Denise Hinn ✉ (University of Kassel) Silja Kotte (Aschaffenburg University of Applied Sciences) Heidi Möller (University of Kassel)
Mentalization is the basis of the human ability to understand interpersonal behaviour and is consid-ered a key competence in psychotherapy research. We apply mentalization theory to workplace coaching and argue for its added value from a conceptual perspective. We illustrate its empirical potential with an exploratory analysis of a coaching process, in which coaching session transcripts are rated using a rating instrument established in psychotherapy (namely the Reflective Functioning Scale; Fonagy et al. 1998). We find indications that the coachee's mentalization changes over the course of the coaching engagement and that mentalization fluctuates considerably within individual coaching sessions.
workplace coaching, reflection, mentalization, reflective functioning scale, rating
Accepted for publication: 11 January 2024 Published online: 01 February 2024
© the Author(s) Published by Oxford Brookes University
Anna, a senior manager in the education sector, engages in a coaching process because she wishes to feel more confident in her leadership role, have better self-esteem, and allow herself to set boundaries with her co-workers. In one coaching session, she addresses her insecurity in dealing with her manager. She describes a situation in which she had a conflict with her manager. Although she generally receives positive feedback from her manager, she reports getting heart palpitations and worries when thinking about the next interaction. She finds these feelings disturbing and is unable to interpret them for herself. When the coach asks her to explain how much stress this situation is currently causing her, she evades the question. She does not answer, but falters, repeats herself, and begins sentences without finishing them.
In this illustration, the coachee is unable to satisfactorily resolve the situation with her manager: even afterwards in her coaching session, she is unable to name her own feelings. As described in this example, in stressful situations and conflicts, emotions or perspective-taking capacities can become inaccessible. Moving beyond a concrete understanding of behaviour in relationships can then become very difficult. Considering our own mental states and those of others, inferring implicit intentions from behaviour, and understanding how mental states change over time is then no longer possible (Chiesa & Fonagy, 2013; Fonagy, Campbell & Allison, 1996; Fonagy, Gergely, Jurist & Target, 2022). The capacity that enables someone to do this is called mentalization. Fonagy et al. (2022) define mentalization as the capacity to be aware of our own internal mental states and to reflect on the thoughts, feelings, and intentions of others. Metaphorically, it is the act of looking at oneself from the outside and at others from the inside (Allen, 2013). Drawing from the traditions of psychoanalytic and attachment theories, this capacity is viewed as essential to the process of making meaning in relation to others and the self and has subsequently been operationalised as reflective functioning (RF) (Fonagy & Bateman, 2007).
We propose that mentalization theory can add value to workplace coaching from three perspectives.
First, mentalization theory can answer the call for more theory in process research. Compared to psychotherapy process research, coaching process research is still relatively young, and the related theory and concepts are poorly developed (Kotte & Bozer, 2022). Thus, a focus on theory-led research is crucial to improve the credibility and quality of coaching research and practice (Kotte & Bozer, 2022). One potential avenue for advancing coaching research is aligning coaching research more closely with psychotherapy research (Kotte & Bozer, 2022; Myers, 2017). Although it is widely acknowledged that there are important differences between psychotherapy and coaching (Peltier, 2001), it can be argued that there are sufficient similarities between these two fields to warrant consideration of the theories and methodologies employed in the more advanced field of psychotherapy process research (McKenna & Davis, 2009).
Second, reflection is an integral part of numerous coaching definitions. For example, in a widely cited definition of workplace coaching by Bozer and Jones (2018), workplace coaching is defined as ‘a one-to-one custom-tailored, learning and development intervention that uses a collaborative, reflective, goal-focused relationship to achieve professional outcomes that are valued by the coachee’ (p. 1). Fostering a coachee’s capacity to reflect is described as a desired outcome of coaching (Jones, Woods & Guillaume, 2016; Greif, 2008). Scientific literature repeatedly claims that coaching promotes reflection and that this has also been scientifically proven (Trager, 2008). However, to date, there are scarcely any empirical studies that show that coaching produces a change in an individual’s capacity to reflect (Bozer & Jones, 2018; Trager, 2008). Therefore, it appears useful to consider how reflection in coaching could be measured and how insights from related areas of research (in this case, psychotherapy research) may be transferred for this purpose.
