Developmental Pathways to Dissociation: Are we forgetting something?

Developmental Pathways to Dissociation: Are we forgetting something?

Written by Andrew M. Leeds

From ESTD Newsletter Volume 3 Number 1, December 2012 > read the original article in our newsletter



Is early trauma the root cause of dissociative disorders?

            Most of the literature and research on pathological forms of dissociation[1] focuses on trauma as the root cause of the DSM IV-TR (American Psychiatric Association, 2000) dissociative disorders, and many studies show an association between dissociation and trauma (Putnam, 1997).  Yet this view by no means tells the entire story as emerging research on attachment, evolutionary psychobiology, and the neurobiology of dissociation continues to influence the field. Recent research emphasizes an interesting observation: “The fact that nontraumatized individuals sometimes demonstrate dissociation and that not all trauma survivors dissociate suggests that there may be more to the etiology and development of dissociation than trauma alone” (Dutra, Bianchi, Siegel, & Lyons-Ruth, 2009, p. 84). If early trauma is not the unique cause of pathological dissociation, then what else could cause it? Although chronic and severe traumatic exposure is central to the development of complex dissociative disorders (Van der Hart, Nijenhuis, & Steele, 2006), we should not ignore emerging data on the relationship between attachment and pathological dissociation. The goal of this brief review is to promote reflection on the significant role of non-traumatic factors in the vulnerability to the development of trait dissociation.

Are dissociative disorders, disorders of attachment?

            Peter Barach (1991) was one of the first to emphasize insecure attachment in the etiology of multiple personality disorder, referred to as dissociative identity disorder (DID) in the DSM-IV-TR (American Psychiatric Association, 2000), and suggested it should be viewed as an attachment disorder. Barach emphasized the idea that emotionally neglectful, detached responses from caretakers lead to chronic emotional detachment and dissociation in those who later go on to develop MPD (DID).  He also described how such neglectful and detached caretaker responses can lead to failures to detect and to protect young children from the impact of later occurring severe trauma. At about the same time, Giovanni Liotti (1992, 2009) emphasized insecure disorganized (type D) attachment and the early formation of multiple internal working models of the self and other as the precursors to the development of severe dissociative disorders later in life. John Bowlby (1973) had previously suggested that infants can internalize unintegrated internal working models of their primary caregivers and themselves.

Can problems with infant attachment be the root cause of pathological dissociation? A growing literature on this subject suggests the answer is “Yes.” However, many clinicians on first exposure to this literature may assume that the focus has merely shifted to “attachment trauma” – such as from persistent infant experiences of frightening or frightened caregivers (Liotti, 1992, 2009) who trigger defensive action systems and disrupt the functioning and development of the attachment system (Van der Hart, Nijenhuis, & Steele, 2006). Yet, as we will see, two recent prospective longitudinal studies suggest that “attachment trauma” is not the factor most associated with the risk of later developing pathological dissociation. In fact, in some cases, it may not be the most significant factor at all.

Wait. Isn’t “attachment trauma” the root of vulnerability for adult pathological dissociation?

            In adults, “trauma” is generally viewed through the lens of the DSM-IV-TR criteria for the development of posttraumatic stress disorder: “direct personal experience of an event that involves actual or threatened death or serious injury, or other threat to one's physical integrity; or witnessing an event that involves death, injury, or a threat to the physical integrity of another person” and “The person's response to the event must involve intense fear, helplessness, or horror (or in children, the response must involve disorganized or agitated behavior)” (APA, 2000, p. 463). Do the infant experiences that lead to a vulnerability to structural dissociation fit these criteria? Most clinicians have tended to assume so, because they view neglectful caretaker behavior as having the potential to induce limbic system (early mammalian brain) mediated states of “abandonment terror” in the attachment system and more primitive brain stem mediated states of “submission-feigned death” in defensive action systems. But what if research from not just one, but from two, longitudinal studies indicated that threat of abandonment and other infant trauma experiences (such as frightened/frightening parental behaviors – Liotti, 1992, 2009) are not the central factor in predicting the later development of adult dissociation? Could we as clinicians and researchers reorganize our thinking to consider the implications or are we already committed to the concept of trauma as the primary cause (and the primary locus for the cure) for all aspects of pathological dissociation?  We need to continuously reflect on the significance of emerging research findings.

