Development of Dissociative Identity Disorder

Children aren’t born with an integrated personality. In the absence of trauma, a child will integrate one coherent self over time as their brain matures and becomes capable of supporting this integration. Dissociative identity disorder (DID) is the result of one sense of self and self history failing to integrate due to trauma. Trauma that’s interpersonal, involves caregivers, begins at a younger age, is more severe, lasts for longer, and isn’t properly acknowledged is more likely interfere with a child's integrative capacity and so to cause DID. Once an individual has one sufficiently integrated sense of self, this process can’t be reversed and the individual can’t develop DID (Nijenhuis, Hart, & Steele, 2010)¹.

Two areas of the brain that are heavily implicated in the development of DID are the hippocampus and the orbitofrontal cortex. The hippocampus, which is responsible for memory and evaluating context, can take the first 3 years of life to mature, and myelination continues through adolescence and then remains unchanged (Nijenhuis, Hart, & Steele, 2010)¹. In contrast, the orbitofrontal cortex-- which plays diverse roles such as emotional processing, self regulation, inhibitory control of information, decision making, “self reflective awareness,” and “integrat[ing] a sense of self across transitions of state” (Schore, 2000)²-- undergoes rapid maturation between ages 6 to 9 (Forrest, 2001)³. Both the hippocampus and the orbitofrontal cortex are associated with processing and integrating information, and their lack of development during the early years of life is one reason why young children do not have strong integrative capacities (Nijenhuis, Hart, & Steele, 2010)¹.

The development of the hippocampus, orbitofrontal cortex, and other areas of the prefrontal cortex are particularly sensitive to trauma. It’s known that during times of stress, neurochemicals such as norepinephrine, epinephrine, glucocorticoids, and endogenous opiates accumulate in these areas and can interfere with the integration of experiences. These effects may be long term. Direct glucocorticoid exposure in particular has been linked to a smaller hippocampal volume, and smaller hippocampal volumes can be associated with posttraumatic stress disorder (PTSD), other specified dissociative disorder subtype 1 (OSDD-1), and DID. It may be that a smaller hippocampal volume is a risk factor for these conditions, but there is evidence that individuals with PTSD, OSDD-1, and DID can regain some hippocampal volume (though not volume of the parahippocampal gyrus) through treatment (Nijenhuis, Hart, & Steele, 2010)¹. As well, in women with DID, hippocampal volume is negatively correlated with the severity of lifetime trauma (Chalavi et al., 2014)⁴. It's also of note that even in healthy individuals, artificial stimulation of the hippocampus has been known to lead to dissociative symptoms (Nijenhuis, Hart, & Steele, 2010)¹. 

Dysfunction of the orbitofrontal cortex is also heavily implicated in DID (Forrest, 2001)³. Sar, Unal, and Ozturk (2007)⁵ found that DID is associated with reduced regional cerebral blood flow to the orbitofrontal cortex and theorized that weakened inhibition of this region allows “distinct mental states” to develop. Chalavi et al. (2015)⁴ show that individuals with DID and with PTSD have significantly smaller orbitofrontal cortices than individuals without these disorders. Nijenhuis, van der Hart, and Steele (2010)¹ also associate DID with the orbitofrontal cortex and other prefrontal regions. Another major prefrontal region that has been implicated in the formation of DID is the medial prefrontal cortex, which is involved in conscious experiences, self reference, and representations of self-concept; its activity has been found to differ between trauma and non-trauma oriented parts when exposed to a trauma-related script (Reinders et al., 2003)⁶.

Early attachment experiences can affect the development of the orbitofrontal cortex (Schore, 2000)². Forrest (2001)³ suggests that when caregivers don’t appropriately respond to an infant’s needs, this results in insecure attachment and an inability to tolerate strong affective states. Disorganized attachment due to maltreatment is especially likely to lead to the child failing to learn to shift states appropriately across contexts or to integrate states. Both avoidant-insecure and disorganized attachment styles have been found to be strong predictors of dissociation (Ogawa, Sroufe, Weinfield, Carlson, & Egeland, 1997)⁷, and it's thought that interaction with caretakers is important to teaching children how to integrate their behavioral states (Nijenhuis, Hart, & Steele, 2010)¹. This is supported by Sar, Unal, and Ozturk (2007)⁵, who suggest that a lack of distress tolerance, which can be the result of inadequate parenting, works to motivate switching between dissociated parts.

