many observers of confabulation in brain-damaged patients have noted a distinction between two different manifestations of the phenomenon: a memory distortion in which actual events from the past are confused in time, and a fantastic or bizarre fabrication with implausible contents based on figments of imagination.16 As Moscovitch15 points out, hpwever, source amnesia is implicated in most instances of confabulation. When confabulations are built on actual events, patients confuse when and where they occurred. In cases of bizarre confabulations, patients may incorporate fragments of dreams, fantasies, or even features of the current environment into their narratives. Yet they fail to recognize the source of these mental contents and attribute them to actual events that never occurred.
Although source amnesia clearly plays an important role in confabulation, it is probably not the entire story, because patients often believe their confabulations even when they are logically inconsistent. Moscovitch15 describes a patient who claimed to have been married only 4 months - yet he acknowledged that he had four children, and he remembered their names and ages relatively accurately (in fact, the patient had been married for over 30 years).
Pertinent evidence is also provided by the confabulatory phenomenon of reduplicative paramnesia, sometimes referred to as Capgras' syndrome, which is sometimes observed in patients with frontal lobe pathology. Such patients believe that a familiar person or place has been "duplicated."19 For example, one patient insisted that his entire family had been replaced with a group of highly similar substitutes. His strong subjective confidence in his beliefs and memories was not undermined by the fact that he was perfectly aware of the logical implausibility of such a state of affairs.19
Thus, it seems clear that high-level cognitive processes that normally monitor and evaluate mental contents are disturbed in confabulating patients, and hypotheses concerning the nature of those deficits have been offered.16,18 For example, Moscovitch15 suggests that confabulation arises when automatic retrieval processes are intact, so that memories frequently spring to mind, but strategic search and monitoring processes are impaired, so that source, plausibility, and other aspects of the memories are not evaluated properly.
In standard laboratory tests of recognition memory, experimental participants study a series of words, objects, or other materials, are later exposed again to the target materials along with distractor or lure items that were not presented previously, and are asked to indicate which items appeared on the study list. During the past few years, several patients have been described who exhibit a pathological tendency toward false recognition - that is, they frequently claim that distractor or lure items, which they had not encountered previously, appeared on the study list. In view of the material covered in previous sections, it is perhaps not surprising that all of these patients are characterized by frontal lobe pathology.
One such patient, R.W., was described by Delbecq-Derouesné, Beauvois, and Shallice.20 R.W. underwent an operation to repair a ruptured anterior communicating artery aneurysm, a condition that is known to produce memory disorders. A CT scan revealed bilateral areas of hypodensity in the medial aspects of the frontal lobes, as well as in regions of the right temporal pole and aspects of the fusiform and parahippocampal gyri. R.W. consistently made an abnormally large number of false recognitions and expressed a great deal of confidence in them.
Another patient who developed memory problems as a consequence of a ruptured anterior communicating artery aneurysm and associated frontal lobe damage, J.B., also made a large number of false recognitions that were accompanied by high confidence - J.B. often said that he was "sure" that he had been exposed to target materials that had never been shown to him previously (A.J. Parkin, C. Bindschaedler, L. Harsent, C. Metzler, unpublished data, 1994). In one experiment, J.B. was shown a list of familiar words and was then given a recognition test. When he thought that a word had appeared on the study list, he was asked to indicate (a) whether he possessed a specific recollection of encountering the word previously, such as an image or association that he made when he studied the word (a "remember" response); or (b) whether he just "knew" that the item appeared on the list, even though he did not have a specific recollection of having encountered it (a "know" response).21 All of J.B.'s false alarms to nonstudied words were accompanied by "know" responses - that is, J.B. felt that these items were familiar, and thus was certain that they had appeared in the study list, but he did not have a specific recollection of having encountered them on the list.
A CASE STUDY OF FALSE MEMORY
We have recently studied a patient, B. G., who exhibits a striking pattern of false recognitions. B.G. is a 65-year-old man who suffered an infarction of the right frontal lobe in December 1993. He performs poorly on neuropsychological tests that are sensitive to frontal lobe pathology, such as the Wisconsin Card Sorting task, but he shows no signs of amnesia, does not offer spontaneous confabulations, is alert, attentive, and cooperative, and has no difficulty understanding or following task instructions.
To investigate recognition memory in B. G., we first exposed him to a fist of familiar words, and then gave him a recognition test for old and new words in which we asked him to assign a "remember" or a "know" response to words that he believed had appeared on the study list. Results indicated that even though B.G.'s memory for previously studied words (i.e., hit rate) was relatively normal, he made many more false alarms than did any of the control subjects. Moreover, B.G. claimed to "remember" nearly 40% of new words that had not been on the study list, whereas control subjects made "remember" responses to only about 5% of new words. Both B.G. and control subjects provided "know" responses to about 10% of new words:
A careful inspection of B.G.'s performance revealed that many of the new words that he claimed to "remember" were associatively related to words that had appeared in the study list. For example, B.G. said that he remembered encountering "cellar" on the study list. Although "cellar" had not appeared on the study list, "basement" had. Cognitive research has shown that normal individuals are more prone to false recognition when new words are associatively related to previously studied words than when they have no associative relation to studied items.22 The general idea is that when a person studies a word on a list, various associations are activated; if a lure item on a recognition test corresponds to one of those associations, false recognition may be elicited. Perhaps B.G. activated too many associations to target words at the time of study, or was unduly influenced by semantic or associative similarity when deciding whether he remembered a particular item.
