Learning from Experience

Michael Pariser, PsyD

How can we make the same mistakes over and over? How can we not learn? You tell me…how? (Elgin Baylor, Vice President of the Los Angeles Clippers, a perennially losing basketball team)


In 1914, Sigmund Freud noted that “it is only by dire experience that mankind ever learns sense” (p. 373), and yet it seems that just as often, mankind fails to learn from experience, no matter how dire or how frequently it occurs. In a discouragingly large percentage of the time, people continue patterns of counterproductive behavior despite massive and continuous aversive feedback from the environment. The obvious question that arises is why? Why do people keep on doing the same thing over and over…and when will they ever learn?

Beginning around the turn of the 20th Century, analysts and other psychological thinkers tried to explicate this phenomenon and identify the internal mechanisms responsible for it. Freud (1920) felt that people were compelled to repeat unanalyzed traumatic events, driven demonically by an innate and unconscious urge that ultimately led back to nature’s most basic biological state: death. Fairbairn (1941, 1943) saw anxiety triggering the return of internalized and repressed bad objects, which now prompt rigid responses in an attempt to maintain relatedness. For Klein (1935, 1946), that same anxiety induced the opening of an internal split, causing a shift from love to hate and from gratitude to envy, thus triggering venomous attacks that produce, ironically, precisely the abandonment that was the source of the anxiety to begin with. Kohut sees the narcissist self trying to maintain psychic equilibrium through the creation of interpersonal relations that provide important selfobject functions, the loss of which activates defensive behavior designed to repair the narcissistic injury. All of these (and others as well) have proven to be clinically useful theories that have guided practitioners for many years, and still do. Unfortunately, despite over 100 years of intense speculation, no universally accepted explanation has arisen that fully accounts for repetitive maladaptive behavior patterns and the failure to learn from experience.

In examining the existing theories, several reasons emerge for the failure to arrive at a consensus explanation. To begin with, the dominance of Freud’s Repetition Compulsion skewed the theoretical debate for many years, so that the only question that could be asked was; how correct is Freud? Then there was the territorial claim on learning theory by Behaviorists and Cognitivists, whose refusal to examine the “black box” of the mind limited their contributions far more than was really helpful. Most of all, however, I believe there were three deficits that robbed these theories of full explanatory power.

The first was the failure to understand the competitive nature of learning. In general, the replacement of one behavior or pattern of behavior with another is not a simple matter of learning something new, but a process of struggle between the pre-existing pattern and the new. And this is not manifest only on the behavioral level, but all the way down to the organic substrate, where the biochemical alterations underlying the learning process predispose an individual to repeat, rather than to learn something now. In other words, it’s very much like a political contest, where the incumbent almost always has the edge.

The second shortcoming of existing theories was the failure to grasp the multiple and critical roles of affect in the learning process; because affect provides both the incentive and the reward for learning, as well as acting as the ultimate arbiter of both personal and interpersonal meaning. In doing so, it generates the impetus to instantiate and to maintain the learning process, and then to encode that learning into long term memory, where it ultimately can replace the previously established patterns.

The central role of affect in learning leads clearly to the third problem, one suggested, but not entirely explicated, by earlier psychoanalytic theories: the impact of the dissociation of affect. Although earlier models acknowledged that defensive dissociation was problematic, they tended to focus on the content of the dissociation – traumatic memories, internal objects, or representations, or relationships – rather than the affective arousal. But if affect is necessary for learning, dissociation prevents its use, and new behavior cannot then survive to replace the maladaptive old.

I would like to address these issues one at a time, utilizing a neuropsychoanalytic perspective. First I will examine the learning as it takes place in its initial phase, looking especially at the role of affect. Then I will explore the competitive nature of replacement learning, in order to show why it is so much more difficult than original learning. Third, I will detail the process of emotional dissociation and its impact on learning failure. Finally, I will talk about an approach to re-associating affect and re-enlivening the emotional learning process.

Before exploring the relationship between affect and learning, however, it is necessary to clarify what will be meant in this paper by learning failure and what will not. By “the failure to learn from experience” I do not mean learning failure as a result of: (1) head trauma, disease, or any other organically caused brain or organ dysfunction; (2) chemically induced alterations of the brain; (3) disorders such as autism or Asperger’s Syndrome; (4) mental retardation; or (5) “learning disorders”. Instead, I am referring to the much more pervasive and mundane experience of repetitious maladaptive behavior and the failure, in the face of negative feedback from the environmental surround, to replace those behaviors with more internally and externally (socially) rewarding ones. This inability to alter archaic organizing principles overlaps considerably with standard definitions of psychopathology as located in the DSM-IV (APA, 1994), including, and perhaps especially, the Axis-II personality disorders.

