University of Groningen A Dynamical Approach to Psychological Resilience Hill, Yannick

(1)

University of Groningen

A Dynamical Approach to Psychological Resilience

Hill, Yannick

DOI:

10.33612/diss.144252644

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Hill, Y. (2020). A Dynamical Approach to Psychological Resilience. University of Groningen. https://doi.org/10.33612/diss.144252644

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

548915-L-bw-Hill 548915-L-bw-Hill 548915-L-bw-Hill 548915-L-bw-Hill Processed on: 30-9-2020 Processed on: 30-9-2020 Processed on: 30-9-2020

Processed on: 30-9-2020 PDF page: 7PDF page: 7PDF page: 7PDF page: 7

CHAPTER 1

(3)

(4)

Introduction

9

1

In our daily lives, we are constantly exposed to stressors. These stressors can represent

everyday hassles that are usually overcome quickly, but may also include severe adverse events, such as the loss of a loved one, which typically requires months or even years to overcome. In the psychological literature, the process by which a person positively adapts to a stressor is called resilience (e.g., Galli & Gonzalez, 2015). The term resilience derives from Latin (“re” – back, “salire” – to leap/jump), and literally means to “bounce back”. Because people have to demonstrate resilience on a daily basis, the topic is of interest across a variety of scientific domains, and has mostly been studied in the field of psychology (Hosseini et al., 2016). In psychology, however, different definitions have been proposed in the past decades which cloud the conceptualization of resilience (see next section). Developing conceptual clarity of what resilience is (not) and how it can be studied in the domain of psychology is important to move the field forward in a unified direction.

In order to achieve such conceptual clarity in psychology, this thesis proposes a dynamical systems approach to resilience, which includes the key properties of resilience across different conceptualizations within psychology as well as recent research findings on other living systems (e.g., eco-systems). In addition, the dynamical systems approach provides a clear and parsimonious conceptualization that comes with a specific analytic toolbox to bring novel insights into resilience in humans. I will exemplify this toolbox in different studies in the context of human motor and sport performance. The insights derived from these studies may be extended to other domains within psychology, such as mental health.

1 | Resilience in the Psychological Literature

Resilience in psychology was originally conceptualized as a personality trait (Block & Block, 1980). This conceptualization assumes that there are consistent individual differences in how a person adapts to stressors in the form of adverse events across domains and time. For example, a resilient person would equally bounce back to their previous level of functioning following stressors occurring in the work life or in the personal life. Moreover, this ability would stay consistent within an individual over the course of their life (see for reviews Fletcher & Sarkar, 2013; Galli & Gonzalez, 2015; Windle, 2011). This trait-conceptualization is still commonly used today in clinical settings to distinguish individuals that are more or less prone to developing psychopathology (e.g., Hu et al., 2015; Smith et al., 2008).

However, resilience may not only be determined by stable characteristics within a per-son. Different stressors and situations may also require different adaptational changes.

(5)

10

Chapter 1

This means that the environment plays a major role in whether an individual can adapt to a stressor or not (Egeland et al., 1993; Luthar et al., 2000). Given the dynamic interplay between the person and the environment to determine resilience, there is growing consensus that resilience in humans may be best conceptualized as a dynamic process (Carver, 1998; Fletcher, 2018; Hill et al., 2018a, 2018b; Masten, 2001; Masten & Obradovic, 2006; Pincus & Metten, 2010; Pincus et al., 2018; Vella & Pai, 2019). This person-environment interaction may additionally change over time. For example, research has shown that humans who have experienced relatively little or relatively much adversity throughout their lifetime may be less resilient than individuals who experienced an intermediate amount (e.g., Seery, 2011; Seery et al., 2010). This means that on the one hand a person needs to learn how to demonstrate resilience by being exposed to stressors, while on the other hand too many stressors may fatigue a person so that they suffer more quickly from a stressor. Thus, the person-environment interaction underlying resilience seems to be malleable as a function of the number of stressors an individual experiences.

