Before I discuss how the infant research might impact on the dominant psychoanalytic models I need to touch on a couple of issues. Firstly, thinking of the dominant psychoanalytic model, there is the question, is there one psychoanalysis or many? The English psychoanalytic traditions of Kleinian and object relations theory have in the past been more open to the infancy research than the analytic traditions in Europe or America. But, while acknowledging that difference, I nevertheless want to try and think in an inclusive way about a body of psychoanalytic theories and clinical knowledge that helps people be more aware of how their minds functions.

Secondly, Andre Greens (Sandler, Sandler & Davies, 2000) trenchant criticism that infant research is of such a different order from psychoanalysis that it is irrelevant to consider a possible contribution from the research is well known. I think, however, that while psychoanalytic theory last century arose independently of other disciplines and crystallised out of the analyst-patient interaction, and I expect that it will continue to develop further insights in exactly that way, there has already been an enriching of each discipline. I think the gains are not only for psychoanalysis but rather that it is a twoway traffic, with validation of clinical hypotheses being reflected back from infant research and clinical work, and further psychoanalytic hypotheses being tested in those fields. Lynne Murray (1989), a researcher with a particular interest in the effect of postnatal depression wrote: I am convinced that developmental psychology is the poorer if it fails to take into account what cannot be directly seen for example the contents of fantasy and that it can be vastly enriched by the generation of theory about the dynamics of interpersonal and emotional processes that arises from the analytic experience. (pp 333-4).

What I shall discuss is how I think that infant research, while it does _refine _psychoanalytic theories of infancy, nevertheless supports much of it, in particular the concepts of containment, and projective identification. I shall then discuss more fully how new infancy thinking about the affects of pride and shame may impact on analytic thinking and also increase our understanding of the therapeutic efficacy of relational factors alongside verbal interpretation. I shall touch on the two-way clinical traffic those working with adults and older children find that familiarity with the infant field can contribute to a deepening awareness of the infantile transference, and in turn infancy workers find psychoanalytic concepts very helpful in understanding and working with troubled families. And lastly I shall make the point that the stress laid by the researchers on how much the infant tries to make meaning of things counterpoints the attempt in psychoanalysis to make things meaningful, particularly the irrational. This is in contrast to the current trend in much mental health work, with its economic pressures, to focus on the present and both minimise the significance of what happens in childhood and of peoples wish to make sense of their lives in a relationship with another person rather than only being given medication or behavioural strategies.

Although I shall not talk so much about the neurobiological research, I think that the next major developments will come from the underwriting of psychoanalysis with the neurobiological research. Kandel (1999), a Nobel prize winner for his work in neuroscience, wrote that psychoanalysis still represents the most coherent and intellectually satisfying view of the mind (p. 505). His hope is that psychoanalysis will join with cognitive neuroscience to develop a new and compelling perspective on the mind. The joint contributions of infant research and neurobiology combined with the evidence of outcome studies (Fonagy, 2000) could herald the re-entry of psychoanalysis as an important influence in the public mental health sector.

The interface with infant research


Attachment theory about the need for infants to feel safe particularly when threatened with separation is very easily operationalised. It has provided a range of findings, including the capacity to predict what the attachment status of a foetus will be when that child is one year old. In other experimental work (Moore, 1994), 6 week old infants, who were developing what would later be assessed when they were a year old as an insecure-disorganised attachment, could detect a look on their mothers face that lasted of a second and respond to it. You and I would not be able to see it unaided unless the videotape was paused. Then it could then be seen that the mothers were looking at their infants in a frightening or fearful way as if the infant reminded them of something traumatic from the past, that is, they were looking with hatred or fear. In response the infants would freeze, look dazed and away, and then look back out of the corner of their eyes.

Young children with insecure-disorganised attachments react to their mothers return after a separation with disorganised and sometimes self-harming behaviour. They may go towards her with their head averted or spin round and bang their head on the wall or sink frozen to the ground. Disorganised attachments are most likely to occur when there is disorienting parental behaviour or abuse. Children with these attachments have a greater rise in cortisol and prolonged cortisol elevations during the strange-situation test, when the mother leaves them for a short time with a stranger present. When children with disorganised attachments were followed for 17 years, they also showed the greatest vulnerability to psychiatric difficulties (Lott, 1998). Fonagy (1995,1998) suggests that an abused or traumatised child may have had the kind of care in which anxiety and pain inhibit the development of the capacity to think reflectively about her self and significant others. The child cannot think about herself in the carers mind because the carer intended her harm. There is decoupling of the mentalising function, with impulsivity and an inhibited capacity for mature relationships, which could contribute to borderline pathology, with the self harming that can already be seen in infancy.