Basic (Initial) Learning

Learning has been defined as “any relatively permanent change in an organism as a result of practice or experience (Chaplin, 1985). Basic, or initial, learning (that is, learning something for the first time) can best be understood by utilizing first an evolutionary perspective. Learning operates from birth (Beebe and Lachmann, 2003; Stern, 1985) and possibly in utero (Piontelli, 1988, 1989). In fact, the ubiquity of learning throughout the lifespan points to its critical adaptive functioning, not just for mankind, but for all species in all niches of the environment; organisms that learn clearly outcompete organism that do not. As Le Doux (1996) observes, learning provides the ability to store information, as well as emotional and behavioral responses, so that elements of the environment do not need to be re-encountered each instance as if it were the first, thus providing an enormous saving of energy. This efficiency is further enhanced by the ability to categorize, which may be thought of as a subset of the learning process. Moreover, learning provides the ability to predict future events, such as the likely location of food sources or the responses of desired reproductive partners. Ultimately, all of these functions promote the ability of the organism or individual to adapt to his or her environment, whether that environment is static or changes over time (Edelman, 1992). Sufficiently adaptable learning even allows for the successful translocation of the organism or the species to a different environment altogether.

The mechanics of basic learning are relatively simple. An orientation to stimuli in the external or internal environment triggers feelings in the individual that push for adaptive action. So you see the lion, you feel fear, you run away; or you experience hunger, you go to the woods (or to the refrigerator) and forage for food. The perceived results of the action also trigger feelings, which are rewarding or punishing, and the experience of those feelings helps determine the meaning of the event: if it worked out, you would be more likely to try it again; if not, probably you wouldn’t. Finally, if it is sufficiently salient, the entire event, along with its affective component, is encoded into long term memory for use in future predictions.

In the neurobiological sphere, the process of learning is complex, and there are changes that take place from the very first instance of an event. This can be seen utilizing three different powers of neurological magnification. On the macroscopic level, transduced sensory input from the environment, along with intrapsychic and somatic information, is fed into reciprocating maps (Edelman, 1992) that are capable of combining differing types of sensory input into overall pictures. The resulting gestalt is a neural activation that triggers arousal of limbic structures such as the amygdala and is experienced consciously as a feeling, which, in turn, prompts adaptive action. The results of the action once again trigger neural activation, this time of the dopamine system from the substantia nigra, whose axons project into the nucleus accumbens, a structure that mediates the feeling of reward. It is here that the ultimate evaluation of the action is made, and to a great extent, the “meaning” of the event is determined. Finally, the neural pattern aroused by that event and its meaning are fed to the hippocampus, where it is encoded into permanent memory (Kandel et al, 2000).

The above processes are underlain microscopically by the biochemical phenomenon of neuroplasticity (Kandel et al., 2000), where experience, as a pattern of neural activation, triggers a biochemical cascade within a nerve cell that culminates in the increased connectivity of that neuron to whichever other neurons are active at that particular time. This increased connection strength leads to the greater likelihood that when any of the neurons in the pattern are active, the other neurons will be aroused as well, so that if neurons A and B are active simultaneously, then, in the future, if either A or B fires, the other is more likely to fire as well. This principle, which was first developed by Canadian neurobiologist Donald Hebb in 1949 and which now bears his name (Hebb’s Axiom or Hebbian Learning), is often translated into the mnemonic phrase: “neurons that fire together, wire together”.

Hebbian learning is the principle that underlies the phenomenon of association, which can be most effectively conceptualized on the “mesoscopic” level (Freeman, 2000), where clusters of neurons operate as tiny networks. There, repetitive patterns of simultaneous activation representing distinct stimuli form powerful biochemical links. So, for example, if a woman has a loving mother, then the neural pattern associated with “mother” will generally be active at the same time as the neural pattern associated with “loving”, and a linking of those two concepts will take place. By contrast, if “mother” is associated with “critical and demanding”, then an altogether different connection will be formed.

An important aspect of this process is that it is cumulative; up to a point, each repetition of the arousal of a network adds to the strength of the active synaptic connections and therefore to the solidity of that pattern of activation. What’s more, as a particular pattern is repeatedly excited, it becomes linked to a greater number of other patterns in an ever widening circle of associated contexts. Hebb’s Axiom tells us that the pattern will then be more likely to be triggered by the activation of all those other patterns, so the net effect is that the pattern will be more easily aroused and more often. From a systems perspective, this heightened frequency of arousal results in the formation of “preferred attractor states”; that is, increasingly deeper “basins” of activation (Thelen and Smith, 1994). These manifest as patterns of thought and behavior, moods, habits, and other commonly experienced modes of being. And the longer and more established they are, the more difficult they become to change.

