There is potentially a rich and productive interface between neuroscience/cognitive science on the one hand and psychoanalysis/psychotherapy on the other. The two traditions, however, have evolved largely independently, based on differing sets of observations and objectives, and tend to use different conceptual frameworks and vocabularies. The following draft of a manuscript is provided by the author and Serendip as a contribution to finding a useful common framework and vocabulary for further exploration of the relations between neuroscience/cognitive science and psychoanalysis/psychotherapy. Comments on this paper are welcome in an on-line forum, by email to the author and/or by sending comments to Serendip.


Joseph W. Slap, M.D.
1601 Walnut Street, Suite 1312
Philadelphia, PA 19102
Telephone: 215-563-8616
February 6, 2004


According to the structural model all perceptions enter the psychic apparatus through a single portal, the conscious ego. Later, in the Outline of Psychoanalysis, Freud asserted that the unconscious is completely shut off from the external world. In 1987 this author argued on clinical and theoretical grounds that there were separate perceptual portals for the ego and the dynamic unconscious. Since 1987 neuroscientific knowledge has become available which strongly supports the contention that the ego and the dynamic unconscious have separate perceptual pathways.


In the various models of the mind he formulated Freud consistently held to the idea that perceptions entered the psychic apparatus through a single portal, a notion which possibly had its origins in Freud's early concept of neurosis arising from an interference in a reflex arc between stimulation and discharge (Breuer and Freud 1895). Thus according to the topographic model (Freud, 1915) perceptions enter the psychic apparatus through the unconscious. But when one turns to the Ego and the Id (Freud, 1923) one reads that all perceptions are conscious from the start. Freud's diagram of the psychic apparatus (p. 24) makes clear that perception is a function of the ego and is linked to consciousness. Later he claimed that the repressed unconscious is part of the id and is "cut off from the external world" (Freud, 1940, p. 198).

It is an everyday observation in analytic practice that individuals react to seductive, hostile, and other behaviors on the part of others without recognizing these behaviors consciously. The events of the outside world stimulate unconscious drive derivatives, guilt, and other elements of intrapsychic conflict. Thus it seems untenable to claim that all perception is conscious from the start and that the unconscious is cut off from the external world.

It has been argued (Slap, 1987) that it would make better clinical and theoretical sense to attribute a perceptual capacity to both the ego and the repressed (except when referring specifically to the structural model the term ego signifies the ego of prestructural theory. It refers to what Freud called the dominant mass of ideas (Breuer and Freud, 1895)). In the years since 1987 neuroscientific knowledge has become available which strongly supports the contention that the ego and the unconscious have separate perceptual pathways.


LeDoux (1996) has shown that raw perceptual data arriving at the thalamus is transmitted to the prefrontal cortex, the amygdala, and the autonomic nervous system. The hippocampus, which is essential for the laying down of conscious, continuous memory is intimately linked to the prefrontal cortex; the prefrontal cortex has attributes which psychoanalysts associate with the ego. The amygdala is the base of a perceptual/mnemic system which has attributes psychoanalysts associate with the dynamic unconscious. Thus according to this neuroscience there are separate pathways to the prefrontal cortex/ego and to the amygdala/dynamic unconscious.

The Prefrontal Cortex and the Ego

Damasio (1984) has found that damage to the prefrontal cortices, specifically the ventromedial prefrontal cortex, causes impairment in decision-making, goal-oriented thinking and social behavior. Damage to the prefrontal cortex changes personality and makes us unlike who we are. As the most specific, neocortical site for information processed in limbic circuits, the prefrontal cortex is the crucial locus of interplay between intellect and feelings.

While as Watt (1990) explains, many ego functions are supported in unimodal cortices and virtually all brain structures, only bilateral damage to the prefrontal areas leads to "severe personality regression and fundamental ego loss" (p. 493). Watt states that there is "evidence that much of what psychoanalysis has termed the executive functions of the ego are supported in the prefrontal cortices" (p. 513). Watt further explores how the functions associated with object-relations are the operations of right hemispheric pathways involving the prefrontal cortex. Given the above, there is clear correspondence between the functions of the prefrontal cortex and the ego of prestructural theory.

