Repressed and Recovered Memories

Posted by permission of Dr. Thomas O’Connor. Copyright by T. O’Connor.

“The ordinary response to atrocities is to banish them from consciousness.” Judith Herman

In psychology, there is no universally agreed upon model of memory, although there is widespread acceptance of the fact that the study of memory is a foundational or central enterprise affecting many subject areas in psychology. The psychology of memory is usually a third year, full semester, undergraduate course, and the subject matter is quite complex. Before taking on a psychology of memory course, it is advisable to have some prerequisite coursework in other areas of psychology (e.g., behavioral, cognitive, and personality). Before taking on the advanced (and controversial) topic of repressed and recovered memory, it is advisable to have some prerequisite knowledge about memory in general, and the following simple primer should suffice (with references and citations kept to a minimum). A forensic psychologist should be able to testify about the nature of memory as we know it today.

A PRIMER ON THE PSYCHOLOGY OF MEMORY

It is customary to note that there are at least three (3) different types of memory: (1) episodic; (2) semantic; and (3) procedural. Episodic memory is about episodes or events of life, such as what a person did last summer. Semantic memory is what a person knows is the correct vocabulary for the world they live in, such as how to answer the question “Are wrenches pets or tools?” Procedural memory is about how things work or how to do things that cannot be expressed in words, such as how to ride a bike. These three types of memory are mutually exclusive, at least theoretically. Sometimes, a distinction is also made between cognitive and emotional memory (Walker & Shapiro 2004), given research in the field of psychoneuroimmunology (PNI) which shows that verbally processed, cognitive memory is stored in the cortex part of the brain, whereas emotional memory is stored in the midbrain area, most likely the hypothalamus, amygdala, adrenalin, and pituitary axis.

There are at least three (3) basicprocesses of memory: (1) encoding; (2) storage; and (3) retrieval. Encoding (sometimes called acquisition) refers to the format of information from the senses, such as whether the information being sensed in acoustic, visual, or semantic. Storage (sometimes called retention) refers to the individual capacity to maintain information in memory over time. Retrieval refers to the individual ability or skill at recognition or recall, where recognition is memory aided by clues (such as with a multiple choice test) and recall is memory that just seems to come easy or automatic for someone (such as with an essay test). Memory via recognition is far more common (and easier) than memory via recall, for most people.

There are, in addition, three (3) basicstages, or systems, of memory: (1) sensory; (2) short-term; and (3) long-term. This notion of three different active systems is known as the Atkinson-Shiffrin Model (Atkinson & Shiffrin 1968) which is also sometimes referred to as the “information processing” theory of memory. Students should realize that 3-4 other, different models (theories) exist in the literature, but that this model is regarded as having the most research support. Sensory memory refers to the physiological “holding” areas, called sensory registers, which exist for each of the five senses in brain regions which control the sense organs by forcing them to slow down and selectively attend to something or speed up and extract some feature out of the ordinary. Only the eyes (iconic memory which lasts one second) and ears (echoic memory which lasts four seconds) have been studied extensively, whereas research into olfactory, tactile, and taste memory is less common.

Short-term memory (STM), which is sometimes called “memory span” or working memory, refers to unrehearsed, nonrepetitive mental representations of information that are stored in the brain for limited amounts of time, usually from eighteen to sixty seconds. STM research has primarily looked into two things: how to improve the span’s “magic number” of seven plus or minus two for “chunking” (the number of meaningful groupings; e.g. TWA NBA CBS CPR CIA, that people can recall perfectly after one presentation); and how to improve the many mistakes and errors people make with STM, especially when encoding acoustic and visual information that is not easy to assimilate because the items are not similar (STM seems driven by an assimilation versus contrast effect). Most of the so-called “mnemonic devices” which are proposed as memory aids (e.g. HOMES to remember the Great Lakes of Huron, Ontario, Michigan, Erie, Superior) are really attempts to expand the use of chunking and hopefully improve STM. The phrase “working memory” technically refers to a theoretical process where something is retrieved from LTM for temporary use, although psychologists disagree over the extent of this two-way mechanism (called the central executive function). Far more promising is research from the working memory approach which suggests that visual-spatial intelligence (the ability to “see” all things, even numbers and formulas, as images) best characterizes the mental representations which comprise effective STM.