Prospective Longitudinal Studies of Predictors of Adult Dissociative Disorders

            The first of these two studies was carried out by Ogawa, Sroufe, Weinfield, Carlson, and Egeland (1997). Dutra, Bianchi, Siegel, and Lyons-Ruth (2009) summarized Ogawa et al. (1997) as: “… a prospective longitudinal study of 126 high-risk children, following this sample from birth to age 19, in an attempt to test Liotti’s model. According to earlier models of pathological dissociation based in trauma theory, one might expect trauma to be the strongest independent predictor of adult dissociation. Ogawa et al.’s (1997) multiple regression analyses indicated, however, that disorganized attachment and psychological unavailability of the caregiver during infancy were the strongest predictors of clinical levels of dissociation, to the extent that these variables alone accounted for approximately one-quarter of the variance in dissociation at age 19. Surprisingly, trauma history did not significantly add to the predictive value of this equation…These findings lead to the question of whether disorganized attachment may be as central to the development of dissociation as trauma itself…It is also notable that Ogawa et al.’s (1997) study demonstrated that some nondisorganized infants in the sample developed dissociative symptomology in young adulthood. This finding suggests that there may be factors above and beyond disorganized attachment that serve to predispose children to the development of dissociation” [Emphasis added] (p. 86).

            These striking findings led Dutra, Bianchi, et al. (2009) to further investigate the association between infant attachment and dissociation in another, more recent prospective, longitudinal study of a high-risk sample from birth to age 19. Their findings partially confirmed and extended the earlier work by Ogawa et al. (1997). However, Dutra, Bianchi, et al. (2009) found early childhood maltreatment did not predict adolescent dissociative symptoms. A range of maternal psychiatric symptoms–including anxiety, depressive, dissociative, and posttraumatic stress disorder symptoms–when caring for children up to nine years old also did not predict adolescent dissociative symptoms. If childhood maltreatment and maternal psychiatric symptoms did not predict adolescent dissociative symptoms, what did? “Quality of maternal communication during infancy” accounted for half of the variance in dissociative symptoms at age 19 (Dutra, Bianchi, et al., 2009, p. 87).

Could disrupted parent-infant dialogue not maltreatment predict adult pathological dissociation?

            Dutra, Bureau, Holmes, Lyubchik, and Lyons-Ruth (2009) used the Observation of Maternal Interaction Rating Scales (HOMIRS; Lyons-Ruth et al., 1987) to assess mother-infant interactions at home at 12 months and the Atypical Maternal Behavior Instrument for Assessment and Classification (AMBIANCE: Lyons-Ruth et al., 1999) to assess maternal disrupted affective communication at 18 months. As they note “[t]rauma theory dictates that maternal hostile (negative-intrusive) and/or disoriented behaviors would likely be the strongest predictors of dissociation in young adulthood, but, surprisingly, this is not what was found. Maternal hostile or intrusive behavior was not significantly related to later dissociation. Instead, lack of positive maternal affective involvement, maternal flatness of affect, and overall disrupted maternal communication were the strongest predictors of dissociation in young adulthood[2]. What is notable about these types of maternal interactions is that they all serve to subtly override or ignore the infant’s needs and attachment signals, but without overt hostility” (Dutra, Bianchi, et al, 2009, p. 87, Emphasis added).

            In Ogawa et al.’s (1997) study approximately 25% of the variance in Dissociative Experiences Scale scores at age 19 was accounted for by the combination of psychological unavailability and infant disorganization. However, 19% of this variance was carried by the psychological unavailability variable alone. In Dutra et al. (2009, p. 88), after controlling for quality of early care, childhood trauma accounted for only 9% of the variance, which was not significant.  After controlling for gender and demographic risk, 50% of the variance in adult dissociation scores was accounted for by quality of early care (p < .001). Within the quality of early care cluster, level of disrupted communication (assessed in the lab – p < .01), mother’s (lack of) positive affective involvement at home (p < .05), and mother’s flatness of affect at home (p < .05) were all significant predictors of dissociative symptoms.

            Early developmental experiences involving parent-infant dialogue and infant disorganization are clearly not the only early vulnerability factors to the later development of pathological adult dissociation. Later occurring traumatic experiences clearly play a role in complex dissociative disorders, as do undoubtedly, genetic, societal and environmental factors. What is so compelling from the latest prospective, longitudinal research from Dutra et al. (2009) is that the portion of the vulnerability for adult dissociation accounted for by disrupted parent-infant dialogue (assessed by AMBIANCE at 12 months), mother’s flatness of affect and lack of mother’s positive affective involvement (assessed by HOMIRS at 18 months) was found to be so much more influential than that of early traumatic experience or mother’s hostile-intrusiveness.

