An enactive perspective on comprehending leadership: A comparative case study approach

2.1 Interpersonal representations

Empirically and at a subjective level, research in leadership has recognized the importance of the relationships between leaders and their subjects [2, 39]. Good performance and positive behaviors were found to be complementary between leaders and their subjects, as noted in transformational leadership [3, 5] and charismatic leadership research [15, 64]. Similarly, research in destructive leadership correlated the leader’s behavior to the employee’s psychological [76] and social well-being [66], and its impact on the organization [19, 55], also called ‘high quality connections [17]. These connections significantly alter behaviors [17], subjective experience [49, 61] and capabilities [8, 63]. Parallel to this, high quality connections offer physiological benefits and have an impact on well-being and performance [4, 28]. Studies have shown how physiological markers [13, 45], emotions and behaviors [6, 22] in a group are contagious. Resonant leadership as a state where there is synchrony in subjective experience, behaviors and physiology between the leader and the group [7]. This formed the motivation for the current work, considering leadership as a phenomenon and recognizing the physiological and behavioral correlations between the leader and the group. Theory of autopoiesis provides the link here as it explains the emergence of a phenomenon taking into consideration all the aspects of human experience, physiological, behavioral andenvironmental.

2.2 Autopoiesis: An enactive perspective on human experience

There exists a large body of knowledge, referred to as autopoietic theory, developed by two Chilean biologists: Humberto Maturana and Francisco Varela (1980). This body of work concerns the dynamics of living systems, asking the questions: What is the definition of a living entity? What does it mean to be alive [48]?

Maturana and Varela define a living entity as a system that produces itself, that is, a system whose output is itself. Such a system is autopoietic, that is, self-producing. An autopoietic (living) system is defined as a network of processes of production of components that produces the components that through their interaction and transformations continuously regenerate the network of processes that produced them [48]. Autopoiesis is basic to the living individual. What happens to the individual is subservient to its autopoietic organization, for as long as it exists the autopoietic organization remains invariant. What this means is that its identity and, therefore, its emergent global properties, are generated through a process of self-organization, within its network of components. However, we must also realise that this process of self-organization is conditioned by a two-way process of local-to-global and global-to-local causation; thus, we need to consider the mutual embeddedness of component dynamics, autopoietic entity and its environment. First, there is the local-to-global determination (‘upward’ causation) through which the entity, with its properties, emerges. Secondly, however, there is global-to-local determination (‘downward’ causation), where global characteristics constrain or direct local interactions between the components. Thus, the internal dynamics of the components (neuronal nets, metabolic nets, energy flow and so on) generate and sustain the global properties of the autopoietic entity. At the same time, however, the global properties (body, consciousness, mind, emotion, and so on) constrain and govern the behavior of the individual components.

This dialectic relationship between local and global levels is described in autopoietic theory as ‘reciprocal causality’ [68, 69]. For example, in organisms with a nervous system, the rules of interactions within the neuronal network are in reciprocal relationship with the overall activity of the autopoietic entity. To a very large extent, behavior is a regulator of perception. We enact our world rather than recognize one [41, 68–70]. Escher’s crystal ball on Fig. 1, gives an artistic representation of the concept of enaction: one could say that the quality of our understanding of what we see, notice, become aware of and take into consideration, is determined by the quality of our being in the world: physical, mental, emotional and relational.

An enactive perspective of understanding and developing leadership will require taking a systemic view and considering the interrelationships between the physical, mental, emotional and relational aspects of being in the world. This perspective could relate to a search for an integrated paradigm of leadership [78] and on resonant leadership [7]. However, in this present paper a different framework is developed, one that focuses on exploring the correlation between the physiological, behavioral and social aspects of leadership.

3.1 Case study methodology

A case study is an informative analysis of an individual or groups [54] that efficiently creates an understanding of an issue or object that is classified as being complex [79]. It can also be utilized to support a theory that has been researched in the past. Case studies place considerable importance on circumstantial analysis of conditions and relationships that result in particular events, assisting in understanding any deep-rooted behaviors and social dynamics. Careful planning is needed to reduce the chances of bias, due to the researcher’s direct involvement in the case. The case study method was used in this research as an exploratory tool in two organizations. The exploratory process of case study research consists of seven core steps [71, 79].

