CLIENTS’ NARRATIVES IN PSYCHOTHERAPY AND THERAPIST’S THEORETICAL ORIENTATION: AN EXPLORATORY ANALYSIS OF GLORIA’S NARRATIVES WITH ROGERS, ELLIS AND PERLS

Paulo MOREIRA¹*, Óscar F. GONÇALVES2, Carla MATIAS3

1Universidade Lusíada do Porto, Porto, Portugal 2Universidade do Minho, Minho, Portugal 3 The Cambian Group, St Pauls Hospital, Essex, UK

 

Abstract

The therapist’s theoretical orientation has been shown to impact the psychotherapy process. However, less is known about the extent to which the therapist’s orientation may impact clients’ narratives. This exploratory study analysed clients’ narrative production in psychoptherapy, when interacting with different therapists. The data consisted of transcripts of Shostrom’s videotaped therapy sessions between the client Gloria and the therapists Carl Rogers, Fritz Perls and Albert Ellis. Gloria’s narratives were analyzed in terms of narrative dimensions: structural coherence, process complexity and content multiplicity. Gloria’s narratives where characterised by higher levels of stuctural coherence, process complexity and content multiplicity when interacting with Carl Rogers. This exploratory study identified the tendency of clients’ narrative production in psychotherapy vary accordingly to the therapist theoretical orientation. Future studies (using more robust methodologies) that contribute to clarify the impact of the therapist theoretical orientation on narrative co-construction in psychotherapy settings are needed.

 

Keywords: clients’ narratives; therapist’s theoretical orientation

 
Introduction

 

Research shows that the therapist’s theoretical orientation has a significant impact on the psychotherapy process (Castañeiras, Garcia, Lo Bianco, & Fernández-Alvarez, 2006; Elliot, Hill, Stiles, et al., 1987; Larrson, Kaldo, & Broberg, 2010). Individual narrative co-construction depends on the audience’s characteristics (Pasupathi & Hoyt, 2009). However, little is known about the extent to which the therapist’s orientation may impact on the clients’ narratives. A recent study found differences in Cathy’s narratives, depending on the therapist she was interacting with (Moreira & Gonçalves, 2010). Although this study identified a tendency, it is restricted the three therapeutic models included in the study (Client-Centered Therapy, Multimodal Therapy and Actualizing Therapy), and further studies with similar methodologies applied to different therapeutic models and therapists are needed. The study aimed at addressing this question by examining the client’s narratives when interacting with three therapists of different theoretical orientation.

One of the main aims of psychotherapy research is to investigate the extent to which therapeutic models differentially impact on several domains of the psychotherapy process and the therapeutic outcomes. There is a large body of research suggesting that the therapist’s theoretical orientation has no significant impact on therapeutic outcomes (Beutler, Crago & Arizmendi, 1986; Elkin, Shea, Watkins, et al., 1989). This evidence has led some authors to consider the therapist’s theoretical orientation as an overrated variable (Strupp, 1978), whereas others assert that common factors such as the therapeutic relationship are more potent than specific therapy ingredients (Messer & Wampold, 2006).

Psychotherapy process research has paid particular attention to the study of language processes, including studies which have explored how the same patient responds (in terms of verbal response modes) to different therapeutic approaches. These studies have shown that the therapist’s response modes were markedly different from one another and were congruent with the therapist’s theoretical orientation. Furthermore, the clients’ verbal response modes were also found to be different according to the therapists’ response modes (Stiles, Shapiro, & Firth-Cozens, 1988). Other studies on the differential impact of therapeutic models on language processes, include analysis of therapist’s responses (Bohart, 1991), language stylistic complexity between therapist and client (Meara, Shannon, & Pepinsky, 1979) and the relation between the therapist’s speech and the existence of therapeutic change (Holzer, Mergenthaler, & Pokorny, 1996).

Narratives in psychotherapy have traditionally been explored in terms of its different dimensions: narrative structure (e.g., McAdams & Janis, 2004; Salvatore, Conti, Fiore., 2006), narrative process (e.g., Angus, Levitt, & Hardtke, 1999), and narrative content (e.g., Detert, Llewellyn, Hardy, Barkham, & Stiles, 2006). Recently, a methodology evaluating the different narrative dimensions in an integrative way – rather than evaluating only some narrative dimensions (structure, process or content) – was developed by Gonçalves and colleagues. This methodology has been shown to be adequate for the evaluation of the different narrative dimensions (structure, process and content) in an integrative way. For example, using this narrative assessment instrument, a study evaluating narrative change during psychotherapy found that clients achieving better outcomes present higher levels of narrative change during the psychotherapy process than clients with poor outcomes (Moreira, Beutler, & Gonçalves, 2008).

Because psychotherapy is a priviliged context for story co-construction, there has been an increased interest in exploring the potential role of narratives in psychotherapy and personality development (Adler, Wagner, & McAdams, 2007).  The speaker’s characteristics impact conversational reconstructions of past events, which depend on the manner in which the speaker remembers these events, but also by the context in which the story is told (Pasupathi, 2001). The latter is influenced by the listener, and the characteristics of each listener (e.g., opinions, interactions) lead them to experience the story in a different way. Evidence showing that the listener’s characteristics impact on the speaker’s narratives has led some authors to suggest that the former are co-narrators (Bavelas, Coates, & Johnson, 2000), because speakers and listeners shape the way that events are narrated in conversation (Pasuphati, 2001). In other words, the manner in which individuals construct stories is to a great extent dependant on their audience, and on the listener’s characteristics in particular (Gonçalves, 2000; Pasupathi, 2001).

