Introduction
Developmental Psychopathology has defined early anorexia as a syndrome characterized by poor feeding and associated failure to thrive or growth stunting where there are no recognizable organic causes. Specifically, 0-3 Diagnostic Classification describes the feeding disorder, whose onset may occur at different stages of infancy, as a child’s difficulty to establish regular patterns as regards feeding with adequate food intake in relation to the states of hunger and satiety (Zero to Three - National Center for Clinical Infant Programs, 1994). This definition allows us to include infantile anorexia within its frame.
The most recent theoretical and clinical contributions to early anorexia research suggest a “multi-factorial etiological model”, according to which many different elements may concur: regulation difficulties, child temperament (difficult state regulation, poor appetite, negativism, stubbornness, dependency), maternal psychopathology (anxiety, depression, feeding disorders), caregiver’s insecure attachment, psychosocial stressors (Benoit, 2000; Chatoor, 1989; 1996; Chatoor, et al., 2000; Kreisler, 1985; Woolston, 1991). This model attempts to identify a plurality of causes which may lead to negative and conflictual parent-child interaction during feeding which, if persistent, may lead to infantile anorexia.
Based upon the few studies available, there is no clear data on the prevalence of infantile anorexia. Generally, feeding disorders in the first three years of life are present in almost 25% of children with typical psycho-physical development (Benoit, 2000; Chatoor, 1996; Lindberg, et al., 1996). Pediatricians show a 5-10% prevalence in the first 15 months of life as regards serious symptoms, i.e. persistent rejection of food and consequent hospitalization and failure to thrive (Lindberg et al., 1996); specifically non-organic failure to thrive has a 50-58% incidence in total cases of failure to thrive (American Psychiatric Association, 1994). Rates change when referring to parental reports, which show a prevalence of 6% in 6-15 months old children and 25-40% in older children, including over-eaters, poor eaters and children with peculiar feeding patterns such as pica and rumination (Benoit, 2000; Chatoor, 1996; Marchi, Cohen, 1990). Data show the lack of specificity as regards clinical descriptions and sources of information; thus, data are confused and contradictory depending both on the varying involvement of people reporting the data (parents, specialists) and the lack of differentiation between temporary disorders, which might be amplified by worried parents, and the defined pathology.
There are differing theories concerning infantile anorexia and they have changed over the course of time. Traditionally, feeding dynamics have been examined according to the analytical “drive model”. Such a perspective links personality and behaviors, including feeding behaviors, to the intensity of drives and the environmental response to them (Abraham, 1916; Freud, 1905; Klein, 1957). Feeding behaviors would be organized by oral drives during infancy, whether in the freudian psycho-biological perspective or in Melanie Klein’s, according to which drives are transformed into unconscious fantasies (Greenberg, Mitchell, 1983). Possible conflicts at the level of orality may lead to fixations and regressions due to specific patterns of response to the drive in which avidity, repeated lack of satisfaction, resentment, envy and dependency have a role in the development of neurosis in children, having as a nutritional effect an abnormal sense of hunger, inhibition of eating or rejection of biting, chewing or swallowing food (Klein, 1957).
The “drive model” has been replaced by the “objet relations model” according to which personality and behaviors are shaped by the Ego relationship with the objects of his environment (Winnicott, 1958; Mahler et al., 1975; Kernberg, 1976). The relationship to the object no longer depends exclusively on libidinal needs but mainly on the real world of adults and real daily interactions (like meals) that have an important impact on early child development. The shift from the drive model to the relational has given space to new theoretical approaches to clinical and developmental issues (Winnicott, 1958; Mahler et al., 1975; Bowlby, 1969-1982), that have also taken into account the early mother-child relationship during feeding.
Starting from the sixties, mainly due to Wolff’s (1966), Sander’s (1964) and Stern’s (197l) work, the number of studies on mother-infant interactions have increased, studies based upon direct and systematic observations (rather than based upon retrospective reconstructions of adult patients), giving birth to the so-called Infant Research that has sharply changed our image of infants (Ammaniti, 2001). The core feature of such a perspective is that, beyond the desire to satisfy the libidinal needs, the child has establishing and maintaining relationships as its main innate tendency (Field, 1994; Sameroff, 1993; Stern, 1985; Trevarthen, 1990). This new knowledge of child development has led researchers to study the infant’s psychological disorders in relation to various developmental stages: Developmental Psychopathology (Cicchetti, 1984), which attempts to describe and classify disorders, to find the causes and outcomes in relation to the biological and relational stages of the infant (Cicchetti, 1984; Sameroff, Emde, 1989; Sroufe, 2000).
Progress in developmental psychology and psychopathology, thus have emphasized the role of affects and care-giving behaviors in the child organization of personality and in the possible transmission of psychopathological risk from parent to child (Ainsworth et al., 1978; Main, 1995). Within this approach, attachment theory has given basic highlights for research and clinical work. The attachment relationship between mother and child has been considered a core experience, as the child builds mental representations of the self and the other, based upon his/her repeated experiences of relations.
Ainsworth and colleagues (Ainsworth et al., 1978) have identified different attachment patterns through the standardized Strange Situation Procedure (the children are stressed by a medium strange condition, that is an unfamiliar room and a brief separation from the mother). Three main attachment patterns emerged: secure (B), avoidant (A) and ambivalent (C) which have been linked to two main care-giving systems: sensitive and insensitive (Cicchetti, Cohen, 1995).
A sensitive caregiver is able to respond to his/her child’s needs and to emotionally communicate with the child. This has led to defining “maternal responsivity” as the emotional availabilityfor understanding and answering to the emotional needs and cognitive and behavioral signals of the infant. Overall, a “sensitive” parent is able to establish a good relationship during feeding (nurturant), pays attention to all the infant’s needs (attentive) and is not controlling (Zeanah, Zeanah, 1989), bearing a certain degree of crisis and interactional disconnection that may be renegotiated and repaired with a renewed sense of trust, efficacy and competence within the dyad (Beebe, Lachmann, 1994).
On the other hand, where there is insensitive care-giving affect and emotions have no shared space between child and parent and the caregiver is not sensitive to the child’s states of mind and intentions. Relational exchanges are incoherent and there is no mutuality, preventing the child from understanding his/her own growing emotional states (Fonagy et al., 1992; Morton, Browne, 1998). This type of interaction features uncoordinated and negative relational patterns between the dyad: messages sent are not read, understood or decoded by the parent nor by the child within the relationship. An insensitive and unpredictable care-giving environment may be disorganizing to the child’s state of mind who may then develop inadequate relational strategies and insecure attachment working models of the caregiver.
