Elaborating the assimilation model: Introduction to a special section on case studies of setbacks within sessions and therapeutic collaboration

Abstract

This article introduces a Special Section of case studies that focus on therapeutic collaboration and setbacks in the process of assimilation with the aim of contributing to the evolution of the assimilation model of therapeutic change. The first study examined setbacks in two depression cases (a good vs. a poor outcome) treated with emotion-focused therapy. The second article traced how therapist activities and positions toward internal voices were associated with setbacks in a case treated with linguistic therapy of evaluation. The third article studied contributions of therapeutic collaboration for both advances and setbacks in assimilation in two contrasting cases treated with emotion-focused therapy. The fourth and final article analyzed the therapeutic collaboration in episodes of ambivalence in two cases of narrative therapy (one good outcome, one poor outcome) reflecting on the implications for the assimilation model’s perspective on the therapeutic relationship. This Introduction concludes by offering some suggestions for theory-building within the assimilation model.

Questo articolo introduce una sezione speciale su studi di casi che hanno come focus la collaborazione terapeutica e le battute d'arresto nel processo di assimilazione allo scopo di contribuire allo sviluppo di un modello di assimilazione del cambiamento terapeutico. Il primo studio ha esaminato le battute d'arresto in due casi di depressione (uno con buon esito rispetto a uno con esito sfavorevole) trattati con una terapia focalizzata sulle emozioni. Il secondo articolo descrive come le attività e le posizioni del terapeuta nei confronti delle voci interne erano associate con battute d'arresto in un caso trattato con una terapia linguistica di valutazione. Il terzo articolo ha studiato i contributi della collaborazione terapeutica sia per i progressi che per le battute d'arresto nell'assimilazione in due casi contrastanti trattati con una terapia focalizzata sulle emozioni. Il quarto e ultimo articolo ha analizzato la collaborazione terapeutica in episodi di ambivalenza in due casi di terapia narrativa (uno con buon esito e l'altro con esito sfavorevole) con riflessioni sulle implicazioni di una prospettiva legata al modello di assimilazione sulla relazione terapeutica. Questa introduzione si conclude offrendo alcuni suggerimenti per theorybuilding all'interno del modello di assimilazione.

Este artigo introduz uma Secção Especial de estudos de caso com foco na colaboração terapêutica e em retrocessos no processo de assimilação, desenvolvidos com o objetivo de contribuir para a evolução do modelo de assimilação da mudança terapêutica. O primeiro estudo analisou retrocessos em dois casos de depressão (um de sucesso e um de insucesso) seguidos em terapia focada nas emoções. O segundo artigo delineou o modo como as atividades e posições do terapeuta face a vozes internas estavam associadas a retrocessos num caso seguido em terapia linguística de avaliação. O terceiro artigo estudou as contribuições da colaboração terapêutica tanto para avanços como para retrocessos na assimilação, em dois casos contrastantes seguidos em terapia focada nas emoções. O quarto e último artigo analisou a colaboração terapêutica em episódios de ambivalência em dois casos de terapia narrativa (um de sucesso e um de insucesso), refletindo acerca das implicações para a perspetiva do modelo de assimilação sobre a relação terapêutica. Esta Introdução conclui com algumas sugestões para a construção de teoria no âmbito do modelo de assimilação.

Dieser Artikel gibt eine Einführung zu einem besonderen Abschnitt von Fallstudien, welche auf der therapeutischen Kollaboration und Rückschritte im Assimilationsprozess fokussiert sind, mit dem Ziel, einen Beitrag zur Weiterentwicklung des Assimilationsmodells der therapeutischen Veränderung zu leisten. Die erste Studie untersuchte Rückschritte in zwei Fällen von Depression (mit gutem vs. schlechtem Ergebnis), welche mit emotionsfokussierter Therapie behandelt wurden. Der zweite Artikel verfolgte, wie Aktivitäten des Therapeuten und Stellungnahmen bezüglich innerer Stimmen mit Rückschritten in einem Fall, welche mit linguistischer Evaluationstherapie behandelt wurde, zusammenhingen. Der dritte Artikel untersuchte Beiträge therapeutischer Kollaboration für sowohl Fortschritte als auch Rückschritte der Assimilation in zwei kontrastierenden Fällen, welche mit emotionsfokussierter Therapie behandelt wurden. Der vierte und letzte Artikel analysierte die therapeutische Kollaboration in Phasen von Ambivalenz in zwei Fällen narrativer Therapie (ein gutes und ein schlechtes Ergebnis) und reflektierte die Implikationen für die Perspektive des Assimilationsmodells auf die therapeutische Beziehung. Diese Einführung endet mit einigen Vorschlägen zur Theorienbildung innerhalb des Assimilationsmodells.

