International Journal of Mental Health & PsychiatryISSN: 2471-4372

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Review Article, Int J Ment Health Psychiatry Vol: 4 Issue: 1

Alter Personalities in Dissociative Identity Disorder: Artefacts or Authentic Entities? Re-Evaluating the Available Evidence

Flavia Spiroiu*

Candidate in Clinical Psychology, Lakehead University, Canada

*Corresponding Author : Flavia Spiroiu, B.A.H.
M.A. Candidate in Clinical Psychology, Lakehead University, Canada
Tel: 1-416-732-3911
E-mail: [email protected]

Received: February 13, 2018 Accepted: February 26, 2018 Published: March 07, 2018

Citation: Spiroiu F (2018) Alter Personalities in Dissociative Identity Disorder: Artefacts or Authentic Entities? Re-Evaluating the Available Evidence. Int J Ment Health Psychiatry 4:1. doi: 10.4172/2471-4372.1000158


An enduring debate has undoubtedly existed with respect to the legitimacy of Dissociative Identity Disorder (DID). While some contend that authentic dissociative experiences are supported by strong associations with a history of childhood abuse, others argue that the purported dissociative experiences are identity enactments sustained and reinforced by states of extreme suggestibility and fantasy proneness. However, the combination of theoretical speculation and empirical findings has thus far failed to provide unequivocal evidence for the existence of alters as either metaphors for different emotional states or truly autonomous entities that are capable of willful action. This article reviews the available evidence from a number of investigations examining the memory performance, behavioral manifestations, and physiological profiles of alters in DID. It concludes that neither memory studies nor psychobiological studies have delivered compelling evidence that alters of DID patients exist in a factual sense. It moreover demonstrates that some studies have suffered from methodological weaknesses, while findings from the overall body of literature are open to multiple interpretations. As such, they do not refute an interpretation of alters in terms of metaphors for disparate affective states. Recommendations for future studies aiming to investigate the phenomenon of alter are discussed.

Keywords: Dissociative identity disorder; DID; Alters; Memory; Psychophysiology


The arguably unparalleled amount of controversy evoked by dissociative identity disorder (DID) is aptly illustrated by the longstanding debate concerning its alleged etiology, phenomenology, and validity as a diagnostic category. On the one hand, proponents of the iatrogenesis/sociocognitive theory, which was advanced by Spanos [1], contend that DID patients are typically attention seeking individuals whose enactment of multiple identities is encouraged and maintained by the suggestions and cuing of clinicians as well as by popular media. On the other hand, the posttraumatic model proposed by Gleaves [2] conceptualizes DID as a legitimate psychiatric condition emanating from overwhelming and persistent childhood trauma. Rather than engaging in enactments of several identities, DID sufferers are believed to develop creative coping mechanisms such as alters as part of an overall survival strategy.

Further complicating these issues is a lack of professional consensus on whether alters are in fact authentic, autonomous, and volitional entities [3]. That is, while some authors claim that they are distinct information processing identities capable of developing compartmentalized memories, convictions, and idiosyncratic behavioural patterns, others assert they are little more than metaphors for divergent affective states [3,4]. Since expert opinion on the nature of alters can have profound implications for legal determinations of culpability in cases involving DID patients, some form of consensus on the status of alters is highly desirable. That being said, it is the objective of this article to review the extant research on alter activity in order to determine whether it has provided compelling evidence for either a literal interpretation of alters as discrete entities or the view that alters are enactments or metaphors for disparate affective states. In hopes of elucidating this position, the limited number of studies on physiological reactivity and implicit/explicit memory performance in DID are examined.

The studies included in this review were retrieved from searches of commonly used psychological and psychiatric databases, such as PsycINFO and MEDLINE (respectively). The studies selected investigated individuals with DID and fulfilled one or more of the following criteria: (1) the studies examined implicit and/or explicit memory transfer across alters; (2) the studies examined unique neurophysiological profiles of alters (e.g., regional cerebral blood flow (rCBF), visual functioning, cerebral hemodynamic activity); (3) DID was confirmed with a valid diagnostic tool; (4) comparisons were made with other samples; and (4) DID was differentiated from other disorders, such as Post-Traumatic Stress Disorder (PTSD).