The Amygdala and the Dynamic Unconscious

Memory is not a unitary process. There are different kinds of memory subserved by different brain structures. The two memory systems essential to our thesis are declarative memory and emotional memory. Neuroscience has demonstrated that these memory systems operate independently. Declarative or explicit memory refers to conscious recollection of past events and experiences; as stated above this faculty is a function of the hippocampus, which is part of the mesial temporal cortex.

Emotional memory is a function of the amygdala. The amygdala makes affective evaluations of perceptual data prior to the processing of that data by the prefrontal cortex. As noted above raw perceptual data arrives at the thalamus and is transmitted to the amygdala, the prefrontal cortex and the autonomic nervous system. Transmitted data arrives at the amygdala earlier than at the prefrontal cortex because the data going to the neocortex must be processed by the sensory and association portions of the neocortex before being available to the prefrontal cortex. The record of such emotionally charged events is unconscious but the affects generated are consciously experienced and provide the emotional tone for conscious experience.

When the amygdala is confronted with an overwhelmingly frightening perception, the amygdala arouses the autonomic nervous system by pathways to the hypothalamus, which then causes release of ACTH from the pituitary. The effect of stress hormones circulating back to the brain is to impair hippocampal and prefrontal cortical activity and to facilitate the functioning of the amygdala. The organism is then more prepared to react to danger rather than to take time thinking about it. In so doing it lays down memories of traumatic events and at the same time tends to dampen the ability of the hippocampus and prefrontal cortex to modify these memories. The memories laid down in the amygdala tend to be difficult to extinguish; the term indelible is frequently used.

Le Doux (1992) notes that the instantaneous activation of the thalamo- amygdala systems is especially useful in life threatening situations such as being approached by a predator. The individual then responds prior to receiving input from the thalamo-cortico-amygdala projection route. In other circumstances the earlier thalamo-amygdala route serves to prepare the amygdala for inputs from the cortex and the amygdala is able to evaluate the significance of complex information such as fully perceived objects and events. Thus amygdala encoded impressions may vary from the crude and fragmentary to the complicated. Le Doux speculates that individuals vary in their constitution "... in the extent to which the thalamic-cortical pathways predominate in the initiation of emotional processing through the amygdala" (p. 277). If so, the variations may help explain the extent to which neurotic conditions may be resolved in different patients; i.e. 'the stickiness of the id' may be harder to change in the case of thalamo-amygdala predominance. Such variations if gene linked, would help explain the hereditary predisposition to post traumatic stress disorder demonstrated by True et al, (1993).

The amygdala matures earlier than the hippocampus. Le Doux (1996) feels that infantile amnesia is better explained by the immaturity of the hippocampus than by the concept of repression. Children of two or three do have memory but are not able to lay down long term memories; continuous memory occurs later. Infantile amnesia may also involve the inhibition of the still nascent, and therefore weak, hippocampus by the amygdala stimulating the pituitary as outlined above.

Accordingly, a child of two or three exposed to the parents' lovemaking may interpret the event as a brutal assault or murder by the father. This impression is laid down as an emotional memory in the amygdala. The still nascent hippocampus is shut down by the stress reaction. The child awakes in the morning without recollection of the event and finds his mother healthy and going about her routine in a normal manner. This observation does nothing to modify the impression laid down a few hours earlier. Years later the child, now an adolescent or young adult, is at risk for sexual dysfunction by the activation of this memory. As Le Doux observes: "Later exposure to stimuli that even remotely resemble those occurring during the trauma would then pass, like greased lightning, over the potentiated pathways to the amygdala, releasing the fear reaction" (Le Doux 1996, p. 258).

Le Doux repeatedly refers to the subcortical, amygdala related perceptual pathway as being 'quick and dirty' (1996, pps, 163, 166, 255, 257); by this he means that there is a rapid assessment of the emotionally significant percept but that the assessment is apt to be inaccurate: a stick may be perceived as a snake. The characterization of the subcortical pathway as quick and dirty provides a consilience with the concept of transference. The perception of person in a patient's current life as a significant childhood figure is characteristic of the dynamic unconscious. Transference is a misperception mediated by the subcortical, amygdala related/pathogenic schema perceptual pathway. The conventional psychoanalytic understanding of transference as displacement or projection has little explanatory power.