Numerous constructs have emerged in the psychotherapy literature to describe reflective processes as a common factor for therapeutic change (Sidis, Moore, Pickard & Deane, 2022). Many of these constructs describe higher-order cognitive processes (Fonagy & Bateman, 2016). The two concepts that are most frequently studied are mentalization and metacognition (Ridenour, Knauss & Hamm, 2019; Sidis et al., 2022). In principle, the terms overlap, as both constructs refer to an individual’s capacity to reflect upon the mental states of self and other. However, the constructs have evolved from different traditions and conceptually differ in terms of a few aspects (Ridenour et al, 2019). We argue that mentalization (compared to metacognition) is particularly well-suited to enrich coaching research and practice for several reasons. One reason is that mentalization has been operationalised as reflective functioning with an established, well-validated assessment instrument —the RF scale (Fonagy, Target, Steele & Steele, 1998). The RF scale is considered a gold-standard instrument and is extensively used across different schools of psychotherapy research (Taubner, 2015). This paves the way for empirically investigating mentalizing in the context of workplace coaching. Psychotherapy research has revealed that a client’s capacity to mentalize can be considered a key competence in psychotherapy. Mentalization contributes to the improvement of an individual’s mental health (Allen, 2013; Taubner, 2015) and enables an individual to adequately cope with external and internal stressors, regulate affect, and form stable interpersonal relationships (Fonagy et al., 1998). These are also important competencies in the workplace. Improved mentalization could help coachees better understand how they themselves and others think, feel, and act, which in turn can lead to increased self-awareness, better interpersonal communication and collaboration, and improved self-regulation under stressful conditions.
Third, mentalization theory (in contrast to metacognition) can help to understand why people are occasionally able to reflect better and occasionally worse. The above-mentioned example, which originates from the coaching case examined in this study, reveals that the coachee has a lower ability to mentalize when experiencing relationship stress. The stress-induced switch model of mentalization (Fonagy & Luyten, 2009, following Mayes, 2006) explicitly theorises that the capacity to mentalize can be disrupted in the context of stress and explains why, in stressful situations, people may be unable to use their otherwise available mentalization capacity. This is highly relevant in the context of workplace coaching. Across numerous professions and industries, employees and managers alike report high levels of work-related stress (Alight, 2022) and conflict; the resolution of such stress is an essential part of workplace collaboration (Vellmer & Vetter, 2022). In addition, conflicts are a frequent subject in coaching (Middendorf & Salamon, 2021). Understanding this as well as why mentalization is affected by stress and is, therefore, likely to vary within and across coaching sessions can help coaches to navigate coaching processes.
In the following, we explain the concept of mentalization in greater detail and describe how mentalization can be operationalised and measured during the coaching process using the RF scale. An exploratory case analysis is then utilised to illustrate the application of mentalization in coaching processes and the benefits of mentalization for coaching research and practice.
In this theory section, we first introduce the concept of mentalization, distinguish it from other related constructs, and elaborate on stress-related mentalization. We then present empirical findings on mentalization and finally introduce the initial links between mentalization theory and coaching.
Mentalization (also called ‘mentalizing’) was defined by Fonagy et al. (2022) as the capacity to be aware of our internal mental states and to reflect on the thoughts, feelings, and intentions of others. It is conceived as the basis of the human ability to understand interpersonal behaviour (Fonagy & Bateman, 2007). Emerging from the psychoanalytic and attachment theory traditions, the capacity to mentalize is considered a key psychological competence that is linked to early relational experiences in that the presence of a caring adult who treats the child as an intentional agent and supports them in naming and describing their emotions helps to develop this capacity (Nazzaro et al., 2017). Mentalization has become the basis of a specific therapeutic approach known as mentalization-based therapy (MBT) (Fonagy & Bateman, 2006) and is conceived as a common factor for therapeutic change across therapeutic schools (Sidis et al., 2022).
The concept of mentalization is understood as both complementary and supplementary to the ‘theory of mind’ and focuses on a special case of social cognition—that is, the manner in which affectively charged or emotionally significant relationships are interpreted (Taubner, 2015). In contrast to classical theories of mind, mentalizing is considered a developmental achievement that depends particularly on the quality of early relationship experiences (Taubner & Kotte, 2022).