Long-term memory (LTM) is a deep level of conscious or unconscious processing where mental schemas (inference sets about what’s going to happen next based upon categorization of mental representations) can be stored indefinitely (five minutes ago or five decades ago; it doesn’t matter). In theory at least, LTM is unlimited, although disagreements can be easily found among psychologists over LTM deterioration (i.e. forgetting) as well as whether “remembering” is something that requires deliberate effort (explicit memory) or something that just comes automatically or unconsciously (implicit memory). Most research points to the importance of semantic encoding, verbal intelligence, and the extent of vocabulary for improving LTM. However, other research (in the field of amnesia, mainly) points to the importance of hidden, implicit memories of which we are not consciously aware (e.g. prejudices and biases). Most of the so-called “subliminal messaging” techniques are really attempts to draw out hidden or implicit memories, as is hypnosis and a variety of other (largely unsuccessful) motivational and sleep-learning tactics. An interesting phenomenon is “photographic” memory (technically called eidetic memory) which occurs in about 5% of school-age children, but almost no adults (Haber 1979). It is important to note that something doesn’t get into LTM unless it is rehearsed in some way, and psychologists usually distinguish between “shallow” and “deep” rehearsal, the latter involving an evaluation of the emotion and reasoning behind the thing to be remembered. It is also significant to note that many psychologists regard LTM as a key feature of self-identity in personality theories.

To complete this primer, a review of research in the field known as memory retrieval is needed. Essentially, this involves the study of remembering, and it is an area which has made many inroads into criminal justice, primarily with the topic ofeyewitness identification and testimony (Loftus 1979). Basically, people remember in one of four ways which make up some standard definitions in the literature: recollection, where pieces of memory are fitted together; recall, where a memory is regenerated without any cue or stimulus; recognition, where a cue or stimulus triggers a memory; and relearning, where a person remembers something forgotten by deliberately attempting to learn about it a second time. It is fair to say that the research on memory retrieval is mixed, with much research quite limited and most theories quite controversial. Memory retrieval is not perfect, for several reasons. For one, it is entirely possible that even though somebody saw or heard something (and probably should remember), they will NOT be able to remember either because of selective attention, the assimilation effect, or bias in their cognitive schema or template (e.g. some people just can’t remember names). Secondly, the so-called “trace” of a weak or partial memory depends heavily upon length of exposure, the age-gender-race characteristics involved, and how significant or prominent the event is to a person. Thirdly, the passage of time and accumulation of additional experiences will significantly alter and change a memory (Ornstein et al. 1998). Finally, memories are easily altered by any suggestive stimulus cues an interviewer provides. AsJeanne Boylan, famous Unabomber sketch artist says: “Ask, Did he have a moustache? and well, he does now, because you’ve implanted that image.”

The most basic theoretical idea about memory retrieval is known as the Tulving Rule (Tulving 1979), which is also called the “encoding specificity principle.” It holds that cues which stimulate or trigger the initial meaning of a long-term memory tend to work best at producing accurate remembering. To accomplish this, the interviewer needs to figure out the cognitive schema by which a person stores things deeply in their conscious mind based on their inference sets or emotional reasoning. For example, the interviewer needs to find out if the memory of a rare bird, for instance, is stored under the category of “rare things,” “animals that are birds,” “living things,” or “things that fly.” Nobody has ever devised a practical way to do this, although a field of study known as “semantic networks” has made progress with such techniques as word trees, concept maps, and computer simulations. Semantic network theories are an emerging area of cognitive psychology (e.g. Medlin & Ross 1992) that make use of other principles, like the idea of spreading activation (thinking about one concept activates a “spread” down all paths related to that concept).

Another group of ideas make up the reconstructive theories of memory, which are also based in cognitive psychology (i.e. cognitive memory is reconstructed), but also based on social psychological research about how college students can be easily tricked into having false memories (e.g. Druckman & Bjork 1994). Reconstructive theories are of a type known as “context-dependent” theories because the remembering is believed to be dependent upon the setting in which the attempted retrieval takes place. For example, if you asked college students to estimate how prevalent marijuana smoking was on their campus, you would find that their answers depend heavily upon who asks the question, why, when, and where, as well as the perceived answers of their peers. If the person asking was an authority figure perceived to be in the process of thinking about a room-to-room search, the obvious student answer would be less than the actual amount of smoking. If the context in which the question was asked offered the possibility of status or gain (especially in front of peer groups), then the answer would be much more than actual. [The reader should clearly see the implications of this for criminal justice courtroom research.] Memory psychologists go a bit farther with this idea, and generally argue that actual memories are reconstructed and/or fabricated (false memories) under certain contextual conditions. There are many names for the phenomenon. Criminologists (in labeling theory) call it retrospective interpretation; sociologists call it the self-fulfilling prophecy; and psychologists call it spontaneous generalization or PDP – parallel distributed processing (when you “know” what a friend might remember). There are some technical differences in definition between these concepts, but a simplification has been made for purposes of understanding.