Furthermore, research has shown how a child, with a sensitive parent who is able to readily and adequately respond to his/her child’s needs and requests, develops a secure attachment pattern (B), i.e. he/she builds up a primary behavioral strategy which allows him/her to look for protection and care and to establish closeness and contact when in need (Ainsworth et al., 1978; Belsky et al., 1984; Egeland, Farber, 1984; Grossman et al., 1985). At the same time, an internal working model is built, in which the parent is represented as available and loving and the infant represented as able to ask for help and worthy of care and love. The child trusts that the mother will be available and helpful in dangerous and frightening situations.
An insensitive caregiver, on the other hand, has a rejecting attitude or is unpredictable and incoherent in responding to the child’s needs, who then develops an insecure attachment pattern (avoidant/A e resistant/C). In particular, children who do not express any distress during separation, do not look for the caregiver and turn away from the parent are defined insecure-avoidant, while children who combine searching for closeness to opposition to contact are defined as insecure-resistant. Both insecure patterns (insecure-avoidant or insecure-resistant) represent 20-25% and 10-15%, respectively, of the normal population (van Ijzendoorn, Kroonenberg, 1988) and are often associated with the appearance of behavioral problems, poor impulse control, poor self-esteem, poor emotional regulation and difficult peer relationships (Sroufe, 2000; Zimmermann, Grossmann, 1994). While insecure children have organized defensive strategies in the face of not-optimal caregiving conditions, these nevertheless have to be considered adaptive strategies functional to the maintenance of the relationship with the caregiver.
More recently, a group of children have shown incoherent and contradictory attachment behaviors with respect to their seriously insensitive mothers (Main, Solomon, 1990). These children develop atypical attachment patterns (disorganized/disoriented-D, avoidant/resistant-A/C, unstable/avoidant-U/A) (Crittenden, 1985; Lyons-Ruth et al., 1987; Main, Solomon, 1990) and show contradictory, not directed and incomplete behaviors because they are not able to organize and plan coherent behaviors in order to look for protection from the parent due to a collapse in adaptive strategies. These atypical attachment patterns have shown to be mostly associated with maternal psychopathology, unresolved losses/trauma in the caregiver, serious family stressors in abusive and violent conditions (Lyons-Ruth, 1996).
The Internal Working Models, representations of the self and the parents, result from experience with the caregiver that generates a specific attachment pattern, creating a complex organization of perceptions, thoughts, feelings and behaviors that effect the quality and pattern of relationships in the course of life. Main and her research team have created the Adult Attachment Interview (AAI) – directed to assess internal working models in the adult, acknowledging the quality of experiences and relationships during childhood (George et al., 1984; Main, Goldwyn, 1991). Different narrative styles in the discussion of past relational experiences have led to define four attachment patterns with respect to specific states of mind and attachment experiences in childhood (Main, Goldwyn, 1991). Free-autonomus (F) adults explore memories, thoughts and feelings coherently and clearly; Dismissing (Ds) adults tend not to acknowledge or to devalue the importance and effects of past attachment experiences; Preoccupied-enmeshed (E) subjects are preoccupied and involved with respect to their relationships in childhood and are not able to process past experiences; unresolved-disorganized (U) adults have a breakdown in their reasoning processes as regards the narration of traumatic events and/or losses, as the loss of an attachment figure and sexual and/or physical abuse episodes. Lastly, the AAI may identify a cannot classify (CC) category (Hesse, 1996), characterized by contradictory representations with inconsistent features belonging to the patterns described above.
Studies have shown significant association between the children’s attachment classification as assessed by the Strange Situation and the adults’ attachment classification as assessed by the AAI; in that there is a close parallel between the organization of the children’s attachment behavior and the adults’ affective and cognitive models (for a review see: van IJzendoorn, 1995; Hesse, 1999).
In this perspective, association between the care-giving system and the protective and risk factors with respect to the infants’ psychopathology has been tested in the last decade. Overall, secure attachment acts as a protective factor in the child’s development (buffering effect), whereas an insecure attachment represents a risk factor and a vulnerability for the development of personality (Belsky, Pasco Fearon, 2002). Moreover, a non-optimal care-giving system may amplify psychopathological risk factors coming from different sources both within the child (i.e., a difficult temperament) and in the environment (i.e., parental psychopathology, poverty, psychosocial stressors, conflictual and unstable marital relationship) (Sroufe, 2000).
Within this framework, some authors have tested the specific association between the child’s and the parents’ attachment patterns and an early anorexia. Several studies have shown that both mothers and children were classified as “insecure”: avoidant/A, resistant/C, or disorganized/D as regards children; dismissing/Ds, preoccupied-enmeshed/E as regards mothers (Benoit et al., 1989; Chatoor et al., 2000; Ward et al., 1993). Furthermore, within a context of early anorexia, insecure attachment may enhance problems and offer the basis for a chronic malnutritional condition (Chatoor et al., 2000; Ward et al., 1993). Specifically, children with a moderate or serious non-organic failure to thrive syndrome are more likely to show atypical attachment patterns (disorganized-disoriented/D, avoidant-resistant/A-C) associated with several psycho-social stressors (poverty, maternal psychiatric pathology, maltreatment and abuse, unstable and conflictual contexts) (Chatoor et al., 2000; Crittenden, 1987; Lindberg et al., 1996; Valenzuela, 1990).
Recently, Ward et al. (2000) have studied a sample of children with failure to thrive belonging to different social levels. Observing such children through the Strange Situation Procedure, a higher rate of anxious and disorganized attachment patterns emerged as compared to the control group. Furthermore, the mothers assessed through the Adult Attachment Interview often showed an unresolved loss and/or traumatic event in their past (unresolved-disorganized /U).
At the present time, empirical data show both mothers and anorexic children to be more often classified within an insecure attachment pattern as compared to control groups (Benoit et al., 1989; Chatoor et al., 2000; Crittenden, 1987; Lindberg et al., 1996; Ward et al., 2000), with rates ranging from 49% to 92% (Benoit et al., 1989; Chatoor et al., 2000; Crittenden, 1987; Lindberg et al., 1996; Ward et al., 2000); and not only in high psychosocial risk samples (extreme poverty, separated families, conditions of maltreatment and abuse), but also in low risk and multisocial level samples (Chatoor et al., 2000; Lindberg et al., 1996; Ward et al., 2000).
Such results point out the importance of exploring attachment models both in children and their parents, since they are good predictors of disfunctional relationships which may lead to infantile anorexia.
Based upon these research guidelines, we have begun clinical collaboration between our research team and the Clinical Nutrition Service of a Pediatric Hospital. This collaboration has enabled us to introduce an assessment procedure administered by a team of pediatricians and psychologists within the context of infantile anorexia. In particular, in addition to medical examinations and nutritional state evaluation, a clinical-psychological assessment is administered, in order to explore the emotional functioning of the child, and the mother’s psychological profile. Moreover, interactional patterns during feeding and attachment models both in the child and the parent are assessed. Research data and clinical observation presented here belong to this experience and team work.