Further, mentalizing is an integrative concept that encompasses several facets such as self-reflection, perspective-taking, or empathy (Allen, 2006; Choi-Kain & Gunderson, 2008) and has strong similarities with other concepts such as emotional intelligence. To delineate and distinguish mentalizing from related concepts, it is helpful to differentiate them along three dimensions: whether the object (or target person) is self or other, whether the content is primarily affective or cognitive, and what level of consciousness or explicitness is assumed (conscious-explicit vs. unconscious-implicit) (Choi-Kain & Gunderson, 2008). Table 1 provides an overview of mentalizing, and examples of related concepts established not only in the clinical field but also in work and organisational psychology; the table also summarises the similarities and differences between these related concepts:
Self-reflection (Grant, Franklin, & Langford, 2002; Greif, 2008) implies applying a mental theory to our own mental states—that is, the object of self-reflection is limited to the self. In contrast to mentalizing, self-reflection is a predominantly explicit process, while mentalizing may also contain automatic and implicit aspects. This implies that, in interpreting others, individuals may make systematic mistakes but without being aware of them—for example, by very quickly attributing hostility to others. Other concepts related to mentalizing which target (only) the other rather than (also) the self, are empathy and perspective-taking. Empathy, like mentalizing, also includes implicit aspects—that is, it can be defined as an automatic ability which entails sharing feelings as an adequate mirroring of affective mental states of another (e.g., Davis, 1983; Decety & Jackson, 2004; Paulus, 2009). While empathy is related to emotional mental states of others, perspective-taking is directed towards the cognitions of others (Paulus, 2009). Further, empathy is also a part of emotional intelligence. Allen (2006, p. 11) defined emotional intelligence and mentalization as ‘conceptual cousins’ in that both constructs pertain to identifying emotions in oneself as well as in other individuals, and using emotions to organize thinking, understanding, and regulating emotions. However, mentalization goes beyond emotion and also integrates cognition.
In summary, mentalizing includes all the above-named dimensions: it implies the mental processes by which individuals implicitly and explicitly interpret the actions of themselves and others as meaningful on the basis of both cognitive and affective aspects (desires, feelings, needs, and beliefs) of mental states (Bateman & Fonagy, 2004; Choi-Kain & Gunderson, 2008).
According to mentalization theory, an individual with highly developed mentalizing capacity can quickly switch from implicit to explicit mentalization when required. However, our mentalization capacity can be impaired by external and internal influences so that it is temporarily unavailable to the individual (Taubner & Kotte, 2022). In this sense, mentalization is understood as a dynamic construct, as theorised in the stress-related (or bio-behavioural) switch model of mentalization (Fonagy & Luyten, 2009, based on Mayes, 2006) which is empirically well-supported (e.g., Nolte et al., 2013; Taubner, Fonagy & Bateman, 2019). With increasing emotional arousal (specifically with attachment-related affective stress), a switch occurs from predominantly pre-frontally accentuated, controlled, and executive functional modes to a predominantly automatic processing of mental states (Taubner & Kotte, 2022). Consequently, the complexity and flexibility of the reflective function decrease, so that with increasing (attachment) stress, the otherwise available mentalizing capacity is reduced. How quickly—that is, at what specific stress level, the switching point is reached varies from individual to individual. Instead of ‘mature’ mentalizing, ‘prementalizing’ modes of thinking become dominant (Allen, Fonagy & Bateman, 2008). These are often accompanied by concretistic thinking, an intolerance towards alternative perspectives, as well as increased pressure to act (Taubner & Kotte, 2022; Taubner et al., 2019).
In summary, this implies that in situations of strong emotional stress (such as a conflict with a partner or co-worker), individuals are no longer able to resort to their otherwise available mentalizing capacities and instead experience a (temporary) ‘mentalizing breakdown’ (Taubner & Kotte, 2022).
There is an increasing amount of research on mentalizing and its role in psychotherapy. Several meta-analyses and literature reviews on mentalization in the clinical context have appeared in recent years (for an overview, see Luyten, Campbell, Allison & Fonagy, 2020). Mentalization is regarded as a success variable of psychotherapeutic transformation (Taubner, 2015). Therefore, numerous studies have investigated changes in mentalization because of psychotherapy. There is evidence that mentalizing (operationalised with the RF scale) can increase over the course of psychotherapy (Fischer-Kern et al., 2015; Katznelson, 2014; Levy et al., 2006) across different schools of psychotherapy (e.g., cognitive behavioural therapy—Babl et al., 2021; Karlsson & Kermott, 2006; Meier et al., 2022; psychodynamic therapy—Levy et al., 2006; Rudden, Milrod, Target, Ackerman & Graf, 2006; Vermote et al., 2010; psychoanalytic therapy—Katznelson et al., 2020; Taubner, Kessler, Bucheim, Kächele & Staun, 2011). However, as results have not been unanimous (Babl et al., 2021; Barber et al., 2020; Boldrini et al., 2018; Lüdemann, Rabung & Andreas, 2021), possible moderators of the degree of change in mentalizing have begun to be explored (e.g., treatment protocols, treatment durations, patient groups) (Barber et al., 2020).