The researcher involved in the data collection realized, however, that another stage is required prior to the data collection stage. This could be called acclimatization as on Fig. 2. This stage was required to establish rapport and understanding between individuals involved in the case study and the researcher. Not only did it increase the confidence of the participants but also created an aspect of trust between the participants and the researcher. The first step involves focusing on the research objective as to what the researcher wants to know from the cases. The design phase of the process involves the researcher making judgments on the selection of case studies. Case studies can be performed at different levels in accordance with the research question, namely, a single case study or a multiple case study. Sometimes, a multiple case study consists of knowledge obtained from separate cases that contribute to the study as a whole. This research work follows such a process. The whole case is defined by the outcomes of two separate cases. The key strength of a case study lies in its ability to involve multiple sources and a wide range of techniques in the data collection process [80]. During the data collection stage, the researcher collects evidence and stores it in a systematic way. This allows the case to be flexible and allows documentation of the changes made, based on the observed phenomenon. The use of field notes is highly recommended as they record attitudes, questions from participants and record stories etcetera [80]. During the analysis phase of the case study, the data is examined using multiple constructs to find links between the research question and the research object. The researchers, however, remain open to novel insights and opportunities. The multiple methods used in the process of data collection and analysis permit triangulation of the data and, thus, the building of a strong case [71, 79]. The final phase then includes reporting the case study. It is essential that the researcher explicitly reports all the experience, findings and evidence [71]. In this research project, the findings from the two cases are examined and compared, making it a comparative case study.

3.2 The cases

Case A and Case B are part of several clinics that come under one umbrella organization. The organization at the top level takes care of marketing, recruiting, business development, material purchase and operations etcetera for the clinics. The clinics are solely responsible for service delivery to their patients, escalation of issues and observance of any policies set by the top management. A separate clinic management structure exists at the clinics, for which the clinic head is responsible. This includes rosters, patient management, clinic issues and costing etcetera. DRSR is the head of Case A and VKDRSR is the head of Case B.

3.3 Case A

Case A has fourteen employees in total, of which eight volunteered to take part in this research work as participants. One of the participants had to be excluded from the investigation as he is a heavy smoker, which will affect the HRV data when considered as a whole. The participant coded as DRSR is the clinic head, a doctor herself. Two senior doctors, two junior doctors, a clinical nurse and an administrative assistant are part of this case study (n = 7). The mean age of Case A at that time was 29, with a maximum of 33 and minimum of 21. The standard deviation was 4.90.

3.4 Case B

Case B also had fourteen employees, out of which seven (n = 7) agreed to participate in this research work. VKDRSR is the clinic head, a doctor himself. In addition to the clinic head, three senior doctors, two junior doctors and a clinical nurse were involved in the study. The mean age at Clinic B at the time of the study was 29.71, with a maximum of 45 and a minimum of 20. Standard deviation was 7.32.

3.5 Ethics

This research was approved by Northumbria University ethics committee and required the informed consent taken of each participant. The identity of the participants was protected using a coding technique and the organizations cannot be named owing to the nature of the research and agreement.

3.6 Heart Rate Variability (HRV)

Harung et al. (2009) had, using EEG measurements in a laboratory, identified that leadership excellence is in correlation with psycho-physiological integration [27]. EEG methods can be invasive and need to be performed within a laboratory, whereas ECG measurements are less invasive and more mobile. HRV is derived from the ECG signal and is a marker of autonomic nervous system function [14, 58]. It is calculated as the time interval between two successive heart beats (R peaks) in the ECG trace Higher HRV is indicative of having greater self-regulatory abilities and lower stress levels [14, 58].

The ECG data was collected using HeartMath Institute’s emWave device, the raw data cleaned [43] and then analyzed using Kubios, an open source HRV analysis software developed by the University of Eastern Finland.