In the last decade or so there has been a dramatic shift in conceptualising that what the infant experiences helps to structure the developing brain. Perry (1995) viewing the brain as a processing template through which all subsequent experience is filtered argues that traumatic events in childhood change the biology of the brain in a particular way. He suggests that when a child remains in a state of hyperarousal or dissociation in response to traumatic events in the early years then these become structured as traits. Schore, the neuroanatomist and psychoanalyst who is probably the best known writer currently suggesting that many analytic concepts about the infant are supported by neurobiological research (1994), elaborates. By regulating affect, the caregiver is also regulating the release of neurohormones in the infants brain. In the inevitable event of distress states in the infant, the caregivers moving in to repair the connection and comfort the infant reduces the levels of cortisol and related stress hormonesWhen there is no interactive repair, when the caregiver is abusive, neglectful, or continually misattuned, infants may remain in chronically negative states, their corticosteroid levels chronically elevated. (Cited in Lott, 1998) This is thought likely to reduce the number of synapses, even the death of neurons, and to undermine the capacity of the brain to regulate emotion.

What if a primary feeling which the baby experiences is one of depression? The neurobiological research suggests that as infants mirror their mothers depressed feelings in their own facial expressions, there are corresponding measurable changes in brain activity. It has been found (Nelson & Bosquet, 2000) that infants as young as 3 months of age, or earlier, who have depressed mothers were more likely to demonstrate the same kind of EEG patterns as those found in adults with depression. Those infants exposed to the most severe depressive symptoms in their mothers exhibited the most extreme negative asymmetry EEG scores. These measures may be crude, and while the possibility that this reflects an endogenous trait cannot be dismissed several researchers have argued, from their reviews of other empirical evidence, that this comes about as a result of repeated exposure to a depressed mother.

So when things go well we have the Trevarthen baby, when the things do not go well over a prolonged time we see something closer to what Melanie Klein hypothesised as the paranoid-schizoid position. Transient moments of being paranoid-schizoid would then become more fixed when there are difficulties between infant and caregiver, and the infant, who feels persecuted, cuts off. It is important to have the Trevarthen baby alongside the clinical baby. Otherwise we may take the reconstructed baby as the norm when what we are seeing is a clinical population. We may then overstate the degree to which all babies need to use desperate projective identification to get rid of what they feel is bad.

**2. Changing views of affects, particularly of pride and shame **

We have come a long way from Freuds affect theory to viewing emotions as joining and integrating minds. Trevarthen believes the newborn seeks to get involved with another person in order to share an experience with them. The infant is looking for enthusiastic companionship in which you enjoy her (Trevarthen, personal communication). He wrote (2001), Being "meaningful" to someone important is what a young child strives for from the first protoconversation (p. 118). The so-called "complex" emotions, the interpersonal sense of "pride" in admired accomplishment, and "shame" in being misunderstood, or disliked, are part of the innate human moral condition. (p.95). He sees pride and shame as primary organisers.

It seems to me that the developmental evidence could support a view that feelings of love, consciousness of self and pride and shame could all arise in the earliest weeks. If when an infant is 6 weeks old, her mother interrupts a dialogue with the infant to keep a still face for, say, 35 seconds, the infant is disconcerted. When the mother can no longer keep it up because of the effect it is having on her baby, the baby bursts into tears as though the mother mattered to her and she was hurt by what the mother did. Reddy, a theorist on joint attention, suggests (personal communication) that you can see selfconsciousness in infants from 2 months onwards, in coy smiles with a curved arm. She understands this as the infants awareness of the self as different from the other, and that momentarily in the interaction the infant has felt overwhelmed. If the baby can be selfconscious from 2 months onwards because she is conscious of her self, and she wants to be intensely related to important others, it is not difficult to see that she might be aware of not having got it right in an interaction.