One may ask: Given that emotional memories laid down in the amygdala are both unconscious and relatively indelible how does psychoanalysis work? Psychoanalysis is an effective treatment for neurosis and it is clear that working with the ego/prefrontal cortex does lead to modifications of the amygdala's power to disrupt the individual. While memories never recorded in the hippocampus may not be recovered, we are able to reconstruct psychic reality through discerning patterns and analyzing derivatives. Le Doux (1996) speculates that psychoanalysis may exert control of the amygdala through "the temporal lobe memory system and other cortical areas involved in conscious awareness" (p. 265). He adds that psychoanalysis may be a lengthy process because the connections from the cortical areas to the amygdala are far weaker than the connections from the amygdala to the cortex.

This neuroscientific understanding finds striking support in a case reported by Solms (2001). He described a patient with a meningioma in the area of the anterior communicating artery who awoke from surgery with a confabulatory amnesia or Korsakoff's psychosis. Solms attempted a psychoanalytical treatment with this patient seeing him six days a week. Solms reported the tenth session in some detail. In prior sessions the patient had represented himself as being strong, at least Dr. Solms' equal. He was an athlete having rowed crew and played soccer; Dr. Solms was a business partner, a client, a colleague.

In the session reported the patient acknowledged that there was something wrong with him but it was something easily fixed with a cartridge, an association to his being an engineer, or with implants, associations to prior successful treatments. He then conceded that there was something wrong with his head but it was only a minor concussion suffered on an athletic field; it would be OK in a few minutes or, in any case, he could go to a top sports medicine specialist; not to worry. Solms persisted in confronting him with his memory loss. Now the patient became agitated and began to talk about explosives. Solms felt threatened. As the patient became aware of his defect he behaved like the weakened and humiliated Samson about to tear the house down.

At this point the patient became acutely concerned that he had lost something important, something that should be in his pants or might rest on the seat of a chair. He searched the pockets of his pants, took them off and searched the trouser legs; he then searched the chair, holding it up and looking under it. At this point Solms broke the session off.

Since the patient had lost his ability to test reality it is clear that his ego had been compromised by the brain damage. His unconscious however continued to function in the usual manner. The brain damage and loss of cognition was interpreted as castration; thus the frenzied search for the missing something in his pants and in the chair. Accordingly it is clear that there are separate perceptual pathways to the ego and the unconscious, the one to the ego having been damaged and the one to the unconscious remaining unscathed.

Since the patients lesion was in the area of the anterior communicating artery like patient's G and H in Kaplan-Solms and Solms (2002), it must have damaged his prefrontal cortex. The thalamus, whose circulation is supplied by the middle and posterior cerebral arteries, and the amygdala, which is embedded in the temporal lobe, would have been spared. Hence the prefrontal cortex/ego was compromised and the amygdala/dynamic unconscious was left intact.


Neuroscience aside, the structural model enjoys a remarkable immunity in that it remains the overarching paradigm for psychoanalysis despite serious theoretical and clinical deficiencies. The id has all but dropped out of clinical discourse; Brenner, (1992) has questioned the validity of the superego concept and it has diminished in its frequency of citation. There is no internally consistent way to define id, ego, and superego. At times these macrostructures appear to be reified, causally effective entities and at other time simply classifications of functions in situations of conflict. There is confusion as to whether unconscious fantasy formations belong to the id or the ego or are a product of id, ego, and superego.

The structural model is not a guide to the conduct of psychoanalytic therapies. Most practitioners appear to employ a bipartite model consisting of the realistic and reasonable part of the patient, analogous to the ego of prestructural theory, and a pathogenic dynamic unconscious. Sandler (1983) has contrasted the structural model to which we pay lip service with the private theories with which we work. A parallel may be found in our institutes where there are two separate curricula. Candidates study the structural model in theoretical courses and learn to ignore it in supervision and clinical seminars.