The Role of Affect in Learning

The basic role of affect within the learning process is reasonably straightforward. Every waking minute, the human sensoria are bombarded by stimuli, far too many to integrate, making it critical for the mind/brain to have a mechanism for discarding the unimportant input and locating the crucial: that which has survival value in the ongoing environmental surround. Affect supplies that evaluative function, acting, by its arousal, to heighten the brain’s attention to certain elements of the total sensory influx, and by its absence, to diminish the rest (Le Doux, 1996).

Neurobiologically, affect performs its stimulus evaluation through the action of summation. Because of the simple fact that there are so many stimuli in the environment, sensory input alone is generally insufficient to cause firing in the neurons further down the line; motor neurons, for example, that would cause an individual to orient and respond physically to a particular stimulus. These neurons, like all neurons, fire on an all-or-nothing basis and require a certain level of input from other neurons in order to do so. So there needs to be additional neuronal stimulation, and that is supplied by the excitatory arousal sent via projecting axons from emotive brain structures such as the amygdala and the cingulate gyrus. The resulting summated stimulation triggers an action potential on the receiving neuron, and the neuron fires. In this way, affect, by augmenting the input to certain neurons and not others, increases attention to some elements of the surround and not others, thus determining the experience of salience, a critical factor in the creation of both personal and interpersonal meaning.

In certain instances, however, the activation profile is not linear. In the case of explicit learning and memory, for example, increased amygdaloid arousal triggers increased hippocampal activation, which produces greater long term potentiation of memory and therefore of learning, but only up to a point. The graph of the effectiveness of this input is actually an inverse-U curve (McEwen, 1999; Siegel, 1997), reflecting the fact that too much or too little input results in a decrease of synaptic activation and a concomitant inhibition of learning. Only when there is a median level of input does hippocampal activation takes place and provide the neuronal context for learning. Translating from the above neurobiology, an important concept emerges: that in order to learn, the associated affect must fall within a usable range; too much or too little emotional arousal are both detrimental to the learning process. The clinical import of this finding is that emotional consciousness is critical for learning; alexithymia is not helpful at all, although flooding with affect is equally useless. As Freud discovered, abreaction and catharsis alone are insufficient for cure.

The provision of reward is another critical role for affect in the learning process. All affects have a hedonic valence, which can be positive (as with happiness, pride, love, and joy), neutral (e.g. interest, surprise), or negative (e.g. sadness, anxiety, fear, disgust, envy, and hate). As stated above, this valence is mediated by an internal reward system (the mesolimbic dopamine pathway to the nucleus accumbens) which provides feelings of either euphoria, in the case of positive arousal, or dysphoria, in the case of negative. It is these basic hedonic experiences that provide the evaluation of human experience: was this something that will be enjoyable or rewarding to repeat? The answer to that question will prompt future action: as a simple formula euphoria triggers a move towards, and dysphoria the opposite, a retreat from, any given stimulus.

Of course, for human adults in complex social systems, the evaluation process is far more complicated. In particular, it has to be recalled that the target behavior that is now such a source of irritation arose in a context in which it was rewarded, not just as a phenomenon in and of itself, but, more importantly, as a relational event connected to important caregivers; so whenever part of that activation profile is triggered, the rest of it is increasingly likely to become triggered as well, including those relational dynamics. In that way, a long history of idiosyncratic reinforcement produces a very particular profile of behavior that is felt to be rewarding. Moreover, this profile can be fluid: the same stimulus can be rewarding or punishing, depending on the overriding affect state at the time. This is why an analyst, making the identical interpretation to two different patients, or the same patient on two different days, may find himself responded to in very different ways.

Replacement Learning

Any system, including a system of neural activation, can only be organized in one way at any given moment in time. If, as the result of long term repetition (and its concomitant Hebbian wiring), a certain pattern of neural activation is activated when an individual is presented by the surround with a certain set of stimuli, then it will be impossible for a differing neural pattern to be induced. In the sphere of learning, therefore, there is an inevitable conflict between pre-existing learning and any new form of learning that is designed to replace it. Schafer (1984) made a similar point in his discussion of “the pursuit of failure” when he commented:

…as a rule, people are more or less unconsciously conflicted with respect to their goals in life. Due to this conflict, what is success from the point of view of one set of aims is failure from the point of view of other aims that conflict with them. Freud (1900/1953, p. 604) established this principle in another connection, that of pleasure, when he observed that what is pleasurable for one psychical system may be painful for another (p. 399).

An important consideration when looking at the competitive nature of learning is that initial learning, by dint of its repetitively and multiply reinforced synaptic bonds, has a distinct advantage. It is the default setting and needs no further input to become activated; the new pattern, in order to trigger attention and learning, needs not the just the normal level of affective boost, but enough power to overcome existing connection strengths, and that much neuronal clout is simply not that easy to muster. As can easily be imagined, this would certainly contribute to the difficulty breaking old habits or changing existing patterns of behavior even in the face of obvious negative feedback from the environment.