Empirical evidence has revealed that increased mentalizing capacity in turn relates to other desirable outcomes. Higher RF scores are associated with psychological well-being (Fonagy et al., 2022; Katznelson, 2014) and function as a protective factor against psychopathology (Berthelot, Lemieux, Garon-Bissonnette, Lacharité & Muzik, 2019). Recently, the potentially health-protective function of mentalizing has been emphasized (e.g., Borelli et al., 2019; Ballespi et al., 2019; Schwarzer & Gingelmaier, 2020). The data suggest that mentalizing can at least partially compensate for the negative influences of aversive experiences such as acute occupational stress and strain and thereby reveals a compensatory, health-promoting influence by positively affecting well-being and mental health (Schwarzer, 2022). Allen (2013) also emphasised that promoting mentalization has a substantial impact on all psychotherapeutic treatments and argued that it can be considered a common factor in psychotherapy. Therefore, to obtain a better understanding of therapeutic change, several authors (Hörz-Sagstetter, Mertens, Isphording, Bucheim & Taubner, 2015; Josephs, Woods & Guillaume, 2004; Kivity et al., 2019; Kornhas, Zettl, Hausschild & Taubner, 2019; Kornhas et al., 2020; Möller et al., 2017) examined mentalizing within psychotherapies across therapy sessions in controlled case studies. The results revealed that the mentalizing capacity of patients fluctuated not only across but also within therapy sessions.
The concept of mentalization has been transferred to organisational and work-related areas, such as mentalization in social work (Kirsch, 2014), mentalization-based management (Döring, 2013), the mentalization-based ‘Professionals in Crisis’ programme (Bleiberg, 2013), the mentalizing community (Twemlow, Fonagy & Sacco, 2005), and mentalization in team supervision (Kotte & Taubner, 2016) and coaching (Goebel & Hinn, 2016; de Haan, 2012; Taubner & Kotte, 2022).
For the coaching context, De Haan (2012) suggested that the mentalization capacity of the coachee may be conducive to the goals of coaching—that is, to bring out hidden potentials and enable new insights and understanding of oneself and others. He posits that coaches would do well to develop this function in themselves and their clients (De Haan, 2012). Taubner and Kotte (2022) further specified the potential practical value of mentalizing for coaching: From a diagnostic perspective, mentalization ‘breakdowns’ can serve as indicators of emotionally significant experiences that are addressed or arise during the coaching engagement. From an intervention perspective, the basic principles of mentalization-based therapy provide guidance regarding which interventions could foster a coachee’s mentalization. Moreover, clients may benefit greatly from becoming familiar with the concept of mentalization, for example, because they become more sensitive and understanding to breakdowns in their own mentalization capacity or that of co-workers. However, empirical studies on the mentalization capacity of coachees have yet to be conducted.
The purpose of our explorative case analysis is to measure a coachee’s mentalization capacity in a workplace coaching process. Psychotherapy research has revealed that mentalizing, operationalised as RF, can increase over the course of psychotherapy (Katznelson, 2014) as well as in shorter cognitive behavioural therapy (CBT) in outpatient settings (Babl et al., 2021). Therefore, we assume that coaching can also foster mentalizing; we apply the RF scale to coaching sessions and seek to investigate the following research question:
Research question 1: Does a coachee’s capacity to mentalize change over the course of a coaching process (i.e., across sessions)?
Psychotherapy research has also revealed that patients’ mentalizing capacity fluctuated not only across but also within therapy sessions (Hörz-Sagstetter et al., 2015; Josephs et al., 2004; Kivity et al., 2019; Kornhas et al., 2019; Kornhas et al., 2020; Möller et al., 2017). In particular, against the background of the stress-related switch model, we seek to investigate the following research question:
Research question 2: Does a coachee’s capacity to mentalize fluctuate within individual coaching sessions?
To the best of our knowledge, the RF scale has not yet been used for assessing mentalization in coaching sessions. Thus, we intend to reveal the applicability and informative value of assessing a coachee’s reflective capacity beyond self-report measures through the explorative analysis of a single coaching case.
