The question of memory, of what we can, cannot or do not want to remember is of central concern to psychoanalytic practitioners and researchers. In his early formulations on the nature of hysteria, Freud understood the hysteric's problem as one of ''suffering from reminiscences'' (Breuer & Freud, 1895: 7). Freud and Breuer (1895) suggested that the source of the hysterical patient's psychic pain was the inability to forget traumatic events that had occurred in childhood but which could not be consciously remembered. The goal of therapy was therefore to bring back to the surface the repressed traumatic events. Although Freud changed his ideas about hysteria later, this early link between disturbances of memory and psychopathology can still be traced in the implicit thinking of some psychoanalytic practitioners who view the excavation of the past as a necessary goal of psychotherapy. As our knowledge of memory has become increasingly more sophisticated, the classical psychoanalytic view of memory and hence of the nature of therapeutic action has been challenged.
A feature of memory that is of special relevance to clinical practice is that memory is by definition always reconstructed and, importantly, influenced by motivation. Memory is influenced as much by present context, mood, beliefs and attitudes, as it is by past events (Brenneis, 1999). Memories are not direct replicas of the facts per se. On the contrary, memory undergoes a complex process of reconstruction during retrieval. This means that memory of some autobiographical events may be reconstructed in ways that differ from the original event or may never be recalled at all. The view that memory is continually being constructed rather than retrieved from storage in original pristine form is consistent with current thinking in cognitive psychology and neurobiology. However, it would be mistaken to infer from this that early memories are mostly inaccurate: research suggests that there is in fact substantial accuracy in early memories (Brewin et al., 1993) even though the more fine-grained details of an experience, even if vividly recalled and reported by the patient, are unlikely to be entirely accurate.
We are now all too aware of the heated debates about so-called false memories. The interest, and indeed controversy, about the reliability and accessibility of early memories gained momentum over ten years ago when the media drew attention to a groundswell of reconstruction of incestuous sexual abuse within the context of psychotherapy. Dreams, puzzling body sensations, specific transference and countertransference patterns and dissociative episodes were taken by many therapists as evidence that their patients had repressed a traumatic experience. This conclusion was predicated on the assumption that analytic data can reconstruct and validate consciously inaccessible historical events. In other words, it reflected a belief that analytic data was ''good enough''. Any of the symptoms listed above, which have been taken as evidence of repressed trauma, may occur in conjunction with trauma, and often do, but they do not occur exclusively with trauma. The danger lies in inferring the nature of unremembered events solely from the contents of any of these repetitive phenomena.
Suggesting that memory is reconstructed does not mean that psychoanalytic reconstructions are necessarily false or that recovered memories are invariably, or mostly, false. It does mean, however, that we must approach notions of ''truth'' based on reconstructions within the context of psychotherapy with some caution. All that we can assert with any certainty is that what our patients believe to be true has important consequences for how they feel and act in the world. Our role as therapists is neither that of an advocate or a jury: we are facilitators of the patient's attempts to understand his internal world and how this impacts on his external relationships and day-to-day functioning. I am not advocating disbelieving what patients say. Patients who have experienced a trauma need to have their traumatic experiences validated. However, all we can validate is their emotional experience of an event and their individual narrative about it. Importantly, we often have to bear the anxiety of not knowing what may have happened so that we can help our patients to bear it too. When our patients have no conscious recollection of any trauma but we, as therapists, infer trauma from their symptomatic presentation, we need to caution against an overeagerness to fill in the unbearable gaps in understanding with the knowing certainty of formulations that may, or may not, be correct. There exists in us and in our patients, as Brenneis suggests, "... a balder desire to locate an original event that unlocks the mysteries of present experience'' (1999:188). This desire can mislead us at times because, as Kris wisely reminds us:
... we are [not], except in rare instances, able to find the events of the afternoon on the staircase where the seduction happened. (1956: 73).
Research on human memory helps us to understand the need for caution in these matters. It suggests that there are different kinds of memory systems and hence different types of memories. Certain sets of memories are consistently reactivated moment by moment. These memories concern the facts of our physical, mental and demographic identity. They orient us in the world. Conventionally, this is variously referred to as declarative or explicit2 or autobiographical memory. Declarative memory - the term I will use from now on - is the underlying organisation that allows us to consciously recall facts and events. It refers to the conscious memory for people, objects and places. It involves symbolic or imaginistic knowledge that allows facts and experiences to be called into conscious awareness in the absence of the things they stand for. This kind of memory includes semantic memory for general and personal facts and knowledge and episodic memory for specific events.
There are also contents of memory that remain submerged for long periods of time, some never to be retrieved. Many aspects of our behaviour rely on us remembering "how to do things'', and we can do this without consciously remembering the details of how to carry out a particular behaviour. This kind of memory is conventionally variously referred to as procedural or implicit or non-declarative memory. It includes primed memory (e.g. for words, sounds or shapes), which facilitates the subsequent identification or recognition of them from reduced cues or fragments, emotional memory and procedural memory, that is, memory for skills, habits and routines.
2"Explicit" and "implicit" refer, respectively, to whether conscious recollection is involved or not in the expression of memory. Long-term memory may be both explicit and implicit. Both involve the permanent storage of information: one type is retrievable (i.e. explicit memories), the other most probably is not (i.e. implicit memories).