There is a model, formulated in 1980, which we believe is congruent with the neuroscience described above. It has appeared in several journal articles (Slap & Saykin, 1983, 1984, Slap, 1986, 1987, Slap & Slap-Shelton, 1994) and as a monograph (Slap & Slap-Shelton, 1991). Here is a brief summary of the model:

According to this model, which owes much to the ideas of Klein (1976), the mind is understood as a unitary schema -- an organization of memories, appetites, moral values, factual knowledge, cognitive skills, and affects -- which, in the ideal condition is generally interconnected. In neurosis, situations or events which cannot be mastered become the nidus for an organization of these traumatic experiences and reactive fantasies with their associated affects, which is also a schema but one that is sequestered from the general or unitary schema. The unitary schema is conceptually analogous to the ego of prestructural theory, and the sequestered schema to the repressed or dynamic unconscious. The sequestered schema may be latent; when it is active and disruptive, the term pathogenic schema is appropriate.

In Piagetian terms, the unitary schema, here referred to as the ego, assimilates and accommodates; that is such a schema is able to interpret fresh data (perceptions) in terms of past experience (assimilation) and is able to adapt to or accommodate to fresh data which differs from past experience (accommodation). The sequestered schema assimilates but does not accommodate; that is, it functions throughout life interpreting current events in terms of templates of past experience. Thus the neurotic goes through life remaking the same film or play with people from current life cast into roles originally created by parents and other important figures of childhood. Transference is the consequence of the assimilation of persons from current life into these roles and the phenomena which led Freud to conceptualize the repetition compulsion are accounted for by the ongoing activity of the pathogenic schema. Analogous to the principle of multiple function (Waelder, 1936) is the understanding that everything one experiences is parallel processed; that is, mental experience and behavior are the resultant of forces of the activity of the unitary and pathogenic schemas.

The mode of cognition of the pathogenic schema is consistent with what Piaget called preoperational thought (Piaget, 1926) by which is meant that diverse elements are intimately associated on the basis of concurrence or juxtaposition: events are registered in loosely connected and disorganized ways; primitive modes of reasoning control current reasoning without recognition of their influence. Thus idiosyncratic schemas of analogy and visual schemas prevail over logical reasoning and deduction. As this is the kind of reasoning which characterizes the child during the years in which infantile neuroses are established, one would expect this type of reasoning to characterize the organized residues of those neuroses. This characterization of the cognitive mode of the pathogenic schema appears to be consilient with Le Doux's (1992) description of thalamo-amygdala projections. He described them as being relatively crude and based on fragments rather than "full-blown perceptions of objects and events" (p. 277).

This is a simple model. Essentially all that is claimed is that the painful impressions and situations of childhood, those which are not mastered by the immature mind, become nidi for the formation of separate organizations of the mind, here called pathogenic schemas. The pathogenic schema is a freeze frame of the mind during the painful periods in its development and encompasses traumatic impressions, reactive fantasies, associated affects, and infantile cognition; over time this organization accretes to itself later experiences which have been perceived egocentrically. The model dispenses with highly abstract, metapsychological concepts and is wedded to the dynamic and genetic points of view. It easily reconciles such concepts as trauma, transference and repetition. It is not troubled by the serious conceptual difficulties, which afflict the structural model.

The schema model is compatible with the neuroscience described above. The ego of the schema model is the clinical manifestation of the functioning of the prefrontal cortex; the pathogenic schema is the clinical manifestation of threatening perceptions registered by the amygdala and their associated affects, fantasies and subsequent assimilated experience.

According to the structural model, drives or drive derivatives arise in the id; defenses are assigned to the ego. The conflictual interface then resides between the ego and the id, or in less abstract language, between the drive derivatives and the defenses provided by the ego. Conceptualized in this way, there is but one conflictual interface. According to the schema model, such conflictual interfaces lie within the pathogenic schema; that is, in the unmastered conflictual situations of the patient's childhood there were drive derivatives that were defended against by the immature ego; in addition a second kind of conflictual interface exists: that between the pathogenic schema/amygdala and the ego/prefrontal cortex.

The analyst who is guided by the schema model will understand that old scenarios are being acted out. He will strive to demonstrate to the patient how the past is being repeated (both within the analytic transference and without), sometimes in exquisite detail, with the same plots and affects, albeit the time, setting, and actors have been replaced. In so doing he will be addressing the interface between the realistic ego and the pathogenic schema. If an analyst, guided by the structural model, perceives the conflictual interface as being between the adult ego on the one hand, and the impulses and painful affects of the repressed on the other, the patient is in the same situation as in the traumatic childhood situation.