A final group of ideas about retrieval are called state-dependent theories of memory (e.g. Eich 1989), and it is important NOT to confuse state-dependent memories with repressed memories (explained below). However, a sizeable number of forensic psychologists frequently cite Van der Kolk & Fisler (1995) for evidence that state-dependent conditions can be associated with repressed memories. As Bartol & Bartol (2004) explain, the basic approach assumes that certain states, such as the experience of traumatic events, may be encoded differently (and more strongly) than ordinary events. Sometimes, the phrase “flashbulb memory” is used where a person experiences a high-trauma event and can remember the slightest of details with remarkable accuracy. However, the strongest research within this approach supports the “mood congruency effect” which holds that recall is more accurate when the memory has been colored by some mood AND conditions at the time of recall involve the same mood (or state of consciousness, as it’s sometimes called). For example, if somebody has learned something under the influence of marijuana or alcohol, they are more likely to have better recall, within limits, for that material when later under the influence of marijuana or alcohol. [Try hiding something when intoxicated, and you’ll likely only find it when you get intoxicated again.] Studies have shown the two most effective moods, however, are anger and sadness (Bernstein & Nash 1999). Some people who experience mood swings, such as those associated with bipolar disorder, tend to demonstrate strong state-dependent memory effects. There are many mysterious things found in studies of state-dependent memory, but for now, this completes our primer on all you need to know about memory in general.

REPRESSION AND REPRESSED MEMORIES

The notion of repressed memories has been around at least since the time of Sigmund Freud (1915-1957), and most readers will instantly recognize the word “repression” as one ofthe nine defense mechanisms in psychoanalysis. The Freudian theory of repression is quite complex, and connected to relatively unsupportable ideas about sexual seduction, libido, verbal slips, and dream symbols. Most forensic psychologists, therefore, subscribe to a non-Freudian definition of “repression” as contrasted below with the Freudian definition:

Freudian Definition of Repression

“The essence of repression lies in turning something away, and keeping it at a distance from the conscious” (an intentional rejection of distressing thoughts and memories from conscious awareness). It also refers to a mode or moment as a defense mechanism (“the foundation stone on which the whole structure of psychoanalysis rests”). As such, it is the mechanism by which the ego keeps itself relatively free of tension and conflict, primarily by separating affect and idea. Since affect cannot become unconscious, only the ideational element can be repressed, such as the instinctual impulses and images. Unlike sublimation, repression runs counter to instinctual drives. Repressed impulses tend not to disappear for long. Rather, they find substitute gratifications and expressions, in compromise formations such as dreams and symptoms of reaction formation. (Jones 1993)

Non-Freudian Definition of Repression

“Repression refers to the psychological process of keeping something out of awareness because of the unpleasant affect connected with it. The “something” may be a memory (or part of a memory), a fantasy, a thought, an idea, a feeling, a wish, an impulse, a connection, and so forth.” (Karon & Widener 1999: 625)

It is important to note that both Freudian and non-Freudian theories of repression allow room for things other than the pictorial memory of an event; e.g., a fantasy, a wish, an impulse, an instinctual impulse, or whatever you want to call it. This means that the mere discovery of a repressed memory does NOT tell whether it really happened or not. Some clinical judgment is called for. A psychologist will generally exercise clinical judgment about theveridicality (correspondence with truthfulness) of a repressed memory based uponan assessment of symptoms, such as how much havoc (or psychological problems) the memory has created across the person’s life span. For ethical reasons, it is impossible to research whether people’s memories of a traumatic event are accurate or not, since such experimental research would require deliberately exposing people to trauma. It is suspected that a “true” repressed memory (as opposed to a pseudomemory) will be accompanied by symptoms of dissociation and other problems with memory (e.g., amnesia, partial amnesia, hypermnesia, delayed recall). However, the APA position on the matter is that no single set of symptoms automatically indicates that a person was a victim of childhood abuse. Research has shown that the type of dissociation (if any) with repressed memories of childhood abuse is quite different from the “flashbulb” memory-type dissociation associated with PTSD (Roediger & Bergman 1998). Analogies to PTSD are therefore inappropriate.

