Read Time: 12 minutes
The clinical puzzle arrives in one of two shapes. The first is the teenager who presents as thoughtful, verbally skilled, and somehow completely unreachable. She answers every question, never misses an appointment, describes her trauma history in language her therapist could publish. And nothing moves. The second is the teenager who swings, often within the same session, from charming engagement to flat shutdown to a flash of contempt that arrives like weather. The clinician is left wondering which teenager is the real one.
Both of them are. What the clinician is meeting in the room is not a personality, stable and whole. It is a system of parts, each of which developed for a reason, and each of which is trying to keep the teenager alive in the way it was built to. This is not a metaphor. It is, increasingly, how the trauma field understands complex presentations in clients of any age, and there are specific reasons to think it is especially true of adolescents.
The observation that the self is multiple, and that trauma fragments what might otherwise integrate, is not a new one. William James wrote about it. Pierre Janet wrote about it in the nineteenth century. But the two frameworks that most shape contemporary parts work in trauma therapy are Internal Family Systems, developed by Richard Schwartz beginning in the 1980s, and the Theory of Structural Dissociation, laid out by Onno van der Hart, Ellert Nijenhuis, and Kathy Steele in The Haunted Self (van der Hart et al., 2006).
The two models use different vocabularies. IFS speaks of protectors (both managers and firefighters) and exiles, all orbiting a core Self that is fundamentally undamaged and capable of leading the system when given the chance. The Theory of Structural Dissociation describes the personality as divided, under conditions of chronic traumatization, into Apparently Normal Parts that handle daily life and Emotional Parts that carry the unintegrated trauma and its defensive responses (Steele et al., 2005). These are not the same taxonomy, and conflating them creates clinical confusion.
But they converge on the same underlying observation. The mind under chronic stress does not stay unitary. It fragments, often predictably, along defensive lines. And the parts that emerge are not pathological inventions. They are adaptations that made sense at the time of the injury, even if they now produce the symptoms the client is in the room to address. The evidence base for parts-based treatment continues to accumulate, with pilot effectiveness studies of IFS for PTSD among survivors of childhood trauma (Hodgdon et al., 2022) and the first randomized controlled trial of a group-based IFS-derived treatment showing significant reductions in PTSD symptoms (Joss et al., 2026). For clinicians working with complex presentations, parts work is no longer peripheral.
There is a developmental argument for taking parts work seriously with any trauma client. There is a stronger one for taking it seriously with teenagers specifically.
Adolescent identity is, by definition, in formation. Erikson placed the central task of adolescence as identity versus role confusion, the work of synthesizing a coherent sense of self out of the disparate identifications, roles, and possibilities a young person is actively experimenting with (Orenstein & Lewis, 2024). That synthesis is not supposed to be complete in adolescence. It is supposed to be in progress.
In a healthy trajectory, this is generative. The teenager tries on identities, tests roles, integrates feedback, and slowly stitches together a sense of who she is. In the presence of complex trauma, that same openness becomes vulnerability. Parts that might have differentiated and then reintegrated in adulthood instead calcify. Protective roles that would have softened as the adolescent developed other coping capacities become load-bearing structures. The eating disorder part, the people-pleasing part, the numbed-out part, the contemptuous part: all of them can become more fixed in adolescence precisely because the nervous system is under the kind of developmental pressure that rewards whatever strategy works right now.
The Theory of Structural Dissociation offers a specific developmental frame for this. Van der Hart, Nijenhuis, and Steele argued that the child’s personality is not innately integrated but gradually synthesized from loosely connected state-based systems over the course of development. Chronic or early trauma disrupts that synthesis (Katerelos et al., 2017).
Adolescents with histories of chronic interpersonal trauma arrive in clinical care mid-synthesis, carrying parts that the standard developmental process would otherwise have integrated but now may not.
This is why parts work with teenagers is not an optional add-on. For many adolescent clients, particularly those with complex or developmental trauma, the parts are the clinical picture.
The part that holds it together is often the one a clinician notices last. She is the one who makes it through school, manages the caseload of social expectations, remembers appointments, appears in session as "fine." In structural dissociation terms, she is the Apparently Normal Part, oriented toward daily functioning and invested in avoiding trauma material that might destabilize her. In IFS terms, she is usually a high-functioning manager. Either way, her job is to keep the trauma out of sight.
The parts that make the clinical picture harder to miss are the ones that intrude when her system cannot hold. The rage that flashes when a parent makes a bid for closeness. The depressive collapse that follows a fight with a friend. The teen who smokes weed every day because the part that wakes up at three in the morning with panic is the one the cannabis settles. The teen who cuts, not because she wants to die but because cutting is the only way one of her parts has found to move emotional pain out of the body.
These are the parts a clinician tends to label as symptoms. They are more usefully understood as protectors. In IFS terms, they are firefighters, parts that emerge when exiled pain threatens to flood the system and whose job is to put the fire out by any means available (Clinical Psychologist scoping review, 2025). In the language of structural dissociation, they are Emotional Parts engaging defensive action systems, fight or flight or freeze, that got locked in during the original traumatization.
