Problems with the Theory

“As an undergraduate student in psychology, I was taught that multiple personalities were a very rare and bizarre disorder. That is all that I was taught on ... It soon became apparent that what I had been taught was simply not true. Not only was I meeting people with multiplicity; these individuals entering my life were normal human beings with much to offer. They were simply people who had endured more than their share of pain in this life and were struggling to make sense of it.”

― Deborah Bray Haddock, The Dissociative Identity Disorder Sourcebook

Problems with the Theory of Structural Dissociation


Research on dissociation is nowhere near conclusive. The theory of structural dissociation is an important theory that sheds light on the connection between dissociative conditions and helps to explain the formation and function of such conditions, but it is not yet without problems, holes, and contradictions. Many of these are not intrinsic problems with the theory but instead arise when it is inconsistently applied instead of being understood as first and foremost an explanation for the formation of complex dissociative disorders. Colin Ross, one of the earliest researchers of dissociative identity disorder (DID), points out some of these problems in a paper published in 2014.


The first problem posited is that many who endorse the theory of structural dissociation have begun excluding depersonalization/derealization disorder from its framework despite including posttraumatic stress disorder (PTSD) and borderline personality disorder (BPD). Research has already shown that all dissociative disorders as classified by the DSM-5 are strongly linked, and, as the most complex dissociative disorder, DID involves symptoms of all other dissociative disorders. Borderline personality disorder, in contrast, is a personality disorder that is not necessarily associated with any dissociative features, and, while DID is almost always comorbid with PTSD, PTSD is not always associated with DSM-5 dissociative disorders.


One possible solution to this problem is to recognize emotional parts (EP) as any collection of dissociated materials that focuses on emotion, memory, sensation, or similar. When an individual experiences depersonalization or derealization, they could then be seen as experiencing a temporary EP. In this case, the EP could be seen to contain whatever the individual is failing to integrate in the moment. If an individual can't recognize themself in a mirror, their ability to recognize themself has been dissociated and may be stored in an inaccessible EP. If an individual doesn't feel connected to their own thoughts or emotions, their thoughts and emotions are dissociated to some extent and are no longer completely integrated with the apparently normal part (ANP). If an individual can't recognize the world around them, the normal cues that allow one to reality check aren't being integrated and so are remaining dissociated within a temporary EP.


It should be noted that this theoretical temporary dissociation is in some ways different from structural dissociation of the personality. For example, individuals may be more vulnerable to temporary dissociation as depersonalization/derealization can be caused by normal daily stress, anxiety, or physical discomfort and do not require trauma. This type of dissociation may also last for only minutes, hours, or days instead of months, years, or decades, indicating that it may be easier to at least temporarily overcome. Finally, the EP would be characterized not by independence or intrusions so much as by the ANP perceiving an absence of the materials that the EP contain. Regardless, this view may help to keep understandings of dissociation cohesive.


Like depersonalization/derealization disorder, dissociative amnesia is another clearly dissociative condition that suffers from the current conceptualization of structural dissociation. It does not involve a clear EP that has its own sense of self. In fact, some might argue that it does not involve an EP at all. Dissociative amnesia is about the absence of one or more traumatic memories. While dissociative intrusions of these memories might occur, it is also possible that the individual with dissociative amnesia may never experience any intrusion related to the traumatic memories up until the memories are regained. It is a disorder of absence, not of alternate presence. Despite this, dissociative amnesia must be dissociative in nature as it highly correlates to all other aspects of dissociation (including the also contested depersonalization and derealization) and fits the basic definition of dissociation as a separation between the self and other aspects of the self, one's surroundings or sense of reality, one's personal narrative, or one's internal processes.


This may sound entirely theoretical, but some individuals have already begun differentiating between dissociative amnesia and the inter-identity amnesia found in dissociative identity disorder although it is thought that the same underlying mechanisms are responsible for both. It is only the way in which the amnesia forms and manifests (i.e., whether or not the dissociated memories are clearly contained within an alter) that differs, and this is not historically or currently a reason to classify symptoms or experiences as medically or psychologically different phenomena.


For these reasons, it may be best to view dissociative amnesia as a type of primary structural dissociation in which a single ANP loses access to a single EP that contains all of the traumatic memories. What differentiates dissociative amnesia from PTSD might be how successfully buried the EP is. If the EP is so dissociated that it fails to intrude at all, it can present as dissociative amnesia. If there are intrusions, PTSD can usually be diagnosed. Overlap is common, and dissociative amnesia is often a feature of PTSD.