Research investigating the impact of the listener’s (therapist) theoretical orientation on the speaker’s (client) narratives requires a methodology that allows for the evaluation of narratives by the same client interacting with different therapists. This information will enable the identification of differences in the client’s narratives whilst controlling for individual differences in narratives. However, direct application of this methodology in therapeutic clinical settings can be difficult to attain given the complex design that is required. The videotapes produced by Shostrom (1966) showing different therapists conducting a therapeutic session with the same client constitute a classic instrument in psychotherapy process research (used for over 30 years by investigators of different theoretical orientations), making it possible to compare the same client interacting with different therapists. Examples of previous research adopting this methodology include the analysis of patterns of verbal language between Rogers and Gloria (Wickman & Campbell, 2003).

A recent study, using videotapes produced by Shostrom (1966)  evaluated Cathy’s narratives with Rogers, Lazarus and Shostrom, found differences in Cathy’s narratives, depending on the therapist she was interacting with (Moreira & Gonçalves, 2010). These results refer only to the three therapeutic models included in the study (Client-centered therapy, Multimodal therapy and Actualizing therapy), and further studies with similar methodologies applied to different therapeutic models and therapists are needed.

The goal of this exploratory study was to identify tendencies in Gloria’s therapeutic narratives dimensions (i.e., structural coherence, process complexity and content multiplicity) when interacting with three therapists (Albert Ellis, Carl Rogers and Fritz Perls) from three major therapeutic models (Client-Centered, Gestalt and Rational-Emotive therapies). Our non-probabilistic hyphotesis was that the Gloria’ narratives scores would differ depending on the therapist she was interacting with.

 

Method

 

Our study analyzed Gloria’s narratives in interaction with each therapist. Carl Rogers (Client-Centered Therapy), Fritz Perls (Gestalt Therapy) and Albert Ellis (Rational-Emotive Therapy) are the therapists and the founders of the relevant therapeutic models (Shostrom, 1966). First Gloria received an intervention with Carl Rogers, then with Fritz Perls and then with Albert Ellis.

 

Gloria

Gloria is a 30-year old European American woman, living in the USA and recently divorced.  At the time of her interviews with the three founders of the therapeutic models, Gloria presented with difficulties relating to her affective and sexual needs. On the one hand she felt the need to be loved again. On the other hand, she struggled to integrate her needs as a woman with what she considered to be her role as a mother. Gloria had many doubts as to how to relate to her own daughter (i.e., should she open up to her daughter and share her feelings with her or should she protect her from the painful process she was going through?). Her value system made it difficult for her to accept her needs (having a relationship) given her current circumstances (being divorced), and being subject to what society and her own daughter would think of her.

 

Therapists

The therapists analyzed in this study are Carl Rogers (demonstrating a prototype session of Client-Centered Therapy), Friederick Perls (demonstrating a prototype session of Gestalt Therapy) and Albert Ellis (demonstrating a prototype session of Rational-Emotive Therapy).

Carl Rogers and Client-Centered Therapy – Client-centered therapy asserts that every human being has the potential for self-actualization, as long as the conditions for self-actualization are provided. The necessary and sufficient self-actualization conditions (genuineness, unconditional positive regard and accurate empathy) are contained within the therapeutic relationship. The aim of Client-Centered Therapy is to promote these conditions in the client. The emphasizes the client’s subjective experience, in an accepting rather than judgemental attitude. The therapist genuinely accepts the clients’ experiences and point of view. Change occurs when the subjective experience of both client and therapist promotes the client’s self-actualization (Raskin & Rogers, 2000; Prochaska & Norcross, 1994).

Albert Ellis and Rational-Emotive Therapy –  Rational-Emotive Therapy suggests that the way human beings deal with and elaborate life events depends on the philosophy of life they construct. Maladjustment occurs when our philosophy of life results in irrational beliefs. These irrational beliefs are mistakes in the client’s thought proceses and these need to be fought against and discouraged. The therapist’s role in changing irrational beliefs is crucial, as he/she develops and implements logic problem solving strategies (using, for example, the test of reality) (Ellis, 2000; Prochaska & Norcross, 1994).

Friederick Perls and Gestalt Therapy – Gestalt Therapy assumes that the individual must achieve an integration of daily experiences in the here and now, and this is the basis for  holistic well-being. According to Gestalt therapy, human beings must accept their primordial biological nature (humans are primarily biological organisms). A human being’s daily goals or end-goals are primarily based on biological needs, which justify the need of the here and now perspective in order to integrate the biological domain with the psychological and social domains. Adjustment is characterized by the way individuals naturally and spontaneously are aware of their organic needs. As human beings are social beings, concerns about social roles, about what is desirable and about others’ expectations, may result in individuals not being able to resist the tendency for homogeneity within those social rules and expectations, resulting in the discounting of biological and organic needs. Disorder and emotional problems are a result of becoming stuck in the process of growth or maturation as  individuals tend to adapt to the demands of society (Yontef & Jacobs, 2000). The therapeutic process aims to promote the individual’s integration of biological, psychological and social needs (Prochaska & Norcross, 1994). The therapist’s role is primarily to promote this process of awareness in the client. To achieve this, the therapist must frustrate the clients’ attempts at protecting desires and social expectations, of escaping unpleasant emotions and denying responsibility for their own choices. The therapist must resist the temptation of “helping” or “saving” the client from their frustration and unpleasant experiences, otherwise he/she would be confirming the clients’ tendency to avoid awareness of their organic needs (Prochaska & Norcross, 1994).

