A Model of Socioemotional Flexibility at Three Time Scales

Abstract

The construct of flexibility has been a focus for research and theory for over 100 years. However, flexibility has not been consistently or adequately defined, leading to obstacles in the interpretation of past research and progress toward enhanced theory. We present a model of socioemotional flexibility—and its counterpart rigidity—at three time scales using dynamic systems modeling. At the real-time scale (micro), moment-to-moment fluctuations in affect are identified as dynamic flexibility. At the next higher meso-time scale, adaptive adjustments to changes in context are characterized as reactive flexibility. At the macro scale is flexibility that occurs across months or years, reflecting flexibility due to developmental or life transitions. Implications of the model and suggestions for future research are discussed.

Keywords

affect dynamics systems dynamics emotion flexibility rigidity

“Mobilis in mobili”

[Changing in the changes]

Motto of Captain Nemo’s Nautilus

Jules Verne, Twenty Thousand Leagues Under the Sea (1870)

Common phrases or idiomatic expressions often reveal societal values. Going with the flow, chilling out, rolling with the punches, etcetera, are all complimentary characteristics, while being a stick in the mud, stuck in a rut, etcetera, are less desirable. Yet the contrary is also true. We decry the wishy-washy, flip-flopping, spineless, and capricious and reward the decisive, intransigent, hard headed, and firm. Hence, the degree to which people yield to the vicissitudes of life is a dimension of some importance for popular notions of social and emotional competence.

As with the colloquial understanding, the scientific conceptualization of flexible and inflexible behavior is an important distinction, but also not so clear. Flexibility—and its counterpart rigidity—is a characteristic of human behavior that has been both a peripheral and focal component in many psychological theories over the past century (Schultz & Searleman, 2002). Early personality theorists (e.g., Cattell, 1946) and gestalt psychologists (e.g., Luchins & Luchins, 1959) attempted to define flexibility in physical, perceptual, attitudinal, or behavioral domains—conspicuously lacking in emotional content. However, these early attempts failed to come up with a unifying definition or delineation of the essential components (Chown, 1959; Leach, 1967), leading to a relative lull in flexibility-focused research. More recently, there has been renewed interest in the flexibility/rigidity construct in relation to a diverse range of emotion-related research topics, including but not limited to: parent–child interactions (Granic, O’Hara, Pepler, & Lewis, 2007; Hollenstein, Granic, Stoolmiller, & Snyder, 2004; Hollenstein & Lewis, 2006; Lichtwarck-Aschoff, Kunnen, & van Geert, 2009; Lunkenheimer, Hollenstein, Wang, & Shields, 2012; Lunkenheimer, Olson, Hollenstein, Sameroff, & Winter, 2011), peer interactions (Lavictoire, Snyder, Stoolmiller, & Hollenstein, 2012), dyadic relationships (Branje, 2008; Gruber-Baldini, Schaie, & Willis, 1995; Paulhus & Martin, 1988), coping (Cheng, 2001, 2009), emotional expressivity (Bonanno, Papa, Lalande, Westphal, & Coifman, 2004; Westphal, Seivert, & Bonanno, 2010), mood variability (Kuppens, Allen, & Sheeber, 2010; Kuppens et al., 2012; Wichers et al., 2010), personality (DeYoung, Peterson, & Higgins, 2002), depression (Deveney & Deldin, 2006), and anxiety (Davis, Montgomery, & Wilson, 2002). These investigations have exposed normative individual differences as well as the problems associated with extreme levels of either flexibility or rigidity.

We present a model of socioemotional flexibility that is an attempt to clarify the term, and provide a unifying framework for focused study of flexibility across the life span. Our focus is emotional because changes in affect and the influence of emotion on cognitions and behavior encompass most, if not all, of the domain of flexibility. Because flexibility is an inherently dynamic process of change over time, we take a dynamic systems (DS) approach to formulating this model. Specifically, we take a developmental DS approach because DS methods and models integrate processes across multiple time scales—from moment-to-moment variability in real time to the vicissitudes of change and stability across the lifespan.