Aims
This study has the following aims:
1) To point out a higher rate of insecure and/or disorganized attachment patterns in the mother-child dyads with a diagnosis of infantile anorexia, as compared to a control group;
2) To show a concordance with respect to security/insecurity between the mother’s and the child’s attachment quality in a group of children with a diagnosis of infantile anorexia, as compared to a control group.

Methodology
Sample
The sample is composed of 45 mother-child dyads with a diagnosis of infantile anorexia (clinical group) and 27 dyads with no physical nor psychiatric disorder in the child and typical development (control group), matched by age and gender.
The control group (40.7% males) attends playschool, the clinical group (22.2% males) in the Pediatric Department. Only children with a diagnosis of anorexia have been selected, excluding children with any organic problems which might explain their difficulties in establishing regular feeding rhythms and food intake, that is, identifying a non-organic failure to thrive of medium to low degree in the total sample. Diagnosis was carried out following the 0-3 Classification System (Zero to Three - National Center for Clinical Infant Programs, 1994) and the DSM-IV (American Psychiatric Association, 1994) criteria.
In both groups, the children’s ages range from 12 to 36 months (mean=18.6; s.d.=7.6), mothers’ ages range from 33 to 40 years (mean=35.7; s.d.=2.0). Gestational age and psychomotor development of all children were in the typical range. Most of the children had been breast fed (control group=65%; clinical group=74%). Most families belonged to a medium or medium-high SES level.

Instruments and procedure
Mother’s attachment quality was assessed by means of the Adult Attachment Interview-AAI; in a total of 18, only 11 mothers were interviewed through the AAI, the other 7 mothers dropped out of the evaluation procedure for several reasons (refusal, marital disagreement with joining the program, family moving). Morevover, all children in the sample were assessed by means of the Strange Situation in order to identify their attachment pattern with respect to the caregiver.
Evaluation of the mothers’ patterns of attachment. The Adult Attachment Interview (Main, Goldwyn, 1991) is an approximately one-hour-long semi-structured interview, divided in 18 questions which explore the relationship with the caregiver and/or other significant figures in childhood, specific memories and, finally, how the adult perceives the effect of past experiences upon the development of his/her own personality. The narrative analysis of the interview, transcribed verbatim, is based upon a 1 to 9 score on several scales which assess both the “probable experience during childhood”, and the “current state of mind with respect to attachment”. Specifically, the “probable experience during childhood” is made of 5 subscales: 1) loving, 2) rejection, 3) involving/role reversal, 4) neglect, and 5) pressure to achieve. The current state of mind with respect to attachment is constituted by 9 subscales: 1) coherence of transcript, 2) idealization, 3) lack of memory, 4) involving anger, 5) passivity of discourse, 6) fear of loss, 7) derogation, 8) metacognitive monitoring, 9) coherence of mind. The coding system leads to the assignment of one of the five attachment classifications as expected in adulthood: secure/autonomus (F), preoccupied/enmeshed (E), dismissing (Ds), unresolved/disorganized (U) and cannot classify (CC). The first three classifications are also subdivided into sub-classifications that allow a more detailed description of the predominant internal working model.
Evaluation of the children’s patterns of attachment. The Strange Situation (Ainsworth et al., 1978) is an assessment procedure composed of 8 episodes, each lasting three minutes, during which the child is subjected to low-level stressful conditions: an unfamiliar room, introduction of a stranger and two separations from the caregiver. The coding system assigns the child to one of the four expected attachment classifications: secure (B), insecure/avoidant (A), insecure/resistant (C), disorganized/disoriented (D). The first three classifications are also subdivided in sub-classifications that allow a more detailed description of the child’s internal working model.