To investigate the matter further, we performed an experiment in which some of the lure items on the recognition test were associatively related to a word that had appeared on the study list, and others were unrelated to the study list words. The key result was that B.G. provided many more "remember" responses to both related and unrelated lure items than did control subjects, although he did so more frequently for related than unrelated lures. Thus, B.G. exhibited false recollections even when a nonstudied lure item had no relation to a previously studied item.
Another possibility is that B.G.'s tendency to "remember" nonstudied words is attributable to the fact that all of the words in our experiments were familiar to him on the basis of pre-experimental knowledge - that is, all of the words were represented in his long-term memory prior to the experiment. Perhaps B.G. mistakenly took the pre-experimental familiarity of a word as evidence that it appeared on the study list, and hence claimed to "remember" nonstudied words. To test this hypothesis, we exposed B.G. to novel pseudowords (e.g., "brap," "spafe") that would not have been familiar to him on the basis of preexperimental knowledge. Results indicated that B.G. still exhibited the same false recognition phenomenon as in previous experiments, claiming to "remember" nonstudied pseudowords much more often than did control subjects.
Is there any way to stop B.G. from claiming that he "remembers" nonstudied items? We were able to nearly eliminate B.G.'s false memories with a simple manipulation. We showed B.G. pictures of inanimate objects from various categories (e.g., furniture, articles of clothing). On a subsequent recognition test, some of the nonstudied lure items came from these categories, other lure items came from miscellaneous categories of inanimate objects that were not represented on the study list, and still other lure items were animate objects (i.e., animals). B.G. claimed to "remember" many of the lure items that were drawn from previously studied categories of inanimate objects. But he almost never claimed to "remember" the lure items that were not members of previously studied categories (A.J. Parkin, C. Bindschaedler, L. Harsent, C. Metzler, unpublished data, 1994).
B.G.'s pattern of false alarms provides an interesting puzzle. On the one hand, B.G. exhibits considerable false recognition even for lure words that have no associative relationship to words that appeared on the study list. On the other hand, he does not exhibit false recognition to lures from nonstudied categories. To understand this puzzle, we find it useful to think in terms of two different levels of knowledge that can be accessed when people attempt to remember past episodes. Following Conway and Rubin,23 we refer to them as general event knowledge and event-specific knowledge. General events refer to high-level episodes, such as going to the movies, whereas event-specific knowledge refers to particular episodes that are nested within the general event, such as spilling my popcorn or being surprised by the end of the film. With respect to a memory experiment, the general event might be represented as seeing a list of words or seeing some pictures of clothing and furniture, whereas event-specific knowledge would refer to memory for the specific items that were presented during the experiment.
We suggest that when a person is trying to remember whether a specific item was shown earlier in an experiment (event-specific knowledge), test items can activate a general event description. The activation of a general event description should in turn signal a search for event-specific knowledge. Under the conditions of our experiments, control subjects typically claimed to "remember" that a word or picture had appeared on a study list only when appropriate event-specific information had been activated (i.e., a memory of the particular word or picture). B. G., by contrast, sometimes accepted activation of a general event description as evidence for event-specific memory, and so offered "remember" responses to some nonstudied items that matched the general event description. Thus, after studying a list of unrelated words, B.G.'s general event description may have been something like "saw a bunch of words."
By our view, B.G. made false recognition responses to lures that were unrelated to previously studied words because these lure words matched the general event description. However, in the categorized pictures experiment, we suggest that B.G.'s general event description may have included category information, such as "saw some pictures of furniture and clothes." Accordingly, when lures were used that did not match a general event description (e.g., when B.G. was shown lure pictures of animals), he no longer made many false recognition responses.
The foregoing observations raise the possibility that frontal cortex, and perhaps right frontal regions in particular, play some role in search processes that access event-specific knowledge from general event knowledge. Note also that recent positron emission tomography studies have revealed that right frontal regions are consistently more activated during episodic retrieval of recently studied items than during encoding or study of those items; the opposite appears to be true of left frontal regions.24 It is conceivable that the activation of right frontal regions during episodic retrieval in normal subjects reflects the search processes that we hypothesize are defective in patient B.G. (For a fuller report, see reference 25.)
Although our ideas are preliminary, they do suggest that future investigations of the relation between general event knowledge and eventspecific knowledge, and the role that frontal regions play in linking them, may provide new insights into the nature of false memories in non-brain-damaged people. It is possible, for example, that false memories sometimes arise when people mistakenly interpret general event knowledge as evidence of a specific event that did not actually happen. Research that examines this hypothesis in normal populations would be desirable.