本文是以專論形式介紹個案研究中的治療合作與同化歷程中的退步經驗，藉此能促進治療改變的同化模式之發展。第一篇研究檢視兩位接受情緒焦點治療憂鬱症的個案（分別有良好與不佳的成效）的退步情形。第二篇是追蹤一位治療師的治療活動與其對案主的內在語言，和一位接受語意評估治療個案退步的經驗有何關連。第三篇係以兩位接受情緒焦點治療的相反類型個案為對象，研究治療合作對於同化模式的進展和退步各有何貢獻。第四篇也是最後一篇則是分析兩位接受敘事治療個案（一位具好成效，另一位成效不佳）在矛盾兩難情境的治療合作情形，反映出同化模式對於治療關係觀點的意涵。最後，這本文提供同化模式於理論建構上的建議。

Keywords:

• assimilation model
• setbacks
• therapeutic collaboration
• ambivalence
• therapist activities

Parole chiave::

• modello di assimilazione; battute d'arresto; collaborazione terapeutica; ambivalenza; attività di terapeuta

Schlüsselwörter::

• Assimilationsmodell
• Rückschritte
• therapeutische Kollaboration
• Ambivalenz
• Aktivitäten des Therapeuten

關鍵詞：:

• 同化模式
• 挫敗
• 治療合作
• 矛盾兩難
• 治療師的活動

The four case studies in this special section on elaborating the assimilation model grew from a panel presented at the 2014 Society for Psychotherapy Research International Meeting in Copenhagen. It includes studies on both good outcome and poor outcome cases treated with several kinds of therapy, including emotion-focused therapy, narrative therapy, and linguistic therapy of evaluation. The first two articles (Caro Gabalda, Stiles, & Pérez Ruiz, 2016; Mendes et al., 2016) address the nature and function of setbacks in the therapeutic process. The second two (Ribeiro, Braga, et al., 2016; Ribeiro, Cunha, et al., 2016) address the process of therapeutic collaboration. Whereas most previous assimilation studies have focused on clients’ processes of assimilation, all of these studies explicitly, attend to the role of the therapist.

The Assimilation Model and Setbacks in Assimilation

The assimilation model (Stiles, 2002, 2011; Stiles et al., 1990, 1991) is a theory of psychological change that has evolved from psychotherapy process research (cf., Llewelyn & Hardy, 2001). The model describes the self as composed of multiple internal voices derived from past experiences, which serve as the person’s cognitive, affective, and behavioral resources, emerging to speak or act when required (Dimaggio & Stiles, 2007). New experiences are normally assimilated to the self and add to the person’s variety of resources. If, however, an experience is too painful, threatening, or inconsistent with the usual self, it may remain unassimilated; nevertheless, it may be addressed by circumstances and emerge to cause pain or maladaptive behavior. We use the term dominant voice to refer to expressions emerging from the assimilated experiences constituting the usual self and we use the term nondominant voice to refer to expressions emerging from unassimilated or partially assimilated problematic experiences.

The model describes a sequence of eight stages anchoring the process of assimilating problematic experiences, summarized in the Assimilation of Problematic Experiences Scale or APES (Stiles, 2002; Stiles et al., 1991). The stages, numbered 0 to 7, describe the changing relation of the nondominant voice to the dominant community of voices: (0) warded off/dissociated, (1) unwanted thoughts/active avoidance, (2) vague awareness/emergence, (3) problem statement/clarification, (4) understanding/insight, (5) application/working through, (6) resourcefulness/problem solution, and (7) integration/mastery. Thus, fully assimilating a problematic experience involves affective (particularly stages 0–2), cognitive (particularly stages 3–4), and behavioral (particularly stages 5–7) processes. In a particular client’s therapy, the process of assimilation may begin or end at any point on the APES, and any overall advance may be considered as therapeutic progress.

Theoretically, all productive work in therapy takes place within or at the edges of a delimited segment of the APES continuum that lies between the APES level the client can reach alone and the APES level the client can achieve in collaboration with the therapist. This segment is called the therapeutic zone of proximal development (ZPD; Leiman & Stiles, 2001), after Vygotsky’s (1978) concept of an analogous working zone in children’s intellectual development. The therapeutic ZPD shifts up the APES as the problematic experience is assimilated.

The assimilation of problematic experiences is associated with positive treatment outcome, both as measured by standard instruments and as observed in case studies (Stiles, 2002, 2006). Once assimilated, the experiences that were initially problematic contribute to making clients’ self richer, with more resources.