We have known for three decades that infants respond to the Still Face situation, with responses of sadness that the researchers see as like depression. But have we been relatively blind to how early shame is there? What I now hear is that some researchers see the infants response as one of shame as well as depression, as you would with a depressed adult. Experimental studies with two-month-olds showed that unresponsive or noncontingent behaviour from the mother precipitated well-organised negative emotional reactions, indicative of frustration, depression, or shame (Murray & Trevarthen, 1985; Tronick, Als, Adamson, Wise & Brazelton, 1978) (Trevarthen 2001, p101). Papousek and Papousek (1975) reported the response of two months old infants to being suddenly left a number of times by their mothers. They illustrate this with photographs of one infant showing first his cheerful reunion with his mother, then his behaviour during a short separation when he looks rather dejected and finally he rejects his mother after repeated separations.

![][1] ![][2]


Trevarthen (1998) published similar photographs taken from video sequence, of a 6 month old infant, Emma who is first shown happily joining in clap-a-handies with her mother. Then when Emma is with a stranger who does not understand her, she suddenly, bravely, tries to show the game to him but when he gives a little sarcastic laugh she stares at the ground. Trevarthen thought that she demonstrated the classic lowering of the head and eyes in shame as reported by Tomkins in 1963, and that when infants are not enjoyed in companionship, they feel incompetent and there is automatic humiliation, and an instant sense of self-debasement.

So perhaps we need to think more widely about all the ways the mother-infant communication can be disturbed, not just when the mother is depressed. Papousek and Papousek (1975) listed a number of reasons why the mother-child interaction could become erratic and in this way incomprehensible for the child: Adverse environmental influences, exhaustion, ambivalence, psychic disorders or physical illnesses on the mothers part, irritation or illness on the part of the infant, even lack of confidence or inexperience on the mothers part. Trevarthen (2001) writes evocatively that the infant hero can suffer shame if submitted to the dull gaze and tuneless voice of indifference, even if kept warm and well fed. (p. 119). Siegel (2001) writes that an adults pride may at times inhibit repair and leave the child isolated in what may be a shameful state of disconnection (p. 79). I think that this has implications for those infants who feel unheld much of the time.

In Infant Observations, we sometimes from very early on hear the mother using shaming comments such as yucky or ratbag, which shape the infants developing sense of self. In one infant observation which I heard about of a three and a half month old infant, he seemed to feel that he often had to wait while his mother attended to others, that is that he could not always entrain her easily. When told by his mother to say hello to the observer he looked furtively, smiled in an almost embarrassed way as if caught out and needing to be polite, and looked away. He repeated this three times.

That shame can be devastating for many child and adult analysands is because, I suggest, shame is linked from the beginning of life with our very being. When we feel competent in our being and our interactions from birth onwards, pride helps structure our very being. When we feel incompetent in our being and our interactions from birth onwards, shame helps structure this. Pride and shame help in the creation of those contents of the internal space that become mothers good girl or bad boy, which contribute to the ideal self and superego. Where the representation is of a self which is disliked or incompetent, that is, of a shameful self, the discrepancy with the ideal self-representation is likely to trigger further shame and rage. It is also thought that infants may be more devastated by difficulties in the mother-infant interaction when they are 2 months old than when they are 6 months old, because they feel more sense of entrapment earlier on.

And where a mother has a severely shaming superego, her unconscious hope is that the infant will restore her internal damage and banish the internal shamer. Where the infant cannot do this, particularly if she evokes those shameful conflicts which the mother has struggled with, the infant is criticised in order to shame her. An internal object then lodges in the infants internal world which is devastating as it is the mothers and cannot be processed by the infant. I think we have sometimes neglected this intergenerational transmission of shame in analysis and the infant research may alert us to this.

Thinking of pride and shame as primary organisers often immediately makes sense to clinicians. If our model as clinicians is modified, we hear things differently and therefore interpret differently and this may change technique. If you do not have prominently in your model an affect such as, for example, shame, it may be harder to hear it in what the analysand says. And knowing that the researchers conclude that they can see evidence of pride and shame from, say two months onwards, would be some confirmation of Kleins view of the early origins of the superego and ideal self.

