ادغام شخصیت، رویدادهای زندگی روزمره و احساسات: نقش اضطراب و عاطفه مثبت در پویایی تنظیم احساسات

We investigated the roles of anxiety and positive affect in emotion regulation, looking simultaneously at personality, daily life events, and affects. We hypothesized that individual differences in the temporal dynamics of affective experience related to trait anxiety would manifest themselves both in affective responsiveness to life events and in homeostatic regulatory forces. Data were collected from 49 adults, who rated their affective state three times a day over a 40-day period. Data were analyzed using a dynamical system model and graphical representations in the form of vector fields. Results showed that anxiety chiefly interacted with home base (attractor) positions as a function of life events. It also influenced the shape of positive affectivity trajectories in response to negative events.

A fundamental characteristic of emotions and affective experiences is that they vary over time. Our lives are characterized by affective ups and downs, changes and fluctuations following the ebb and flow of daily life. Understanding the nature of the temporal dynamics of affect and emotion, and the processes that underpin them, as well as individual differences in the patterns and regularities characterizing affect dynamics (Kuppens, Oravecz, & Tuerlinckx, 2010) remains one of the most important challenges in the study of emotion (Scherer, 2009). It is important to study the dynamics of emotional fluctuations, as this allows us to predict observable behaviors more accurately (e.g., Eid and Langeheine, 1999, Ghisletta et al., 2002, Nesselroade, 1988 and Nesselroade, 2001). A better understanding of the mechanisms that underpin emotion regulation could help us gain a clearer idea of individual trajectories and of the long-term impact of these mechanisms on psychological health and well-being (Dodge & Garber, 1991). Given that their impairment can account for various personality disorders, including depression and anxiety, they constitute key factors in numerous psychiatric diagnoses (Murray et al., 2002 and Russell et al., 2007). There has been a growing interest in the dynamics of emotion regulation processes (John and Gross, 2007 and Vansteelandt et al., 2005) and more and more researchers are now starting to examine the patterns and regularities that drive the dynamics of affect (Kuppens et al., 2010). The main aim of the present study was to undertake the simultaneous investigation of affect, personality and daily life events (Nezlek & Kuppens, 2008), and more specifically to study the role of anxiety in variations in positive and negative affect in reaction to events, within the framework of a model of affect dynamics (DynAffect; Kuppens et al., 2010). This model formalizes three processes involved in affective fluctuations and seems to offer a heuristic conceptual framework for exploring individual differences. We refer to this framework throughout our paper. After describing the DynAffect model (Kuppens et al., 2010) in some detail, we tackle the role of personality in affective fluctuations, focusing on trait anxiety and its links with the perception of daily life events. We then attempt to pinpoint the role of positive affect in emotional dynamics. 1.1. A dynamical system model for the study of individual differences in affective fluctuations The DynAffect model developed by Kuppens et al. (2010) treats the affect system as an open, dynamic system featuring three main sources of interindividual variations: the coordinates of the home base – a baseline attractor state or benchmark –, the range of affective fluctuations around this home base, and the strength of the system’s homeostatic attraction force, which curbs these fluctuations brought about by internal or external processes. This model considers affect in a two-dimensional space, with valence along the x-axis and arousal along the y-axis. The home base constitutes an equilibrium point in this two-dimensional system, serving as a specific attractor for each individual, around which the latter’s affective state fluctuates. Particularly wide affective fluctuations constitute discomfort zones, motivating the individual to engage regulation processes in order to restore equilibrium and return to the home base ( Russell, 2003). The basic idea, therefore, is that our affective state fluctuates around an equilibrium point, which serves as a baseline for the affective system, reflecting its expected state given the characteristics of its environment in a given period. It reflects the average emotional experience of a person in a given period. It can also be viewed as the point where the affective state would stabilize itself in a steady and homogeneous environment. In the DynAffect model, the home base position is essentially an individual characteristic. In our view, the home base position is linked to the appraisal that an individual makes of their environment. It is thus influenced by both individual and environmental characteristics. If the first process is the affective home base, the second process is variability, referring to affective changes and fluctuations. Being an open system, our affective state is subject to dynamic-stochastic variability (Russell, 2003 and Russell, 2009) resulting from the many internal and external events that influence our core affect at any given time. The extent of these variations depends on the individual. Some of us experience important emotional changes, react more strongly to the event or encounter