As the articles in this special section emphasize, however, progress through the assimilation sequence is typically not smooth or regular but is frequently interrupted by setbacks, in which one client expression is at a lower APES level than the preceding expression (Caro Gabalda & Stiles, 2009, 2013). Clients in nondirective therapies seem to make relatively smooth APES progress, whereas clients in directive therapies seem to show bursts of rapid forward progress in assimilation punctuated by setbacks, a pattern described as sawtoothed (Caro Gabalda, 2006, 2008; Goodridge & Hardy, 2009; Osatuke et al., 2005). However, all clients seem to have setbacks.

What are the reasons for setbacks? Why are they so common? What are therapists doing to produce or respond to APES advances and setbacks? Should they be doing something different? Answering such questions requires a detailed understanding of the moment-by-moment process of how the therapeutic collaboration leads to advances and setbacks. As background, out of an a-priori list of nine possible reasons for setbacks (Stiles, 2005), essentially all of the setbacks observed in several previous case studies could be classified as attributable to one of three reasons (Caro Gabalda & Stiles, 2009, 2013). (i) In a balance strategy setback, the therapist directs the client’s attention to material at a lower APES stage than the preceding passage, for example, in order to increase client’s awareness and promote an alternative construction. (ii) In an exceeding the therapeutic ZPD setback, the client is pushed to the upper limit of the therapeutic ZPD for the current material and switches to related material at a lower APES stage. (iii) In spontaneous switches, a residual category the client shifts to less assimilated strands of the problem for reasons that were not attributable to therapist intervention.

The Articles in This Special Section

Mendes et al. (2016) focused on setbacks in two depressive clients treated with emotion-focused therapy (EFT): A good outcome and a poor outcome case. Assimilation progress in both cases was sawtoothed (note that EFT therapists are directive about the process of therapy, though not about the client’s life outside the session), though the overall APES trend was upward only in the good outcome case. The main reasons for setbacks replicated the previous studies of reasons for setbacks (Caro Gabalda & Stiles, 2009, 2013), balance strategy, exceeding the therapeutic ZPD, and spontaneous switches. Mendes et al. (2016) added a refinement to this classification by distinguishing setbacks that could be explained by exceeding the therapeutic ZPD induced by the therapist (ZPD-T), or induced by the client (ZPD-C). Most setbacks in the good outcome case were attributable to the balance strategy and spontaneous switches, whereas most setbacks in the poor-outcome case were attributable to ZPD-T and ZPD-C. The difference could reflect different widths of the therapeutic ZPD in the two cases, which may constrain the opportunity for therapeutic work (cf., Zonzi et al., 2014). That is, the poor-outcome case may have had a narrower therapeutic ZPD and so been more likely to encounter its upper limit in the course of therapeutic work.

Caro Gabalda et al. (2016) examined two classes of therapist activities, interventions and positions toward the client’s internal voices, immediately preceding the different kinds of setbacks in a good-outcome case treated with linguistic therapy of evaluation (LTE, a directive cognitive therapy). A large majority (70%) of the setbacks in this case were attributed to balance strategy, and most of the rest (25%) were attributed to exceeding the therapeutic ZPD. The most frequent pre-setback categories of interventions, “Exploring” and “Looking for an extensional orientation” (essentially, adjusting words to facts, a category specific to the clinical aims of LTE) tended to be associated, respectively, with the two most frequent categories of positions toward the client’s internal voices, “Empathizing with the nondominant voice” and “Defying the dominant voice.” Analysis showed different configurations of these therapeutic activities and setback types depending on the APES level at which the dyad was working. In essence, in the lower APES stages, the therapist was more likely to be actively listening, and setbacks reflected pushing a theme beyond the upper limit of the therapeutic ZPD, whereas in the higher APES stages, the therapist more actively directed clients to consider alternatives, following the LTE agenda, and setbacks tended to reflect following these directives toward more risky or difficult material.

Ribeiro, Cunha, et al. (2016) also examined therapist’s contribution, but focusing more specifically on the dyad in a two cases of EFT, using the Therapeutic Collaboration Coding System (TCCS; Ribeiro, Ribeiro, Gonçalves, Horvath, & Stiles, 2013). Most of the time, the therapists intervened within the clients’ therapeutic ZPD, staying closer to the clients’ actual developmental level than to their potential level. Congruently, both clients tend to confirm therapist interventions by showing validation and safety in their responses. In the strongly good-outcome case, the therapist used a balance of supporting interventions (closer to the clients’ actual level) and challenging interventions (closer to the clients’ potential level), whereas in the case with a less strongly positive outcome, the therapist used mainly supporting interventions. Systematic changes in interventions across phases of therapy as the problems’ ZPD advanced through the APES sequence were observed in the strongly good-outcome case. The most frequent therapist interventions were supporting the dominant voice in the initial phase, challenging in the middle phase, and supporting the nondominant voice of innovation in the final phase. These collaborative actions seemed to reflect a sequence of clinical tasks: Building an empathic connection, exploring the client’s difficulties, then challenging the client’s constructions, and finally supporting a new integration of the dominant and nondominant voices. The other case moved more slowly, and the therapist remained mainly supportive throughout. APES advances and setbacks occurred in all phases and showed few associations with particular types of therapist–client exchanges, consistent with the view that the therapist calibrated his interventions to stay within the client’s therapeutic ZPD in both cases.