3.7 HRV recording and analysis

The average recording time for each participant was 16 minutes. A five minute sample from the tachogram (between 6 and 11th minute), was analyzed to calculate the HRV measure. The timing of the sample ensured that the participant had already relaxed into the interview. Greater HRV values represent a resting yet focused state and better psychophysiological well-being [14, 58]. The HRV measures for every individual in the clinics were taken and then analyzed using SPSS statistics. The means between the clinics were compared and also the means of the Heads of clinics were compared separately to investigate whether there were any differences. Normal distribution and paired t tests were performed as an investigative measure, although it was expected that significance could not be achieved owing to a low sample size.

3.8 Rickter interviews

The Rickter Scale is a unique interview method developed by Rick Hutchinson and Keith Stead during the early 1990s [29]. It involves a visual analogue scale (VAS), for an individual to self-assess themselves against factors relevant to their personal/professional lives. In the case of this project, the factors in consideration were the attributes of embodied leadership [37], as described in Table 1. The Rickter interviewing method was used in this research as a means of evaluating embodied leadership capability [37].

The Rickter method’s theoretical grounding is in appreciative inquiry, emotional intelligence, applied positive psychology and solution-focused motivational interviewing. It gives utmost importance to a process called ‘anchoring’, whereby participants are encouraged to hold the slider on the Rickter board (the Visual Analogue Scale is presented as a set of physical sliders on a board), reflect and notice their emotions, while moving the slider across present, past and desired state for the assessment of each attribute. It is particularly suitable for assessing embodied leadership attributes as it facilitates self-assessment through connecting with one’s reasoning and emotions, and exploring personal experience of past and present while linking this experience with desired future states [60]. In addition to self-assessment, the Rickter process requires exploring the meaning of the self-assessment through encouraging the interviewee to share narratives of lived past and present experience and imagined desired states. This assisted the researchers in making sense of the data and in the data analysis [36, 37].

3.9 Analysis of rickter interviews

The interviews were analyzed based on a visual analogue scale and narrative transcripts. SPSS was used for statistical analysis of the VAS data [61] and the interviews were analyzed applying a grounded theory based analysis process, by bringing forward themes identified from the transcripts. It should be clarified that this is not a grounded theory study; rather, it uses the coding technique, where the results are grounded in the data [73]. This method has been used successfully in the past for coding and identification of themes in case studies [72]. It is important to mention here that the themes are built not only on the basis of the researcher’s interpretations, but are also based upon the research participants’ perspective, which ensures credibility and rigour. The analysis follows a Straussian approach [75] until the level of identifying the open-ended codes. Open coding was performed to understand the concepts of the data. This was followed by a review of all the available data, which resulted in eliciting the embodied leadership attributes that were then used in the case studies for further exploration. Memos were taken while analyzing the themes to keep track of the subtle semantic nature of those themes.

3.10 Leadership Behavior Development Questionnaire (LBDQ)

The LBDQ is a three-part questionnaire that is used by researchers to understand the behavior ofa leader from the perspective of the subordinates, the leadership qualities that subordinates desire, and the behavior of a leader from the perspective of the leader, via a behavioral self-assessment questionnaire that the leader answers her/himself [74] as in Table 2. The work on LBDQ first started in 1945, as there were then no theories describing leadership behavior. Years of research led to the first version of the questionnaire in 1957.

After a general application of the questionnaire, a few more attributes with regard to self-definition and courtesy were added to the questionnaire and the latest version was released in 1963. It is still being used today as a means to analyze leadership behavior. When administered carefully, the questionnaire can provide a holistic measure of leadership within an organization. It has been successfully applied in different organizations in the past and data validation of the same proved to be successful as well. This method also takes up a socio-constructivist approach and is thus in accord with paradigm of the present project: understanding leadership as a phenomenon in a social context. With regard to this research project, the LBDQ was administered in the two case study organizations.

LBDQ is a measure that involves analysis in addition to describing a trait. It also takes real situations into account, thus, suitable for application in this study. The LBDQ analyzes twelve categories, each consisting of either five or ten sub-categories. The participant’s scores (in brackets) in each sub-category are represented alphabetically: A (5), B (4), C (3), D (2) and E (1). For a few sub-categories, the scoring system works in reverse order and they are starred in the scoring sheet for the researcher’s reference. A visual analogue scale approach to analysis is followed and the final scores which are summated are analyzed using statistics in SPSS.