For example, a young attorney was extremely sensitive to any situation that could be perceived as his being displaced by a younger rival. When he was two years old, the patient's mother required a hospitalization of several months following an auto accident. He reacted poorly to this separation and was unfriendly to his mother when she returned. She soon became pregnant and delivered a brother whom the patient despised. Another brother, born a few years later, was better tolerated by the patient.

After the birth of the first brother, the patient was trapped in a painful situation for which he could not find a resolution. His brother was there to stay, and his rage at this sibling and at his parents for having him was not acceptable to his parents and engendered guilt and fears of punishment. He was unable to resolve the matter and the configuration of traumatic distress, fantasies of murder and punishment, and associated affects of anger, fear, and remorse remained as a separate organization within the patient's psyche and constituted a freeze frame of this period of his life. The activity of this organization in processing events and relationships in his adult life was abundantly clear.

When given the files of several law students who were applying for a summer position in the firm, for which he worked, the patient left them in the locker room of his squash club. He had been impressed with the background and achievements of these prospects and felt that he would be overshadowed were they to join the firm. When he was asked if he would mind if a space he intended to move into were given instead to two paralegals, he created a storm which hurt his standing in the firm. When a junior associate complained about an assignment that would encroach on what otherwise would have been a long holiday weekend, the patient became enraged and screamed at the complainer, calling him, among other things, a spoiled brat. When the entire city was excited about the local team playing in and winning the World Series, the patient secretly rooted against them; they were, after all, everyone's darlings. When his wife became pregnant, he dreamed of encountering a stranger in his favorite childhood play spot and fighting with him. Thus his adult situation replicated his childhood in that he could not tolerate a situation in which he was being displaced by a younger rival and he could ill afford to express his rage at being forced to do so.

It is not unusual for patients over time to oscillate between untenable positions. Thus a patient may take an aggressive attitude in dealing with a superior at work or in an academic setting only to worry that he is pushing things too far and may suffer unpleasant consequences. In response to this anxiety, he may adopt a meek, conciliatory posture in the hope of appeasing the now dreaded authority and escaping retribution. In time, feeling that such self-castrating behavior is unbearable, he reacts by once again taking an aggressive attitude. Such patients frequently say they wish to find a 'middle position.' Some analysts may understand such seesaw situations as instinctual conflicts between for example activity and passivity or between heterosexuality and homosexuality. Arlow (1963) wrote;

Where there appears to be a conflict between opposing drives, the ego takes sides with the expression of one set of drives, i.e., permits the discharge of the cathexis of one drive, in order to fend off the expression of the other, the more objectionable drive representation. Conflicts of this type ... in actuality represent a conflict between the ego and the id, with the superego apparently in accord with the defensive position taken by the ego (p. 16).

Such a view of how the neurosis is structured is not apt to be helpful because there is no escape. If the patient adopts a passive, submissive attitude out of fear of the consequences of his aggressiveness, one might say the ego is taking sides with the passive drive derivatives; and if the passivity becomes intolerable one might say the ego is taking sides with the aggressive drives in order to defend against the passivity. It is understand-able that the patient who swings back and forth between such polarities should wish he could find a middle position.

The analyst who is guided by the schema model will view the situation differently. He will understand that the aggressive and passive fantasies and behavior and the reasons for oscillating between them are all within the pathogenic schema. He will work to understand what elements of this organization are being manifested in these oscillations. Thus, he may discover that as a child a patient had reacted to his father's absences with fantasies of his father's death and taking his father's position with his mother, only to become terrified when his father returned. Given this perspective, the patient will no longer be trapped into alternating between the positions, but will have the alternative of seeing both positions, the reasons for shifting from one to the other, and the associated affects as elements of an anachronistic, pathogenic residue. He will then have the option of striving to deal with the situation from the perspective of mature reason and judgment.


If psychoanalysis were a science the demonstration that the structural model was incompatible with hard, cold, scientific evidence would be a staggering development. Textbooks would be rewritten, curricula revised. There would be an intellectual ferment such as occurred in 1889 when it was shown that pancreectomized dogs developed diabetes. This led to intense research on the pancreas resulting in the isolation of insulin in 1921. The clinical manifestations of diabetes had been known for millennia; when this knowledge was joined by laboratory research, the understanding and treatment of the disease improved immensely. Sadly, the psychoanalytic establishment is reactionary in character. They may be likened to the cardinals of the Catholic Church who held on to another erroneous model, the theory that the earth was the center of the universe, for four centuries after Copernicus had proven otherwise.