The coaching case that we explore for this study is taken from a comprehensive process-outcome study on workplace coaching. For this explorative single case analysis, we selected a coaching process that could be considered typical for the entire study sample and comparable to process characteristics described in surveys on the German-speaking coaching market (Middendorf & Salamon, 2021)—for example, in terms of coach and coachee characteristics and process duration. The selected coaching process was conducted face-to-face and consisted of five sessions with an overall duration of 8 hours and 18 minutes. The coach is male, 48 years of age, has worked for 10 years as a coach and earned a university degree as his highest level of education. The coach has completed a certified coach training (systemic coach), is a member of a coaching association (German Association for Supervision and Coaching (DGSv) and aligns himself with a systemic and psychodynamic coaching approach. The coachee of this case study has been a senior manager in the field of education for three years and is the line manager for 30 employees. She is female, 49 years of age, and holds a university degree. The coachee reports persistent exhaustion and a tendency to suffer from colds. When defining her goals for the coaching process, the coachee indicates that she wants to feel more confident about her senior management role, improve her self-esteem, and allow herself to make mistakes.
The RF scale (5th edition) (Fonagy et al., 1998) is an empirically based measure for dimensional and categorical assessment of mentalizing and is considered the ‘gold standard instrument that has operationalised the full range of the mentalizing concept and made it empirically tangible’ (Taubner, 2015, p. 32). The RF scale is used in all schools of psychotherapy and is an established instrument for measuring higher-order cognitive processes (Babl et al., 2021; Sidis et al., 2022). It entails an elaborate coding procedure, thereby determining the extent to which the client can reflect on his or her attachment-related experiences regarding various mental states (Fonagy et al., 1998). By using a scoring manual, the transcripts are coded on an 11-point scale from -1 (anti-reflective) to 9 (exceptional RF). This leads to the distinguishing of two subgroups: The low to negative reflective function (values -1 to 3) and the medium to high reflective function (values five to 9), with the value of 4 being the cut-off (Taubner, White, Zimmermann, Fonagy & Nolte, 2013). In non-clinical populations, the reflective capacity—as measured using the RF scale—is indicated by an expected mean value of five (‘Ordinary’ RF) (Chiesa & Fonagy, 2013; Fonagy et al., 1996). The RF scale was originally developed as an evaluation instrument for the ‘adult attachment interview’ (AAI) (George, Kaplan & Main, 1996), a semi-standardised interview for assessing attachment styles, and has been validated multiple times (Fonagy et al., 1998). The application of the RF scale to transcripts of psychotherapy sessions is also well established (Hörz-Sagstetter et al., 2015; Josephs et al., 2004; Karlsson & Kermott, 2006; Talia et al., 2018). For the analysis of psychotherapy sessions, instead of AAI interview responses, psychotherapy transcripts are divided into 150-word blocks and rated individually. The rating itself is conducted in exactly the same manner. For trained raters, the interrater reliability of the RF scale is considered ‘good’ (intraclass correlation coefficient, ICC) when the value is between 0.70 and 0.91 (Fonagy et al., 1996; Taubner et al., 2013). Further, the discriminant validity and convergent validity were independently substantiated (Bouchard, St-Jaques, Robillard & Renaud, 2008; Müller, Kaufhold, Overbeck & Grabhorn, 2006).
The use of trained raters is a method commonly used in psychotherapy research to increase interrater reliability and improve the validity of assessments of psychotherapeutic interventions (Mörtl & Gelo, 2015). In this process, raters are specially trained to employ a particular method of assessing therapy sessions. In a second step, their reliability is measured against that of established expert raters. Once they achieve a predefined level of interrater reliability, they are authorized to conduct ratings. In the current study, the rating was conducted by the first author of this study, who is a certified and reliable RF rater. She completed two three-day training workshops and achieved reliability on an extensive set of transcripts. Her interrater reliability is Cronbach’s α = .91. The transcripts of the coaching sessions were analysed according to the manual (Version 5; Fonagy et al., 1998) in the following manner: First, each transcript was divided into 150-word blocks. Second, an RF rating was assigned for each block (Talia et al., 2018). When coding each section, attention is paid to the presence of the following items: (a) if the client is making an explicit reflection or only relating a narrative (e.g., without naming mental states); (b) if the reflection leads to outlining a mental state, underlying behaviour, or other mental states (‘successful’ reflective functioning); and (c) whether qualitative markers for reflective functioning are present (Fonagy et al., 1998). These qualitative markers include (c1) the awareness of the speaker regarding the nature of mental states; (c2) the explicit effort made by the speaker to tease out mental states underlying certain behaviour; (c3) the speaker’s recognition of the developmental nature of mental states; (c4) the recognition of the probable mental states in relation to the interaction partner; and (d) the presence of evidence of impaired reflective functioning. A global RF rating is then assigned according to the rules given in the manual. In this study, the rating was produced blindly—that is, the rater was not able to determine at what point in the process the sessions took place. It was only after the blinded evaluations had been unmasked that the progress of the RF values per session and throughout the coaching process were analysed.