Despite the growing consensus on the person-environment interaction, the specific way in which the process is conceptualized is not consistent across different studies and subdisciplines of psychology. Specifically, different process-like conceptualizations and operationalizations of resilience exist, which are sometimes at odds with each other and with the original conceptualization of resilience as the ability to bounce back. For instance, Woods (2015) pointed out that resilience is often defined as the process by which change is resisted and a healthy state is maintained despite perturbations (cf. Wieland & Wallenburg, 2012). This means that a person would withstand potential negative effects that a given stressor would usually cause. Note that, in psychology this may be defined as immunity, which is a different property than resilience. Thus, I propose that, only when the process of adaptation is mapped over time following a stressor, one is able to accurately map out the process of bouncing back which marks resilience. If responses to stressors over time are taken into account, another common misconception can be clearly distinguished from resilience, namely, the ability to functionally change and grow following a stressor (Woods, 2015). Although the psychological literature contains different terms, such as thriving (Carver, 1998), to clearly distinguish this stress-related growth from resilience, many resilience studies still operationalize improvements in the level of functioning or overall facilitative response to stressors as resilience (e.g., winning an Olympic gold medal despite adverse events, Fletcher & Sarkar, 2012). Conceptually, this permanent (adaptive) change in the person in response to the stressor is called plasticity, which may actually go beyond resilience (Taleb, 2012). Instead of returning to the previous state, plastic deformation may imply beneficial changes for a system (cf. Agrawal, 2001). If the structural or behavioral changes that need to occur in response to

(6)

Introduction

11

1

a stressor form a more adaptive state, phenotypic plasticity has occurred in an organism

(Kiefer et al., 2018). To illustrate, in vaccinations a small dosage of a potentially infectious agent is injected into the system, which “trains” the system to undergo functional changes to immunize the organism against larger dosages of the same infectious agent. This plastic change in response to stress, occurs frequently in biological systems (Agathokleous et al., 2018; Calabrese 2005a, 2005b; Costantini et al., 2010; Hill et al., 2020; Kiefer et al., 2018; Mattson & Calabrese, 2011; Southam & Ehrlich, 1943) and goes beyond a simple return to the previous level of functioning (i.e., resilience).

Similarly, a system also does not demonstrate resilience when it enters a state that is less adaptive than the previous following a stressor. Although in clinical psychology resilience may be operationalized as the absence of psychopathology following a traumatic event that is statically associated with maladaptation (Luthar et al., 2000), the absence of psychopathological symptoms is not equivalent to returning to the previous functioning. In order to distinguish this outcome from resilience Carver (1998) defined it as survival

with impairment.

Summarizing the current literature on resilience, there is consensus that resilience is a dynamical process which develops from person-environment interactions. However,

Figure 1 | Chapter overview of the thesis. By mapping out how a person adapts over time

following a stressor (grey lightning bolt), we can detect the deformation taking place to determine a precise measure of resilience (i.e., area under the curve, Chapter 4), the temporal patterns that precede resilience (Chapter 5), and whether a person grows from the stressor (Chapter 6). Furthermore, in Chapter 3, the temporal patterns of the factors that are assumed to underlie the process are focused on.

Performance

Time

Chapter 6 Growing from stress

Chapter 4 Measuring bouncing back Chapter 5

Predicting changes in resilience

Chapter 3 Factors protecting an individual

(7)

12

Chapter 1

a coherent theoretical approach that guides researchers how to study this process is lacking. Such a theoretical framework needs to capture the dynamics of the person-environment interaction, explain how these dynamics can change over time with repeated stressors, and differentiate the resilience process from related, but different concepts (e.g., plasticity). In this thesis I present a dynamical systems approach to psychological resilience, which provides a parsimonious theoretical framework that is explicitly built on these key features of resilience (Chapter 2). I applied the analytic toolbox of dynamical systems to several studies in the context of human performance (Chapters 3-6, see Figure 1).