Memory Performance Research

One of several attempts to address the paucity of objective documentation on DID was initialy made by Lewis, Yeager, Swica, Pincus, and Lewis [5],who investigated the link between extreme childhood abuse and dissociative symptoms in 11 males and one female diagnosed with the disorder and convicted of murder. In order to determine what, if any, dissociative symptoms subjects experienced in childhood and/or adulthood, researchers conducted psychiatric evaluations wherein questions were posed about trances, amnesic states, lifelike imaginary companions, alterations in voice, demeanour, and penmanship, time loss, and hallucinations. Handwriting samples were obtained for analysis from subjects’ letters, journals, and other handwritten documents, all of which were produced long before the psychiatric interviews were administered. A substantial quantity of corroborative data purportedly supporting the existence of DID symptoms was procured not only from the descriptive accounts of friends and family members, but also from police documents and audiotapes, trial transcripts, school, medical, and social service records, as well as the observations of co-workers, cell mates, jail guards, a legal assistant and a school teacher. Aside from the finding that all 12 subjects endured a tremendous amount of physical and sexual torture throughout childhood, it was determined that they exhibited partial or complete amnesia for their childhood abuse, trance-like states, and strikingly noticeable fluctuations in voice, demeanour, and handwritings during adulthood. These fluctuations, in particular, were interpreted by the authors as objective evidence for the emergence of distinct alters in DID [5].

One might certainly entertain the possibility that patients in the abovementioned study did not engage in malingering. It is quite unlikely that they contrived the details of their murder and commenced to feign dissociative symptoms years in advance with the mere anticipation of someday perpetrating the criminal act and shifting culpability onto an alter. Or so the argument might go. However, let us consider the performance demands of an actress assigned to play two or more characters in a theatrical play. The audience’s ability to distinguish between the various roles depends predominantly on her adeptness in modifying her demeanour and vocal qualities such as tempo and inflection. Since the findings of Lewis and colleagues [5] might reflect the possible adoption of these and other superficial differences by DID patients and not abnormalities in their procedural motor responses, we are precluded from concluding that alters are genuine entities. Furthermore, as the following study demonstrates, methodologically sound experiments equipped with adequate control groups have generated results that are even less amenable to literal interpretations of alters.

Following several claims that compartmentalization of procedural learning and memory is facilitated by inter-identity amnesia in DID, Huntjens, Postma, Woertman, van der Hart and Peters [6] systematically investigated procedural memory functioning in 31 DID patients and 50 controls. Half of the control participants were further assigned to a ‘simulators’ group, whose task was to mimic DID, and half remained in a control group. Two identities, each of which was entirely amnesic for the personal biography and actions of the other, were selected by DID subjects to engage in the experiment, whereas simulating controls were required to invent a fictitious amnesic identity. All subjects participated in a serial reaction time (SRT) task containing 8 blocks of 120 trials each, with blocks 1 and 8 adhering to a random sequence, and the remaining blocks following a repeating sequence. Upon completion of the first 4 blocks in one alter, DID sufferers were explicitly instructed to switch to the second, presumably amnesic identity. In contrast, simulators performed the first 4 blocks as themselves, and were subsequently instructed to assume their imaginary identity and perform blocks 5-8 as thought they possessed no knowledge of the learning material just viewed. It was hypothesized that, following the transition to a second identity, inter-identity amnesia in DID subjects would manifest itself as an increase in reaction times that is indicative of a failure to remember previous learning of the repeating sequence. Simulating controls, however, were expected to be unsuccessful in their attempts to mimic this specific pattern of amnesia in their imaginatively produced alter.

Although results confirmed the absence of inter-identity transfer of implicit memory in DID sufferers, simulators effectively imitated this pattern of amnesia between identities despite being devoid of explicit knowledge about the repeating sequence they were exposed to during the SRT procedure. The inability of both DID patients and simulating controls to identify and record the correct ordered sequence revealed that explicit processing of the sequence itself was largely nonexistent among these individuals. Nevertheless, in view of simulators’ capacity to mimic interalter amnesia, the apparent failure of patients to transfer implicit-procedural memory across identities cannot (without ambiguity) be interpreted as evidence for the existence of discrete and compartmentalized alters. In fact, patients’ performance could just as well signify deliberate enactment of interalter amnesia. But perhaps even more importantly, Huntjens et al. study [6] demonstrates that simulation of implicit memory impairments can be executed even in the absence of explicit stimulus knowledge, insofar as subjects possess detailed information about the amnesic profile associated with DID.