The controversial “repressed memory syndrome” involves a “forgetting” of a painful experience which is then “remembered” through psychological treatment, and where the “remembered” experience often involves being the victim of a crime, such as child abuse, rape, witchcraft, or witnessing a murder, and where the “revealed” or recovered memory becomes the basis for criminal or civil charges against some parties, such as parents, relatives, caretakers, or priests. Courts and legal experts frequently refer to “repressed-then-recovered” memories as “delayed discovery” where delayed discovery is a type of tolling doctrine which extends the statute of limitations. Normally, the year-and-a-day rule is followed in criminal law, except for murder, for which the time period does not expire. However, delayed discovery is a part of common law (Judge-made law) increasingly finding its way into state statutes where the statute of limitations begins on the date that the memory was recovered.

During the 1990s, several professional associations released reports from their “working groups” regarding memories of childhood abuse. Based on a comparison of these reports, there is widespread disagreement among psychologists. It is perhaps beyond the scope of this lecture to provide a detailed comparison, but some brief commentary on the positions taken can be excerpted as follows:

American Psychological Association (1993)
There are gaps in our knowledge about how to distinguish accurate from inaccurate recollections of childhood abuse. It is possible that questionable therapeutic practices implant pseudo-memories for events that never occurred. Ethical practitioners should maintain an empathic, non-judgmental, neutral stance towards reported memories of sexual abuse, neither encouraging nor discouraging specialized techniques (e.g., hypnosis or amytal) which might assist in the full emergence of such memories and/or result in a course of litigation.

American Medical Association (1994)
The medical profession generally agrees with the APA position and further questions the scientific validity of various memory enhancement methods (e.g. amytal and age regression) used in cases of possible childhood sexual abuse. The use of hypnosis should be limited to the investigative process, and the AMA encourages further research into ways to assist in improving the collection and/or assessment of valid corroborating information.

Australian Psychological Society (1994)
Psychologists should not avoid asking clients about the possibility of sexual or other abusive occurrences in their past. Confidential and careful investigation ought to take place into the typically destructive effects of childhood abuse on mental health, self esteem, and personal relationships. Special care should take place in dealing with clients, their family members, and the wider community when allegations of past abuse are made. Psychologists should evaluate critically their assumptions or biases about attempts to recover memories of trauma-related events. Equally, psychologists should assist clients to understand any assumptions that they have about repressed or recovered memories.

British Psychological Society (1995)
The clarity and detail of repressed memories depend heavily on a number of factors, including the age at which the event occurred. Nothing can be recalled accurately from before the first birthday and little from before the second. Poor memory from before the 4th birthday is normal. The psychologist should be alert to a range of possibilities, for example that a recovered memory may be literally true, metaphorically true or may derive from fantasy or dream material. Hypnosis makes memory more confident but less reliable, it is always important to take the client seriously, but avoid drawing premature conclusions about the truth of a recovered memory. Guidelines can and should be laid down for good practice.

Canadian Psychiatric Association (1996)
Developmental psychology casts doubt upon the reliability of recovered memories from early childhood. Certain psychologists have used a list of symptoms that are said to indicate the likelihood of individuals having been abused. Common symptoms such as depression, anxiety, anorexia or overeating, poorly explained pains, and other bodily complaints have all been used as proof of alleged sexual abuse, and there is no support for such propositions. There are no well-defined symptoms or groups of symptoms that are specific to any type of abuse. Routine inquiry into past and present experiences of all types of abuse should be a regular part of psychiatric assessment, but great caution should be exercised before acceptance in the absence of solid corroboration.

NEW DIRECTIONS THAT ARE NEEDED

At the risk of sounding presumptuous, it seems time to move beyond the debates and controversies. Far too much time has been wasted over arguing whether we have erred on the side of too much caution or too much intervention, and far too many forensic psychologists are simply being used as experts on suggestibility. A solid, well-established set of “best practices” would be helpful, but only the British seem inclined to move in that direction. Here are some other ideas. It might be helpful to go back to Freud, and assess, once and for all, if the Freudian concept of repression can be validated or not. It might be helpful to expand basic research on memory, especially kinesiological memory and where such memory is stored in the brain. It might be helpful to come up with a series of interview tests (like in the research area of amnesia or the area of priming on implicit memory tests) where repeated trials and/or a team approach (involving multiple psychologists) might be based on the “best” retrieval tools for reviving such memories. Further research must continue on those histories of childhood abuse that are associated with some adult psychiatric disorders, and although we are a long way from it, cross-cultural research is vastly needed.