John McClean (personal communication) has suggested that where there is a failure in the earliest roots of competence, where the infant does not feel that she has a good effect on her caregiver, her experience is of a fundamental futility and despair. He thinks that this may show up in work with adults through projective identification, in the form of generating despair, which may look like destructiveness or sadism, but may be more helpfully understood in this way.

3. Widening the theory of change in clinical work

Perhaps one of the most profound debts to infant research is acknowledged by Anne Alvarez (1992), whose knowledge of the research changed her thinking about Robbie, her patient with autistic defences. In turn this changed her way of working with him so that she became more active in challenging the defences. She viewed her changed technique as a reclaiming, similar to the way in which the good enough mother would attempt to pull a withdrawn child into relating. Alvarez in turn changed the thinking of her own supervisor, Frances Tustin, and that of many supervisors and supervisees at the Tavistock Clinic.

A Study Group in Boston of analysts and developmentalists which includes Daniel Stern has for several years studied analytic material and infant research in order to understand how therapeutic change comes about is anything else needed in addition to interpretation? They think that therapeutic change comes about as a result of the moment of meeting, that moment of authentic person-to-person connection with another. They conceptualise that what happens between an analyst and analysand is similar to what happens between mothers and infants.

Tronick (1998), a member of the Group, describes mother-infant interaction in this way. The caregiver provides the infant with regulatory input, (a) scaffolding (which) is emotionalwhich can expand the complexity and coherence of the infants state of brain organisation. Creation of this dyadic system necessitates that the infant and mother apprehend elements of the others state of consciousness Thus, a principle governing the human dyadic system is that successful mutual regulation of social interactions requires a mutual mapping of (some of) the elements of each partners state of consciousness into the other partners brain At this moment of forming a dyadic state of consciousness,there must be something akin to a powerful experience of fulfilment when a dyadic state of consciousness is achieved there is a restructuring and change of the infants present and past mental organisation. (p. 295).

I shall give an example from work with infants. Beatrice Beebe (cited in Davis,1999), was videotaped sitting with depressed babies. She said that she felt that, in order not to fail the baby, she had to sit and do nothing, just to be with the baby and not encourage her to be something other than what she was, a baby in despair. As she attuned to the babies despair they could be seen on video perking up within minutes, as presumably they felt met and could become hopeful again.

In therapeutic work Beebe said (1998), it is the moment of meeting, of meaning, that carries the therapeutic action, the power to change each persons mental organisation, at a procedural level. The Boston Study Group considers that there can be changes in unconscious procedural knowledge, such as those that occur during a moment of meaning, that is, that there can be changes in the analysands unconscious internal representations which need not be directly related to conscious insight, which lead to progress. I think that we can easily see these changes taking place in work with infants and with children. I think it may not be so easy to be convinced that moments of meeting on their own would often bring about change in an analysis of an adult, but certainly they might enable some interpretations to be heard which would not otherwise be heard. If in our conceptual model we also privilege interpreting above all else, the Boston view may not fit into our schema but at least we should be aware of it.

4. Deepening awareness of and receptivity to the infantile transference

A number of psychoanalysts feel that with their personal experience of the interplay of psychoanalysis and knowledge about infants comes a deepening awareness of the infantile transference. They are more aware at times when there is a regression in analysis of a quality in the room that is like being with a very young infant or a raging or oedipal toddler. They then feel that when they interpret out of a greater receptivity to this, there is more conviction for the analyst and for the analysand.

Ruth Safier (personal communication), a psychoanalyst working with infants, describes how she feels able to interpret more helpfully with some patients, for example, those with borderline personality disorder who might otherwise have felt wounded or offended by what she said. When the mother and baby are both in the room, it often seems easier to address the mother at the right time and the baby at the right time without offending or wounding either. Then sometimes with her adult patient Ruth Safier thinks, Who would I be talking to if there were two people here? For myself, from my work with infants, I would talk less about the baby part to adults in analysis as I feel that may come across a bit cliched on my part. But I may be more likely to say that I have a sense of a baby or that the baby in the room seems to feel such and such.