But there is another reason as well for the persistence of established emotional and behavioral responses, one that has to do with their place within an invisible, but intrinsically rewarding system. It is by now a common observation that a sense of self-cohesion is provided by the known, even if that known is negative. Why is that?

A deeper look at the phenomenon of familiarity reveals again the role of affect. All unfamiliar stimuli produce some measure of sympathetic arousal. Evolutionarily, this functions to provide maximum survival possibilities; it is safest to be vigilant until the new stimulus has been fully assessed and found harmless. Over time, the repeated presentation of the stimulus without danger induces the process of habituation, in which neurons respond at an increasingly lower intensity. This reduction in activation allows for a decrease of vigilance and anxiety, an emotional event which is, in itself, intrinsically rewarding. The new, by contrast, still triggers anxiety, thus further handicapping it in the competition. To further complicate matters, much of the competition takes place unconsciously, along with the ongoing rewards and punishments.

So new learning faces not just an uphill battle, but one that must proceed at night, into the teeth of hidden enemy guns.

Affect Regulation and Dysregulation

As Le Doux (1996) suggests, emotions function by making an individual feel good or bad, thus prompting adaptive action; but in a complex and communal society, however, it is not always most useful or beneficial to express unregulated affect, so a system needs to be in place to adjust and tune levels of emotional arousal in response to ongoing events in the individual’s internal and external world. This system is called affect regulation, which:

acts as a recovery mechanism that efficiently autoregulates the duration, frequency, and intensity of not only positive, but also negative affect states (mood regulation). This emergent function, in turn, enables the individual to recover from disruptions of state and to integrate a sense of self across transitions of state, thereby allowing for a continuity of experience (Schore, 2003, p. 24).

With relation to learning, which requires emotional arousal to fall within the medial part of the curve in order to be usable, the regulation of affect brings outlying levels of affect into that useful range.

Early in a child’s life, emotional regulation is difficult to achieve, because the “wetware” necessary for its operation, the right orbitofrontal cortex, is not yet completely online. It is only when that area of the brain is fully myelinated, at around eighteen months of age, that effective auto-regulation can be completely operationalized (Schore, 1994). Before that time, the child achieves affective control through a mutual regulatory process co-created with his or her caregiver (Beebe and Lachmann, 2002). As time goes on, however, this process becomes increasingly internalized, a developmental achievement with useful clinical implications. Schore (1994, 2003) observes that many psychopathological disorders can be seen as problems of affect regulation; and from that perspective, therefore, attainment of the capacity for emotional self-regulation is an important milestone in therapeutic cure.

If affect regulation is the process of tuning and adjusting levels of emotional arousal, affect dysregulation can be thought of as the failure to do so in an effective way. Dysregulation, however, is not synonymous with non-regulation; in fact, it can be either under-regulation, in which the mind is flooded with feeling, or over-regulation, in which the useful emotional response is simply “regulated” out of existence. The precise pattern of regulation and dysregulation is unique to each individual and is determined by both a distinctive interpersonal history, including caregivers’ patterns of tolerance and intolerance for the expression of specific emotions, and the current triggers in the environmental surround. Fosha (2000), from an attachment perspective, supports this observations:

Both feeling but not dealing (being overwhelmed with feeling and unable to cope) and dealing but not feeling (to “go on automatic,” eradicating feelings in order to cope) are products of defensive strategies. Closeness is achieved at the cost of either compromised external functioning (resistant pattern of attachment) or loss of internal suppleness and aliveness (avoidant pattern of attachment) (p. 42).

Over the years, dysregulation has adopted many names, most of which entered the psychoanalytic lexicon with either Sigmund or Anna Freud (A. Freud, 1937/1966; S. Freud, 1900/1965) including repression, suppression, denial, splitting off, warding off, etc. etc. The ultimate mechanism, however, as Schore (2003) and Siegel (2003, personal communication) suggest, is a massive inhibition of amygdaloid arousal and/or the functional separation of the links between the amygdala and the dorsolateral prefrontal cortex, where consciousness and working memory reside. Therefore, I agree with Bromberg (1999) that the term that best describes this neural severing is dissociation, because through this dysregulating process, emotion is dis-associated from the events to which it is both the response and evaluation.

It is that destruction of the links between an event and its emotional meaning that produces the greatest impact of dissociation on learning. If affect is not present to generate an evaluation of social cues, no effective assessment can be made of the complexities of a situation or the possible consequences of various actions. If specific emotional arousal is absent, then no associations can be made to memories of past events in which that emotion was generated. If emotions are not present to provide reward and punishment, the feedback generated by old patterns of behavior is not connected in a way that would allow the negative results of an event to be experienced as punishing. Even when there is, to a certain extent, an aversive experience, the dissociation can be sufficient so that the positive reward generated by the familiar pattern will override it.