From a justice standpoint, there is a need for tort reform and changes in the malpractice insurance business. It can safely be said that these are issues blocking any new developments and the involvement of good professionals. Victims of false allegation are certainly entitled to their rights, and the False Memory Syndrome Foundation, among others, have been good advocates for the legal interests of those who have been so accused. There will always be a need to expose charlatans who pass off pseudoscience as therapeutic practice. However, nitpicking at the validity and reliability of studies and/or the status of social science in general is not going to be very productive. Not all repressed memory experts are charlatans, and if the “politics of the memory war” as well as the media’s seemingly prurient interest in sex can be toned down, it is almost inevitable that the strength of the medico-psycho-legal system will eventually sort out a fair and balanced way to deal with the issue.

REFERENCES
Arrigo, B. (2000). Introduction to Forensic Psychology: Issues and Controversies in Crime and Justice. NY: Academic Press.

Atkinson, R. & Shiffrin, R. (1968). “Human Memory: A Proposed System and its Control Processes.” In K. Spence & J. Spence (eds.) The Psychology of Learning and Motivation, vol. 2. New York: Academic Press.

Bartol, C. & Bartol, A. (2004). Introduction to Forensic Psychology. Thousand Oaks, CA: Sage.

Bernstein, D. & Nash, P. (1999). Essentials of Psychology. Boston: Houghton Mifflin.

Bettelheim, B. (1983). Freud and Man’s Soul. NY: Knopf.

Druckman, D. & Bjork, R. (1994). Learning, Remembering, Believing. Washington DC: Learning Academy Press.

Eich, E. (1989). “Theoretical Issues in State Dependent Memory.” In H. Roediger & F. Craik (eds.) Varieties of Memory and Consciousness. Hillsdale, NJ: Lawrence Erlbaum.

Gazzaniga, M. (Ed.) (1988). Perspectives in Memory Research. Cambridge, MA: MIT Press.

Haber, R. (1979). “Twenty Years of Haunting Eidetic Imagery: Where’s the Ghost?”Behavioral and Brain Sciences 2: 583-629.

Jones, B. (1993). “Repression: The Evolution of a Psychoanalytic Concept from the 1890s to the 1990s.” Journal of the American Psychoanalytic Association 41: 63-93.

Karon, B. & Widener, A. (1999). “Repressed Memories: Just the Facts.” Professional Psychology: Research and Practice 30: 625-626.

Loftus, E. (1979). Eyewitness Testimony. Cambridge, MA: Harvard University.

Medlin, D. & Ross, B. (1992). Cognitive Psychology. Ft. Worth: Harcourt Brace Jovanovich.

Ofshe, R. & Watters, E. (1994). Making Monsters: False Memories, Psychotherapy, and Sexual Hysteria. NY: Scribner’s.

Ornstein, P., Ceci, S. & Loftus, E. (1998). “Adult Recollections of Childhood Abuse.” Psychology, Public Policy, and Law 4: 1025-1051.

Ramsland, K. (2002). The Criminal Mind: A Writer’s Guide to Forensic Psychology. Cincinnati: Writer’s Digest Books.

Ricoeur, P. (1977). “The Question of Proof in Freud’s Psychoanalytic Writings.”Journal of the American Psychoanalytic Association 25: 835-871.

Roediger, H. & Bergman, E. (1998). “The Controversy Over Recovered Memories.” Psychology, Public Policy, and Law 4: 1091-1109.

Towl, G. & D. Crighton. (1996). The Handbook of Psychology for Forensic Practitioners. NY: Routledge.

Tulving, E. (1979). “Relation between Encoding Specificity and Levels of Processing.” In L. Cermak & F. Craik (eds.) Levels of Processing in Human Memory. Hillsdale, NJ: Lawrence Erlbaum.

Van der Kolk, B. & Fisler, R. (1995). “Dissociation and the Fragmentary Nature of Traumatic Memories.” Journal of Traumatic Stress 8: 505-525.

Walker, L. & Shapiro, D. (2004). Introduction to Forensic Psychology: Clinical and Social Psychological Perspectives. NY: Kluwer Academic/Plenum.

Whitfield, C., Silberg, J. & Fink, P. (2002). Misinformation Concerning Child Sexual Abuse and Adult Survivors. Binghamton, NY: Haworth.

Wrightsman, L., M. Taflinger & M. Linsenman. (2000). Forensic Psychology. Belmont, CA: Wadsworth.

Online Resources