Revising our understanding of the foundations of adult dissociative disorders

            The findings from Dutra, Bureau, et al. (2009), extending the earlier work from Ogawa et al. (1997), should serve to shift our attention from early models of vulnerability for adult pathological dissociation as being founded solely on traumatic experiences, to consider the central role for what Dutra et al. (2009) refer to as disruptions of parent-infant dialogue. “In contrast to a more discrete traumatic event, the child’s fear of remaining unseen and unheard by his caregiver, resulting in unmet needs, is worked into the fabric of identity from a very early age” (Dutra, Bianchi, et al., 2009, p. 91). The role of disruptions in maternal infant dialogue and the absence of maternal-infant play should be included in future studies to evaluate the relative contribution of different factors in the development of pathological dissociation as well as in dissociative disorders.

            Some may be tempted to narrow the clinical implications of the findings from Dutra, Bureau, et al. (2009) to emphasize the potential impact on brain stem mediated defensive systems for flight and submission based on the notion of fear states induced by a parental “still face” as described by Tronick (1989, 1998; Tronick & Weinberg, 1997). However such a narrowing of attention to fear mediated defensive subsystems would fail to give sufficient attention to the parallel and equally significant under-development of the social engagement systems for activating and organizing attachment, cooperation, play, and competition. The persistent absence of a healthy “reflective gaze” from the caregiver may help to explain the lack of development of metacognitive, high order mental functions in survivors of neglect (Putnam, 1997; Schore, 2003). For individuals with such histories of neglect, the effects of traumatizing events – which trigger defensive action systems -- could be much more destructive, leading lesser adverse events to have more severe consequences for the development of subsequent psychopathology.

            These findings challenge us to rework our views of what lies at the root of the multiple, complex problems of patients with severe, adult dissociative disorders. Perhaps the implications of these findings will help clinicians broaden what has, at times, become an overly narrow focus on uncovering and metabolizing the content of autobiographical memories of “what happened,” and support a shift 1) to a consideration of the intersubjective and affective clinical “dialogue” between client and clinician, and 2) to deeper, neuroaffective strategies that facilitate activation, development and integration of neurobiologically mediated systems for social engagement with higher cortical systems for self-representation, such as advanced EMDR interventions (Gonzalez & Mosquera, 2012; Knipe, 1995, 2008).

            To succeed as the medium for developmental repair, such psychotherapeutic interventions must address and activate neurobiological circuits that are fundamentally both pre-verbal and encoded in early forming subcortical, brain stem and limbic circuits in addition to later developing cortical and cognitive maps in higher regions of the brain. The model of early parent-child dialogue described by Dutra, Bureau, et al. (2009) may help us understand why individuals vary so much in their response to later traumatic experiences as we consider the neuro-protective role of such early nutritive attachment experiences for the development of multiple social engagement action systems and the internalization of a cohesive model of self as both worthy of self-care and self-compassion, as well as capable of obtaining and assimilating social supports and interpersonal care at times of stress. Hopefully these remarkable findings will lead to more work to address longstanding gaps in the literature and better inform the development, dissemination and acceptance of effective approaches that clinicians can bring to their work with the full range of dissociative disorders.



American Psychiatric Association (2000). Diagnostic and statistical manual of mental       disorders: DSM-IV-TR. Washington, DC : American Psychiatric Association.

Barach, P. M. (1991). Multiple personality disorder as an attachment disorder. Dissociation,           4, 117-123.

Bowlby, J. (1973). Attachment and loss: Vol. 2: Separation: Anxiety and anger. London:           Penguin Books.

Dell, P. F. & O'Neil, J. A. (Eds.) (2009). Dissociation and the dissociative disorders: DSM-V          and beyond. New York: Routledge.

Dutra, L., Bianchi, B., Siegel, D. J., & Lyons-Ruth, K. (2009). The relational context of   dissociative phenomena. In P. F. Dell & J. O'Neil (Eds.), Dissociation and the dissociative disorders: DSM-V and beyond (pp. 83-92). New York: Routledge.

Dutra, L., Bureau, J. F., Holmes, B., Lyubchik, A., & Lyons-Ruth, K. (2009). Quality of early         care and childhood trauma: A prospective study of developmental pathways to dissociation. Journal of  Nervous and Mental Disease, 197, 383-90.

Gonzalez, A. & Mosquera, D. (Eds.) (2012). EMDR and dissociation: The progressive          approach. [English edition.] Charleston, SC: Amazon Imprint.

Knipe, J. (1995). Targeting defensive avoidance and dissociated numbing. EMDR Network     Newsletter, 5(2), 6-7.

Knipe, J. (2008). Loving eyes: Procedures to therapeutically reverse dissociative processes            while preserving emotional safety. In C. Forgash & M. Copeley (Eds.), Healing the   heart of trauma and dissociation with EMDR and ego state therapy (pp. 181-    225). New York, NY: Springer Publishing Company.