2 | Chapter 2 – A Dynamical Approach to

Psychological Resilience

The foundation of the dynamical approach to resilience is described in Chapter 2. A dynamical system can be defined as a set of elements, which are in constant dynamic interactions and undergo change over time (e.g., Kelso, 1995; Nowak & Vallacher, 1998; Vallacher & Nowak, 1997; Van Geert, 1994). The behavior of such a system emerges from the ongoing interactions between its elements (i.e., interaction-dominance) and unfolds in the dimension of time, rather than being driven by static components that can be assessed at a given point in time. Because of this interaction-dominance, the system as a whole can achieve better performance than the sum of the constituent elements. For example, a human consists of thousands of cells, which form tissues, which in turn form movements systems that ultimately make up the person. When a person moves, the higher order movement systems order the underlying structures, while at the same time react to the feedback that is returned by the lower-order cells (Balagué et al., 2017). Thereby, a circular feedback loop between the elements of the system emerges. However, because performance never occurs in a vacuum, an athlete is also constantly interacting with the environment in which they perform in. Thus, during the performance, the environment constantly influences the athlete, while the athlete’s performance also shapes and changes the environment. Consequently, this theoretical approach captures the person-environment dynamics by capturing how the system changes over time. The ongoing change in the system caused by the interactions of the constituent elements and the environment form a so-called iterative process (e.g., Den Hartigh, Van Dijk et al., 2016; Gernigon et al., 2015; Nowak & Vallacher, 1998; Vallacher et al., 2015; Van Geert, 1991). During an iterative process, a given state of the system develops out of the system’s previous state, while simultaneously serving as a starting point for any possible future iterations. This iterative process unfolds in idiosyncratic trajectories (e.g.,

(8)

Introduction

13

1

Den Hartigh, Hill et al., 2018). In the first part of Chapter 2, I connect the properties of

dynamical systems with what we know about resilience. This has triggered a fruitful academic discussion on the application of a dynamical systems approach to studying resilience with experts from the field. Indeed, our paper was published as a target article, to which different scholars have commented. The abstracts of the commentaries are included (second part of Chapter 2). In turn, we responded to the commentaries, and aimed to provide the groundwork for future resilience studies (third part of Chapter 2).

3 | Chapter 3 – Idiosyncrasy of Resilience and its

Underlying Factors

One important first avenue for resilience research is to check some of the assumptions, or consequences, of a dynamical systems approach. In Chapter 3, I assessed the property of idiosyncrasy which is often connected to the notion of iterative processes. In the case of sports, this means that each athlete changes in a unique way over time and that this development cannot be adequately represented by the average change that a larger sample of athletes would undergo. Moreover, averaging the trajectories of several individuals may mask important dynamics that could be derived from the temporal

Performance

Time

Figure 2 | Example of non-ergodicity in mathematically determined trajectories. The nine

grey trajectories represent mathematically simulated logistic growth curves that depict a hypothetical change of performance in different athletes over time. The black solid line represents the mean score of each simulated trajectory per time point. The group-level mean trajectory approaches a linear growth curve and thus does not represent the underlying mathematical model of the individual trajectories.

(9)

14

Chapter 1

patterns that are observable on the individual level (Liu et al., 2006). This lack of group-to-individuals generalizability is called non-ergodicity (Fisher et al., 2018; Hamaker, 2012; Hamaker et al., 2005; Medaglia et al., 2011; Molenaar & Campbell, 2009). This effect is neatly illustrated in Simpson’s (1951) paradox: a given association may appear in all individuals (or subgroups) of a sample, but disappear or even become reversed when the individuals (or subgroups) are combined. Although non-ergodicity represents a commonphenomenon in dynamical processes and can be found in mathematically generated time-series (see Figure 2), the use of group-level data and inferences is still the dominating approach in psychology. Therefore, we test the ergodicity assumption for phenomenon in dynamical processes and can be found in mathematically resilience in Chapter 3 to identify whether studies and analyses of this process need to be tailored to the individual.