Clinical Implications

From a neurobiological perspective, an understanding of the competitive nature of learning, the critical roles of affect, and the pernicious effects of dissociation provide a clue as to the effectiveness of widely disparate treatment modalities, including at the one end, cognitive and behavioral approaches, and at the other, psychoanalysis. All provide systems for overriding the activation of existing neural networks and the establishment of new patterns in their place. Moreover, all provide some reward for doing so, although in the case of the more concrete approaches, that reward tends to be primarily extrinsic. In my opinion, however, this same understanding provides compelling reasons why psychoanalysis is, in the long run, the most effective solution to widespread learning failure.

To begin with, if the primary problem centers on affect, affect must take center stage. As Fosha (2000) points out, “The experience of vital affects in the context of an attached relationship is the primary agent of emotional transformation in life and – a fortiori – in treatment” (p. 5). In contrast with other therapeutic approaches, more and more, affect is being understood as being at the heart of psychoanalytic investigation (Stolorow, 1997).

Specifically, in order to counter the strength of established neural pathways the dissociated affect needs to be elucidated and re-associated to the event from which it originally arose. This reintegration of affect and experience in the context of a safe holding environment, overrides the fear normally triggered by those feelings in the past and provides the patient with an experience of a useful and tolerable emotional response. This is the beginning of affect regulation.

Technically, in order to reassociate the initial emotion, it is helpful to do what I call, “stretching the moment”, a process in which the instant between the precipitating event and the first conscious emotional response is extended and micro-analyzed. Almost always, I have found that despite the almost instant awareness on the patient’s part of a particular emotional response that then inevitably triggered an unwanted behavior pattern, there is a chain reaction of affect and dissociation that may be many steps long. As the moment is stretched, analyst and patient attempt to retrace the steps of the patient’s affective sequence, working backwards and forwards and filling in the emotional blanks, until a complete picture of that affective chain reaction emerges, including the all-important initial unconscious response, which can then be investigated to determine the reason it was so potentially destabilizing to the prevailing affective-relational dynamics. In addition, the causal links of the chain can be analyzed, in order to bring to light the means by which the patient traveled from one inevitable and irresistible step to the other.

The second reason that psychoanalysis provides the best treatment for learning failure in general, is that although all treatment modalities provide some reward, the extrinsic focus in more behavioral approaches tends not to address the power of the interpersonal connection or to examine the hidden reward system built up over the years. In order to rob the existing system of its subjective sense of reward, it is critical to understand the nature of the patient’s archaic relational goals, those that are in conflict with the stated goals in the present. As these long-established interpersonal dynamics are analyzed and worked through, they begin to lose their reinforcing power; and when intrinsic reinforcement for old organizing principles is reduced, the resistance to new ways of structuring experience will be lessened as well.

Finally, an aspect of psychoanalysis often criticized by others – the length of treatment – is precisely what is needed to provide sufficient repetitions of the desired new approach, so that, through Hebbian connectionism, it becomes the default setting. I call this (with apologies to the Firesign Theater) “regrooving”: a process in which the adaptive spiral builds up so much synaptic strength that it is more likely to be activated than the old, vicious spiral. This is a process that permits the expansion of one’s “world horizons” (Stolorow et al, 2001) to include an ever-widening view of life and of the self.

Clinical Example #1

Ms. J is a 47 year-old woman with a history of mental illness and an air of emotional fragility. At the time, she lived on disability payments and reported that she was unable to work. Historically, Ms. J’s home life had been chaotic and abusive, and school was the only emotionally safe place for her. Accordingly, she strove to excel there: she had been a straight-A student from the first grade and had accumulated two master’s degrees; she also became what she describes as “the teacher’s pet” in all her classes. However, she has a long history of failure in the workplace and has not held any job for more than a couple of months. Nonetheless, she engaged, in treatment, in a series of career plans, all of which had an unreal quality in which everything would work out spectacularly well.

After about a year in analysis, she decided that she, too, would like to become a psychoanalyst, a goal that, considering her financial and emotional states, was unrealistic. I interpreted the wish to be like me, the hope to be big instead of small, the fantasy of not being the vulnerable and dependent one, etc., etc. but to no avail; no amount of interpretation would deter her from her new-found goal. Accordingly, she signed up for pre-requisite classes at the local community college.

One day, she came in very upset and announced that she was dropping out of her statistics class because the teacher was prejudiced against her. This seemed unlikely to me, so I began to inquire. I discovered that what had occurred was that she had made a silly mistake on a test and had called the teacher at his office to request a chance to correct it. When the teacher hadn’t gotten back to her, she became furious and decided to quit.