In a more recent study, Huntjens et al. [7] investigated autobiographical memory specificity in 12 female patients diagnosed with DID, 27 female PTSD patients with a reported history of sexual and/or physical abuse, 26 female participants simulating alters, and 31 healthy control participants who were also female. They examined autobiographical memory specificity for trauma versus apparently normal alters, as well as autobiographical memory specificity for DID patients compared with PTSD patients and healthy controls. It was found that although DID patients retrieved trauma-related memories faster in their trauma state than in their seemingly normal state, the faster responding in this state was a general finding apparent on all retrieval trials. This was purported to reflect a faster responding strategy aimed at precluding the recall of trauma-related memories triggered by the test cue words. Contrary to the authors’ hypothesis of differential alter responding, patients did not retrieve fewer specific memories in their purported normal identities relative to their trauma alters, nor were there any differences between alters in the type of overgeneral memories retrieved. Overgeneral memories reflected a whole class of events and consisted of extended memories that referred to an extended period of time, categorical memories (e.g., “every Time I had to sit in the basement”), and semantic associates (e.g., “I am a clumsy person”) [7]. There were likewise no differences between DID patients’ trauma alters and their apparently normal alters in the retrieval of negative or older specific memories. In contrast, the simulators included in the study performed as instructed and retrieved a greater number of trauma-related memories in their trauma alter while dating these memories as older in comparison to their ostensibly normal alter. Irrespective of alter, DID patients demonstrated an absence of memory specificity that very closely resembled the lack of memory specificity exhibited by PTSD. Given the similarities found between DID and PTSD patients in symptomatology and memory functioning, the authors suggested the possibility of a joint (complex) PTSD dimensional classification that functions on a continuum from nondissociative to dissociative—what we might today describe as a dissociative subtype of PTSD [7,8]. Despite their findings, it should be noted that the authors’ study did not adequately control for other comorbid psychopathology such as depression, which has been shown to contribute independently to reduced autobiographical memory specificity [9].

Evidently, the abovementioned discoveries in no way attenuate the disagreement among academics about the exact type of memory disturbance plaguing (the alters of) DID sufferers. For instance, in a study containing 12 individuals diagnosed with DID [10], investigators found that explicit memory for neutral and emotionally valenced stimulus words was significantly reduced between alter identities, while implicit memory performance was wholly maintained. Other experimental studies have reported inter-identity transfer, or at the very least leakage of both explicit and implicit memory, thereby contributing to the inconsistent pattern of findings on memory functioning across alters [6,10-14].

Psychophysiological Research

Among those espousing a biological approach to the understanding of alters, Matthew and colleagues [13], set out to investigate patterns of regional cerebral blood flow (rCBF) in a DID patient with 3 different identities and 3 non-disordered controls. Although rCBF measurements could not be obtained for the patient’s third alter due to its infrequent manifestation, they were carried out for her first pre-integration alter and, after a half-hour interval, for her second pre-integration alter. Eight weeks later and prior to discharge from the hospital, it was determined that the 3 alters had fused into a single, relatively stable post-integration personality characterized by an acceptance of the patient’s adverse childhood experiences. At this point, the patient’s CBF was measured twice more at half-hour intervals, and her patterns of CBF distribution were compared with those displayed by normal controls. It was determined that in her second identity, which was a timorous young girl tormented by visions of her sadistic stepfather, parieto-occipital regions of the patient’s cortex exhibited modest increases in flow values (hyperperfusion) [13].

Curiously, the researchers attributed this increase in cerebral circulation to mechanisms of perception rather than to the presumed “switch” to an alter with a different physiological profile. In contrast to controls, a marked hyperperfusion in the patient’s right temporal lobe was also detected when she assumed the second identity. It was thus surmised that since the temporal lobe plays a vital role in memory processes, its hyperperfusion in the DID patient might reflect a resurgence of affectively valenced memories of childhood sexual abuse in the second “fearful child” alter. While interesting, this study nonetheless failed to include any statistical tests that would have enabled researchers to make inferences concerning the significance of changes in rCBF. Even if such tests were performed and revealed statistically significant results, the evidence put forth by the investigators would remain insufficiently compelling to undermine the view that alters are metaphors for various affective states [13].

With respect to sensory phenomena, Miller S sought to ascertain whether patients would displayed greater inter-identity variability in visual functioning than control participants simulating the disorder [14]. Nine patients attended an initial appointment wherein a psychiatrist identified, interviewed, and videotaped 3 discrete alters, as well as a second appointment in which each alter was instructed to complete a detailed ophthalmic examination without interference from other alters. After viewing the film Sybil in order to gain a more comprehensive picture of the disorder, healthy controls watched a videotaped interview of a DID patient, whose 3 identities they were required to impersonate during the subsequent ophthalmic assessments [14,15]. All participants and their corresponding alters underwent the same ophthalmological tests intended to measure foveal and peripheral vision, refraction of light, pupillary size, and eye-muscle balance and coordination. Univariate analyses of variance revealed that, in comparison to simulating controls, patients evinced significantly greater variability in optical functioning across alters on ophthalmic measures of corrected and uncorrected visual acuity, peripheral visual fields, ocular muscle balance, and light refraction. However, unlike antecedent research that indicated clinically significant differences between distinct alters on pupillary size measures, Miller’s sample of DID patients manifested no interidentity alterations in pupil dimensions [14].