Results
The children belonging to the clinical group, with respect to the attachment classifications as assessed by the Strange Situation, show the following distribution: 4 children are secure (B) (22.2%), 7 children are insecure/avoidant (A) (38.9%) and 7 are disorganized (D) (38%). Children in the control group, on the other hand, have the following distribution: 15 children are secure (B) (55.6%), 5 insecure/avoidant (A) (18.5%), 2 insecure/resistant (C) (7.4%) and 5 children are disorganized (D) (18.5%). The two distributions differ significantly [c2 (3 fd)=7.53, p<.05] with a prevalence of insecure attachment patterns among the clinical children (Fig.1).
Within the 11 mothers of the clinical children assessed through the Adult Attachment Interview, the classification distribution was: 2 Secure/Autonomus (F) (18.2%), 1 Dismissing (Ds) (9.1%), 7 Preoccupied/Enmeshed (E) (63.6%), 1 Cannot Classify (CC) (9.1%). The difference with respect to the control mothers: 20 Secure/Autonomus (F) (87.0%), 2 Dismissing (Ds) (8.7%) and 1 Preoccupied/Enmeshed (E) (4.3%) is highly significant [c2 (3 fd)=18.64, p<.001] (Fig. 2).
Furthermore, in the clinical sample there is a high correlation between the child’s and the mother’s attachment quality with respect to security/insecurity: 90.9% (k=.62, p<.02). The concordance between the main classifications of patterns of attachment of the children and of the mothers, on the other hand, is very low since all the Preoccupied/Enmeshed mothers had an avoidant or disorganized child. Concordance is also low with respect to security/insecurity (52.2%, k=.07, n.s.).

Discussion
In line with previous studies, our research points out how insecure attachment patterns between child and caregiver are often associated with the onset of infantile anorexia.
Specifically, regarding the first aim, the research showed a higher rate of insecure and or disorganized attachment patterns in the mother-child dyads with a diagnosis of infantile anorexia, who differ significantly from the control dyads. Moreover, regarding the second aim, our research has pointed out a significant correlation between the caregiver’s and the child’s attachment patterns with respect to security/insecurity within the clinical group, as compared to the control dyads. This data supports other clinical studies focused on the exploration of mother-infant attachment classification within the context of infantile anorexia (Benoit et al., 1989; Benoit et al., 1997; Chatoor et al., 1998; Chatoor et al., 2000; Coolbear, Benoit, 1999; Ward et al., 2000).
Our research has also identify a higher percentage of preoccupied/enmeshed mothers (E) (63.6%), within the clinical group. A majority of preoccupied/enmeshed mothers (61%) has been also seen in a previous study conducted by Benoit and colleagues, in which the mother and her child with a diagnosis of anorexia were assessed with respect to their attachment classification (Benoit et al., 1989). Other studies have also identified such a classification among the mothers (Benoit et al. 1997; Chatoor et al., 2000; Coolbear, Benoit, 1999). All the preoccupied/enmeshed mothers in our sample have avoidant or disorganized children.
As many authors have already pointed out (Fonagy et al., 1992; Tronick, 1989), “avoidant defensive strategies” may play a central role in the mental functioning of children “forced” to precociously develop mental mechanisms that limit the interaction with the primary objects, as an effect of non-optimal object relations, such as when a preoccupied mother shows anxiety, confusion, excessive preoccupation or anger linked to parental care. Avoidant responses by the child may represent adaptive and defensive strategies aimed to maintain his/her relationship with a caregiver featuring poor emotional regulation and difficulties in bearing conflicts and uneasiness in the relationship with the child.
Clinically, it is also interesting to notice that in our sample, insecure children with disorganized aspects often show a “controlling” behavior which corresponds to a maternal passive attachment, typical of preoccupied/enmeshed adults (E1), as if the child has the need to control through his/her behavior the relationship with a poorly responsive, confused, emotionally insensitive and incoherent caregiver. The need to control the relationship has as an effect a restriction in the exploration and autonomy of the child. Such clinical observations are in line with research that has studied the relationship between mothers and their anorexic child, pointing out difficult communication exchanges in the developmental stage of the child’s emerging autonomy. Specifically, such dyads have been shown to be negatively involved during feeding in rigid and unsuccessful communication patterns in which the child is intensely oppositional and disengaged and his/her mother intrusive and controlling, unable to set limits and rules and to negotiate control over the feeding situation (Ammaniti et al., 2003; Chatoor, 1989, 1998; Kreisler, 1985; Lindberg et al., 1996; Woolston, 1991). Our results, then, may well support data from other studies according to which precocious disorganized strategies with respect to attachment may be correlated to controlling behaviors toward the parent at age six (Main, Cassidy, 1988) and to oppositional and hostile and aggressive attitudes at age seven and problems as peer relationships concerns (Jacobvitz, Hazen, 1999).
In conclusion, based upon our results, in order to understand infantile feeding disorders, it appears suitable to take into account the complex interplay among the child’s characteristics, the mother’s and their relationship’s and the developmental tasks they both have to face, considering the separation-individuation processes and the growing autonomy of the child.
Gradually, the balance between attachment to the mother and emerging autonomy should be mirrored by the parent’s ability to balance protective behaviors with “letting go” behaviors, which stimulate feeding self-regulatory abilities, autonomous initiatives and the self-reliance of the child (Ammaniti, 2001; Lieberman, Slade, 2001; Speranza, 2001). On the contrary, insecure attachment patterns between mothers and their children with infantile anorexia seem to effect not only the communicative pattern and the feeding regulation, but also give rise to opposition, negativism and disengagement of the child, which may be a setback to the development of exploration, autonomy, individuation and the ability to relate to the others.
Our contribution, while needing to be supported by testing on a more representative sample, stresses the importance of closely examining the assessment of the care-giving system and functioning in order to formulate a diagnosis and intervention strategies specific to feeding disorders and non-organic failure to thrive.