Emotional memory is the conditioned learning of emotional responses to a situation and is mediated by the amygdala. There is a difference between emotional memory, that is, a conditioned emotional reaction formed in response to a particular event, and declarative memory of an emotional situation, that is, the recall of events felt to be of emotional significance. Classically conditioned emotional responses (e.g. classically conditioned expectations, preferences, desires) constitute the affective colouring of our lives. They orient us unconsciously to aspects of our environment and to particular types of relationships. Often, there is no conscious memory connected with this learning. LeDoux (1994) suggests that a focal point for cognition - the hippocampus - can be involved in the activation of emotions before cognitive processes take place. His research indicates that emotions can bypass the cortex via alternative pathways leading from the thalamus to the amygdala. This makes it possible for emotionally charged schemas to be repeated without the mediation of consciousness.
Like emotional memory, procedural memory is unconscious and is evident in performance rather than in conscious recall. This type of memory refers to the acquisition of skills, maps and rule-governed adaptive responses that are manifest in behaviour but remain otherwise unconscious. It includes routinised patterns or ways of being with others. For example, we may have a coordinated procedural system for ''how to ask for help''. In turn, these procedures shape, organise and influence a person's unconscious selection of particular interpersonal environments. Moreover, emotionally charged events are particularly prone to repetition when events of a similar nature are anticipated.
Neuropsychology has demonstrated complete independence of the declarative and procedural memory systems. Declarative memory is located in the hippocampus and the temporal lobes. Procedural memory is located in sub-cortical structures such as the basal ganglia and the cerebellum. The declarative and procedural memory systems are relatively independent of each other. Studies of amnesic patients provide evidence for the potential dissociability of the two forms of knowledge contained within these memory systems: amnesic patients, for example, demonstrate evidence of prior learning of words, as shown in a word-recognition task, but display no conscious recollection of whether they had ever seen the word before. This suggests that procedural knowledge was acquired in the absence of any conscious recall of the learning experience. This finding suggests that a change in procedural forms of learning may thus come about through different mechanisms than a change in conscious, declarative forms of knowledge. As we shall see later in this chapter, this has important implications for psychotherapy.
In normal adult development, both declarative and procedural memory systems overlap and are used together. Constant repetition, for example, can transform a declarative memory into a procedural one. Likewise, repeated avoidance of particular thoughts or feelings may result in the associated behaviour becoming automated, thus resulting in a so-called "repression". Procedural memory influences experience and behaviour without representing the past in symbolic form; it is rarely translated into language. Whilst we can say that procedural memories operate completely outside of conscious awareness (i.e. they are unconscious), they are not repressed memories or otherwise dynamically unconscious. This means that they cannot be directly translated into conscious memory and then into words: they can only be known indirectly by inference.
In the very early years of childhood, declarative memory is impaired because of the immaturity of the prefrontal cortex and hippocampus, whereas the basal ganglia and amygdala are well developed at birth. During the first two to three years, the child relies primarily on her procedural memory system. Both in humans and in animals, declarative memory develops later. In other words, a child learns how to do things before she is able to recall an actual event in her past. Research suggests that it is highly unlikely that we can remember events predating our third or fourth year of life. This means that there may be procedural memories for infantile experiences in the absence of declarative memories. Indeed, amongst many analytic therapists there is a shared assumption that pre-verbal experiences are expressed indirectly and can only be grasped through the skilled use of the countertransference.
Declarative memories emerge around three years in line with the increasing maturity of the relevant brain systems. This finding suggests that the infantile amnesia Freud spoke of may have less to do with the repression of memory during the resolution of the Oedipus complex, as he suggested; rather, it may reflect the slow development of the declarative memory system. Lack of verbal access to early experiences may therefore have little to do with repression as an unconscious defence process. On the contrary, it probably results from the fact that these early experiences are encoded in a pre-verbal form and are expressed indirectly, for example, through somatic symptoms. In this sense, it is both true to say that we do not forget and that we cannot remember very early events, thereby explaining their continued hold over us in the absence of conscious recollection of the formative experiences in our early childhood.
The very early events that may exert a profound influence on the development of the psyche are most probably encoded in procedural memory. Procedural memory stores a lot of knowledge, but the experiences out of which such knowledge is born are seldom retrievable. In procedural memory, we thus find a biological example of one component of unconscious mental life: the procedural unconscious. This is an unconscious system that is not the result of repression in the dynamic sense (i.e. it is not concerned with drives and conflicts), but it is nevertheless inaccessible to consciousness. By contrast, the world of the psychoanalytic unconscious, in its dynamic sense, has its roots most probably in the neural systems that support declarative memory. Repression can occur here, but it is a process that can only act on events that are experienced at a developmental stage when encoding into declarative memory is possible.
Taken as a whole, our current understanding of perception and memory points to a fundamental fact, namely, as Gedo put it, "What is most meaningful in life is not necessarily encoded in words" (1986: 206). This, as we shall see in the next section, has important implications for how we might understand the process of change in psychoanalytic therapy.
Was this article helpful?