Common teen protector patterns include the sarcasm that makes the therapist feel briefly small, the intellectualization that turns every clinical question into an essay question, the people-pleasing that presents as maturity, the aggression that appears to be character, and the substance use that appears to be choice. None of these are who the teenager is. They are what she does to manage what she is carrying.
Most therapists trained in parts work learned language developed for adult clients. The challenge in adolescent sessions is to keep the clinical precision of the framework while letting the language meet the teenager where she actually is.
The mapping itself can be straightforward. The clinician asks the teenager to name the parts most familiar to her and describe what each one does. Some teens are fluent in this immediately, because they have grown up in an internet culture that already speaks in parts. They talk about their anxious part, their avoidant part, their rage part. The framework is not foreign; it is almost already theirs.
Others resist the concept entirely. Parts language can feel weird, too intimate, or uncomfortably close to what a teenager might be actively working to avoid acknowledging about herself. In those cases, the clinician does not need the teen to use the word "part" at all. She can use roles, characters, or versions. She can ask which version of the client shows up at home versus at school versus in the therapy room. She can notice out loud when the teenager seems to shift, and ask with curiosity what just happened. The work is the same. The vocabulary adjusts.
Two clinical principles hold regardless of vocabulary. The first is that every part is trying to help. This is a non-negotiable IFS premise and it applies to the part that cuts, the part that lies to parents, the part that appears to want therapy to fail. Parts develop extreme roles not because they are malicious but because they were pushed into those roles by circumstances that gave them no other option (Schwartz & Sweezy, 2020). When a clinician extends curiosity rather than correction toward a part the teenager is ashamed of, something shifts quickly.
The second principle is that mapping is diagnostic, not interventional. The clinician is not trying to fix parts in this phase. She is learning the system so that EMDR preparation, target selection, and processing can be designed around what is actually there. This parallels how ego state work integrates across the eight phases of EMDR in adult practice, with the added translation of keeping the language developmentally congruent.
Internal Healing Dialogues, a technique Jon Roberts teaches in his adolescent EMDR work, are an application of parts work that deserves specific attention. The dialogue is a structured conversation, facilitated by the clinician, between two parts of the teenager’s system: often a protector and the younger, exiled part the protector is guarding. The work of the dialogue is not to resolve the trauma. It is to shift the internal relationship between parts enough that processing becomes possible.
In IFS terms, the dialogue asks the protector for permission to work with the part it is protecting. Until that permission is granted, processing the exiled material is premature. The protector will sabotage the work, not out of defiance but because its job is to keep the exile hidden. In structural dissociation terms, the dialogue begins to resolve what van der Hart, Nijenhuis, and Steele called trauma-related phobias, specifically the phobia of dissociative parts and the phobia of traumatic memories that together prevent integration (Steele et al., 2005).
With teenagers, how the technique is introduced matters as much as the technique itself. A clinician who announces she is going to facilitate a dialogue between parts may lose the room. A clinician who says, "Can I check in with the part of you that gets angry when we talk about your mom, and see what it needs from us before we keep going?" is doing the same work without tripping the adolescent’s sensitivity to performativity. The technique is identical. The packaging is adapted for the brain in the room.
What often surprises clinicians new to this work is how quickly teenagers take to it once the framing is right. Adolescents have been negotiating internal conflict every day of their lives. Given a framework that honors what they are already doing, many move through parts work faster than their adult counterparts. The developmental openness that makes them vulnerable to fragmentation also makes them unusually responsive to integration when the conditions are right.
One of the most practically important things parts work offers adolescent EMDR is a better way to generate targets. In standard EMDR training, target selection often proceeds from presenting symptoms or trauma history. That approach works for many clients. It frequently falls short with teenagers whose symptoms are the protective output of parts whose actual material is still unmapped.
Parts work reframes target generation as a question about which part holds what. The client is not "anxious." One of her parts is. When that part is contacted, it often turns out to hold a specific memory or set of memories that the presenting self does not consciously connect to the anxiety. Those memories become the targets. The work becomes more precise because it is aimed at the right part of the system.
Trauma-related phobias, as described in the structural dissociation literature, are a useful target-generation frame. Van der Hart and colleagues identified several: the phobia of inner experience, the phobia of attachment, the phobia of attachment loss, and the phobia of traumatic memories themselves. Each shows up in adolescent presentations in specific patterns. The teen who cannot tolerate being alone with her own emotions has a phobia of inner experience. The teen who cycles between clinginess and withdrawal in her friendships has a phobia of both attachment and attachment loss. Naming the phobia gives the clinician a clear entry point.
Can parts work be done with teens who aren’t familiar with IFS? Yes. Prior familiarity is not required. What is required is that the clinician has a working framework and can translate it into language the teen can meet. Many adolescents pick up the concept within one session. Others never use the word "part" and still engage through roles, characters, or observations of how they change across contexts.