3.11 Researcher’s diary

A researcher’s diary was maintained throughout the time spent in the clinics. This helped the researcher to describe events and experiences on a day-to-day basis, recording the new knowledge, feelings and reactions of not only the researcher, but also other individuals connected to the clinic. The diary was maintained in electronic format and is a clear primary data source offering qualitative evidence. Reflexivity was practised [50] after time spent in the clinic, whereby the researcher came back from a day of work and described what he felt and experienced in the clinics. The diary consists of factual accounts of the tasks performed by the researcher, thoughts, questions for exploration, observation, reports and action plans, etcetera. The experiences described in the diary helped in the analysis. The diary was maintained separately for both clinics and thus comparisons could be drawn.

4.1 HRV results

The people working at Clinic A had greater average measures of HRV than those working at Clinic B [59]. As there is a high level of correlation between HRV and cognitive behavioral attributes, it may be argued that Clinic A could exhibit leadership behaviors pertaining to embodied attributes. Figure 3 summarizes the time and frequency measures of HRV in both the clinics.

4.2 Rickter interview findings

This section analyzes the difference in the scores of the visual analogue scales and the narratives with respect to the leadership attributes that were included in the Rickter interviews. This research considers the present situation of the clinic and the desired or future state. The present state is considered as it represents the employee state and the leader’s effect on the clinic. The future or desired state is considered as it represents the aspirations of reaching a higher state of the particular attributes in the clinic. The past state could be considered to acknowledge the progress that has been made by the employees from what they were as they perceived themselves in the past. There are a few sections in the narratives where some of the participants have either credited the leader with their personal development and well-being or blamed the leader for their decline. Table 3 represents the means of the attributes from all the participants in the clinics. Table 4 represents the attribute scores of the leaders in their respective clinics. The scores that have been starred in Table 3 represent a higher attribute score, which characterizes more effective embodied leadership. It is evident that Clinic A is leading in scale on both present and future or desired states in all the attributes except Congruence and Decision Making.

It is interesting to see that Clinic A is behind Clinic B with regards to the situation in the past. The attributes of uncertainty, congruence, intuition, reflective practice, decision making, meaning/presence, intention and authentic presence are either equal or lower in Clinic A. This could imply that collectively the clinic had managed to overcome its tough past. From Table 3, it could be implied that, overall, Clinic A is doing a better job than Clinic B, i.e. has scored high on the scales of the attributes of embodied leadership.

Table 4 shows the scores of the attributes that the leaders of the clinics have scored themselves. The present situation scores support the leader of Clinic A. When it came to the desired future state, the leader of Clinic B aspired higher than that of his counterpart. Another surprising aspect of Table 4 is that the leader of Clinic A had lower past scores than the leader of Clinic B, except on the attributes of decision making and authentic presence. This might indicate that the leader of Clinic A had worked more effectively and understood the complexities involved in managing a clinic better than her counterpart. Additionally, the visual analogue scales indicate a clear advantage to Clinic A over Clinic B in the leadership attributes.

Additionally, the following comparison will bring forth any difference in operations or behaviors found in the clinics through the descriptions of the participants during the Rickter interviews. Table 5 shows the different themes of opinions of the participants about the functioning of their clinic. These opinions were extracted from the interviews.

4.3 Comparison of LBDQ results

The LBDQ questionnaire was used to understand the behavior of the leader from the perspective of that leader’s subordinates: their expectations and whether their leader fulfils their expectations. From Table 6, it is observed that the staff of Clinic A have higher expectations of their leader than the staff of Clinic B in all characteristics except in tolerance of uncertainty, tolerance of freedom, predictive accuracy, and superior orientation. What is also interesting to see is that the leader of Clinic A reaches the expectations of her staff only in the characteristics of tolerance of uncertainty, whereas, the leader of Clinic B was able to reach the expectations of his staff in the characteristics of representation, tolerance of freedom and consideration.

Previous findings in this research have given an impression that the leader of Clinic B expects his staff to take ownership and responsibility of their own decisions and actions. This is reflected here in the observation of tolerance of freedom, where the leader meets the expectations of his staff. With regard to Clinic A, there seems to occur a reverse Pygmalion effect [18], where the good performance of the leader has raised the expectations of the staff and thus the leader is trying to meet those expectations. The LBDQ thus presents mixed results, favoring neither Clinic A nor Clinic B.