Dynamic Systems

With the identification of core characteristics of naturally occurring, complex, open, and adaptive systems (i.e., dynamic systems) established by mathematicians and physicists in the last quarter of the 20th century, behavioral scientists have been better able to articulate and evaluate the systemic properties of human behavior. In particular, developmental science, with its fundamental focus on change and stability, has been a particularly fruitful arena for the application of dynamic systems (DS) principles (see Hollenstein, 2011; Lewis, 2011; van Geert, 2011, for reviews). DS approaches function as a metatheory that incorporates the domain-specific models of psychology and other disciplines and as such functions as the realization of the general systems theory espoused by von Bertalanffy (1968) and others. We will refrain from a complete description of DS principles and mechanisms (see comprehensive descriptions in Granic & Hollenstein, 2006; Thelen & Smith, 1994; van Geert, 1994) and instead focus here on two concepts central to our model of flexibility: variability and interacting time scales.

From a DS perspective, system variability at every time scale is the signal that provides crucial information and not noise to be ignored (Thelen & Smith, 1994). This is because system variability reveals the nature of the interactions among system elements, signifies the emergence and breaking-down of structure, and shows how the system adapts to perturbations and how it functions over long stretches of time. One of the interpretations of system variability is the relative flexibility or rigidity of that system at a particular time scale. For instance, a system’s variability can be used to detect the underlying structure of attractors and repellors that functionally constrain the trajectory of a system’s state changes. The stronger the “pull” of these attractors, the more limited the variability. Examples of emotional attractors include depression (Johnson & Nowak, 2002) and mutually hostile parent–child interactions (Granic & Lamey, 2002). Additionally, changes in system variability can reveal transitional periods of system-wide reorganization that are accompanied by temporary increases in (emotional) variability and unpredictability (Granic, Hollenstein, Dishion, & Patterson, 2003; Lichtwarck-Aschoff, Hasselman, Cox, Peppler, & Granic, 2012).

Within a particular time scale, interactions among system elements are reciprocal, reflecting positive and negative feedback processes. These mutually causal processes are known as self-organization due to the fact that there is no proscriptive or external causation leading to the emergence of certain structures (Haken, 1983; Kelso, 1995; Witherington, 2011). Furthermore, these (within time scale) processes at lower levels create stable patterns or structures at the next higher levels. For instance, emotions create moods which form personality structures over longer periods of time (Lewis, 2000, 2005). Thus, a hierarchical, nested structure emerges over time as interactions at one level beget structures at the next, which then become the interacting elements for the next level, and so on. Consequently, in the hierarchy, higher order structures take more time to develop and hence emerge at longer time scales than the lower order interactions. Importantly, higher order structures also exert influence on lower order structures: they constrain the interactions among the lower order interactions. This interaction between different time scales in the form of bottom–up and top–down processes is known as circular causality (Haken, 1983).

In the emotion realm, the combination of these two processes, reciprocal and circular causality, is best illustrated in the relations between emotions and moods and personality (Lewis, 1997, 2000; Lewis & Ferrari, 2001; Lichtwarck-Aschoff, van Geert, Bosma, &, Kunnen, 2008). Emotional processes that occur in real time are reciprocal interactions among system components (e.g., appraisals, physiology, expression) that, when they cohere for longer stretches of time, become moods. Moods are longer lasting, higher order structures that constrain the emotional processes that occur in real time (e.g., an irritable mood makes anger more probable and positive emotions less probable), thus completing the top–down process of circular causality. At the next more macro scale, this pattern of relations continues as the structure of personality that develops over long time spans constrains the moods and emotional states that gave rise to the personality in the first place. Taken together, these primary DS concepts of variability and multiple interacting time scales enable a more precise modeling of flexibility. Borrowing from the pioneering modeling of Lewis and colleagues (Lewis, 1997, 2000, 2005; Lewis & Ferrari, 2001; Lewis & Liu, 2011; Lewis & Todd, 2007), we propose our model of flexibility at three time scales.