I was being supervised by a Kleinian analyst at the time, and so I considered trying to interpret projective identification. I thought perhaps of bringing this into the transference by allowing her to locate me as an additional source of unfairness. I also considered a more empathic response, something along the lines of how hard it must be when she feels as if the world is so unfair. But I felt that both of these interventions missed the main point somehow, so instead, I tried to retrace the emotional steps, which produced the following chain of events (presented in a forward direction for clarity’s sake):

• The mistake produced disappointment

• The disappointment was felt to be intolerable because it was associated with failure and activated memories of other disappointments and failures, so it was dissociated

• At the same time, however, the memories of failure triggered an overpowering sense of despair and doom in which this one mistake would begin a domino-like collapse that would involve not just this test and this class, but her whole plan to become a psychoanalyst and, ultimately, her entire life

• This threat also activated the potential destruction of the ideal mentor-student relationship she values as so perfect: she could no longer be the teacher’s pet, and she was thrown into the kind of chaotic relational dynamic that characterized her non-school life, full of emotional pain and suffering, but close to her obsessive mother and her alcoholic father. She began to panic.

• When she thought of calling the teacher, a desperate hope arose in her mind that the Hell she had entered could be escaped with a single stroke, but when he didn’t call her back, it was confirmation of her true badness and her bleak and empty future.

• These feelings were also intolerable and needed to be dissociated. That left her with no explanation for the situation in which she found herself – a perfect student with a career-threatening problem – so her interpretive left hemisphere (Gazzaniga, 1998) created a “story” in which it was the teacher who was at fault, and the reason was that he was prejudiced against her because she was disabled, and “no one likes disabled people”.

• It was a short step then to anger at the unfairness, and then the only logical solution in her mind was to drop out of the class and take it at some later time with another “nicer” teacher, or, better still, develop yet another (unrealistic) career plan.

Reconstructing this affective cascade enabled us to identify the original dissociated emotion and to explore the reasons it was so intolerable to her. Reassociating it, we were then able to address the underlying dynamic in the transference, and instead of the issue of unfairness or prejudice or the projection of her fears onto the outside world, we focused on her wished-for ideal relationship with the perfect teacher (me), and we were able to discover how, though she imagined that relationship as leading to perfect happiness and success in life, it nonetheless neglected important aspects of reality that also needed to be considered in creating a successful career. Shortly after that, she let go of the fantasy of becoming a psychoanalyst, and took a job as a peer advocate at a mentally ill women’s shelter, where she is, little by little, becoming well respected for the work she is doing.

Clinical Example #2

Mr. M is a 30 year old computer programmer who works for a medium sized entertainment firm. His childhood history with a critical and religious mother left him with a perfectionism that drove him to attempt to achieve a Christ-like goodness in all phases of his life. His failure to do so created an ongoing depression and an overwhelming sense of hopelessness.

He was in treatment with me on a three-times-a-week basis for about a year, when, one day, an interpretation I made concerning his role in a relationship problem was met with an immediate and sudden silence. When I inquired as to what had happened, Mr. M did not answer, but only stared up at the ceiling. This was highly unusual for him; he is normally an outgoing, talkative person and an eager analysand. After some time, I was able to get from him that my interpretation had triggered an overwhelming reaction of shame. This response was so massive, that he could not speak for the remainder of the hour, and it was not until the following session that we were able to look into what had happened to him.

At first, he was able to say only that he had heard my words and then instantly felt ashamed; however, by stretching the moment between the event – my interpretation – and the patient’s emotional reaction – shame – we were able to reconstruct the following dynamic:

• My interpretation was received by Mr. M as criticism

• Criticism triggered feelings of hurt, the subjective experience of abandonment and the obliteration of his illusion of Christ-like perfection; this replicated experiences he had had with his critical mother

• Hurt and abandonment triggered anger, which was historically not tolerated, so it was dissociated, along with the constellation of feelings that gave rise to it, leaving the situation – bad feelings with a good therapist – to be explained

• The situation was explained as there being something wrong with patient, which thus preserved me as the good parent (reminiscent of Fairbairn’s famous [1943] dictum that it is better to be bad in a good world than good in a bad world)

• Since I am normally experienced by him as non-critical and supportive, in fact, as ideal, the reinstatement of me as perfectly good made him perfectly bad, thus triggering a massive reaction of shame.

Once all the links in this chain had been connected, it was then possible to reassociate the disowned feelings of hurt, so that he was able to acknowledge the pain my comments had caused him. When I then apologized for hurting him, it became possible for him to perceive us both as well-meaning men with flaws, rather than as Christ-like figures in a perfect landscape. Moreover, he also began to experience himself as having a real sense of agency and the ability to make a substantive impact on an important person in his life.