4.4 Researcher’s experience of the working profiles

At this point, it is important to discuss the difference in the profiles and working methods of the two Heads of clinics as there is a great difference in the working cultures of the clinics. The Heads of both clinics share the common ground of a strong work ethic and a good work-life balance. One of the critical factors that exists in their methods is that the head of Clinic A believes in a guided or mentored learning approach, sharing knowledge proactively with the clinic staff, whereas the head of Clinic B believes in independent learning and reactive dissemination of knowledge. This was visible clearly in the case of Clinic B during the compression incident where the staff did not know how to operate the compression unit and the clinic head had to come and fix the problem. Another key difference is that the head of Clinic A took personal responsibility for training any staff in the clinic, whereas the head of Clinic B delegated training duties to other staff that he deemed as knowledgeable. Both clinics were centered on clear performance indicators, that is, quality and treatment standards for their staff, apart from a few restrictions that were enforced in Clinic B, for example, a ‘staff room for doctors only’ rule, which seemed divisive. Another difference between the clinic heads is that the head of Clinic B had financial interests in the clinic and had made an investment in the clinic. Additionally, he had owned a clinic in the past and had the experience of running an independent clinic. The head of Clinic A had started her career working in clinics and this present position was the first in which she assumed responsibility for an entire clinic.

The head of clinic B is very knowledgeable and experienced, whereas the head of Clinic A is knowledgeable, but her experience is limited. Demographically, there is a 12- year age difference between the two clinic heads, in addition to the gender difference. Another observation that was made is that the head of Clinic A exuded empathic intelligence [30], which was clearly missing in the head of Clinic B. Some of the staff of Clinic A had mentioned that they had observed a change in the behavior of their clinic head, her behavior becoming more appreciative. This change in behavior could be attributed to the maternal instincts of the clinic head, who had in the recent past given birth [32]. An observation of behavior of mothers at workplaces pre and post-natal, reported a shift of mothers behaviour towards becoming more compassionate [32]. This is attributed to oxytocin that is produced during childbirth and lactation [46] and that increases parasympathetic activity [50]. Raising the parasympathetic activity is linked to increased HRV [43, 57] and to the reduction of anxiety or stress. From what has been observed above, the head of clinic A clearly has an advantage over her counterpart. One of the key points to consider is that although independent and responsible learning conditions may be good, it is necessary to understand what the staff wants in the first place. From the experience of the researcher, it seems that the head of Clinic B needs to have a dialogue with his staff regarding the functioning and needs of the clinic. These are the key differences in profiles and working styles between the heads of these two clinics, a summary of which is tabulated in Table 7.

4.5 Difference in performance parameters

The clinic performance factors were collected from the HR and Finance department. Table 8 indicates a clear advantage as regards training period, income, unscheduled leaves and equipment wastage.

5.1 Linearity in findings

From the VAS results of the Rickter interviews it was noticed that the participants in Clinic A exhibited higher scores of embodied leadership attributes than their counterparts in clinic B. Similarly, the leader of clinic A exhibited higher scores than the leader of clinic B. From the narratives of the Rickter interviews, clinic A was able to garner more positive themes than clinic B. Correlation between physiology and subjective experience (embodied leadership attributes) was achieved when the HRV results showed that participants of clinic B were more stressed than the participants of clinic A. There was also a significant difference between the HRV data of the leaders of both clinics, where the leader of clinic A was less stressed than her counterpart in clinic B. The LBDQ questionnaire revealed that the subordinates were satisfied with their respective leaders. However, in clinic A, a reverse Pygmalion effect occurred whereby the subordinates expected more from their leader, which was due to her good performance in the recent past.

The themes from the researcher’s experience in the clinics also favoured clinic A and the themes gathered from the researcher’s diary closely matched the themes from the narratives of the Rickter interviews. Other aspects brought forward were the working styles and beliefs of the leaders in both clinics. A key revelation was that the head of clinic A followed a mentoring approach in the clinic, taking ownership of the development of her staff at every step. The head of clinic B took a self-sufficiency approachin which he believed that the staff was mature enough to understand the world and its members could be left to fend for themselves. Lastly, it could be concluded that clinic A was physiologically and cognitively more efficient than clinic B; there exist significant correlation between the physiological, behavioral and social domains of leadership.