The Flex3 Model of Flexibility

Flexibility is a process by which a state or behavioral pattern that has persisted for some time transforms into a different state or pattern in response to a perturbation, such as a change in context, modification of goals, or shift in environmental demands. Rigidity, flexibility’s counterpart, refers to a lack of responsivity and a persistence of the original state or behavioral patterns despite a perturbation. See Figure 1 for a schematic illustration. Flexibility and rigidity therefore lie along a single dimension for which they are opposite poles. This operationalization of flexibility is quite general, encompassing most, if not all, definitions proffered in the past. Importantly, there are three key aspects that complete this most basic definition. First, there is a preexisting state that serves as a baseline point of comparison to what follows the perturbation. This is typically considered a neutral state, but that is not a necessary condition. The “baseline” could be a negatively valenced condition, for example (Granic et al., 2007). Second, there is some perturbing event that influences change. As the full model will show, the nature of these perturbations varies by time scale. For example, a conversation can be considered a series of smaller perturbations that occur moment-by-moment, whereas the death of a spouse is a perturbation at a larger scale. For the present simplified illustration, however, the specifics of the perturbation do not matter. Third, because flexibility is about change, it is necessarily a process that unfolds over time. Affective dynamics have been a central part of emotion theory for a long time (e.g., Tomkins, 1962). For instance, in Thompson’s list of seven core features of emotion dynamics (Thompson, 1990), four relate to the basic rise and fall of an emotion episode (latency, rise time, intensity, and recovery), but the remaining three relate to key aspects of flexibility/rigidity (lability, range, and persistence). It is this inherent temporality of flexibility which requires a dynamic approach.

Figure 1.

Basic conceptualization of flexibility and rigidity.

Bringing it all together, we propose the Flex3 model by which this state–perturbation–state process can reflect the fluctuations of real-time dynamics, the adaptive adjustment to qualitative shifts in environmental demands, or the impact of large-scale changes in context across the lifespan. Therefore, as shown in Figure 2, the Flex3 model is separated into three distinct levels: micro, meso, and macro. These levels correspond to separate aspects of flexibility that reflect a particular time scale. What is classified as a state or pattern and what functions as a perturbation is different at each scale. Moreover, although each scale is distinct, there are meaningful interactions across these scales.

Figure 2.

The Flex3 model: Flexibility at the micro, meso, and macro scales.

Dynamic Flexibility: Micro Scale of Within-Context Transitions

At the micro level are the moment-to-moment changes that occur within one context. These changes may be, for example, the variety of conversational utterances or a range of emotional states within an interpersonal interaction. In the Flex3 model, this real-time variability is identified as dynamic flexibility. Behavior within any given context can be described in terms of this variability and be deemed more or less flexible. These are the micro adjustments made to maintain balance and engagement in the situation at hand.

In previous studies, we have considered dynamic flexibility as a way to reveal the structure of emotional behavior (i.e., patterns of state changes regardless of content) in concert with and in contrast to the emotional content (e.g., anger, sadness, joy). The rationale for this approach is that it is less important whether negative emotional states occur than whether individuals get stuck in these states without shifting out of them easily (Granic & Hollenstein, 2003; Granic et al., 2007). For example, Hollenstein et al. (2004) analyzed the dyadic affective states of parents and kindergarten children observed for 2 hours. Rigidity (a limited range of dyadic states, few transitions between states, and high mean durations of states) predicted both internalizing and externalizing problems in the children. Specifically, those with problems had more rigid interactions, even with controlling for negative affect. Thus, the structure of these interactions predicted psychopathology above and beyond the affective content of those interactions.

Using the same measures with a sample of clinically aggressive children, Granic et al. (2007) showed that affective flexibility was associated with successful treatment. In this study, children were treated with a cognitive behavioral therapy (CBT)-like behavioral inhibition program while their parents underwent parent management training, a combination associated with some of the best treatment outcomes for children of this age (Brestan & Eyberg, 1998; Eyberg, Nelson, & Boggs, 2008; Kazdin, 1997). Parent–child dyads with children who improved over the course of a 12-week treatment period also showed an increase in flexibility pre- to post-treatment. In contrast, dyads with children who did not improve actually became less flexible. Thus, rigidity is not only characteristic of childhood behavior problems, but an increase in flexibility is associated with the amelioration of those problems.

One of the questions that arise in this line of research is whether valence interacts with flexibility. Lunkenheimer et al. (2011) sought to answer this question by examining statistical interactions between positive affect and dynamic flexibility in parent–toddler interactions. Results showed main effects of positive affect and dynamic flexibility (affective range and transitions), as well as the interaction effect of positive affect with flexibility. However, the direction of these effects differed for interactions with mothers and interactions with fathers. For father–toddler interactions, greater overall dynamic flexibility and the positive affect–flexibility interaction predicted lower externalizing 2 years later. For mothers, the results were a bit more complex. Dynamic flexibility predicted higher externalizing levels at age 5, but the positive affect–flexibility interaction predicted lower externalizing. These results may reflect a developmental shift in dynamic flexibility that occurs some time after the age of 3 that may further depend on the valence and which parent is involved in the exchange. Certainly more research has to be completed in this area. Nevertheless, there is growing evidence that dynamic flexibility of dyadic affect is an important individual difference factor related to healthy social development.