Ribeiro, Braga, et al. (2016) used the TCCS to study episodes of client ambivalence about therapist interventions in two cases of narrative therapy, one good outcome and one poor outcome, both seen by the same therapist. Ambivalence episodes were identified as a return to the problem marker (RPM; Gonçalves et al., 2011), a passage in which the client returns to safety of the usual (but problematic) self-narrative immediately following an innovative moment (IM), in which he or she used an alternative, innovative self-narrative. In assimilation terms, an IM can be understood as an expression by the nondominant voice and an RPM as a subsequent expression by the dominant voice. An RPM suggests that the client retreats toward safety after an expression that is near the upper limit of the therapeutic ZPD.

Both the good- and poor-outcome cases were selected for having a high incidence of RPMs. Most of the RPMs in both cases occurred following therapist challenging or supporting of the alternative, innovative narrative (the nondominant voice), both of which suggest attempts to extend the upper limit of the therapeutic ZPD. In the good-outcome case, the client tended to accept or go beyond the therapist’s intervention in the subsequent turn, by speaking again from the alternative narrative (the nondominant voice). In the poor-outcome case, however, the client usually did not fully validate the therapist’s challenging interventions, lagging behind the level proposed by the therapist. The difference in response may have reflected differences in the two clients’ readiness for change or the therapist’s differential sensitivity to their readiness.

What Can We Learn from These Case Studies?

These case studies show how assimilation in psychotherapy involves a collaboration between therapist and client. Close examination of the work of the dyad shows how the observed sequences of assimilation, such as the sawtoothed pattern of change, are rooted in therapist–client relationship and responsiveness. For example, successful implementation of the balance strategy requires that a therapist be sensitively attuned to the client, especially when the focus is on less assimilated, more problematic material, or developing alternative constructions. Sensitively used, such directiveness seems to promote progress in therapy even when the therapist is pushing the client away from his/her comfort zone.

These studies have begun to unpack how therapists and clients work productively within the therapeutic ZPD. In the relatively good-outcome cases, therapists appeared to calibrate their interventions to stay within the therapeutic ZPD (Ribeiro, Cunha, et al., 2016), even while they were challenging the client to extend it (Ribeiro, Braga, et al., 2016). Elaborating previous suggestions (Caro Gabalda & Stiles, 2013), these studies indicated that setbacks are not necessarily detrimental; on the contrary, they appear to be integral to the treatment in these directive therapies (EFT, LTE, narrative therapy). Therapists test and push the zone’s upper limits, leading sometimes to ZPD-T setbacks, and they actively direct clients to more difficult material, leading to balance strategy setbacks (Caro Gabalda et al., 2016; Mendes et al., 2016).

Observations in the relatively poor-outcome cases (Mendes et al., 2016; Ribeiro, Braga, et al., 2016; Ribeiro, Cunha, et al., 2016) converge with previous work (Zonzi et al., 2014) in suggesting that these clients’ therapeutic ZPD had relatively limited breadth. Interventions seemingly similar to those that stimulated gradual advances (albeit irregular ones) in the good outcome cases, failed to do so in the poor-outcome cases, though in the Ribeiro, Cunha, et al. EFT case (which really had a less good rather than poor outcome), the therapist seemed to adapt to a slower pace, using relatively less challenging. Of course, it is not useful for researchers to blame clients for their predisposition or to blame therapists for not sensing it. Rather, it points to a problem that therapy theory, research, and practice needs to learn how to detect and solve.

Thus, the therapeutic ZPD seems to be fulfilling the original expectations (Leiman & Stiles, 2001) that it would be a useful concept to understand progress and difficulties in therapy. More generally, the work reported in this special section illustrates the promise of the theory-building approach (Stiles, 2007) used here and elsewhere in constructing the assimilation model. It shows that case studies can be fruitful even though observations on a single case do not, by themselves, justify generalizations. Instead, the theory specifies its own range of generality, and the theoretical fit with the rich details of the case lend confidence to the theory.