Figure 1 depicts the global score for the coachee’s mentalizing capacity for all five sessions of the coaching process.
To answer research question 1, if the coachee’s capacity to mentalize changes over the course of a coaching process, the following picture emerges: The coachee begins with an RF score of 3 (low level of mentalization) in the first session, increases to an RF score of six (high level of mentalization) in the second session, decreases to an RF score of four in the third and fourth sessions, and increases to an RF score of five in the fifth session.
The first coaching session is characterized by an agreement on the goals of the coaching process. In this session, the coachee shows a low mentalizing capacity, and the session was assigned a global value of 3. This implies that the coachee does have a partial understanding of her feelings but tends to express herself superficially and use banalities, generalities, and clichés in her conversation. Complex aspects such as conflicts or ambivalence are not clarified.
In the second coaching session, the coachee reports being increasingly tired and exhausted as well as having crossed her boundaries. The coach and coachee explore the possible reasons for this together, such as that prioritizing her own needs is accompanied by feelings of guilt, rage, and fear of punishment. The coachee then speaks of her childhood and her relationship with her parents. In this session, the coachee shows a higher mentalizing capacity. Thus, the session was assigned a global value of five, which corresponds to the average mental capacity in non-clinical populations (Chiesa & Fonagy, 2013). In contrast to the first session, there are now persuasive examples that the coachee possesses a conceptualisation regarding the inner states, thought processes, and wishes of others. This also applies to the conceptions of her own inner world of thoughts and feelings. Further, the coachee shows an average competence to embed her experiences in a sense of purpose. The understanding of inner mental states is higher in certain instances, but this level cannot be maintained in specific problem areas.
In the third coaching session, the coachee speaks about her concern that her leadership position might constantly be questioned by her team. For this session, the coachee’s mentalizing capacity was assigned a global value of four. While there are certain statements with qualitative markers, the statements overall remain rather vague. The mental concepts are generally not elaborated upon or not embedded in a comprehensible example. For example, the coachee discusses a difficult performance appraisal. Although she voices her anger towards her manager, the statement remains rather vague, and the descriptions are not sufficiently detailed to be comprehensible, despite the coach asking for more information.
In the fourth coaching session, the coachee decides to address her relationship with her manager. The coachee’s overall mentalizing capacity does not change. She was again assigned a global value of four for this session. Once more, the coachee shows limited mentalizing capacity. She addresses her conflicts—that is, statements with qualitative markers are found—but they remain vague and the coachee is not able to explicitly formulate her feelings and thoughts.
In the fifth and final coaching session, the coach and coachee reflect upon the coaching process they have gone through. The coachee displays a consistent manner of reflecting upon feelings that is comprehensible to the rater. The overall session was assigned a value of five. The coachee is able to refer explicitly to attachment relationships and describe specific mental states that are relevant to the respective situation. However, her statements are not particularly complex or elaborate.
After analysing the changes in the coachee’s mentalization over the course of the coaching process, we now turn to fluctuations within individual sessions (Research Question 2). Figure 2 depicts the within-session fluctuations of RF values across the five sessions of the analysed coaching process.
In the first session, the RF values varied between three and five, in the second session between three and six, in the third session between three and five, in the fourth session between one and five, and in the fifth session between four and six. The expected fluctuations of RF within each session were found in all the analysed sessions. Figure 3 presents the frequency of the individual RF ratings (ratings of one, three, four, five, and six were assigned) for all five coaching sessions. In session one, the RF value of three was most frequently assigned (19 times). In session two, four was most frequently assigned (16 times), closely followed by five (14 times) as well as a few cases of six (seven times). In sessions three-five also, four was most frequently assigned (16, 21, and 13 times, respectively), although session five also included a substantial frequency of assigning five.