Scientific understanding of false memories will require research from a variety of perspectives that examine biological, psychological, psychiatric, social, and cultural aspects of remembering. Cognitive neuroscience approaches have begun to elucidate key psychological and neurological aspects of source amnesia, confabulation, and false recognition. Hopefully, this research can help to establish a foundation for basic understanding that may someday shed light on the real-world cases of recovered memories that are currently at the center of so much controversy. Understanding the relation between false and true recovered memories is not only an important scientific challenge, but is also essential in order to protect innocent people from unjust accusations of abuse that are based on false recovered memories, and to ensure the credibility of abuse victims who recover accurate memories.
We are optimistic that the cognitive neuroscience approach can play a useful role in this emerging enterprise by helping to tease apart components of the remembering process that contribute to false memories, and by linking these component processes to the networks and systems of the brain.
1. Schacter DL. Memory distortion: history and current status. In: Schacter DL, Coyle JT, Fischbach GD, Mesulam MM, Sullivan LE, eds. Memory Distortion Cambridge, MA: Harvard University Press. 1995.
2. Lindsay DS, Read JD. Psychotherapy and memories of childhood sexual abuse: a cognitive perspective. Applied Cognitive Psychology. 1994; 8:281-338.
3. Loftus EF. The reality of repressed memories. Am Psychol. 1993; 48:518-537.
4. Kosslyn SM, Koenig O. Wet Mind: The New Cognitive Neuroscience. New York, NY: The Free Press; 1992.
5. Schacter DL, Harbluk JL, McLachlan DR. Retrieval without recollection: an experimental analysis of source amnesia. Journal of Verbal Learning and Verbal Behavior. 1984;23:593-611.
6. Johnson MK, Hashtroudi S, Lindsay DS. Source monitoring. Psychol Bull. 1993; 114:3-28.
7. Jacoby LL, Whitehouse K An illusion of memory: false recognition influenced by unconscious perception. J Exp Psychol Gen. 1989; 118:126-135.
8. Zaragoza MS, Lane SM. Source misattributions and the suggestibility of eyewitness memory. J Exp Psychol Learn Mem Cogn. 1994; 20:934-945.
9. Ceci SJ, Loftus EF. "Memory work": a royal road to false memories? Applied Cognitive Psychology. 1994; 8:351-364.
10. Milner B, Petrides M, Smith ML. Frontal lobes and the temporal organization of memory. Human Neurobiology. 1985;4:137-142.
11. Schacter DL. Implicit memory: history and current status. J Exp Psychol Learn Mem Cogn. 1987; 13:501-518.
12. Shimamura AP, Squire LR. A neuropsychological study of fact memory and source amnesia. J Exp Psychol Learn Mem Cogn. 1987; 13:464-473.
13. Janowsky JS, Shimamura AP, Squire LR. Memory and metamemory: comparisons between patients with frontal lobe lesions and amnesic patients. Psychobiology. 1989; 17:3-11.
14. Craik FIM, Morris LW, Morris RG, Loewen ER. Relations between source amnesia and frontal lobe functioning in older adults. Psychology and Aging. 1990; 5: 148-151.
15. Moscovitch M. Confabulation. In: Schacter DL, Coyle JT, Fischbach GD, Mesulam MM, Sullivan LE, eds. Memory Distortion. Cambridge, MA: Harvard University Press. In press.
16. Johnson MK Reality monitoring: evidence from confabulation in organic brain disease patients. In: Prigatano GP, Schacter DL, eds. Awareness of Deficit After Brain Injury: Clinical and Theoretical Issues. New York, NY: Oxford University Press; 1991.
17. DeLuca J. Predicting neurobehavioral patterns following anterior communicating artery aneurysm. Cortex. 1993; 29:639-647.
18. Dalla Barba G. Confabulation: knowledge and recollection experience. Cognitive Neuropsychology. 1993; 10(1): 120.
19. Stuss DT. Disturbance of self-awareness after frontal system damage. In: Prigatano GP, Schacter DL, eds. Awareness of Deficit After Brain Injury: Clinical and Theoretical Issues. New York, NY: Oxford University Press; 1991.
20. Delbecq-Derouesné J, Beauvois MF, Shallice T. Preserved recall versus impaired recognition. Brain. 1990; 113:1045-1074.
21. Gardiner JM, Java RI. Recognizing and remembering. In: Collins AF, Gathercole SE, Conway MA, Morris PE, eds. Theories of Memory. Hove, United Kingdom: Lawrence Erlbaum Associates; 1993.
22. Roediger HL III, McDermott KB. Creating false memories: remembering words not presented in lists. J Exp Psychol Learn, Mem Cogn. 1995; 21:803-814.
23 . Conway MA, Rubin DC. The structure of autobiographical memory. In: Collins AF, Gathercole SE, Conway MA, Morris PE, eds. Theories of Memory. Hove, United Kingdom: Lawrence Erlbaum Associates; 1993.
24. Tulving E, Kapur S, Craik FIM, Moscovitch M, Houle S. Hemispheric encoding/retrieval asymmetry in episodic memory: positron emission tomography findings. Proc Natl Acad Sci. 1994;91:2016-2020.
25. Schacter DL, Curran T, Galluccio L, Milberg W, Bates J. False recognition and the right frontal lobe: a case study. Neuropsychologia. In press.