Reactive Flexibility: Meso Scale across Contexts

At the next time scale, flexibility is defined differently. This adaptive flexibility is the correspondence between variability or change in behavior and a change of context or contextual demands (e.g., Cheng, 2001). Although this change can be reflected in differences in dynamic flexibility from one context to the next, reactive flexibility is more generally the process of shifting from one “set,” orientation, approach, or behavioral pattern to another because the conditions of the situation have changed. It is at this scale that behavior could be characterized as being perseverative if there is no flexible adaptation to the new conditions. Reactive flexibility thus reflects the most colloquial sense of flexibility and the meaning most often employed in models of cognitive flexibility that rely on measures of set shifting (e.g., Wisconsin Card Sort Task), flexible problem solving (e.g., Einstellung Water-Jar Task), and other performance tasks (e.g., Stroop tasks). Reactive flexibility also occurs in the interpersonal domain when the demands of the context change (e.g., having a conflict vs. doing something fun).

Flexibility at this time scale reflects the nature of open systems. A perturbation is what reveals a system’s adaptive response patterns. Without that perturbation, the functional properties might not be known. A rigid system with few and/or strong attractors might appear very similar to a flexible system with many and/or weak attractors prior to a perturbation. After the perturbation, however, the system dynamics reveal the relative adaptability of each. Granic and Lamey (2002), for example, showed that externalizing children who were not comorbid with internalizing problems (pure externalizers) had parent–child interactions that were similar to dyads with externalizing children who also had internalizing problems (comorbid group). However, after 4 minutes of discussing a conflict they had been having at home, the experimenter knocked on the door to signal that they should wrap up, resolve the conflict, and end on a good note within 2 minutes. This perturbation initiated two different response patterns, such that the pure externalizer dyads did not shift out of their parentally permissive interaction pattern (parent positive, child negative), but the dyads with comorbid children shifted into a more mutually hostile pattern.

Several studies to date have employed an A–B–A design to examine changes from positive to negative and back to positive contexts (Granic et al., 2007; Hollenstein, 2007; Hollenstein & Lewis, 2006). In these studies, parents and children first discussed a fun topic (e.g., winning the lottery), then a self-identified conflict or issue they had been having at home, followed by a second positive discussion. With a sample of adolescent girls and their mothers, Hollenstein and Lewis (2006) showed that dynamic flexibility (affective range and transitions) changed by discussion—it was high in the positive context and low in the negative context—reactive flexibility. With reactive flexibility, it is also possible to examine the changes in affective content. As expected, these mother–daughter dyads expressed the most negative affect during the conflict. In this way, greater dynamic flexibility during more positive contexts gave way to reduced dynamic flexibility combined with greater negative affect during the conflict context. However, the greater emotional challenge is to return to positive states immediately after negative ones. That is, a shift from a conflict discussion laden with negative affect to a positive discussion would require the most adaptively flexible shifts in affect. To examine this aspect of flexibility more closely, Granic et al. (2007) operationalized repair as the absence of negative affect during that last positive discussion. Recall that in this study described earlier, dynamic flexibility was associated with improvements in externalizing problems due to treatment. In terms of reactive flexibility, those dyads with children who improved due to treatment for externalizing behavior were significantly more likely to show repair after the conflict discussion than those who showed no improvement due to treatment. These studies reflect the nested nature of dynamic and reactive flexibility mentioned earlier. Being dynamically flexible at the micro scale enables the flexible adjustment to shifting environmental demands at the meso scale, while the degree of contextual sensitivity at the meso scale constrains real-time flexibility within a given context.

Reactive flexibility in the socioemotional domain can be seen across repeated situations of the same type. Lichtwarck-Aschoff et al. (2009) gathered diary data from adolescent girls in six 2-week periods over the course of a year. The diary reports included the girls’ emotional states (almost always, several specific emotions were reported at each time point) during and after conflicts (with their mothers) that had occurred between them each day. There was a quadratic relationship between the number of conflicts and emotional states. Those who reported consistent emotional states across conflicts had either a high number of conflicts or very few; those dyads with moderate numbers of conflicts reported more changes in emotional states from conflict to conflict. More importantly, those who were the most rigid in terms of unchanging emotional states across conflicts also had conflicts about a greater number of topics. Thus, those who experienced more interpersonal problems by dint of the larger number of conflicts perseverated in terms of repeating the same emotional reactions regardless of the topic. This “emotional context insensitivity” (Bylsma, Morris, & Rottenberg, 2008; Rottenberg, 2005; Rottenberg, Gross, & Gotlib, 2005) is a strong indication of diminished reactive flexibility.