Clinical vignette

We accompany the empirical data of our work with a clinical description of a clinical case of a girl with infantile anorexia in day hospital at the Clinical Nutrition Service of a Pediatric Hospital we collaborate with.

Case description: N
N, the 13-months-old second child of Mrs. and Mr. S, is brought to Clinical Nutrition Service of a Pediatric Hospital. Her parents report that the child “vomits” during meals; the problem has started about six months before with the introduction of the first baby food. Vomiting does not occur when she is bottle-fed and thus the mother tends to bottle-feed her with milk. Independently from the food intake strategy – bottle or spoon – the child’s feeding patterns are irregular both in terms of rhythm and daily food intake. Almost all nights N awakes crying; the mother interprets the child’s behavior a signal of hunger since the child falls asleep after being bottle-fed.
Mrs. S thinks her daughter might have a disease that causes her vomiting, nevertheless she thinks she may have committed a mistake interrupting breast-feeding to begin weaning the child; she thinks the child has not accepted the change. She describes N as a “stubborn”, “overbearing” child, “if she does not want to eat there is no way to convince her”; she is “always in movement”, “she wants to touch everything”, “we need always to control what she’s doing”.
N’s father states the daughter is just like him who has always been thin, when he was a child he would also not enjoy eating, he has gained weight since he got married, but he still never eats much. He tends to stress the “normality” of his daughter: “she is joyful” and is “normal”, he points out positively that N “goes with anybody” showing it to the pediatrician when the child goes from the mother’s arms, to the pediatrician’s to the father’s and then again to the mother’s. Since the feeding problem as described by the parents (vomiting during meals) may be a symptom of a gastroesophageous reflux and the child is also under weight with respect to her age a brief hospitalization is suggested. Together with the medical controls and the nutritional state evaluation, a psychological counselling is offered to the parents.

Pediatric evaluation
Medical tests exclude a gastroesophageous reflux diagnosis; a low thrive in the child has emerged as an effect of malnutrition; the child has a regular psycho-motor development.

Psychological evaluation
Mother-child feeding interaction was observed by means of the Feeding Scale (Chatoor et al., 1997), a standardized procedure that examines during a twenty minutes video-recording the quality of the communicative exchanges through the evaluation of the mother’s and the dyad’s emotional states, the frequency and intensity of food rejection and disengagement of the child during the meal.
In the case of N, the social and communicative exchanges are very restricted; there are few positive affects and poor pleasure in the interaction; the mother shows an intrusive behavior and the child defends herself through an avoidant behavior. There is also a remarkable rejection of the food when the mother forces her to eat; here are expressions of discomfort and anger during which the child avoids the mother’s gaze, becomes stiff and arches her back while in the mother’s arms; finally, there is a poor response on the mother’s part to the child’s attempts at autonomy, such as reaching for the dish and touching the food.
Afterward, we interviewed the parents together and individually. Mrs. S is 30 years old, Mr. S 35. They have met in late adolescence, their marriage has occurred within a strained and difficult context, mainly for Mrs. S, who was in serious conflict with her own parents.
Mr. S is a skilled workman and Mrs. S has worked as a shop assistant before marriage but decided to leave work to devote herself to the family. He couple has jointly decided to have the first child, the elder daughter is 6-year-old. The second pregnancy, which produced N, was chosen at the moment when the couple felt “united” and engaged in facing an significant health problem in the elder daughter (repeated fever convulsions). Moreover, during that period, Mrs. S became certain about something she had always believed: her father is not her biological father. Mrs. S mother refuses to reveal the daughter her father’s real identity. Mrs. S since then (2 years ago) is extremely stressed by her personal history. She is still unaware of her biological father identity and in conflict with her parents. She also reports to have been institutionalized in her first 3-4 months of life, then she has lived her first 4/5 years of life with her grandmother, when her mother got married; childhood memories are confused and unclear and here stay at the Institute may eventually have been longer. Mrs. S, as a matter of fact has come to know on her own and later on that her mother was a single mother. She describes a difficult childhood featuring by rejection and neglect from both parents and by a harsh and violent discipline with strong pressures toward autonomy.
Mr. S has lost his father traumatically when he was 5 years old – the father has died falling down the stairs while he was working at home. The loss and the lack of a paternal figure seem to be a core problem in the affective life of Mr. S.
N’s parents are in a crisis due to the woman’s suspicion of an husband’s betrayal: S has expressed violent anger toward his wife which has decreased with S has threatened to leave him.
Then, we decide to administer the Strange Situation with respect to the mother-child attachment interaction and the Adult Attachment Interview to both parents to assess the quality of attachment with respect to each member.