 

Measures

The three coding systems used to assess three narrative dimensions (structure, process and content) were The Narrative Structural Coherence Coding System, The Process Complexity Coding System and The Content Multiplicity Coding System. These narrative dimensions were selected on the basis of the empirical evidence for their clinical significance  and the coding systems were selected because they have shown to allow for an integrative assessment of the different narrative dimensions (e.g., Gonçalves, Henriques, Alves, & Soares, 2002).

The Narrative Structural Coherence Coding System was developed by Gonçalves and colleagues (Gonçalves, Henriques, & Cardoso, 2001) to assess narrative structure coherence. Narrative Structural Coherence refers to the way in which different aspects of experience relate to one another, engendering coherent feelings with one’s self. The Narrative Structural Coherence Coding Manual is a measure based on the narrative structure models proposed by Labov and colleagues (Labov & Waletsky, 1967) and by Ferreira-Alves and Gonçalves (1999). The Narrative Structural Coherence Coding System evaluates the structure and narrative coherence according to four subdimensions: Orientation, Evaluative Commitment, Structural Coherence and Integration. The Orientation sub-dimension gives information about the characters and the social context, time and space, and personal characteristics that influence behavior. It can also include important recent events that have influenced the present moment.  Essentially, these events are the circumstances that surround the episode.  Circumstances are the preceding and succeeding elements to the event and not only the specific circumstances of its occurrence. In other words, orientation defines the context – all circumstances and historical, social, and cultural factors that facilitate, interfere with or determine the production and reception of a theme in the narrative process.  In the case of personal circumstances, orientation also allows one to respond to the questions: “Who?” , “When?”, “Where?”, and “In what personal circumstances?”. The Structural sequence subdimension refers to a series of events that are defined by the temporal sequence of an experience at the precise moment it occurred. The sequence, or structure narrative, allows the speaker to answer the fundamental question “then what happened?”. That sequence or narrative structure consists of several elements that follow a sequence in a specific temporary structure: (1) an initial event; (2) an internal response to this event (objectives, plans, thoughts, or feelings); (3) an action; and, finally, (4) consequences.  The Evaluative commitment subdimension refers to the degree of involvement or the narrator’s dramatic behavior with the narrative.  In other words, it refers to the value the client/narrator gives to their narrative.  The central question regarding evaluative behaviour is “to what extent does the client involve himself/herself in the story he/she is telling?”.  The Integration subdimension refers to the degree of diffusion or integration among various elements or stories in order to produce a meaning that binds the elements or stories together. This dimension evaluates the extent to which the story contains a main linking thread (Gonçalves et al., 2001). Each dimension is coded using a five-point, anchored Likert scale (1=absent or vague; 2=little; 3=moderate; 4=high; 5= very much).  The Narrative Structural Coherence Coding System presents a high level of inter-observer fidelity (i.e.,  96%) and internal consistency (alpha values between .79 and .92) (Gonçalves et al., 2002), and is available from the authors (Gonçalves, Henriques & Monteiro, 2001).

The Narrative Process Complexity Coding System was developed by Gonçalves and colleagues (Gonvalves, Henriques, Alves, & Monteiro, 2001) to assess narrative process complexity. Narrative Process Complexity refers to the individual’s degree of openness to experiences, evidenced by the quality, variety and complexity of the narrative process, in sensorial, emotional, cognitive and meaning terms. The evaluation of the Narrative Process Complexity includes four subdimensions: Objectifying, Emotional Subjectifying, Cognitive Subjectifying, and Metaphorizing. The Objectifying subdimension refers to the diversity of elements in the sensorial experience that are present in the narrative (e.g., vision, hearing smell, taste and physical sensations). The Emotional Subjectifying subdimension evaluates the degree to which the narrative presents a diversity of emotional experiences (e.g., emotions, feelings). The Cognitive Subjectifying subdimension concerns the degree to which the client includes and integrates several elements of his/her cognitive experience in his/her narrative (e.g., thoughts, beliefs). Finally,  the Metaphorizing subdimension refers to the diversity of meta-cognitive elements and meanings present in the narrative. Metaphorizing is defined as the construction of meanings or significance based on experience: “It supposes a meta-analysis of the situation, frequently expressed through metaphors that condense the meanings that the subject infers from the experience.  The Metaphorising subdimension evaluates the reflexive attitude developed during the process of building multiple meanings for his or her experiences” (Gonçalves et al., 2001, pg. 7).  Each subdimension in the coding system is rated using a five-point anchored Likert scale (1=absent or vague; 2=little; 3=moderate; 4=high; 5=very much).  The Narrative Process Complexity Coding System presents high levels of fidelity among inter-observers (i.e., 89%), and internal consistence (alpha values between .66 and .87) (Gonçalves, et al., 2002), and is available from the authors.