4 | Chapter 4 – Measuring the Bouncing Back

The notion of iterativity also entails that the current state of the system depends on its previous states. In terms of resilience, this means that whether an athlete is able to positively adapt to a stressor or not depends on their performance history and how they dealt with stressors in the past (Hill et al., 2018a, 2018b). Recent research on resilience of eco-systems has shown how changes in temporal patterns of a system can signal resilience losses as a result of repeated stressors (Scheffer et al., 2009, 2012). For example, before a sudden, drastic loss in biomass (i.e., phase transition) occurs in an eco-system, the system experiences a loss of resilience. This loss of resilience is expressed by an increasing sensitivity to stressors and increasing time the system requires to return to its previous level of functioning after its state has been perturbed by a stressor. This period of resilience loss is called critical slowing down (e.g., Scheffer et al., 2009, 2012, 2018; Van de Leemput et al., 2014).

Critical slowing down typically occurs when a system is exposed to repeated stressors within a relatively short timeframe. For example, psychopathological states may emerge from a series of relatively mild stressors during which resilience is lost up to a point where a single mild stressor can cause a drastic shift in well-being, rather than being caused by a single major life event (Van de Leemput et al., 2014). Thus, in Chapter 4, we introduce a novel measure to determine resilience by combining the basic properties of critical slowing down: the sensitivity to a stressor and how long the system takes to return to its previous functioning. Using this measure, resilience losses in human motor performance in response to repeated stressors are explored.

(10)

Introduction

15

1

5 | Chapter 5 – Early Warning Signals of

Resilience Losses

Another tool to detect critical changes in a system that may indicate resilience losses is explored in Chapter 5. Here, the focus is placed on the link between complexity and resilience. Adaptive biological systems show a combination of flexibility and stability when tracked over time. This temporal pattern is referred to as complexity and has been found in healthy motor systems (Davids et al., 2003; Den Hartigh et al., 2015; Kiefer & Myer, 2015), cardiovascular systems (Goldberger et al., 2002) and psychological systems (e.g., self-esteem, Delignieres et al., 2004; De Ruiter et al., 2015). The combination of flexibility and adaptability enables the system to adapt to stressors (flexibility) while simultaneously maintaining global functioning (stability). Similar to the notion of the autocorrelations, a deviation from this state to either extreme (i.e., either extreme stability or extreme flexibility) would indicate a loss of the adaptability and therefore diminished resilience in the system (Kiefer et al., 2018; Pincus et al., 2014, 2018, 2019; Pincus & Metten, 2010). For example, an overly stable athlete would reside within a rigid behavioral pattern and thus could not adapt to stressors, while an overly flexible athlete could not maintain necessary functioning when adapting to stressors. Therefore, we assess in Chapter 5 whether changes in the complexity of a system are indeed linked with resilience losses in team performance.

6 | Chapter 6 – Growing from Stress

The dynamical systems approach also offers a precise theoretical account how and under what conditions, humans may benefit from stress and demonstrate growth. This was outlined in one of the commentaries in Chapter 2 (Kiefer et al., 2018). Specifically, the structural and behavioral changes in the system triggered by a stressor may yield lasting, more adaptive states (i.e., plasticity). Thus, due to its structural re-organization in response to a stressor, a biological system can demonstrate growth from stressors that goes beyond resilience (i.e., returning to the previous level of functioning), termed

antifragility (Taleb, 2012). As illustrated in the example of vaccinations, it is essential that

the applied dosage of stress is large enough to trigger the desired changes to develop an immune response, while at the same time not being too large to have a toxic effect on the system (Calabrese 2005a, 2005b). Thus, the beneficial immune response depends on the dosage of the stressor the system is exposed to. In collaboration with authors of one of the commentaries in Chapter 2, I conducted a study in Chapter 6 to move beyond resilience by exploring antifragility in athletes.

(11)