Assessment of the attachment pattern in the child: Strange Situation Procedure
Attachment classification D-B2/A2 disorganized features
N shows avoidant behaviors at the first reunion, she seeks for mother’s proximity at the second reunion; while close to the mother she shows moderate avoidant signals as when N is in her mother’s arms and goes back to normality, to exploration. We thus observe simultaneous expression of contradictorial attachment patterns: avoidant behavior, proximity seeking and resistance to contact. The caregiver is unable to attune mutually to her child’s communication; N has still a poorly articulated speech and the mother is not engaged in understanding her daughter's talk.

Assessment of the attachment pattern in the mother: Adult Attachment Interview
Attachment classification: CC/Ds4/E2
In Mrs. S state of mind with respect to attachment, an avoidant defensive organization as emerges from a high rating in the fear of loss scale coexisting with a preoccupied, enmeshed and angry state of mind. Specifically, in Ds4 subcategory the fear of loss of the child is not linked to a realistic source; such fear has been highly correlated to the insecure/avoidant children. Ds4 adults often show in their history the loss of a close figure where the circumstances have been kept secret; in Mrs. S transcript, fear of separation from her daughters is also associated with fear of maltreatment perpetrated also by strangers. The subcategory E2 has instead been coded as a consequence of a high score on the involving anger scale with respect to her mother, a rejecting, neglectful, and critical caregiver who has also pushed her towards early autonomy and a role reversal. Mrs. S. is preoccupied, angry and overwhelmed by her past attachment experiences.

Assessment of the attachment pattern in the father: Adult Attachment Interview
Attachment classification: U/F4b
An uresolved loss, preoccupation as regards separations and losses, incoherence and confusions with respect to attachment relations emerge; in the probable experiences there is role reversal on behalf of the mother. There seems to be a genuine awareness of his past difficulties, nevertheless, the unresolved loss and the consequent vulnerable sense of the self and inconsistent autonomy do not allow to be assigned the F3 classification.

Comment
Data as collected by means of the standardized instruments are in agreement with the clinical interviews. A diagnosis of infantile anorexia was given at the end of the evaluation process based upon the 0-3 Classification (Zero to Three - National Center for Clinical Infant Programs, 1994) and DSM-IV (American Psychiatric Association, 1994) criteria. Disregulated patterns in the child’s sleep and feeding patterns are strictly linked and interplay with the disfunctional caregiving system. Maternal difficulties as concerns emotional and communicative attunement are at the onset of a problematic mother-child interaction. The controlling and intrusive behavior of the mother and her difficulties in setting limits and rules contrasts with N’s temperament: a lively child, attracted by environmental stimuli. For instance the mother gives poor response to N’s loquacity, and states: “she has her own way of talking, and I cannot understand it at all”. The past history of the mother marked by rejection and neglect seems to be re-proposed in the current relationship with the child, unable to offer a sensitive caregiving. The sub-clinical levels of a feeding disorder in the mother as shown by the administration of specific test may suggest an early onset of the difficulty to empathically recognize the daughter’s affective signals and self-regulating rhythms within the feeding context. In the current developmental stage, the hypercriticism of the mother, the strong control and the intrusiveness are remarkable, as well as the difficulty to give an emotional and protective support in the process of N’s separation and individuation.
N’s father is poorly affectively present, possibly as an effect of his emotional withdrawal due to his unresolved issues. As a consequence he cannot assume a paternal role in order to help Mrs. S in her relationship with the daughter, supporting N’s autonomy. In fact, the couple is not capable of emotionally communicating; the parents do not share their expectations and feelings about N’s rearing, thus, their daughter’s separation-individuation process is impaired.
The clinical evaluation reported here has stressed how infantile anorexia and the linked difficulties within the caregiver-child relationship emerge from a complex interplay of several variables: the individual characteristics of the child and the caregivers, the internal working models of the relationship, the quality of the family interactions and of the co-parenting system. Data and clinical evidence has allowed us to offer to N’s family an intervention program focused both on parental support and on interactional strategies within the feeding context, hat is to promote play and autonomy during meals, re-establishing competence and trust in the caregiver, involving both parents.
Moreover, a psychological intervention has also been suggested to support the parents both individually and jointly as a couple.