The Narrative Content Multiplicity Coding System was developed by Gonçalves and colleagues (Gonçalves et al., 2001) to assess narrative content multiplicity. Narrative Content Multiplicity refers to the degree to which there is diversity of content in the individual’s narrative. The Narrative Content Multiplicity Coding System assesses the degree to which the individual’s narratives are characterized by diverse content.  Narrative content multiplicity is assessed according to four subdimensions: Themes, Events, Scenarios and Characters. The Themes subdimension concerns the diversity and multiplicity of themes present in the narrative (e.g., a school year, marriage difficulties); the Events subdimension refers to the diversity and multiplicity of events (e.g., entering the office, someone arriving); the Scenarios subdimension analyses the diversity and multiplicity of scenarios (e.g., a city, a house); and the Characters subdimension evaluates the diversity and multiplicity of characters (e.g., my mother, her husband).  Each subdimension in the coding system is rated using a five point anchored Likert scale (1=absent or vague; 2=little; 3=moderate; 4=high; 5=very much). The Narrative Content Multiplicity Coding System presents high levels of inter-observers fidelity (i.e., 94%) and internal consistence (alpha values between .86 and .90) (Gonçalves et al., 2002), and is available from the authors.

 

Procedure

The therapeutic sessions which constituted the object of analysis of the present study were transcribed and then coded independently by two pairs of judges, blind to the study hypthotesis (narratives were given to raters without identification of who the therapist or the client were).

The judges were psychologists who had graduated a 5-year Psychology program (pre-Bologna system, the equivalent of the Master degree in the Bologna system), including 2 years specialization in Clinical Psychology. Judges received their Psychology trainning in different Northen Portugal Universities. All judges were less than 30 years old and were enroled in a pos-graduated trainning at the moment of the invitation to participate in the study.

Judges had 30 hours of training in each coding system. After the initial 30-hour training, in which the judges were introduced to the coding concepts and methodology, ten therapeutic sessions were evaluated, as a training strategy. Ten more therapeutic sessions were distributed and rated in order to evaluate fidelity between judges. Inter-judge agreement was established for each narrative dimension (structural coherence, process complexity and content multiplicity), with two pairs of judges evaluating separately each narrative dimension. Only when inter-judges agreement was equal or superior to 80% were the pair of judges allowed to initiate the coding of the sessions used in this study.  Narratives were then coded by pairs of similarly trained judges presenting high levels of agreement (reliability of rating on the actual sample was superior to 80% agreement). The different therapist-client dyads were evaluated by the same pairs of raters.  Also each one of the narrative dimensions were evaluated by the same pair of judges (i.e., the same pair of judges evaluated the structural coherence dimension in the three therapist-client dyads, another pair of judges evaluated the process complexity dimension in the three therapist-client dyads and another pair of judges evaluated the content multiplicity in the three therapist-client dyads). Each pair of judges was specialized in the evaluation of the respective narrative dimension, and had evaluated the same narrative dimensions in previous studies (Moreira et al., 2008).

 

Data Analysis Procedures

Percentages were calculated regarding the difference between Gloria’s narrative dimension and subdimension scores while interacting with each of the 3 therapists. The maximum score Gloria could obtain with each therapist was 5, and the minimum score was 1.  The maximum difference rating Gloria could obtain with the three therapists was 4, i.e. the difference between the maximum score of 5 that she could obtain with therapist A and the minimum score of 1 she could obtain with therapist B. The percentage of the difference score between two therapists was calculated via conversion of the difference percentage score.  That is, the maximum difference score (5-1=4) corresponds to 100%. The values of each difference score are calculated using a simple rule in which the maximum difference score (md = 4) corresponds to 100% and this allows calculation of the percentage of the difference. The difference in percentage score obtained with therapist A comparative to the score obtained with therapist B is calculated using the following formula: ed*100/md(md=4).

 

Results

 

Gloria’s narratives registered different scores, depending on the therapist she was interacting with. Gloria achieved the highest narrative score (T=3.08) with Carl Rogers (25% higher than Perls and 18.25% higher than Ellis); the second score was with Ellis (T=2.33) (Ellis presented a rating 18.75% lower than Carl Rogers and 12.5%  higher than Fritz Perls), and the lowest score with Perls (T=2.08) ( 25% lower than Rogers and 12.5% lower than Ellis). Table 1 contains these data broken down:

 

Table 1. Gloria’s mean narrative scores with Carl Rogers, Fritz Perls and Albert Ellis

 

Dimension	Subdimension	Carl Rogers	Fritz Perls	Albert Ellis
Structural Coherence	Orientation	2	2	2
	E.Commitment	3	3	3
	Str. Sequence	5	4	4
	Integration	4	4	4
	Total	14	13	13
	Mean	3.5	3.25	3.25
	St.Deviation	1.11	0.82	0.82
Process Complexity	Objectifying	2	1	1
	E.Subjectifying	4	2	4
	C.Subjectifying	4	2	3
	Metaphorizing	4	2	1
	Total	14	7	9
	Mean	3.5	1.75	2.25
	St. Deviation	0.86	0.43	1.29
Content Multiplicity	Characters	2	2	2
	Scenarios	3	1	1
	Events	1	1	1
	Themes	3	1	2
	Total	9	5	6
	Mean	2.25	1.25	1.5
	St. Deviation	0.82	0.43	0.5
Total narrative dimensions	Total	37	25	28
	Mean	3.08	2.08	2.33
	St. Deviation	0.58	0.84	0.71

 

 

There was a half a point difference (18.75%) between Gloria’s average narrative score with Rogers and the average score with Ellis while the difference between Gloria’s average narrative scores with Rogers and Perls is of one point (25%). When comparing outcomes for each narrative dimension, the highest narrative score with the three therapists concerned structural coherence (structural coherence and process complexity scored the same in the interaction with Rogers), followed by process complexity and content multiplicity. Refer to Figure 1 for this data.