The right hemispheric processing of affective information at unconscious levels is implicated in the communication and reception of emotional states (Wexler et al, 1992). Transmission of emotional information is very rapid. The implicit appraisal of facially expressed emotional cues can take place with 2 milliseconds (Niedenthal, 1990) which is outside conscious awareness. This provides a way of understanding how the concept of the therapists projective identification works in practice, that is, how the therapist can be in touch with emotional states in the patient of which the patient is hardly aware. Then when the patient feels safe enough to explore regressed states in therapy in order to master them or when there is a transference-countertransference rupture more deeply unconscious psychobiological states are expressed in the transference (Schore, 2001, in press). These may include affective re-experiencing of the infant having been aware of the caregivers emotional state as disorienting, depressed, shaming or abusive. An interactive brain perspective thus emphasises environmental factors in psychopathology and we are now in a better position to appreciate how this will mesh with any constitutional or genetic vulnerability which the infant may have, and affect the developing structure of the brain accordingly. Conversely, when the patient can form a close trusting relationship with the therapist, there is considerable possibility of change. Nowhere is the possibility of a change from insecure to secure attachment status more evident than in sessions with mothers who are experiencing difficulties in the early stages of a relationship with a young infant.

5. Psychoanalytic thinking helps most in infancy work

Now I want to move to work with infants. Infancy workers who work with very troubled infants and their parents tell us that some key psychoanalytic concepts such as containment, projective identification and the unavoidability of hate in relationships provide the most containment for the workers. Only psychoanalytic thinking with its understanding of the unconscious and the meaning behind apparently irrational thinking can help those working with troubled families understand the meaning of a parents ambivalence, why a wanted child can also be a hated child and how these ghosts contribute to a mothers depression.

In the following example (in Thomson-Salo et al, 1999), the clinicians thinking about transitional objects is what facilitates a way of thinking about 8-month old Colin, who had been re-admitted with severe failure to thrive. His head lag was like that of a neonate. The medical staff attributed the failure to thrive to organic factors of unknown aetiology. The mother was keen to wean but Colin would not allow this. He had had three previous admissions to an Early Parenting Centre and to the Royal Childrens Hospital which had not helped with the difficulties. Colin, while not looking at his mothers worried eyes, was keen to engage with her, reaching out a little anxiously to her, touching her, smiling and vocalising a little. She waved a small toy in front of his face to try and distract him so she could turn away and it seemed that all she wanted was to get Colin off her breast but did not allow him appropriate play opportunities. She did not understand why he could not bear to be on his own and she had no clear idea what to do. The clinician, Brigid Jordan, observed Colin sucking a torn vinyl strip on his pusher and therefore found him a toy which could be looped onto his pusher with a plastic strap. Her thinking, which she explained to the parents and the nurses, was that before he could give up the breast he needed to create symbolic links to the breast and his mother as the breastfeeding was his only way of relating to his mother. There was immediate improvement in Colins capacity to play and in the mother-infant relationship, and he rapidly made up much of the physical delay.

In the following example, the clinicians being able to think about, in a containing way, what was communicated of the parents hate of their child was the therapeutic factor. Belinda Keatinge (in Andrees & Keatinge, 1998) describes how Joe was regularly putting 6-mths-old Tony on the open window ledge of their 2nd floor unit, with his legs dangling out, holding him only by the elastic of his pants. The therapist tried all sorts of waysbut Joe said, "Tony is all right." Together with the team, she was able to think about what Joe might be trying to communicate. On the next visit, she was able to put into words, how he must have felt, he was being held over the ledge, in his life. Only then, and seemingly unconsciously, did he take the baby off the ledge, and began to talk about often he felt like throwing himself out of the window, and how he had not been kept safe as a child.

Psychoanalytic infant observation can facilitate the development of thinking analytically about the experience of the baby and of her inner world. This is all the more striking when maternal and child health nurses with 20 years of experience report that their practice has changed sometimes within as little as 3 weeks of starting infant observation.

To be this helpful, to have such an impact, is a sort of a test of psychoanalysis.

6.A shared search for meaning

Researchers have thought for a long time that infants were processing information from birth, or before birth. There is considerable stress in the current infant research literature on the newborn as seeking to be "meaningful" to someone (Trevarthen, 2001). Being meaningful means to matter and for there to be meaning. Such a stress in infant research on the meaning the infant endows things with from birth confirms psychoanalysis search to make meaning of the irrational.