What do I do when a part refuses to communicate? Treat the refusal as information. A part that will not communicate usually has a good reason, often protective. Rather than trying to bypass the refusal, the clinician asks what the part is guarding against. What would it be worried would happen if it engaged? This reframes the work from coercion to collaboration, which is generally more productive clinically.
Is parts work safe with adolescents who have active high-risk behaviors? Parts work is typically indicated for these presentations, not contraindicated. The firefighter or Emotional Part driving self-harm or substance use is not going to respond to behavioral intervention alone because the behavior is serving a protective function. What shifts the system is working with that part in a way that honors its role and begins to offer alternatives. Specialized protocols like the Feeling-State Protocol, DeTUR, and the LOU-A are forms of parts-informed work designed specifically for this population.
How does parts work fit with EMDR fidelity? It fits inside the Adaptive Information Processing model and the eight-phase structure. Parts work happens in preparation (Phase 2), shapes target selection and assessment (Phase 3), and informs how the clinician responds during desensitization and installation (Phases 4 and 5). The protocol stays intact. The clinical lens becomes more accurate.
Clinicians who begin working with adolescent parts consistently report the same shift. Sessions that used to feel like they were happening to a teenager who could not quite participate start to feel collaborative. The teen becomes an active interpreter of her own internal system. The clinician is no longer asking her to perform insight she does not yet have; they are mapping something together. Preparation takes longer than a protocol-only approach would suggest, and the processing that follows is more stable because the preparation was honest.
For teens whose presentations include dissociation, fragmentation, or high-risk behaviors, the parts framework is often the difference between EMDR that stalls and EMDR that moves. It does not replace clinical skill, and it does not replace training in the specific adaptations adolescent work requires. But it gives a clinician a way to understand what she is meeting when she sits across from a sixteen year old carrying more than one self in the same body, and who needs someone in the room who knows how to work with all of them.
Build your adolescent parts work skills.
EMDR with Adolescents: Parts Work, Modified Processing, and High-Risk Behaviors is a five-module on-demand training with Jon Roberts, LCSW, CAS. It walks clinicians through mapping parts in teen systems, working with structural dissociation and IFS frameworks together, and using Internal Healing Dialogues to unblock stuck processing. 5 CEs (EMDRIA, ASWB ACE, NBCC). Available July 30, 2026.
I’m Ready to Work With Parts, Not Around Them →
Hodgdon, H. B., Anderson, F. G., Southwell, E., Hrubec, W., & Schwartz, R. C. (2022). Internal family systems (IFS) therapy for posttraumatic stress disorder (PTSD) among survivors of multiple childhood trauma: A pilot effectiveness study. Journal of Aggression, Maltreatment & Trauma, 31(1), 22-43. https://www.tandfonline.com/doi/full/10.1080/10926771.2021.2013375
Joss, D., Comeau, A., Chevannes, R., Parry, G., Rea, H. S., Barria, J., Bumpus, C., Rector, A., Rajan, A., Rosansky, J., Rice, F. K., Ward, M. C., Tobiasz Veltz, L., Ally, D., Rosenberg, L. G., Sweezy, M., Lovas, D., & Schuman-Olivier, Z. (2026). A randomized controlled trial of an online group-based internal family systems treatment for posttraumatic stress disorder: The Program for Alleviating and Resolving Trauma and Stress (PARTS) study. Psychological Trauma: Theory, Research, Practice, and Policy. Advance online publication. https://doi.org/10.1037/tra0002089
Katerelos, M., Le Breton, H., Lockhart, M., & Vallianatos, H. (2017). EMDR therapy and the theory of structural dissociation of the personality in severe interpersonal trauma of young adolescents. Journal of EMDR Practice and Research, 11(3). https://pmc.ncbi.nlm.nih.gov/articles/PMC5632787/
Orenstein, G. A., & Lewis, L. (2024). Eriksons stages of psychosocial development. In StatPearls. StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK556096/
Redfern, E., O’Brien, G., Plester, H., & Arnold, C. (2025). Exploring the evidence for Internal Family Systems therapy: A scoping review of current research, gaps, and future directions. Clinical Psychologist. https://www.tandfonline.com/doi/full/10.1080/13284207.2025.2533127
Schwartz, R. C., & Sweezy, M. (2020). Internal family systems therapy (2nd ed.). The Guilford Press.
Steele, K., van der Hart, O., & Nijenhuis, E. R. S. (2005). Phase-oriented treatment of structural dissociation in complex traumatization: Overcoming trauma-related phobias. Journal of Trauma & Dissociation, 6(3), 11-53. https://pubmed.ncbi.nlm.nih.gov/16172081/
van der Hart, O., Nijenhuis, E. R. S., & Steele, K. (2006). The haunted self: Structural dissociation and the treatment of chronic traumatization. W. W. Norton.