 

 

Table 2 presents the difference in percentage scores between the three therapists concerning each of  the narrative dimensions and subdimensions. Values in parenthesis refer to the percentage to which Gloria’s narratives differ depending on the therapist. For example, (25%) means that Gloria registered a score in a given narrative subdimension 25% superior to that obtained in the same narrative subdimension with another therapist.

 

 

 

 

Table 2. Comparison of Gloria’s narratives scores (%) with Rogers, Perls and Ellis

 

		Rogers		Perls		Ellis
Dimension	Subdimension	Rogers and Perls	Rogers and Ellis	Perls and Rogers	Perls and Ellis	Ellis and Rogers	Ellis and Perls
Structural Coherence	Orientation	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
	E.Commitment	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
	Str. Sequence	1(25%)	1(25%)	-1(-25%)	0(0%)	-1(-25%)	0(0%)
	Integration	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
	Total	.25 (6.25%)	.25 (6.25%)	-.25 (-6.25%)	0(0%)	-.25 (-6.25)	0(0%)
Process Complexity	Objectifying	1(25%)	1(25%)	-1(-25%)	0(0%)	-1(-25%)	0(0%)
	E.Subjectifying	2(50%)	0(0%)	-2(-50%)	-2(-50%)	0(0%)	2(50%)
	C.Subjectifying	2(50%)	1(25%)	-2(-50%)	-1(-25%)	-1(-25%)	-1(-25%)
	Metaphorizing	2(50%)	3(75%)	-2(-50%)	-1(-25%)	-3(-75%)	1(25%)
	Total	1.75 (43.75%)	1.25 (31.25%)	-1.75 (-43.75%)	-.5 (-12.5%)	-1.25 (-31.25%)	.5 (12.5%)
Content Multiplicity	Characters	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
	Scenarios	2(50%)	2(50%)	-2(-50%)	0(0%)	-2(-50%)	0(0%)
	Events	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
	Themes	2(50%)	1(25%)	-2(-50%)	-1(-25%)	-1(-25%)	1(-25%)
	Total	1 (25%)	.75 (18.75%)	-1 (-25%)	-.25 (-12.5%)	-.75 (-18.75%)	.25 (12.5%)
Total narrative dimensions	Total	1 (25%)	.75 (18.75%)	-1 (-25%)	-.25 (-12.5%)	-.75 (-18.75%)	.25 (12.5%)

 

 

The most similar scores in Gloria’s narratives with the three therapists concerned the structural coherence dimension. Yet, Gloria’s narrative with Rogers obtained a higher score (T=3.5) (i.e., 6.25% higher) than those obtained with the other two therapists, which registered the same value (T=3.25).

In what the process complexity dimension is concerned, Gloria’s narrative score with Ellis (T=2.25) was lower than with Rogers (1.25 points lower,  i.e., 31.25% less). Gloria’s narrative with Perls registered the lowest score (T=1.75 which is 43.75% lower than the score with Rogers). There was a half a point (12.5% lower) difference between Gloria’s narrative scores with Ellis and Perls.

Gloria’s narrative scores with each therapist also registered different values in terms of the process complexity subdimension. Gloria’s narratives with Rogers and Ellis got the same score on the emotional subjectifying subdimension (T=4), which was twice (50% higher) that Gloria’s narrative with Perls (T=2). Regarding the cognitive subjectifying subdimension, Gloria’s narrative score with Rogers obtained, once again, the highest score (T=4), followed by Gloria’s narrative with Ellis (T=3) and by Gloria’s narrative with Perls (T=2). In summary, these results show a tendency for differences between Gloria’s narratives with the 3 therapists (Rogers 50% higher than Perls, and 25% higher than Ellis). Gloria’s narrative with Ellis scored the lowest in terms of the Metaphorizing subdimension (T=1). Gloria’s narrative score with Perls (T=2) was twice higher than that obtained with Ellis, while her narrative score with Rogers was higher than the narrative scores obtained with the other two therapists (T=4; i.e., 50% higher than Perls and 75% higher than Ellis).

Finally, the lowest narrative scores referred to the content multiplicity subdimension, regardless of which therapist Gloria interacted with. Gloria’s narrative with Rogers once again achieved the highest score (T=2.25), followed by Gloria’s narrative with Ellis (T=1.5) and by Gloria’s narrative with Perls (T=1.25). There was a one point difference between Gloria’s narrative score with Rogers and with Perls (i.e., the former score was 25% higher than the latter), while the difference between Gloria’s narrative score with Rogers and with Ellis was lower (.75, that is 18.75% lower in Ellis’ case).

Regarding the themes subdimension, Gloria’s narrative score with Rogers was 50% higher than the one with Perls and 25% higher than the one with Ellis. It is worth noting that Gloria’s narrative with Perls scored the lowest in all subdimensions, except for the characters subdimension, where the same scores were obtained. By contrast, Gloria’s narrative with Rogers only scored the lowest (T=1) in the events subdimension, for which the same score was obtained by the other two therapists.

Because of the exploratory and qualitative characteristics of the sudy, excerpts ilustrating the narrative subdimensions of Gloria’s narratives with each therapist are presented below. Table 3 ilustrates how the average length of Glorias’ narrative with Rogers is much longer (containing more narrative elements) than her narratives with the other two therapists.