more striking events, while others experience a life more stable emotionally. The third process is the force exerted by the attractor, or home base. If, after a perturbation, the affective state of a person is far away from its current home base, this state will move gradually toward the home base driven by the attraction force of this home base. One can imagine this attraction, by the force exerted by a spring attached to the home base. The spring gradually returns to its initial state after being stretched, suggesting regulation processes. The intensity of this force depends on the distance between the current emotional state and the home base. The further the affective state moves away from the home base, the greater the attraction force. Whenever events open up too great a gap, this self-regulation process undertakes to redirect affect toward the system’s equilibrium point. Its purpose is thus to prevent the system from reaching extreme values and, by so doing, reduce the affective fluctuations that disturb the individual’s equilibrium and, by extension, his or her psychological wellbeing. The intensity of the attraction also depends on the thickness of the spring which could vary depending on the subject and relies on dispositional characteristics. A person with a hight attraction strength returns more easily to his home base. This model has shown its ability to account for emotional fluctuations in a longitudinal protocol. The model used in this study basically replicates the framework developed by Kuppens et al. (2010), albeit with three modifications. The first difference concerns the two axes of affective space. Whereas the DynAffect model relies on the distinction between valence and arousal, we decided to take positive affect (PA) and negative affect (NA) as its two axes. This choice raised the question of the independence or bipolarity of PA and NA, which has been the subject of hot debate in the literature (Russell and Carroll, 1999 and Watson and Tellegen, 1999). One of the present study’s objectives was to analyze combined changes in PA and NA in reaction to daily life events and, more specifically, the likelihood of asynchrony and uncoupling between PA and NA. This amounted to assuming that there is a degree of leeway in NA–PA bipolarity and a relative independence in certain conditions (Reich et al., 2003 and Zautra et al., 2005). For this reason, we believed it was important to collect PA and NA data separately and to make them the main axes of affective space. This meant that we had to neglect variance linked to arousal to some extent, even though it could well be relevant here (Kuppens, Van Mechelen, Nezlek, Dossche, & Timmermans, 2007). A more comprehensive approach would consist in considering three dimensional affective space (PA, NA and arousal), as Stanley and Meyer (2009) recently suggested, but this would result in a far more complex model and go far beyond the scope of our research. The second contribution deals with the concept of home base and its relation with life events. We believed that this notion could be extended, by regarding it as the result of environmental factors, as well as individual characteristics. For example, an individual might have a home base in one position corresponding to a welcoming environment characterized by a succession of positive events (e.g., a week’s vacation) and in another position corresponding to a hostile environment characterized by overwhelmingly negative life events (period of considerable stress at work). In each case, therefore, the system would stabilize itself or fluctuate around a different equilibrium point with different coordinates. We therefore decided to turn the home base into a continuum, rather than a fixed point – a curve in affective space where each section would correspond to life events of a particular valence. Some of the DynAffect model’s variability parameter was therefore represented by this affective “moving target”. In order to track this variability, our protocol provided for the recording of daily life events at each observation. The first set of hypotheses we tested therefore concerned shifts in the home base according to daily life events and trait anxiety. The third contribution was to take eventual coupling effect between PA and NA into account to describe individual trajectories in the affective space. In the present study, PA and NA were assumed to be governed by two distinct but connected entities. This connection could take the shape of a lateral inhibition of one on the other. Empirical results show that changes in PA and NA are negatively correlated when they are studied in dynamics (Vautier, Steyer, Jmel, & Raufaste, 2005) and this correlation is increased when the interval between observations is shorter (Diener, Smith, & Fujita, 1995). The study of eventual coupling (Zautra et al., 2005) effects between AP and AN also allows to examine the role of positive affects in the regulation of negative emotions, which is a major objective of this research. More particularly, it is assumed that, during the recovery phase after a negative event, the occurrence of PA might contribute to a reduction of NA (Ong, Bergeman, & Bisconti, 2006). Some of us may be able to use PA in order to curb the increase in NA in the recuperation phase, this idea is detailed in the Section 1.3. The coupling is not always complete between PA and NA. The coexistence of positive and negative affects has been shown in literature (Cacioppo, Larsen, Smith, & Berntson, 2004). The model that we used is flexible enough to highlight the effects of coupling while allowing certain independence between PA and NA and taking into account the coexistence of high levels of PA and NA found in literature (e.g., being happy and sad at the same time). Technically, this kind of coupling effect can be modeled using cross-lagged regression coefficients. The next two sections first relate the role of anxiety in the differences in affect regulation and the role of positive affect in the regulation of negative affect. 