The survival of psychoanalysis may hinge on its willingness to meet the call for the new intellectual framework in psychiatry issued by Kandel (1998); he observed that the activity of the mind is actually a range of functions carried out by the brain and went on to declare that our great challenge is to delineate the relationship between the brain and mental processes in terms satisfying to both the neural scientist and the psychiatrist. In so doing it would be well to heed a principle enunciated by Yovell (2000) which states that while psychoanalytic theory can and must go beyond current neurobiological understanding, it should never contradict it.


Arlow, J.A. (1963). Conflict, regression, and symptom formation. Int. J. Psychoanalysis, 44: 12-22.

Brenner, C. (1992). The mind as conflict and compromise formation. Paper delivered at the Fall meeting of the American Psychoanalytic Association, December.

Breuer, J. and Freud, S. (1895). Studies on Hysteria, S.E. 2.

Damasio, A. (1994). Descartes' Error, New York: Grosset/Putnam.

Freud, S. (1915). The Unconscious. S.E. 14.

_____ (1923). The Ego and the Id. S.E. 19.

_____ (1940). An Outline of Psychoanalysis. S.E. 23.

Kandel, E.R. (1998). A new intellectual framework for psychiatry. Am. J. Psychiatry. 155:4 (p. 457-468).

Kaplan-Solms, K. and Solms, M. (2002). Clinical Studies in Neuro-Psychoanalysis. 2nd Edition. London & New York; Karnac.

Klein, G.S. (1976). Psychoanalytic Theory: an Exploration of Essentials. New York: International Universities Press.

Le Doux, J. (1992). Emotion as memory: Anatomical systems underlying indelible neural traces. In S. - A. Christianson (ED.), Handbook of Emotion and Memory. Hillsdale, NJ: Erlbaum.

____ (1996). The Emotional Brain, New York: Simon & Schuster.

Piaget, J. (1926). The Language and Thought of the Child, New York: Basic Books.

Sandler, J. (1983). Reflections on some relations between psychoanalytic concepts and psychoanalytic practice. Int. J. Psycho-analysis 64: 34-40.

Slap, J.W. (1986). Some Problems with the Structural Model and a Remedy. Psychoanalytic Psychology, 3: 47-58.

Slap, J.W. (1987). Implications for the Structural Model of Freud's Assumptions About Perception. J. of Am. Psychoanalytic Association, 35: 629-646.

___ & Saykin, A.J. (1983). The Schema: Basic Concept in a Non-Metapsychological Model of the Mind. Psychoanalysis & Contemporary Thought, 6: 305-325.

___ (1984). On the Nature and the Organization of the Repressed. Psychoanalytic Inq., 4: 107-124.

___ & Slap-Shelton, L. (1991). The Schema in Clinical Psychoanalysis. Hillsdale: The Analytic Press.

___ (1994). The Schema Model: A proposed Replacement Paradigm for Psychoanalysis. Psychoanalytic Review. 81: 677-693.

Solms, M. (2001). An example of neuro-psychoanalytic research: Korsakoff's syndrome. Unpublished paper presented to the British Psycho-Analytical Society on March 7, 2001.

True, W. R., Rice, J., Eisen, S. A., Heath, A. F., Goldberg, J., Lyons, M.J., & Nowak, J. (1993). A twin study of genetic and environmental contributions to liability for post-traumatic stress symptoms. Arch. Gen. Psychiatry, 50: 257-264.

Waelder, R. (1936). The Principle of Multiple Function: Observations on Overdetermination. Psychoanalytic Quarterly, 5:45-62.

Watt, D.F. (1990). Higher cortical functions and the ego; Explorations of the boundary between behavioral neurology, neuropsychology and psychoanalysis. Psychoanalytic Psychology, 7: 487-527.

Yovell, Y. (2000). From hysteria to past traumatic stress disorder: Psychoanalysis and the neurobiology of traumatic memories, Neuro-Psychoanalysis. 2: 171-182 [an error occurred while processing this directive]