Recently, Kuppens and colleagues have explored context insensitivity through the construct of emotional inertia (Suls, Green, & Hillis, 1998) in adolescence (Kuppens et al., 2010; Kuppens et al., 2012). Emotional inertia reflects the tendency to perseverate in a particular emotional state despite shifting contextual demands. In these studies, the degree of autocorrelation of emotional states over time was used to operationalize inertia–the more that someone persisted in an emotional state, the higher the autocorrelation. Within multilevel models, inertia was found to be related to low self-esteem and depression. Moreover, this relation was found for inertia in real time across several different interaction contexts as well as across 14 days of experience sampling reports. Thus, emotional inertia is consistent with both dynamic rigidity at the micro-time scale and reactive rigidity at the meso-time scale in the Flex3 model.

As with the Kuppens et al. (2012) study, it is important to note that there have been a range of studies employing the experience sampling method (ESM, also known as ecological momentary assessment) that have exposed the importance of affective variability at the meso-time scale over the course of days or weeks. The results at these time ranges show a different pattern than with reactive flexibility across temporally adjacent contexts. For example, measuring positive affect (PA) and negative affect (NA) 10 times a day over five 5-day periods, Wichers et al. (2010) showed that NA but not PA variability (measured as the within-subject standard deviation) predicted symptoms of depression and anxiety, but did not predict the recurrence of a depressive episode in previously depressed subjects. Similarly, Eid and Deiner (1999) showed that affect variability was uniquely positively associated with neuroticism. However, daily variability is not necessarily equivalent to reactive flexibility because the measured changes in affect are not empirically connected to situations or contextual demands. Thus, although daily variability of affect is likely to be related to flexibility, it is not yet clear how.

Trait/Developmental Flexibility: Macro Scale of Traits and Transitions

At the longer time scale of months and years, there are at least two ways of conceptualizing flexibility. First, rather than identifying changes in behaviors or emotional states, there is the identification of individuals as being either flexible or rigid. This trait flexibility is the flexibility construct described by personality theorists (e.g., Catell, 1946; DeYoung et al., 2002) and theories of ego development and resilience (e.g., Block, 2002), often most closely associated with neuroticism (Eid & Deiner, 1999). Flexibility can then be viewed as a stable trait of the individual that has emerged bottom–up from the dynamic (i.e., behavioral) and reactive (i.e., situational) flexibility at the shorter time scales. With the top–down portion of circular causality, trait flexibility constrains these micro and meso processes to instantiate the stability of flexibility within the individual. In some circumstances it may be efficacious to identify a person as flexible or rigid, but, as noted by Paulhus and Martin (1988), possessing the trait does not ensure the appropriate deployment of the affective behavior to a situation. Similarly, although trait flexibility is a useful framework and consistent with our model, it is also static and does not directly reflect the flexibility dynamics that might occur at the developmental-time scale. Therefore, we also conceptualize flexibility at the macro scale with perturbations or transitions that occur across the lifespan.

As an individual grows through the lessons of childhood, the tumult of adolescence, the novel independence of young adulthood, and the varied experiences of work, relationships, parenting, and illness across adulthood, there is an accumulation of experiences with which one needs to cope, flexibly or rigidly (e.g., Cheng, 2001). This is developmental flexibility at the macro time scale that reflects responses to two sources of perturbation: (a) the exogenous or idiosyncratic perturbations unique to each individual and (b) the endogenous perturbations of normative development. Exogenous perturbations are major life events which subsume a wide range of contexts and interpersonal interactions. Examples of these perturbations include the death of a spouse or parent, surviving a natural disaster, moving to a new city, or becoming a university professor. Here the change in contexts, goals, and demands is so significant that it has an impact on all aspects of daily life. Flexible adaptation at this level certainly requires dynamic and reactive flexibility as well, but is not simply a static summary of these micro and meso processes a la trait flexibility. Uprooting one’s life and moving to a new city requires the formation of new routines and habits as well as new relationships. One can adapt to this major life event rigidly (by persisting in previously formed habits and routines and not cultivating new relationships) or flexibly (by adapting behaviors to fit the new circumstances). This type of flexibility may be appropriately termed life-transition flexibility to connote both the time scale and scope of the perturbation.