References
Abraham K. (1916). Ricerche sul primissimo stadio pregenitale della libido. In Opere, vol. 1, Boringhieri, Torino 1975.
Ainsworth M.D.S., Blehar M., Waters E., Wall S. (1978). Patterns of attachment: a psychological study of the Strange Situation. Lawrence Erlbaum Associates, Hillsdale, NJ.
American Psychiatric Association (1994). DSM-IV. Manuale Diagnostico e Statistico dei Disturbi Mentali. Masson, Milano, 1996.
Ammaniti M. (2001). Manuale di psicopatologia dell’infanzia. Raffaello Cortina Editore, Milano.
Ammaniti M., Lucarelli L., Cimino S. (2003). Intergenerational transmission: childhood feeding disorders and maternal psychopathology. Paper presentato al 12th International Congress European Society for Child and Adolescent Psychiatry, Parigi, 28 Settembre-1 Ottobre 2003.
Beebe B., Lachmann F.M. (1994). Representation and internalization in infancy: three principles of salience. Psychoanalytic Psychology, 11, [127-165].
Belsky J., Pasco Fearon R.M. (2002). Infant-mother attachment security, contextual risk, and early development: A moderational analysis. Development and Psychopathology, 14, [293-310].
Belsky J., Rovine M., Taylor D.J. (1984). The Pennsylvania infant and family development project, III. The origins of individual differences in infant-mother attachment: Maternal and infant contributions. Child Development, 55, [718-728].
Benoit D. (2000). Feeding disorders, failure to thrive, and obesity. In Zeanah C.H. (a cura di). Handbook of infant mental health. Guilford Press, New York-London.
Benoit D., Zeanah C.H., Barton M.L. (1989). Maternal attachment disturbances in failure to thrive. Infant Mental Health Journal, 10, [185-202].
Benoit, D., Zeanah, C. H., Parker, K. C., Nicholson, E., Coolbear, J. (1997). “Working Model of the Child Interview”: Infant Clinical Status Related to Maternal Perceptions. Infant Mental Health Journal, 18, [107-121].
Bowlby J. (1969-1982). Attachment and loss. Volume I: Attachment. Basic Books, New York.
Chatoor I. (1989). Infantile Anorexia Nervosa: A developmental disorder of separation and individuation. Journal of the American Academy of Psychoanalysis, 17, [43-64].
Chatoor I. (1996). Feeding and other disorders of infancy or early childhood. In Tasman A., Kay J., Lieberman J. (a cura di). Psychiatry. W.B. Saunders: Philadelphia.
Chatoor I., Ganiban J., Colin V., Plummer N., Harmony R. J. (1998). Attachment and feeding problems: a reexamination of non organic failure to thrive and attachment insecurity. Journal of the American Academy of Child and Adolescent Psychiatry, 37, [1217-1224].
Chatoor I, Loeffler C, McGee M, Menvielle E (1998) (a cura di), Observational Scale for Mother-Infant Interaction During Feeding. Manual, 2nd edition. Washington DC: Children’s National Medical Center.
Chatoor I., Ganiban J., Hirsh R., Borman-Spurrell E., Mrazek D.A. (2000). Maternal characteristics and toddler temperament in infantile anorexia. Journal of the American Academy of Child and Adolescent Psychiatry, 39, [743-751].
Cicchetti D. (1984). The emergence of developmental psychopathology. Child Development, 55, [1-7].
Cicchetti D., Cohen D.J. (1995) (a cura di), Developmental Psychopathology, Vol. 1: Theory and Methods. New York: J. Wiley & Sons.
Coolbear J., Benoit D. (1999). Failure to thrive: risk for clinical disturbance of attachment? Infant Mental Health Journal, 20, [87-104].
Crittenden P.M. (1985). Maltreated infants: Vulnerability and resilience. Journal of Child Psychology and Psychiatry, 26, [85-96].
Crittenden P.M. (1987). Non organic failure to thrive: deprivation or distortion?. Infant Mental Health Journal, 8, [51-64].
Egeland B., Faber E.A. (1984). Infant-mother attachment: factors related to its development and changes over time. Child Development, 59, [1080-1088].
Field T. (1994). The effects of mother’s physical and emotional unavailability emotional regulation. In Fox A. (a cura di). The development of emotion regulation. Biological and behavioral consideration. Monographs of the Society for Research in Child Development, 59, [208-227].
Fonagy P., Moran G., Steele M., Steele H. (1992). L’integrazione della teoria psicoanalitica e del lavoro sull’attaccamento; la prospettiva intergenerazionale. In Ammaniti M., Stern D. (a cura di). Attaccamento e psicoanalisi. Laterza, Bari.
Freud S. (1905). Tre saggi sulla teorie sessuale. In Opere, vol. IV, Boringhieri, Torino 1967-1980.
George C., Kaplan N., Main M. (1984). Adult attachment Interview. Unpublished manuscript, University of California, Berkeley.
Greenberg J.R., Mitchell S.A. (1983). Le relazioni oggettuali nella teoria psicoanalitica. Il Mulino, Bologna 1986.
Grossman K., Grossman K.E., Spangler G., Suess G., Unzner L. (1985). Maternal sensitivity and newborns’ orientation responses as related to quality of attachment in Northern Germany. In I. Bretherton, E. Waters (a cura di). Growing Points in Attachment Theory and research. Monographs of the Society for Research in Child Development, 50, [233-256].
Hesse E. (1996). Discourse, memory, and the Adult Attachment Interview: A note with emphasis on the emerging cannot classify category. Infant Mental Health Journal, 17, [4-11].
Hesse E. (1999). The Adult Attachment Interview: Historical and Current Perspectives. In Cassidy J., Shaver P.R. (a cura di). Handbook of attachment. The Guilford Press, New York.
Jacobvitz D., Hazen N. (1999). Developmental pathways from infant disorganization to childhood peer relationships. In Salomon J., George C. (a cura di). Attachment Disorganization. Guilford Press, New York.