The problem of learning failure and the repetition of maladaptive behavior patterns is one of the most vexing for people in all walks of life. Over the years, many theorists, analytic and otherwise, have attempted to address this issue; however, due to the scientific constraints of the time, explanations were either limited (in the case of Learning Theorists) or metaphorical (in the case of psychoanalysts). In all cases, theories were delimited by the isolated mind approach that has recently come under telling criticism (Mitchell, 1993; Orange, 1995). More contemporary neurobiological research (e.g. Le Doux, 1996; Kandel et al, 2000; Schore, 2003), along with newer relational psychoanalytic models, (Benjamin, 2003; Coburn, 2001; Greenberg, 1999; Mitchell, 1993) have provided perspectives that make possible a different approach to the problem, one focused on the centrality of affect and affect regulation.

Specifically, the critical pieces of the puzzle are the strengthened synaptic connections that form the basis of learning (Hebb, 1949), the reward that comes with that learning (both intrapsychically and interpersonally), the maintenance of archaic relational ties, the intense competition between established structures and newer patterns of organization, and the dysregulation of the affect that would be necessary for replacement learning to prevail. Through the process of locating that dissociated affect and reconnecting it to the original event, the analyst can help the patient to perturb the rigid psychic system that maintains the maladaptive behaviors and begin to establish more useful and constructive approaches to himself and the world around him.


American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th Ed. Washington, DC: American Psychiatric Association

Beebe, B. and Lachmann, F. (2002). Infant Research and Adult Treatment. Hillsdale, N.J.: The Analytic Press

Benjamin, J. (2003). Thirdness. Unpublished paper presented at a conference in Los Angeles entitled: “Beyond Doer and Done To: Impasses, Enactments, and the Intersubjective Third.

Bromberg, P. (1999). Shadow and Substance: a relational perspective on a clinical process, in S. Mitchell and L. Aron (Eds.), Relational Psychoanalysis, the Emergence of a Tradition. Hillsdale, N.J.: The Analytic Press

Buckley, P. (1986). Essential Papers on Object Relations. NYC: New York University Press

Chaplin, J.P (1985). Dictionary of Psychology (2nd rev. ed.). NYC: Laurel Books

Coburn, W. (2001). Subjectivity, emotional resonance, and a sense of the real. Psychoanalytic Psychology, v. 18 (2).

Coburn, W. (2002). A World of Systems: The Role of Systemic Patterns of Experience in the Therapeutic Process. Psychoanalytic Inquiry

Damasio, A. (1994). Descartes’ Error. NYC: G. P. Putnam’s Sons, Inc.

Damasio, A. (1999). The Feeling of What Happens. San Diego: Harcourt, Inc.

Edelman, G. (1992). Bright Air, Brilliant Fire: On the Matter of the Mind. NYC: Basic Books

Ellenberger, H. (1970). The Discovery of the Unconscious: The History and Evolution of Dynamic Psychology. NYC: Basic Books

Fairbairn, D. (1941). A revised psychopathology of the psychoses and psychoneuroses. In P. Buckley, ed. Essential Papers on Object Relations. NYC: New York University Press

Fairbairn, D. (1943). The repression and the return of bad objects (with special reference to the war neuroses). In: P. Buckley (Ed.). Essential Papers on Object Relations. NYC: New York University Press

Fosha, D. (2000). The Transforming Power of Affect. NYC: Basic Books

Freeman, W. (2000). How Brains Make Up Their Minds. NYC: Columbia University Press

Freud, A. (1937/1966). The Writings of Anna Freud, V. II: The Ego and the Mechanisms of Defense. Madison, WI: International Universities Press

Freud, S. (1900/1965). The Interpretation of Dreams. NYC: Avon Books

Freud, S. (1905 (1901)/1950). A fragment of an analysis of a case of hysteria. In The Collected Papers of Sigmund Freud, vol 3. p. 13-136. London: The Hogarth Press and the Institute of Psychoanalysis

Freud, S. (1914/1950). Remembering, repeating, and working through. In The Collected Papers of Sigmund Freud, vol 2. p. 366-376. London: The Hogarth Press and the Institute of Psychoanalysis

Freud, S. (1920). Beyond the Pleasure Principle

Gazzaniga, M. (1998). The Mind’s Past. Berkeley, CA.: University of California Press

Hebb, D. O. (1949/1988). Organization of Behavior. NYC: John Wiley, Inc.

Hunt, M. (1993). The Story of Psychology. NYC: Anchor Books

Kandel, E., Schwartz, J., and Jessell, T. (2000). Principles of Neural Science, 4th Ed. NYC: McGraw Hill, Inc.