5.2 Strengths and limitations

This work has had its own set of challenges. Owing to time and financial restrictions, only two clinics were studied. Cultural differences could affect the generalizability. The authors believe that a wider selection of cases would have provided richer data. Although case studies are useful in understanding the culture and behavior of groups, one of the primary limitations is the sample size involved in the case. Most case studies have a general sample size of four participants and are extremely limited [47]. As fourteen participants in total volunteered to take part in the present research, the sample was limited to seven participants from each clinic. Additionally, baseline HRV measures were not taken as this was performed as a pilot experiment. A longitudinal study using similar methods has high potential to bring forward richer data and thus generalizability of this research. The sample size also restricted the statistical analysis. There is the opportunity to use statistical functions to analyze the data set by applying a logarithmicfunction, but this compromises the integrity of the data by data transformations, which further skews an already small sample. Thus triangulation and comparison of means was used during analysis of the cases, which is the heart of mixed methods [27] and research was driven by its aims [11].

The key strength of this research is the integrated framework of study and mixed methodology. In all aspects of data analysis, statistical manipulations were avoided and were replaced by triangulations and differentiating the means. Triangulations were carried out between the relationships of HRV and leadership attributes in leaders and their subordinates. This is the perspective grounded in the theory of autopoiesis and a generative approach to leadership. Although this is a short-term study, it provides a gateway for further research to be performed using longitudinal methodologies. Most case studies in organizations involve an external agent coming in and performing research. This may present a threat or engender a lack of trust with the employees and the data collected could thus be biased or inaccurate. In this case, however, the researcher had time to establish rapport with the organization, thus creating a sense of open dialogue with the participants and, as expected, the quality of the data was therefore consistent and reliable. The framework of study was able to identify the differences in the leadership landscape and performance of the two clinics.

5.3 Implications

Theory of autopoiesis proposes that the quality of human experience, cognition and action, are determined by the interplay between the internal dynamics (biological processes) and the environment (social and other) of an active situated human agent, and thus offers an alternative perspective to interpreting and enacting leadership capability [48, 69]. Leadership appears as both deeply personal and inherently collective and may be defined as shaping ‘life-enhancing’ conditions [70]. This requires focus on the ‘common good’ for the individual human agent to flourish [23, 48].

From the perspective of autopoiesis and theory of complex systems, the phenomenal domain of human enterprises is realized through the network of interactions between the human actors [41, 48]. Such networks through the interactions of local agents are capable of spontaneous self-organization, to produce emergent patterns of behaviors of the networkwithout any prior comprehensive, system wide blueprint for the evolution of the system [69]. The immediate local ‘intentions’ of the interacting human actors are continually emerging in a context.

This paper suggests that the quality of ‘intentions’ and quality of ‘interactions’ are interlinked, and effective leadership is distributed and impacted by both the structure of the system and its emergent properties, i.e. trust, wellness, sustainability, as well as, by the individual intent, morality, creativity and motivation.

Research into conditions and models of governance that promote collegial decision making also implies that both quality of awareness, emotional intelligence and morality of individual actors, as well as, the structure of the interactions, will need to be attended in parallel for catalyzing leadership [23, 82].

This paper also brings into focus the ‘embodied’ aspect of leadership [71]. Behavior is interlinked with both biological processes and responses to environmental triggers [68]. It is thus important to take into account measures of the ‘invisible’ aspects of cognition that are impacted by the function of the nervous system, i.e. Heart Rate Variability [14, 58], etc.

Additional research using similar methods and indicators, but a larger and more diverse sample could improve the understanding of the dynamics of leadership behavior and produce statistical results that are significant. While this study uses Heart Rate Variability as a physiological measure, it would be interesting to look at cortisol profile as stress marker and also identify different physiological indicators that can recognize its relationship with leadership behaviors.

The team welcomes an open dialogue with other researchers in order to further develop a holistic framework of understanding leadership in organizations.