Liotti, G. (1992). Disorganized/disoriented attachment in the etiology of the dissociative           disorders. Dissociation, 5(4), 196-204.

Liotti, G. (2009). Attachment and dissociation. In P. F. Dell & J. O'Neil (Eds.), Dissociation    and the dissociative disorders: DSM-V and beyond (pp. 53-65). New York:         Routledge.

Ludwig, A. M. (1966). Altered states of consciousness. Archives of  General Psychiatry,   15(3), 225-234.

Lyons-Ruth K, Bronfman E, & Parsons E. (1999). Maternal frightened, frightening, or            atypical behavior and disorganized infant attachment patterns. In J. I. Vondra & D.     Barnett (Eds.), Atypical attachment in infancy and early childhood among   children at             developmental risk. Monographs of the Society for Research in Child   Development, 64, 67–96.

Lyons-Ruth K, Connell D, Zoll D, & Stahl J.(1987). Infants at social risk: Relations among            infant maltreatment, maternal behavior, and infant attachment behavior.   Developmental Psycholology, 23, 223–232.

Ogawa, J. R., Sroufe, L. A., Weinfield, N. S., Carlson, E. A., & Egeland, B. (1997).       Development and the fragmented self: Longitudinal study of dissociative            symptomatology in a nonclinical sample. Developmental Psychopatholology, 9, 855-  79.

Putnam, F. W. (1997). Dissociation in children and adolescents: A developmental            perspective. New York: Guilford Press.

Schore, A. N. (2003). Affect dysregulation & disorders of the self (pp. xvii, 403 p.). New      York:   W.W. Norton.

Tart, C. T. (1969). Altered states of consciousness: A book of readings. New York: Wiley.

Tronick, E. (1989). Emotions and emotional communication in infants. American        Psychologist, 44, 112-119.

Tronick, E. (1998). Dyadically expanded states of consciousness. Infant Mental Health            Journal, 19(3), 295.

Tronick, E.Z., & Weinberg, M.K. (1997). Depressed mothers and infants: Failure to form           dyadic states of consciousness. In L. Murray & P.J. Cooper (Eds.), Postpartum    depression in child development (pp. 54–81). New York: Guilford Press.

Van der Hart, O., & Dorahy, M. J. (2009). History of the concept of dissociation. In P. F. Dell      & J. A. O'Neil (Eds.), Dissociation and the dissociative disorders: DSM-V and   beyond (pp. 3-26). New York: Routledge.

Van der Hart, O., Nijenhuis, E. R. S., & Steele, K. (2006). The haunted self: Structural           dissociation and the treatment of chronic traumatization. New York: Norton.

Waller, N. G., Putnam, F. W., & Carlson, E. B. (1996). Types of dissociation and dissociative types: A taxometric analysis of dissociative experiences. Psychological Methods, 1,        300-321.


[1] An adequate discussion of the many meanings and dimensions of the construct “dissociation” is beyond the scope of this short article. The reader is directed to Dell and O’Neil’s (2009) comprehensive Dissociation and the dissociative disorders: DSM-V and beyond in which several chapters explore alternate perspectives on this issue. In this article, when using the term “dissociation” I will be referring to two pathological dimensions of dissociation: 1) brain stem mediated experiences of depersonalization and derealization; and 2) cortically mediated deficiencies in connectivity giving rise to experiences of identity confusion and a failure to develop cohesion across all aspects of the personality – referred to as structural dissociation of the personality by Van der Hart, Nijenhuis, and Steele (2006). I prefer to exclude normative experiences such as absorption and fantasy-proneness from the construct of pathological “dissociation.”

[2] Chronic parental “acts of omission” define neglect. Some of these “acts of omission” clearly constitute “traumatic” experiences such as failures to protect infants and children from overt acts of harm by others, or failures to provide basic medical care when infants and children are suffering from treatable medical conditions. Work by Tronick and others (Tronick, 1989, 1998; Tronick and Weinberg, 1997) shows that failures to form dyadic states of consciousness has severe consequences for infant development and can in many cases induce fear states and the triggering of defensive reactions. Such findings lead to the pervasive concept of “attachment trauma.” It is possible that some forms of neglect involving insufficient maternal-infant play and inadequate “maternal-infant dialogue” may be “non-traumatic” in that they may not induce fear states nor defensive reactions.  Using a metaphor from nourishment, feeding an infant only spoiled food that tastes bad and that triggers persistent diarrhea is clearly traumatic. Feeding an infant high caloric food of low nutritional value can lead to malnourishment, but the child might not be “traumatized” by this diet. While we would consider this neglect, it would not be a traumatic experience.