In abusive environments, there can be advantages to a child not having an integrated self. Knowledge, feelings, reactions, and abilities that may be helpful or even necessary in one context may be unhelpful or dangerous in another, necessitating that the child keep these materials separate. They do this by failing to integrate the parts of themself that contain these opposing materials. Not only do they not develop an overarching "global me" that encompasses all of these materials, they must inhibit conflicting states of self so as to have the right materials active at the right times. This may be managed by the orbitofrontal cortex developing to favor context-specific understandings of self as opposed to one integrated self that transcends context. Beyond simply struggling to integrate one self, it learns that maintaining multiple self states that can be activated in turn is the only way to survive an unpredictable, uncontrollable, and unsafe environment (Forrest, 2001)³. 

Many of these self states  must contain more than basic dissociated materials because the behavior expected from the child at any time could require more organized behavior than a basic part could successfully perform in a situation where switching between parts could be difficult or dangerous. This is what causes dissociated parts to become more differentiated and developed alters. There may still be basic overlap between alters when information does not evoke conflict between them and so does not require isolation of any particular parts. However, there is enough differentiation that when there is a situation that requires a specific self state to be expressed, the associated alter must be able to handle the situation without bringing to consciousness any other alters that may contain contradictory materials that could endanger the overall individual's ability to correctly react (Forrest, 2001)³. 

Once trauma and attachment disturbances have lead to a lack of integration, classical conditioning may contribute to the maintenance of this fragmentation and to continued switching behavior. When an individual is exposed to reminders of the trauma that they experienced, they're forced to recall highly unpleasant events, emotions, sensations, thoughts, and so on. During the trauma, the individual learned to rely on dissociation and failure of integration in order to protect themselves from these negative intrusions, and so they continue to use this behavior even once safe. As a result, they successfully avoid the negative materials and so are rewarded by a lack of overwhelming negative affect. When they cannot successfully avoid the negative materials, they experience the resulting traumatic intrusions, which is a highly aversive experience. This reinforces their desire to continue to avoid traumatic stimuli and any internal representations that it brings forth and so reinforces their dissociative response to such stimuli. Avoidance can also be encouraged when others who are aware of the trauma refuse to acknowledge it or encourage the traumatized individual to be ashamed of having survived the trauma or of their responses to the trauma. This may be one reason why child sexual abuse in particular is linked with dissociative amnesia and why both PTSD and dissociative disorders are associated with a lack of support following trauma (Nijenhuis, Hart, & Steele, 2010)¹.

For individuals with DID in particular, avoidance of traumatic materials can manifest as switching between parts so that the alters that deal with daily life do not have to be exposed to any reminders of their past. Even different trauma oriented parts may learn to avoid each other in order to avoid being exposed to traumatic materials that they were not meant to handle. This avoidance may be deliberate (such as involving conscious avoidance of potential reminders of trauma or emotionally intense experiences, substance use or abuse, or distraction and absorption into non-upsetting tasks) or may be experienced as automatic and uncontrollable (Nijenhuis, Hart, & Steele, 2010)¹.

State-dependent learning may also play an important role in the functioning of posttraumatic intrusions, repressed memories, and switching between dissociative parts. Norepinephrine has been linked with flashbacks in individuals with chronic PTSD and so may play a role in activating specific states, such as trauma-oriented states, in structurally dissociated individuals. That is, acute stress that leads to norepinephrine release may be what's needed to activate an emotional part (Nijenhuis, Hart, & Steele, 2010)¹.