Perturbations at the macro scale may also be endogenous, reflecting the inevitable changes of physiology (e.g., puberty) and cognition (e.g., formal operations) that transpire across the lifespan. The progression of development is sequential, frequently conceptualized as a series of stages (e.g., Case, 1985; Piaget, 1954), and between stable periods are periods of transformation and change (Granic et al., 2003; Hartelman, van der Maas, & Molenaar, 1998; Lewis, Zimmerman, Hollenstein, & Lamey, 2004). In DS terms, this macro level of flexibility may reflect developmental phase transitions. Phase transitions are common to all complex and adaptive systems and are periods in which stability and predictability are decreased, variability is increased, and the system is most vulnerable to perturbations (Granic et al., 2003; Hartelman et al., 1998; Lewis & Granic, 1999; Lewis et al., 2004; Lichtwarck-Aschoff et al., 2012; Thelen & Smith, 1994; van der Maas & Hopkins, 1998; van der Maas & Molenaar, 1992; van Geert, 1998). For example, the onset of the 18–20-month transition is a developmentally normative transition period during which the cognitive and emotional advances of the second year of life instigate less predictable, more unstable, more variable, and more flexible behavior (Lewis et al., 2004). Colloquially termed the “terrible twos,” this period can be marked by radical changes in sleeping and eating habits, defiance, aggression, and tantrums (Lewis & Granic, 2000, 2009; Lewis et al., 2004). In a 10-year longitudinal study of parent–boy interactions across adolescence, Granic et al. (2003) showed that the number of affective states and transitions between these states (i.e., dynamic flexibility) peaked at the onset of early adolescence. According to the developmental phase transition hypothesis (Granic et al., 2003; Lewis et al., 2004), this developmental phase transition renders the adolescent system open for change, for better or worse, as a period of vulnerability or opportunity (Dahl, 2004). Across the lifespan, developmental flexibility can manifest through normative transitions (e.g., adolescence) as well as idiosyncratic transitions (e.g., relocation to a distant city). Accordingly, this temporary increase in flexibility during a phase transition enables a system (e.g., person) to actualize its adaptive potential by exploring new patterns of behavior.

In summary, the Flex3 model forms the basis of a more coherent and integrated conceptualization of flexibility and rigidity. These three levels of micro, meso, and macro are distinct though interrelated, and we have offered a terminology and structure through which to view previous research and to guide future investigations. Furthermore, we have focused here on affect to most clearly illustrate the model, but of course flexibility of cognitions and other behaviors could also be modeled at these three scales. Therefore, we next briefly review some implications and extensions of the Flex3 model.

Implications and Extensions of the Flex3 Model

There are several important implications and extensions of the Flex3 model. First, the separation of flexibility into three distinct levels also allows for rigid flexibility and flexible rigidity. These are best considered at the micro and meso scales. A person could remain dynamically flexible despite changes in context that would require more adaptive adjustment of real-time behavior. This would be exemplified by the indecisive or “wishy-washy” individual (Paulhus & Martin, 1988). In contrast, there is the possibility of flexible dynamic rigidity—getting stuck in specific states (e.g., anger) within a context (e.g., conflict), but also being able to adaptively switch to a new, yet still dynamically rigid, state (e.g., elation) in a subsequent context.

Second, these possibilities then lead to a more general implication that both rigidity and flexibility can be beneficial or deleterious depending on the context(s) or overall pattern. The positive aspects of flexibility are not difficult to imagine. Learning, creativity, social competence, and openness to experience, to name a few, are all enhanced by flexibility in one way or another. Similarly, rigidity is a key component of functional fixedness, obsession/compulsion, and many other mental health disorders. However, this “flexibility = good” and “rigidity = bad” conceptualization is overly simplistic and does not reflect real-world flexibility. The negative aspects of flexibility include the lowered predictability of behavior, especially in a social context (Paulhus & Martin, 1988), and limited persistence toward achievement of personal goals. Attention deficits may beget a flexibility that is problematic, for example. Rigidity, on the other hand, through routine, habits, and conditioning, is efficient by dint of the conservation of energy and effort. Without a certain amount of rigid perseveration, difficult social, cognitive, or performance tasks would not be resolved. This bivalent merit of flexibility has led to the hypothesis that the relation of flexibility to healthy socioemotional functioning is curvilinear (Lichtwarck-Aschoff et al., 2009; Lunkenheimer et al., 2011). That is, modest levels of dynamic and reactive flexibility may be associated with desirable outcomes, whereas excessively low or high flexibility may be associated with poorer outcomes. Further research is needed to test this hypothesis more directly.