Kernberg O. (1976). Teoria della relazione oggettuale e clinica psicoanalitica. Boringhieri, Torino 1990.
Klein M. (1957). Invidia e gratitudine. Martinelli, Firenze 1969.
Kreisler L. (1985). Conduites alimentaires déviantes du bébé. a. L’anorexie mentale.b. La rumination ou mérycisme. c. Les vomissements psychogènes. In Lebovici S., Diatkine R., Soulé M. (a cura di). Nouveau traité de psychiatrie de l’enfant et de l’adolescent. PUF, Quadrige, Paris, 1999.
Lieberman A.F., Slade A. (2001). Parenting toddlers: developmental and clinical considerations. In Osofsky J.D., Fitzgerald H.E. (a cura di). WAIMH Handbook of Infant Mental Health. Vol. 3, Parenting and Child Care. Wiley & Sons, New York 2000.
Lindberg L., Bohlin G., Hagekull B., Palmérus K. (1996). Interactions between mothers and infants showing food refusal. Infant Mental Health Journal, 17, [334-347].
Lyons-Ruth K., Connell D.B., Zoll D., Stahl J. (1987). Infants at social risk: Relationship among infant maltreatment maternal behavior and infant attachment behavior. Developmental Psychopathology, 23, [223-232].
Lyons-Ruth K. (1996). Attachment relationship among children with aggressive behavior problems: The role of disorganized early attachment patterns. Journal of Consulting and Clinical Psychology, 64, [64-73].
Mahler M., Pine F., Bergman A. (1975). La nascita psicologica del bambino. Bollati Boringhieri Editore, Torino 1978.
Main M.(1995). Recent studies in attachment: Overview with selected implications for clinical work. In Goldberg S., Muir R., Kerr J. (a cura di). Attachment Theory: Social, Developmental and Clinical Perspectives. The Analytic Press, Hillsdale.
Main M., Cassidy J. (1988). Categories of response to reunion with the parent at age six: Predictability from infant attachment classifications and stable across a one-month period. Developmental Psychology, 24, [415-426].
Main M., Goldwyn R. (1991). Adult attachment classification system. Unpublished manuscript. Department of Psychology, Berkeley, University of California.
Main M., Solomon J. (1990). Procedures for identifyng infants as disorganized/disoriented during the Ainsworth Strange Situation. In Greenberg M., Cicchetti D., Cummings M. (a cura di). Attachment in the preschool years: Theory, research and intervention. University of Chicago Press, Chicago.
Marchi M., Cohen P. (1990). Early childhood eating behaviors and adolescent eating disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 29, [112-117].
Morton N., Browne K.D. (1998). Theory and observation of attachment and its relation to child maltreatment: A review. Child Abuse & Neglect, 22, [1093-1104].
Sameroff A.J. (1993). Modelli di sviluppo e rischio evolutivo. In Zeanah C.H. (a cura di). Manuale di salute mentale infantile. Masson, Milano 1996.
Sameroff A.J., Emde R.N. (1989). I disturbi delle relazioni nella prima infanzia. Bollati Boringhieri, Torino 1991.
Sander L.W. (1964). Adaptive relationships in early mother-child interaction. Journal of American Academy of Child Psychiatry, 3, [231-264].
Speranza A.M. (2001). I disturbi dell’attaccamento. In Ammaniti M. (a cura di). Manuale di psicopatologia dell’infanzia. Raffaello Cortina Editore, Milano.
Sroufe L.A. (2000). Early relationships and the development of children. Infant Mental Health Journal, 21, [67-74].
Stern D.N. (1971). Una microanalisi dell’interazione madre-bambino. Le interazioni madre-bambino. Raffello Cortina Editore, Milano, 1998.
Stern D.N. (1985). Il mondo interpersonale del bambino. Bollati Boringhieri, Torino 1987.
Trevarthen C. (1990). Le emozioni intuitive: l’evoluzione del loro ruolo nella comunicazione tra madre e bambino. In Ammaniti M., Dazzi N. (a cura di). Affetti. Laterza, Bari.
Tronick E.Z. (1989). Emotions and emotional communication in infants. American Psychologist, 44, [112-119].
Valenzuela M. (1990). Attachment in chronically underweight young children. Child Development, 61, [1984-1996].
van IJzendoorn M.H., Kroonenberg P.M. (1988). Cross-cultural patterns of attachment: A meta-analysis of the Strange Situation. Child Development, 59, [147-156].
van IJzendoorn M.H. (1995). Adult attachment representations, parental responsiveness and infant attachment: a meta-analysis on the predictive validity of the Adult Attachment Interview. Psychological Bulletin, 117, [387-403].
Ward M.J., Lee S.S., Lipper E.G. (2000). Failure-to-thrive is associated with disorganized infant-mother attachment and unresolved maternal attachment. Infant Mental Health Journal, 21, [428-442].
Winnicott D.W. (1958). Dalla pediatria alla psicoanalisi. Martinelli Editore, Firenze 1975.
Wolff P.H. (1966). The causes, controls and organization of behavior in the neonate. Psychological Issues, 17.
Woolston J.L. (1991). Eating and growth disorders in infants and children. Sage Publications, Newbury Park (CA) – London - New Delhi.
Zeanah C.H., Zeanah P.D. (1989). Intergenerational transmission of maltreatment: insights from attachment theory and research. Psychiatry, 52, [177-196].
Zero to Three - National Center for Clinical Infant Programs (1994). Classificazione diagnostica: 0-3. Classificazione diagnostica della salute mentale e dei disturbi di sviluppo nell’infanzia. Masson, Milano 1997.
Zimmermann P., Grossmann K.E. (1994). Attaccamento, emozioni e comportamento aggressivo. Età Evolutiva, 47, [92-98].

Fig.1 Distribution of attachment classification in the children

Fig.2 Distribution of attachment classification in the mothers

.