Table 4 ilustrates how Gloria’s narratives (of similar length with each therapist) with Rogers also scored higher than the narratives with Pearls and Ellis in terms of coherence, structure, process complexity and content multiplicity

 

Discussion and conclusions

 

Results of this exploratory study revel tendency for Gloria’s narratives scores to vary  according to the therapist she was interacting with. In other words, there seems to be a variability in the narratives of the same client when interacting with different therapists of contrasting theoretical orientations. In fact, in all of the narrative dimensions studied, Gloria scored higher when interacting with Rogers than when interacting with Ellis and Perls. This is consistent with the notion that the client’s narrative varies according to the therapist’s characteristics (including his/her theoretical orientation), lending support to previous research on the impact of the listeners’ characteristics in stories co-construction (Pasupathi, 2001).

 

Table 3. Examples of narrative dimensions of Gloria’s longest narratives with the three therapists

		Rogers	Perls	Ellis
Strutural coherence	Orientation	“I got divorced…”; “I had alrwady sought therapy before”; “I went through many changes” “trying to adapt to my life as a single woman” (…)	——————-	“Generally I think…”
	Structural sequence	“she saw a girl who was single but was pregnant” (the initial event); “and the conversation was going smoothly, and I felt totally at ease …” (an internal response to the event); “and she asked me…and I lied to her” (an action); “I feel guilty for having lied to her…” (respective consequences)”	“ Just because I smile when I’m embarrassed or backed into a corner, that doesn’t mean I’m fake”	“when I want to (…) but when (…) then I  (…) I’ve spoilt it all over again”
	Evaluative commitment	“I wish I could stop shaking…”	“I admit it!” “But gosh!”	“I am… am”
	Integration	I have lied to her…I haven’t been able to forget that”	“ Just because I smile when I’m embarrassed or backed into a corner, that doesn’t mean I’m fake”	“I’ve spoilt it all over again”
Process complexity	Objectifying	“she saw...”	——————-	——————-
	Emotional subjectifying	“nothing to upset her” “shock her” “for her to accept me” “… I feel guilty”; “hurt her”	“It is very difficult for me” “I hate feeling embarrassed”	“I’m afraid”
	Cognitive subjectifying	“that most worries me” “always on my mind …”; “ anything to upset her…”; “I’m very consciouse”; “I want her…” “I don’t want”	“it offends me when you say I’m fake”	“I also thought about it”; “I want to change… “I want to show” “I think”
	Methaphorizing	“I’m conscious of her problems”	“I’m backed into a corner” “it doesn’t mean I’m fake”	——————-
Content multiplicity	Characters	“daughter Her father…”  “men, when they visit me at home …”, “girl who is single but pregnant”	I (Gloria) Therapist	I (Gloria) This man
	Scenarios	Home	——————-	——————-
	Events	“divorced”; “lied”, “asked me”,	——————-	——————-
	Themes	divorce, sex, guilt	To be or not to be fake	Losing opportunities with men

Table 4. Narrative sub-dimensions of Gloria’s narratives of similar length with each therapist

 

		Rogers	Perls	Ellis
Strutural coherence	Orientation	“when I was a child”	——————-	——————
	Structural sequence	“when I was a child… I found out… I felt…”	“Could have been hurt… but you wouldn’t easily show it”	“if (…) I could (…) he likes (…) I’m only giving him the worst in me”
	Evaluative commitment	“and I don’t know…I….”	——————-	“anyway”
	Integration	“I have to be careful”	“wouldn’t easily show it”	“I’m only giving him the worst in me”
Process complexity	Objectifying	“it was dirty”	——————-	——————
	Emotional subjectifying	“I feel” “I didn’t like her anymore” “I felt that”	“put on a brave face” “fragile inside” “get hurt”	“anxious”
	Cognitive subjectifying	“I remember” “I don’t want”	“I think”	——————
	Metaphorizing	“I found out…” “I felt it was dirty…”	——————-	“Trying to get this man”
Content multiplicity	Characters	Mother Father, Pammy	I (Gloria) Therapist	I (Gloria)
	Scenarios	——————-	——————-	——————
	Events	“For the first time I knew”	——————-	——————
	Themes	Tell her daughter the truth	Show your feelings	Anxiety stops you from being truthful

 

 

Therapists’ atitudes impact on the therapeutic process and outcomes (Sandel et al., 2007). However, evidence attesting for the impact of the therapists’ behaviour and atitudes on the client’s narratives is less well established. The current study provides two key indicators of the variation in Gloria’s narratives with the three therapists. The first indicator refers to the average length of each therapeutic narrative and the second indicator pertains to the variability in narrative quality dimensions (i.e., coeherence, structure, process complexity and content multiplicity). Therapy narratives result from the sharing of aims and objectives between therapist and client. When faced with the same dilemma, different therapists prioritise different aims. For example, the therapist with a psychodynamic orientation tends to prioritise the identification of past patterns which impact the client’s current psychological functioning. Within a behaviorist orientation, the therapist will prioritise the analysis of behavioural contingencies affecting the client’s present difficulties. In summary, the concept of theoretical orientation encompasses therapeutic objectives, behaviours and atitudes. Therefore, it is expected that therapists with different therapeutic orientations will have different attitudes and will behave differently when interacting with a particular client. These assumptions were confirmed by previous studies (Sandel et al., 2007), and are conguent with the tendencies found in this exploratory study.