State dependent learning has also been linked to different neural conditions during memory formation. One study found that when memories were encoded in a glutamate receptor mediated state versus GABA receptor mediated state (as enforced by gaboxadol injections in the hippocampus) in mice, fear conditioning associations and memories seemed to be only accessible in the state in which the fear conditioning occurred. That is, mice that were injected with gaboxadol only before fear conditioning or only before memory testing did not exhibit the conditioned fear response during memory testing while those that were injected before both the fear conditioning and the memory testing did. Fear conditioning involving gaboxadol was found to enhance PKCβII phosphorylation, decrease miR-33 as opposed to increase it as non-gaboxadol associated fear conditioning does, and enhance early growth factor 1 responses in subcortical but not cortical hippocampal targets. A cortical structure which is required for retrieving most memories, the retrosplenial cortex, was found to suppress the retrieval of state-dependent memories (Jovasevic et al., 2015)⁸. In short, memories can be encoded and stored in completely different neuronal circuits even on a genetic level when the memories are made under different neural states, and this may be what allows some traumatic memories to be inaccessible under normal conditions (Paul, 2015)⁹. This may be a key mechanism of how PTSD and dissociative disorders function.

All of this is important not only for understanding how DID forms but also why it can only form due to childhood trauma. The associated neurology in particular is important. Consider again the orbitofrontal cortex. It has a critical period of growth from 10 to 12 months, it undergoes another rapid period of growth between 6 to 9 years, and complete myelination occurs around age 12 (Forrest, 2001)³. It’s thought that the orbitofrontal cortex does not have to be completely finished developing in order for an individual to have a sufficiently integrated sense of self. In short, in the absence of trauma, a child will likely emerge from the rapid period of growth between 6 and 9 with a coherent sense of self. It’s likely that for most children, this coherent sense of self would have been present from closer to age 6 and would not have needed the entire period of growth to form (Dell & O'Neil, 2009)¹⁰. However, estimating the range at which DID can form as before ages 6 to 9 accounts for any individuals who may have taken slightly longer to form a sense of self.

It should be noted that an individual may be able to develop OSDD-1 slightly later than they would be able to develop DID. There is not enough research on the development of OSDD-1 specifically to say for sure when it must develop by. It can be assumed that OSDD-1 likely can form slightly later than DID because OSDD-1 doesn’t require the same strength of dissociative barriers between parts. However, it’s extremely unlikely that even OSDD-1 could form in adulthood. By then, a central sense of self would be too strong for any alternate identities to form in the absence of existing DID or OSDD-1.

Consider this discussion in regards to the theory of structural dissociation. Individuals who experience trauma at any age can develop PTSD, which is a type of primary structural dissociation. Individuals with PTSD (and without comorbid DID or OSDD-1) have one central self, an apparently normal part (ANP), that contains all of the aspects of their self that were present before the trauma and that are not now associated with the trauma. These individuals also have an emotional part (EP), a trauma-oriented part that contains all of the thoughts, memories, feelings, impulses, and other materials that are associated with the trauma and that the individual hasn’t yet managed to integrate into their overall sense of self and self narrative. The individual with PTSD still has one functioning identity, their ANP. That they failed to integrate traumatic materials does not change this. That some trauma-oriented materials are stored in a different state and are not accessible under normal conditions does not say anything about their sense of self, which remains intact (Hart, Nijenhuis, & Steele, 2006)¹¹.

Similar occurs in two of the types of secondary structural dissociation. Individuals with complex posttraumatic stress disorder (C-PTSD) and with borderline personality disorder (BPD) (and without comorbid DID or OSDD-1) have one central ANP and multiple EP. In these cases, the different EP contain different aspects of the traumas that the individuals have experienced. This may mean that the ANP contains less because more is contained by other parts. However, there is still one central self. The individual may experience this self as hollow, absent, uncertain, or malleable, but it is still present and is still singular (Hart, Nijenhuis, & Steele, 2006)¹¹. Even if childhood trauma has led the individual to have a more disjointed, context-dependent organization of behavior, their parts are not differentiated enough to have competing senses of self. One identity is overarching even if attachment styles, patterns of affect, and understanding of self and others are not (Forrest, 2001)³. 