While these results might be construed by some as evidence for the notion that alters are independent entities, the fact remains that the specificity of these fluctuations in visual function to DID has not been demonstrated by this or any other studies. Since no comparison groups of patients with other psychiatric disorders were incorporated in the study’s design to help determine the existence (or lack thereof) of such specificity, we are left free to speculate that the reported pattern of visual functioning might also occur in, say, borderline personality disorder or schizophrenia. Moreover, keeping in mind that the alter activity characteristic of DID has often been postulated to be the result of a predisposition to self-hypnosis, it is worth noting that hypnotic alterations in visual acuity have been revealed by several studies [16-19]. As such, the possibility that spontaneous self-hypnosis was partly responsible for the observed variability cannot be dismissed.

Touted as a substantial contribution to the physiobiological underpinnings of alters at the time of its publication, the fMRI study conducted by Tsai and colleagues on one female patient with DID also appeared to endorse literal conceptualizations of alters [20]. Their subject underwent a procedure wherein she repeatedly “switched” from her native (adult) alter to her sexually abused young girl alter (Guardian) and then back to her native state. The volitionally induced switches occurred in response to the instruction “switch to... now,” and were indicated by the subject via a specific button press. Throughout this process, her cerebral hemodynamic activity was measured by fMRI. Interestingly, results showed that the switch from the native alter to Guardian was accompanied by significantly less activity in both hippocampi, medial temporal, parahippocampal, and pallidal regions, as well as tiny regions of the substantia nigra. Conversely, the shift back to the native alter involved solely an increase in hippocampal activity. In light of these findings, it was tentatively concluded that the switches between alters in DID might result from temporohippocampal or hippocampal-mediated cognitive processes, with both declarative and nondeclarative memory structures possibly involved in the switching phenomenon. However, rather than elucidating how the proposed declarative/nondeclarative memory structures might influence the switching process, the researchers merely cited a number of studies indicating either that the hippocampus and structures in the medial temporal lobe are critical for declarative memory, or that the nigrostriatal system has implications for certain procedural memory functions [20]. The question of whether the changes in brain activity are specific to DID moreover remained unanswered because the patient was diagnosed with comorbid Post Traumatic Stress Disorder (PTSD) and DID.

Limitations and Concluding Remarks

Notwithstanding the drawbacks of this study, it must be acknowledged that the preponderance of research on the physiological reactivity of alters suffers from methodological weaknesses that render the making of any definitive generalizations about their status difficult, if not impossible. Aside from a failure to implement rigorous designs that assess the specificity of neurophysiologic, autonomic, or sensory phenomena to DID, these studies generally have demonstrated an excessive reliance on single case study designs, a lack of experimental blinds and (adequate) controls, and a taken-for-granted assumption that self-initiated and self-controlled “switching” is a capacity possessed by all patients with DID [21-25]. Studies of explicit and implicit memory, on the other hand, have been conducted in the absence of well-formulated theories about memory function in DID and have generated inconclusive findings that do not particularly favour any side of the interpretive debate [6,7,26-29]. Thus, two courses of action ought to be taken to determine whether alters are indeed autonomous entities with distinct memory processes and physiological profiles or merely metaphors for divergent affective states. Firstly, greater attempts ought to be made to produce clearly articulated theories about memory functioning and the precise type of memory disturbance (e.g., autobiographical, semantic, procedural) that characterizes the disorder [24]. Secondly, and with these theories in mind, future memory and psychobiological studies should incorporate double-blind designs with control groups composed of other psychiatric populations, subjects high on hypnotic suggestibility and/or fantasy proneness, normal controls with a history of severe childhood trauma, and adequate adjustments for other comorbid psychopathology. This will enable us to determine whether any significant differences exist between these populations on factors such as authentic and/or simulated dissociative identity states. Conducting such methodologically sound research is critical to determine whether alters are genuine entities in DID, particularly given the implications of this diagnosis for clinical decision-making and mental state defenses such as insanity and diminished capacity. Until this occurs, the ongoing debate about the clinical validity and nosological status of alter personalities is likely to continue. It concludes that neither memory studies nor psychobiological studies have delivered compelling evidence that alters of DID patients exist in a factual sense. As such, there is no refusal for interpretation of alters in terms of metaphors for disparate affective states.


Track Your Manuscript