1.2. Anxiety, response to daily life events and emotion regulation The factors involved in emotional fluctuations (subjective assessment of events, biological and environmental factors) are numerous and interconnected, and it is the outcome of this complex combination that determines affective variability over time (Fok, Hui, Bond, Matsumoto, & Yoo, 2008). Whereas the relationship between personality and affective responses has been investigated on many occasions (see, for example, Diener et al., 1995, Larsen and Ketelaar, 1989 and Yik and Russell, 2001), the relationship between intraindividual affective variability and personality has rarely been considered (Eid & Diener, 1999). Although the usefulness of studying the individual * situation transaction is underscored in numerous models (Mischel & Shoda, 1998), approaches have tended to focus either on the role of events or on personality, and only rarely on the two together. The aim of the present study was thus to highlight the interaction between personality, daily life events and affective fluctuations. Fluctuation predictors mentioned in the literature include neuroticism and extraversion, anxious and depressive characteristics, self-esteem and coping strategies (e.g. Beck, 1985, Eid and Diener, 1999, Kuppens and Van Mechelen, 2007, Kuppens et al., 2007 and Larsen and Ketelaar, 1989). However, we chose to focus on anxiety, as this is the dimension with the most convergent results. Trait anxiety is one of the most insidious facets of personality when it comes to emotional management, giving rise to poorly adjusted emotion regulation processes (Zelenski & Larsen, 2002). It is defined as a stable characteristic that predisposes individuals to perceive situational information as a threat or potential danger (Beck, 1985) and react by displaying disproportionately intense anxiety, given the degree of objective danger (Spielberger, 1966). It is also characterized by heightened sensitivity to threat stimuli, resulting in an excessive tendency to summon NA (Gray, 1987 and Zelenski and Larsen, 1999). Anxious subjects ruminate for longer on negative events (Muris, Roelofs, & Rassin, 2005) and find it hard to put them into perspective (Avila et al., 1999, Corr et al., 1995, Gupta and Shukla, 1989 and Zinbarg and Mohlman, 1998). Researchers who have triangulated personality, events and emotions by undertaking multilevel analyses have shown that the occurrence of threatening events is an important factor in the increase of NA, but that personality also plays a non-negligible role in regulating the individual * situation transaction (De Beurs et al., 2005). This type of approach makes it possible to identify the particular relationship that exists between personality, situation and behavior, and to bring discussions about personality into sharper conceptual focus (Mischel & Shoda, 1995). It was within this framework that the present study was conducted. Our first set of hypotheses was that anxiety exerts a major influence on home base position and on the dynamics of emotional trajectories over time, more specifically the force of attraction and its direction. For example, the advent of a negative event leads to the activation of behavioral repertoires at the intraindividual level that may differ according to the level of that person’s trait anxiety and therefore generate specific trajectories in affective space. These affective responses may, in turn, have an impact on home base position; the home bases of anxious individuals would be more negative than those of their non-anxious peers. We also probed the influence of PA on the regulation of NA, as set out in the following section. 1.3. The role of positive affect in emotion regulation: the relevance of studying the dynamic interaction between positive and negative affect An increasing number of studies have highlighted the contribution of PA to affect regulation (Lyubomirsky et al., 2005 and Tugade et al., 2004) and sought to elucidate the role it plays in adaptive behavior in stressful situations (Zautra et al., 2005). More importantly still, as far as we are concerned, these studies have suggested that PA serves to replenish the individual’s resources, thereby enabling the latter to recover from negative events (Ong et al., 2006). PA appears to broaden thought-action repertoires and enhance problem-solving (Fredrickson and Branigan, 2005 and Isen et al., 1987). It counterbalances the negative physiological effects of NA (Ong & Allaire, 2005) and promotes the use of suitable coping strategies (Folkman & Moskowitz, 2000). It also facilitates the implementation and management of resources in order to deal with a negative event (Tugade et al., 2004) and allows for more rapid recovery following stressful events (Fredrickson et al., 2003 and Zautra et al., 2005). Thus, whereas NA is governed by the sympathetic nervous system, which narrows attention in order to direct it toward relatively primal actions (fight or flight), PA offers a means of reducing the automatic activation of the sympathetic nervous system