However, caution is warranted in interpreting the findings of this study. The exploratory analyses conducted did not allow for the control of potential confounding factors. These factors include therapist and client related variables. The therapist’s theoretical orientation is included as a therapist-related variable in Beutler and colleagues’ typology (Beutler et al., 1994) and has been shown to impact on the therapeutic process  (Elliot et al., 1987; Gomes-Schwartz, 1978). However, the mechanism explaining the interaction between these two variables remains unclear. Although some of the therapists’ characteristics derive from their theoretical orientation (e.g., therapeutic aims and expectations), therapists with the same theoretical background can still diverge from one another. A key area of divergence refers to the therapists’ attitudes, also shown to influence treatment outcome in psychotherapy (Sandel et al., 2007). Future studies regarding narratives in psychotherapy should take into account variables other than the psychotherapist’s theoretical orientation. It is particularly important to analyse the main factors impacting on narratives whilst controlling for variables that may distinguish therapists that share the same theoretical orientation, such as attitudes and experience.

Speaker characteristics also influence the co-construction of situated stories (McLean, Pasupathi, & Pals, 2007) and impact on the therapeutic process (Beutler & Harwood, 2000). Furthermore, the compatibility between the therapeutic treatment and the client’s characteristics, such as his/her personality (e.g., coping styles, resistance) promote therapeutic change (Beutler & Harwood, 2000). For example, story co-construction by a highly resistant client could benefit from a less directive therapist, whereas a highly cooperative client could benefit from a more directive therapist. Although the client’s characteristics remained constant in this study (i.e., the same client interacted with the three therapists), further studies should explore the interaction between different clients and therapists and how this affects story co-construction.

Psychotherapy is a privileged context for narrative construction. Equally, the psychotherapy process provides a unique opportunity for goal-oriented self development. The present study suggests that the listeners’ more abstract characteristics such as beliefs, perceptions and meaning-making are potentially important factors impacting on the client’s narrative construction. Therefore, these characteristics could play an important role in storytelling and self-development during adulthood. This topic should be further explored in studies integrating both domains – story co-construction as an element of personality and ego-development in the psychotherapy context.

In this study, the therapeutic models under investigation refer to classical approaches, and although many current therapies follow one of these three models, it is highly unlikely that at present most therapists employ these therapies in their original format. This is because these models have evolved in time, and there is currently a tendency for a more eclectic approach to treatment and theoretical integration. Therefore, the models analysed in this study are not necessarily representative of the current theoretical approaches. Nevertheless, a key methodological consideration of this investigation was to ensure the therapist’s fidelity to the therapeutic model. This is why we opted for having the same client (Gloria) exposed to the three therapists who were the founders of these therapeutic models and therefore considered as the most representative of each theoretical approach (i.e., Rogers, Ellis and Perls).  Further research should include different therapeutic models than the ones analysed in this study as well as current and new approaches to treatment.

 

REFERENCES

 

Adler, J. M., Wagner, J. W. & McAdams, D. P. (2007). Personality and the coherence of psychotherapy narratives. Journal of Research in Personality, 41, 1179-1198.

Angus L., Levitt H., & Hardtke K. (1999). The narrative processes coding system: research applications and implications for psychotherapy practice. Journal of Clinical Psychology, 55, 1255-1270.

Bavelas, J. B., Coates, L., & Johnson, T. (2000). Listeners as conarrators. Journal of Personality and Social Psychology, 79, 941-951.

Beutler, L. E., Crago, M., & Arizmendi, T. G. (1986). Therapist variables in psychotherapy process and outcome. In S. L. Garfield, & A. E. Bergin (Eds.), Handbook of psychotherapy and behaviour change (3^rd edition), pp. 257-310. New York: Wiley.

Castañeiras, C., García, F., Lo Bianco, J., & Fernández-Alvarez, H. (2006). Modulating effect of experience and theoretical–technical orientation on the personal style of the therapist. Psychotherapy Research, 16, 595-603.

Detert, N. B., Llewellyn, S. L., Hardy, G. E., Barkham, M., & Stiles, W. B. (2006). Assimilation in good- and poor-outcome cases of very brief psychotherapy for mild depression: An initial comparison. Psychotherapy Research, 16, 393-407.

Elkin, I., Shea, M. T.,  Watkins, J. T.,  Imber, S. D.,  Sotsky, S. M., Collins, J. F., et al., (1989). National Institute of Mental Health Treatment of Depression Colaborative Research Program: General effectiveness of treatments. Archives of General Psychiatry, 46, 971-982.

Elliot, R., Hill, C., Stiles, W. B., Friedlander, W., Maher, M. M., Magison, A. R., et al. (1987).Primary therapist response modes comparison of six rating systems. Journal of Consulting and Clinical Psychology, 55, 218-223.

Ellis, A. (2000). Rational Emotive Behavior Therapy. In R. J. Corsini & D. Wedding (Eds.) Current Psychotherapies (6^th edition), pp.168-204. Itasca, Illinois: P. E. Peacock Publishers, Inc.

Ferreira-Alves, J., & Gonçalves, O. F. (1999). Manual de avaliação da coerência narrativa. Braga: Universidade do Minho.

Gonçalves, O. F. (2000). Viver narrativamente: A psicoterapia como adjectivação da experiência. Coimbra: Quarteto Editora.