In order to better understand these results, session four is selected as an example and explained in more detail through Figure 4. At the beginning of the session, the coachee is easily able to describe her goals for the session and reflects upon her feelings towards her manager. However, the statements are lacking in complexity and specificity (thus, an RF value of four). The coachee begins this session with the wish to address her insecurity in dealing with her manager as the topic for the day. When the coach asks her how important this topic has been for her recently, the initial decline in her capacity to mentalize becomes apparent. The coachee’s mentalizing capacity is largely absent and she provides evasive and incomprehensible responses. Empathy and reflection are substantially inhibited at that time. However, the coach persists and asks for concrete scenarios and feelings. Then, the coachee slowly engages herself and increases her mentalizing capacity again. In the seventh word block, she is able to even adopt an intergenerational perspective and establish a connection between herself and her mother (RF rating of five). Since then, the remaining RF ratings fluctuate between the values three and four, except for when the coachee once again experiences a breakdown of her mentalization capacity in the 29th word block. In this word block, the coach returns to a previously described situation between the coachee and her manager and asks the coachee to specify her insecurity. The coach attempts to explore the coachee’s feelings and the reason for her insecurity. However, the coachee blocks this, answers in monosyllables, and dismisses it with a simple explanation (‘I don’t like that’). These breakdowns lead to an effort on the part of the coach to understand the coachee’s mental states so that the coachee does not withdraw into herself or offer pseudo-explanations too quickly.
Our exploratory case analysis investigates the level of mentalization over the course of an entire coaching process. In the following, we suggest initial interpretations of our analysis and directions for future research as well as practical implications.
First, the analysis of the entire coaching process reveals that the coachee’s mentalizing capacity most notably increases when the first and last sessions are compared, with the mentalizing capacity increasing from a low (RF value of three) to an average mentalization level (RF value of five). The mentalization level achieved by the coachee at the end of the coaching process corresponds to the expected value of five in non-clinical populations (e.g., Chiesa & Fonagy, 2013; Fonagy et al., 1996). One possible interpretation of this pattern is that coaching leads to an increase in mentalization. To the best of our knowledge, this is the first empirical indication—apart from self-report measures—that a coachee’s reflection can improve over the course of a coaching process. Further, an increase in mentalization could be seen both as a desired outcome of coaching and as a process factor conducive to achieving coaching goals. However, an alternative interpretation for the observed changes in mentalizing capacity across sessions is based on the nature of the coaching process. The first coaching session is often about contracting (Bachmann & Steinke, 2019) as well as clarifying boundary conditions and goals, which is likely to require and provide less space for mentalizing than subsequent coaching sessions. In addition, at the beginning of a coaching engagement, the bond between the coach and coachee still needs to be built. Therefore, it can be assumed that a secure attachment to the coach is only just developing, which in turn leads to more relational stress and, thus, less mentalization, based on the stress-related switch model (Fonagy & Luyten, 2009; Mayes, 2006). Fonagy et al. (2019) suggested that improvement in mentalization can only take place in a secure, trustworthy therapeutic relationship. Therefore, the coachee’s mentalizing capacity may be lower initially and then return to the coachee’s ‘normal’ mentalizing capacity of between four and six in the subsequent sessions.
Second, we find substantial fluctuations of RF levels within the coaching sessions. In psychotherapy research, fluctuations of mentalization within therapy sessions have already been investigated (Hörz-Sagstetter et al., 2015; Kornhas et al., 2020). Following the idea of stress-related temporary breakdowns of mentalization, this could imply that highly emotional situations for the coachee can emerge during the coaching process or that stressful work situations that are addressed are re-experienced during coaching. Once coachees experience a certain stress level, they can no longer access their otherwise available mentalizing capacities. In our case study, the coachee’s mentalization capacity consistently collapsed in connection with the coachee’s struggles and frustration in dealing with her direct manager. When working on this issue, the coachee did not (yet) appear to be able to deal with the conflictual situation and required the specific support of her coach to reflect on her behaviour and feelings towards her manager in these situations.
In summary, the application of the RF scale to coaching sessions appears to be a promising addition to the research on reflective processes in workplace coaching. Although our explorative case analysis is only a starting point, it is a first step that leads the way to empirically investigating mentalization in coaching. It does so by means of a validated rating system rather than relying on assessing reflection via a self-report measure. Therefore, it meaningfully contributes to advancing workplace coaching by paving the way to coaching process research that employs rigorous methods of data collection for a construct—reflective capacity—that is considered to be at the core of numerous coaching definitions (De Haan, 2012; Greif, 2008; Smither, 2011).
In order to systematically examine whether the patterns we observed over the course of the coaching engagement and within sessions hold across coaches and coachees as well as to empirically test the alternative explanations we suggest, process studies assessing RF in a larger sample of coaching processes need to be conducted. In addition, such studies could also test whether mentalizing functions as a mediating or moderating factor for coaching success.