triggered by NA and of broadening attentional capacity, thought and behavioral repertoires (Fredrickson, 2001). Therefore, people with a high level of positive affectivity have easier access to a range of behavioral repertoires that allow them to adjust more successfully to stress (Ryff et al., 1998 and Staudinger et al., 1995). Our second set of hypotheses was therefore based on these studies and, more particularly, on the research conducted by Cacioppo et al., 2004 and Zautra et al., 2001, which assumes that the ability to maintain PA and strike a balance between PA and NA reflects a potential for adaptation and flexibility. Bergeman and Wallace (2006) studied the role of PA when faced with a stressful event. Multilevel analyses showed that PA helped to moderate reactions to a given event. Resilient participants with a high level of PA seemed better able to recover when faced with daily life stressors and more efficient in the way they tackled, managed and transformed stressful events. This point can be translated in our model using the idea of inhibitory coupling effect between PA and NA. The coupling effect can be deleterious just after a negative event when a rising of NA inhibits PA, but it can also have a positive effect if a rebound of PA can curb the increase of NA. It seems that more resilient people are able to use this last property more efficiently. We expected that for these people, a stronger coupling effect would appear. The intensity of the coupling could then vary among subjects as a function of trait anxiety. Our aim was to demonstrate how far interactions between personality dispositions, studied from the angle of trait anxiety, and daily life events influence home base position and homeostatic forces (attractor strength), by investigating the role of PA in the regulation process. To this end, we developed four specific hypotheses, two concerning the position of the home base and two concerning its attraction strength and the coupling of PA and NA. Our first hypothesis was that daily life events are capable of modifying the coordinates of individual home bases. As such, we expected individual differences to be expressed in the shape of several different home bases corresponding to events of different valence. Our second hypothesis concerned the effect of trait anxiety on home base position, predicting that the home bases of the most anxious participants would be more negative on the whole, with greater reactivity to negative events. Our third hypothesis was that the less anxious individuals should exhibit stronger attraction strength. A person with a hight attraction strength returns more easily to his home base reflecting more efficient emotion regulation mechanisms (Kuppens et al., 2010). Our fourth and final hypothesis concerned the existence of a partial and negative coupling of PA and NA. We assumed that the ability to feel PA could allow, in addition to the attraction strength, to regulate negative emotions. This effect could be of different intensity depending on the trait anxiety.

Using a model of two-dimensional change encouraged us to stop seeing emotional reaction simply in terms of a causal relationship and instead to view it as a dynamic process of transaction between an individual and a situation. The bivariate difference score model allowed us to fit a dynamical system model using a standard statistical package and provided us with a means of integrating explanatory variables, such as trait anxiety and the perception of events, and the dynamics between positive and negative affectivity. As such, it enabled us to take our interpretation one step further. Adding arousal to PA and NA to create a three dimensional affective space (Stanley & Meyer, 2009) is one of the prospects for future research. Another original feature of the present study relates to the use of graphical representations, which we borrowed from the field of dynamical systems with a view to gaining a more accurate picture of the dynamic nature of these interactions and understand what happens over time. We were able to identify predictable, characteristic patterns of variations in individual behaviors across different situations as a function of trait anxiety, whilst integrating the dynamic relationship between positive and negative affectivity, which was presented in a two-dimensional affective space. The present study opens up several directions for future research, notably the asymmetrical nature of attractor strength and the impact of the individual’s affective state on the perception of ongoing or future life events. Regarding the protocol, it would be worthwhile building in the individuals’ subjective appraisals of the situation (other-blame, self-blame, danger/threat, loss/helplessness, achievement, positive encounters) to gain a more fine-grained understanding of the individual * situation transaction (Nezlek, Vansteelandt, Van Mechelen, & Kuppens, 2008) and the stabilization of personality traits. From a more theoretical point of view, this would provide a means of elucidating the relationship between trait emotions and functional characteristics linked to the dynamic aspects of affective dysfunction. This type of model could provide a useful heuristic for characterizing the emotional phenomenology of psychopathological disorders which, at their core, are characterized by dysfonction. The different forms in which this affective dysfonction expresses itself may be associated with distinct combinations of the dynamic properties of affective home bases and attractor strength.