I think that if we accept the infants wish to be meaningful to someone, and out of this matrix all experiences are construed and responded to, this would rebut the current trend in mental health work to minimise the significance of the childhood experience of adult patients and to treat adult disorders only as a biological dysfunction. I think that when people come to view themselves as having faulty thoughts which have to be blocked this may lead to a feeling that there is something defective and shameful about them that they produce such faulty thoughts, rather than understanding that there may be some meaning to these thoughts. And when people want to change things about themselves they often want to do it in a relationship with someone; they need the effect of a relationship on their external and internal relationships. Medication and the other cognitive approaches alone cannot change these permanently in the way that psychoanalysis and psychoanalytic therapy can.

One of the criticisms of psychoanalysis is that it has not proven its cost effectiveness and how it would currently contribute to the public mental health sector. If the bridges between it and neuroscience were to be joined, the outcome studies concerning psychoanalytic treatment, against hearing which there is resistance, would have to be heard. Then the insights which psychoanalysis offers could have a more widely acknowledged part to play in the public sector. This could be in treatment planning particularly in preventive work and in providing professional development for workers, and perhaps even funding for selected research psychoanalytic treatments.

There has recently been some mention in the literature of presentations by patients that are not well understood yet, by Danielle Quinodoz (2001) of what she calls the heterogenous patient, who functions with a number of splits, also by Gail Reed (2001) of more clinically problematic forms of shame-contempt interchange occurring in what she calls widening scope patients. She thinks that they tend to have had less than optimal interactions with primary objects. If there are new presentations of people who want help, then psychoanalysis could be recognised as the treatment of choice to elucidate the aetiology, which could then provide hypotheses for best helping them.

Concluding discussion

It promises to be a very interesting time: neurological research into infant brain development and parent-infant interaction is in unchartered territory. Neurologists think that they are just beginning to discover the details about a possible neurological origin of conscience (Siegel, 2001, p.84). Schore (personal communication) pointed out that while there is no evidence on the neurobiological effects of the way in which fathers provide a containing function for their infants there is now dense array EEG technology that can localize deep limbic structures to study this. He would expect to see both quantitative and qualitative differences in the effects of adult female and male brains on those of the infant, with an initial heavier influence in the first years from the mother, and then there will be the question of these effects on different infant gender. We have not been able to look at this in psychoanalysis in a systematic way and it would be interesting to have this data.

Neuroscience can offer an explanation of how emotionally attuned communication helps the brain develop and how it is the psychotherapy has been found in the multi-site NIMH trials in USA (DiMascio, Weissman, Prusoff et al 1979) to be effective in the first few weeks in a different way from antidepressant medication. Neuro-imaging techniques are being used to ascertain brain function before and after psychotherapy (Lonie, 1999). Neuroscience will give us the scientific explanation of how processes already identified clinically actually take place. It may even generate new hypotheses for psychoanalysis to consider. But I think that it will still take the clinicians sensitivity and fine-grained analysis of what works in the consulting room to put words to personal experience and reach new ways of helping people. We still need the clinical insights of psychoanalysis to make sense of how, for example, a person who has been abused may unconsciously enact or help the internal abuser time and again in order to ensnare the analyst and defeat the analysands attempts to get relief.

If in the public sector psychoanalysis and neuroscience could be seen to join, then the insights of psychoanalysis could be used particularly in early interventions, for example, with troubled infants and their families, and where there had been early trauma. The voice of psychoanalysis could also be heard in speaking up for those who need help over a longer time period.

If we return to the interface between the research about infants and mother-infant interactions, we can say the following.

  • Infancy research confirms and refines rather than refutes psychoanalytic theory.
  • Infancy research is helpful in the work with analysands (recognising infantile transferences and conceptualising how changes comes about)
  • Psychoanalytic theory is helpful to workers seeing infants and mothers troubled by ghosts in the nursery
  • There is continual cross-fertilisation at the interface. If you see something differently there is a transitional space in which new thoughts can be thought.

If we can say all that, then these can be recognised as ways of testing psychoanalysis. They may take us confidently into the twenty-first century.