Kaplan, H. and Sadock, B., (1998). Synopsis of Psychiatry (8th Ed.). Baltimore, MD: Williams and Wilkins

Kaplan-Solms, K. and Solms, M. (2000). Clinical Studies in Neuro-Psychoanalysis: Introduction to a Depth Neuropsychology. London: Karnac Books

Kernberg, O. (1966). Structural Derivatives of Object Relationships. In P. Buckley (Ed.), Essential Papers on Object Relations (pp. 350-384). NYC: New York University Press

Kernberg, O. (1967). Borderline Personality Organization. In M. Stone (Ed.) Essential Papers on Borderline Disorders (pp. 279-319). NYC: New York University Press

Klein, M. (1935). A Contribution to the Psychogenesis of Manic-Depressive States, in P. Buckley, (Ed.) Essential Papers on Object Relations (pp. 40-70). NYC: New York University Press

Klein, M. (1946). Notes on Some Schizoid Processes. International Journal of Psychoanalysis (v. 27).

Kohut, H. (1984). How Does Analysis Cure? Hillsdale, NJ: The Analytic Press

Kohut, H. and Wolf, E. (1978). Disorders of the Self and Their Treatment: An Outline. International Journal of Psychoanalysis (v. 59: 413-425).

Krystal, H. (1988). Integration and Self-Healing: Affect-Trauma-Alexithymia. Hillsdale, NJ: Analytic Press

Le Doux, J. (1996). The Emotional Brain. NYC: Touchstone Books

McEwen, B. (1999). Stress and the Brain. In R. Conlan (Ed.), States of Mind (pp. 81-101). NYC: John Wiley & Sons, Inc.

Mitchell, S. (1993). Hope and Dread in Psychoanalysis. NYC: Basic Books

Orange, D. (1995). Emotional Understanding: Studies in Psychoanalytic Epistemology. NYC: Guilford Press

Orange, D., Atwood, G. and Stolorow, R. (1997). Working Intersubjectively: Contextualism in Psychoanalytic Practice. Hillsdale, NJ: The Analytic Press

Pally, R. (2000). The Mind-Brain Relationship. London: Karnac Books

Piontelli, A. (1988). Pre-natal Life and Birth as Reflected in the Analysis of a 2-year-old Psychotic Girl. In: International Revue of Psychoanalysis, v. 15, p. 73-81.

Piontelli, A. (1989). A Study of Twins Before and After Birth. In: International Revue of Psychoanalysis, v. 16, p. 413-435.

Schafer, R. (1984). The Pursuit of Failure and the Idealization of Unhappiness. In: American Psychologist, v.39 (4): p. 398-405.

Schore, A. (1994). Affect Regulation and the Origin of the Self; The Neurobiology of Emotional Development. Hillsdale, N.J.: Lawrence Erlbaum Associates

Schore, A. (2003). Affect Dysregulation and Disorders of the Self. NYC: W.W. Norton and Co., Inc.

Siegel, D. (1997). The Developing Mind; Toward a Neurobiology of Interpersonal Experience. NYC: The Guilford Press

Siegel, D. (2001). Toward and Interpersonal Neurobiology of the Developing Mind: Attachment Relationships, “Mindsight,” and Neural Integration. Infant Mental Health Journal v.22 (1-2), 67-94

Spitzer, M. (1999). The Mind Within the Net. Cambridge, MA: MIT Press

Stern, D. (1985). The Interpersonal World of the Infant. NYC: Basic Books

Stolorow, R. (1997). Dynamic, Dyadic, Intersubjective Systems: An Evolving Paradigm for Psychoanalysis. Psychoanalytic Psychology v.14 (3), p. 339-346

Stolorow, R. and Atwood, G. (1992). Contexts of Being: The Intersubjective Foundations of Psychological Life. Hillsdale, NJ: The Analytic Press

Stolorow, R., Atwood, G., and Brandchaft, B. (1994). The Intersubjective Perspective. Northvale, NJ: Jason Aronson, Inc.

Stolorow, R., Brandchaft, B., and Atwood, G., with Stolorow, D. (1987). Affects and Selfobjects. In: Psychoanalytic Treatment: An Intersubjective Approach. Hillsdale, NJ: Analytic Press

Stolorow, R., Orange, D., and Atwood, G. (2001). World Horizons; a post-Cartesian alternative to the Freudian unconscious. Contemporary Psychoanalysis v.37(1), 43-61

Thelen, E. and Smith, L. (1994). A Dynamic Systems Approach to the Development of Cognition and Action. Cambridge, MA.: MIT Press

Vaillant, G. (1993). The Wisdom of the Ego. Cambridge, MA.: Harvard University Press

Wilson, E. O. (1998). Consilience: The Unity of Knowledge. NYC: Alfred Knopf, Inc.

Winnicott, D. W. (1969). The Theory of the Parent-Infant Relationship. In: P. Buckley (Ed.), Essential Papers on Object Relations (pp. 350-384). NYC: New York University Press

Comments are closed