OSDD-1 is usually characterized as another type of secondary structural dissociation, but this is where the pattern begins to become disrupted. Even if an individual with OSDD-1 only has multiple EP and still retains a single ANP, their EP may be significantly more developed than the EP associated with the previous three disorders. Individuals with OSDD-1 may experience their parts in various ways, ranging from complete alters to “me but not me,” but a disruption of identity is always present. It is not just that the individual has failed to integrate traumatic materials. Childhood trauma affected their development so significantly that their sense of self is now fragmented. They do not and cannot retain a single sense self across transitions of state (Hart, Nijenhuis, & Steele, 2006)¹¹.

DID is the most extreme example of this. Most cases of DID are cases of tertiary structural dissociation, or structural dissociation that involves both multiple EP and multiple ANP. Individuals with DID experienced childhood trauma that was so detrimental to their development that they never developed a coherent sense of self across any domain of their life. Their dissociated parts can be highly differentiated, and the dissociative barriers between parts can be so strong that parts can be completely unaware of each other or each others’ activities. This is not something that could be acquired later in life. This represents a failure of the brain to develop in a way that supports sufficient integration of self and self history. The first 6 years of life in particular have been implicated as a critical period for the development of a coherent self that can integrate stressful experiences as opposed to the development of a self that must rely on state-dependent functioning and dissociative parts of the personality (Hart, Nijenhuis, & Steele, 2006)¹¹. That is to say, when one takes into account the complex interplay of neurological and environmental factors that lead to the development of DID, it becomes impossible to argue that DID could form after around age 9 because so many vital developmental processes would have already been sufficiently completed to lead to one sufficiently integrated sense of self.

¹ Nijenhuis, E.R.S., Van der Hart, O., & Steele, K. (2010). Trauma-related structural dissociation of the personality. Activitas Nervosa Superior, 52(1), 1-23.

² Schore, A. N. (2000). Attachment and the regulation of the right brain. Attachment & Human Development, 2(1), 23-47. doi:10.1080/146167300361309

³ Forrest, K. A. (2001). Toward an etiology of dissociative identity disorder: A neurodevelopmental approach. Consciousness and Cognition, 10(3), 259–293. doi:10.1006/ccog.2001.0493

⁴ Chalavi, S., Vissia, E.M., Giesen, M.E., Nijenhuis, E.R.S., Draijer, N., Barker, G.J., . . . Reinders, A.A.T.S. (2015). Similar cortical but not subcortical gray matter abnormalities in women with posttraumatic stress disorder with versus without dissociative identity disorder. Psychiatry Research: Neuroimaging, 231(3), 308-319. doi:10.1016/j.pscychresns.2015.01.014

⁵ Sar, V., Unal, S.N., & Ozturk, E. (2007). Frontal and occipital perfusion changes in dissociative identity disorder. Psychiatry Research, 156(3), 217-223. doi:10.1016/j.pscychresns.2006.12.017

⁶ Reinders, A.A.T. S., Nijenhuis, E.R.S., Paans, A.A.J., Korf, J., Willemsen, A.T.M., & den Boer J. A. (2003). One brain, two selves. NeuroImage, 20(4), 2119-2125. doi:10.1016/j.neuroimage.2003.08.021

⁷ 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 [Abstract]. Development and Psychopathology, 9(4), 855-879. doi:10.1017/S0954579497001478

⁸ Jovasevic, V., Corcoran, K. A., Leaderbrand, K., Yamawaki, N., Guedea, A. L., Chen, H. J., . . . Radulovic, J. (2015). GABAergic mechanisms regulated by miR-33 encode state-dependent fear. Natural Neuroscience, 18(9), 1265–1271. doi:10.1038/nn.4084

⁹ Paul, M. (2015). How traumatic memories hide in the brain, and how to retrieve them. Northwestern. Retrieved from http://www.northwestern.edu/newscenter/stories/2015/08/traumatic-memories-hide-retrieve-them.html

¹⁰ Dell, P. F., & O'Neil, J. A. (2009). Dissociation and the dissociative disorders: DSM-V and beyond. New York: Routledge.

¹¹ Hart, O., Nijenhuis, E. R. S., & Steele, K. (2006). The haunted self: Structural dissociation and the treatment of chronic traumatization. New York: W.W. Norton.