Gonçalves, O. F., Henriques, M. R., Alves, A., & Rocha (2001a). The Narrative Process Complexity Coding System. Unpublished manuscript. University of Minho, Portugal.

Gonçalves, O. F., Henriques, M. R., Soares, L. & Monteiro, A. (2001b). The Narrative Content Multiplicity Coding System. Unpublished manuscript. University of Minho, Portugal.

Gonçalves, O. F., Henriques, M. R., & Cardoso, G. (2001c). The Narrative Structural Coherence Coding System. Unpublished manuscript. University of Minho, Portugal.

Gonçalves, O. F., Henriques, M., Alves, A., & Soares, L. (2002). Analyzing structure, process and content in narratives of clients diagnosed with agoraphobia. International Journal of Clinical and Health Psychology, 2, 389-406.

Gonçalves, O. F., Pérez, A., Henriques, M., Prieto, M., Lima, M. R., Siebert, M. F., & Sousa, N. (2004). Funcionamento Cognitivo e produção narrativa no Síndrome de Williams: Congruência ou Dissociação Neurocognitiva? International Journal of Clinical and Health Psychology, 4, 623-638.

Holzer, M., Mergenthaler, E., & Pokorny, D. H. (1996). Vocabulary measures for the evaluation of therapy outcomes: Re-studying transcripts from the Penn Psychotherapy Project. Psychotherapy Research, 6, 95-108.

Labov, W., & Waletsky, J. (1967). Narrative analysis: oral versions of personal experience. In J. Helm (Ed.), Essays on the verbal and visual arts. Seattle: American Ethnological Society.

Larsson, B. P. M., Kaldo, V., & Broberg, A. G. (2010). Theoretical orientation and therapists’ attitudes to important components of therapy: A study based on the valuable elements in psychotherapy questionnaire. Cognitive Behaviour Therapy, 39, 161-172.

McAdams, D. P., & Janis, L. (2004). Narrative Identity and narrative therapy. In L. E. Angus & J. McLeod (Eds.) The Handbook of Narrative and Psychotherapy – Practice, Theory and Research, pp. 159-174. London: Sage.

McLean, K. C., Pasupathi, M., & Pals, J. L. (2007). Selves creating stories creating selves: A process model of self-development. Personality and Social Psychology Review, 11, 262-278.

Meara, N. M., Shannon, J. W., & Pepinsky, H. B. (1979). Comparison of the stylistic complexity of the language of counselor and client across three theoretical orientations. Journal of Counseling Psychology, 26, 181-189.

Messer, S. B., & Wampold, B. E. (2006). Let’s face facts: common factors are more potent than specific therapy ingredients. Clinical Psychology: Science and Practice, 9, 21-25.

Moreira, P., & Gonçalves, O. F. (2010). Therapist’s theoretical orientation and patients’ narrative production: Lazarus, Rogers, Shostrom and Cathy revisited. International Journal of Psychology and Psychological Therapy, 10, 227-244.

Moreira, P. A., Beutler, L. E., & Gonçalves, O.F. (2008). Narrative Change in psychotherapy: differences between good and bad outcome cases in cognitive, narrative and prescriptive therapies. Journal of Clinical Psychology, 64, 1181-1194.

Pasupathi, M., & Hoyt, T. (2009). The development of narrative identity in late adolescence and emergent adulthood: The continued importance of listeners. Developmental Psychology, 45, 558-574.

Pasupathi, M. (2001). The social construction of personal past and its implications for adult development. Psychological Bulletin, 127, 651-672.

Prochaska, J. O., & Norcross, J. C. (1994). Systems of Psychotherapy – a Transtheoretical analysis (3^rd edition). Pacific Grove: Brooks/Cole Publishing Company.

Raskin, N. J., & Rogers, C. R. (2000). Person-centered therapy. In R. J. Corsini & D. Wedding (Eds.), Current Psychotherapies (6^th edition), pp.133-167. Itasca, Illinois: P. E. Peacock Publishers, Inc.

Salvatore, G., Conti, L., Fiore, D., Carcione, M.,  Dimaggio, G., & Semerari, A. (2006). Disorganized narratives: Problems in treatment and therapist treatment intervention hierarchy.  Journal of Constructivist Psychology, 19, 191-207.

Sandel, R., Lazar, A., Grant, J., Carlsson, J., Schubert, J., & Brogerg, J. (2007). Therapist attitudes and patient outcomes: II. Therapist attitudes influence change during treatment. Psychotherapy Research, 17, 201-211.

Shostrom, E. L. (1966). Three approaches to psychotherapy II (Film). Santa Ana, CA: Psychological Films.

Stiles, W. B., Shapiro, D. A., & Firth-Cozens, J. A. (1988). Do sessions of different treatments have different impacts? Journal of Counseling Psychology, 35, 391-396.

Strupp, H. H. (1978). The therapist’s theoretical orientation: An overrated variable. Psychotherapy: Theory, Research and Practice, 15, 314-317.

Wickman, S. A., & Campbell, C. (2003). An Analysis of how Carl Rogers enacted client-centered conversation with Gloria. Journal of Counseling & Development, 81, 178-184.

Yontef, G. M., & Jacobs, L. (2008). Gestalt therapy. In R. J. Corsini & D. Wedding (Eds.), Current Psychotherapies (8^th edition), pp.328-367. Itasca, Illinois: P. E. Peacock Publishers, Inc.