One limitation of rating RF based on coaching sessions is that it may be confounded with the structure and progress of the coaching engagement. For example, the RF level of a coachee at the beginning of a coaching engagement may be underestimated because RF may be impaired due to the initial lack of trust and the often highly structured first session. In order to adequately attribute pre–post changes in RF, a measure independent of the coaching sessions themselves would be required. This would enable the assessment of a potential increase in RF over the course of the coaching process without confounding factors. Analogous to psychotherapy studies that measure RF by means of an AAI interview before the beginning and after the end of therapy, an interview suited for the work context could be developed.
A further step for future research could be to examine the relationship between coach interventions or behaviour and coachee mentalizing capacity. Existing psychotherapy studies have examined fluctuations in mentalization in relation to therapist interventions and the thematic focus of sessions (Hörz-Sagstetter et al., 2015; Kornhas et al., 2019, 2020). Initial findings suggest that mentalization is explicitly stimulated by mentalization-promoting questions (Kivity et al., 2019; Kornhas et al., 2020; Möller et al., 2017). In contrast, no specific topics have yet been identified that are associated with changes in the level of mentalization (Kornhas et al., 2020). Coaching process studies, which examine phenomena that may be related to mentalizing—such as experiences of insight and moments of awareness in coaching (Lightfoot 2019)—may also indicate coach interventions that may promote mentalizing. For example, Lightfoot (2019) revealed that an insight event for the coachee was generally preceded by the following steps: 1) the coach provides a summary; 2) the coachee reflects verbally; 3) a pause occurs; 4) the coachee’s insight manifests, thereby suggesting that reflection can be considered a response to a coach action. Future research should explicitly examine which types of coach interventions promote and which types reduce mentalization.
Further, with the help of the RF scale, reflection in coaching processes can be measured independently of self-reports. This is an important contribution to coaching research, since reflection is a core concept in coaching research, and self-report data have been shown to lead to an overestimation of our own performance (Athanasopoulou & Dopson, 2018) and thereby to an overestimation of coaching effectiveness (e.g., Bozer & Jones, 2018; De Meuse, Guangrong, & Robert, 2009). Moreover, Fontes and Dello Russo (2021) indicated that three types of change in reflectivity can be observed in coaching: Alpha change refers to a substantive actual change in reflection (i.e., an increase or decrease), beta (or response shift bias) to a recalibration of our assessment of reflection, and gamma to a re-conceptualization of reflection. They show that measurement by self-reporting alone can cause a change to not be measured or the difference between two measurements not reflecting the actual variance. Therefore, beta and gamma changes may confound the assessment of within-person changes in self-report measures of reflectivity, thereby making it critical to assess reflectivity through other-assessments, such as the RF scale. While the RF scale is a validated and proven assessment method, it obviously cannot capture subjective internal processes. Therefore, combining RF assessments with self-reports of reflexivity may be a promising avenue for future research to develop the capturing of different types of changes in reflexivity that coaching may elicit.
Our study reveals that despite possible overall improvements in mentalization throughout the coaching process, fluctuations within coaching sessions are to be expected and were indeed found in our explorative case analysis as well. Therefore, it is important for the coach to serve as a ‘seismograph’ and to pay attention to ‘tremors’ in the coachee’s mentalizing capacity.
These ‘tremors’ can serve as hints for the coach to identify emotionally ‘hot’ topics for the coachee. In particular, mentalizing breakdowns can serve as important indications of significant moments or themes in the coaching process, which in turn are relevant to the success of the coaching. Taubner and Kotte (2022) suggested that, in the work context, mentalizing capacity can be influenced—that is, impaired or strengthened—by role-related factors (e.g. role ambiguity and role conflict), team-related factors (e.g. cohesion and psychological safety), and the organisation as a whole (e.g. insecurity through major change and organisational restructuring) in addition to individual factors (e.g. a vulnerability in the sense of early switching points or resilience in the sense of secure attachment). Thus, fluctuations in coachee mentalization can be conceived as a diagnostic ‘instrument’ for the coach to better understand the coachee and navigate the coaching process accordingly.
Denise Hinn is a research associate at University of Kassel, Germany. Her research focuses on coaching in the workplace, particularly on the topic of mentalizing in coaching.
Silja Kotte is a professor of business psychology at the University of Applied Sciences Aschaffenburg, Germany. Her research focuses on coaching, human resource development and leadership.
Heidi Möller is a professor of counselling psychology at University of Kassel, Germany. Her research focuses on workplace coaching, coaching supervision, and leadership development.