Complicating the question of dissociative amnesia is that dissociative fugue is classified as a type of dissociative amnesia, but dissociative fugue as a full blown disorder could be said to involve two ANP and no EP. One personality has replaced another, and neither personality takes ownership of anything belonging to the other. In some cases, the second ANP may be less developed than the first, but it could not accurately be classified as an EP. This configuration is not one that the theory of structural dissociation currently accounts for. One solution is to assume that all of the missing memories are held by an EP that also holds all missing autobiographical information. This could force the single ANP to assume a new identity in order to try to make sense of what has gone missing. It could also be seen as the ANP trying to avoid associating with anything that might activate the EP and so cause intrusions.


Another problem presented is the question of how differentiated emotional parts must be. Nijenhuis, one of the original and current researchers behind the theory of structural dissociation, stated that EP must have a sense of subjective selfhood in order to differentiate conditions that involve structural dissociation from conditions that do not. While such a sense of self is clearly present in EP for most individuals with other specified dissociative disorder subtype 1 (OSDD-1) and DID, a unique sense of self cannot be proven to exist within EP for individuals with PTSD, complex PTSD (C-PTSD), or BPD. In fact, there is no clinical evidence that suggests that EP for individuals with PTSD, C-PTSD, or BPD can be significantly differentiated in any way. It is not typical for individuals with PTSD or C-PTSD  to identify themselves by a different name or even age while experiencing a flashback, let alone present with different skills or traits. Complete flashbacks are not even a requirement of DSM-5 PTSD! As for BPD, while identity confusion is common, actual switching behavior is not. EP for individuals with BPD clearly function more as dissociated aspects than as alternate selves. When one or more EP are noticeably differentiated, OSDD-1 or DID is usually diagnosed as well.


This makes it difficult to say that all individuals with structural dissociation of the personality have a separate part of their personality that has its own sense of self. On the other hand, if EP do not require a subjective sense of self, some have argued that almost all disorders of the DSM-5 could be said to be dissociative conditions. Any condition that involved an ego-alien thought, impulse, affect, sensation, or perception could be argued to qualify. Ross suggests that this would be a good thing, but it would in reality destroy the link between structural dissociation and trauma, dissociative symptoms, and dissociative intrusions. In short, it would undermine the entire theory.


For this reason, it may be best to accept the following. EP are dissociated collections of materials such as emotions, memories, urges, thought patterns, body awareness, or similar. EP are dissociated from the ANP and, when present, are experienced as intrusions. When EP are absent, the ANP does not have access to what the EP contain. EP become more elaborate as they are present in different situations, and repeated use during early childhood trauma leads to the most emancipation and elaboration. The most emancipated EP have a sense of autonomy, and the most elaborated EP may have many individual characteristics. Not all EP are equally emancipated or elaborated, especially across disorders; EP in DID can be much more emancipated and elaborated compared to those in PTSD, C-PTSD, or BPD. EP in different disorders can preferentially hold different materials; EP in BPD are more likely to hold attachment needs whereas those in PTSD are more likely to be pure trauma memories. Not all internal disconnections are EP; an individual may feel disconnected from OCD-related compulsions, but this does not mean that OCD compulsions are coming from an EP. EP must not be better explained by non-dissociative mechanisms. In this way, structural dissociation can remain a more selective and so more useful theory.


Finally, one might question the distinction between primary, secondary, and tertiary structural dissociation. Consider the distinction between OSDD-1 and DID. Is the distinction solely related to the number and types of parts present (as defined by the theory of structural dissociation) or does it relate to the degree of differentiation of parts and the presence of inter-identity amnesia (as defined by the DSM-5)? If all that matters is whether or not there are one or more EP and one or more ANP, OSDD-1 and DID do not fit neatly within their designated categories. An individual who would fit the clinical criteria for OSDD-1 due to a lack of amnesia between their alters or a lack of sufficiently differentiated alters could be labeled DID by the theory of structural dissociation if their alters included two or more ANP. Conversely, an individual who would fit the clinical criteria for DID due to amnesia between two or more sufficiently differentiated alters could be labeled OSDD-1 by the theory of structural dissociation if the individual had only one ANP regardless of their number of EP. However, if the presence or absence of amnesia is what differentiates OSDD-1 from DID, problems arise when EP for those with PTSD, C-PTSD, or BPD can fully take control of the individual in a manner that cases the ANP to black out and lose time. A solution is to define structural dissociation solely by the number of ANP and EP and accept that OSDD-1 and DID can represent either secondary or tertiary structural dissociation depending on the individual.


Overall, though the theory of structural dissociation is an important step in progressing towards an understanding of dissociative symptoms and disorders, it is not yet entirely applicable outside of a theoretical setting. It may at current be prudent to view it more as an origin theory than as an explanation for how dissociative disorders function.



Ross, C. A. (2014). Unresolved problems in the theory of structural dissociation. Psichiatria E Psicoterapia, 33(3), 285-292.


*The author wishes to reiterate that while they are using Dr. Ross's concerns as a springboard and to organize their argument, they do not endorse all of Dr. Ross's conclusions in this nor in many other matters.