Third, flexibility and the Flex3 model are descriptive. It will be important to understand the underlying mechanisms that might give rise to more or less flexible behavior. One likely candidate for affective flexibility is the temperament construct of self-regulation: effortful control (Rothbart, 2007). In order to transition from one emotional state to another, the first must be inhibited before the next can manifest. Thus, greater self-regulation via control of inhibition, attention, and arousal enables flexibility. Physiological fluctuations governed by the parasympathetic nervous system are another possible mechanism underlying flexibility (Friedman, 2007) and fit well with the Flex3 model. The moment-to-moment or breath-to-breath dynamics of resting respiratory sinus arrhythmia (RSA) that indicate preparedness for adaptation are consistent with dynamic flexibility. The actual adaptive reactions of RSA changes due to socioemotional challenges are also consistent with changes at the meso scale and reactive flexibility. Direct investigations of these regulatory mechanisms will eventually enable the descriptive Flex3 model to be transformed into a more explanatory one.

Conclusion

Flexibility is a dynamic process. That is, it necessarily includes a temporal component as flexibility can only be observed as a change (or lack of change) over time. We have presented a conceptual model of flexibility at three time scales—micro, meso, and macro—in order to provide a definitional framework for flexibility theory and research. Consideration of different time scales provides an account of flexibility (a) within a given context or situation, (b) as an adaptation to changing demands across different contexts, and (c) as a metadescriptive characteristic or trait. Thus, our Flex3 model accounts for both contextual and temporal parameters.

The success of human evolution has depended on flexible adaptations to shifting environmental demands. Indeed, it is not controversial to claim that humans are the most flexibly adaptive species on earth. This is true ontogenetically as well as phylogenetically; the development of individuals depends on flexibility as well. The process of development from birth through adulthood can be characterized by a series of challenges that require ever more sophisticated methods of adaptation including learning, self-regulation, and metacognition. We hope that our model will facilitate and strengthen inquiry in this area to better understand the functional complexity of human behavior.

References

Block

(2002). Personality as an affect-processing system. Mahwah, NJ: Lawrence Erlbaum Associates.

Bonanno

G. A.

Papa

Lalande

Westphal

Coifman

(2004). The importance of being flexible. Psychological Science, 15, 482–487.

Branje

S. J. T.

(2008). Conflict management in mother–daughter interactions in early adolescence. Behaviour, 145, 1627–1651.

Brestan

E. V.

Eyberg

S. M.

(1998). Effective psychosocial treatments of conduct disordered children and adolescents: 29 years, 82 studies, and 5,272 kids. Journal of Clinical Child Psychology, 27, 180–189.

Bylsma

L. M.

Morris

B. H.

Rottenberg

(2008). A meta-analysis of emotional reactivity in major depressive disorder. Clinical Psychology Review, 28, 676–691.

Case

(1985). Intellectual development: Birth to adulthood. Orlando, FL: Academic Press.

Cattell

R. B.

(1946). Personality structure and measurement II: The determination and utility of trait modality. British Journal of Psychology, 36, 159–174.

Cheng

(2001). Assessing coping flexibility in real-life and laboratory settings: A multimethod approach. Journal of Personality and Social Psychology, 80, 814–833.

Cheng

(2009). Dialectical thinking and coping flexibility: A multimethod approach. Journal of Personality, 77, 471–494.

10.

Chown

S. M.

(1959). Rigidity: A flexible concept. Psychological Bulletin, 56, 195–223.

11.

Dahl

R. E.

(2004). Adolescent brain development: A period of vulnerabilities and opportunities. Annals of the New York Academy of Sciences, 1021, 1–22.

12.

Davis

Montgomery

Wilson

(2002). Worry and heart rate variables: Autonomic rigidity under challenge. Journal of Anxiety Disorders, 16, 639–659.

13.

Deveney

C. M.

Deldin

P. J.

(2006). A preliminary investigation of cognitive flexibility for emotional information in major depressive disorder and non-psychiatric controls. Emotion, 6, 429–437.

14.

DeYoung

C. G.

Peterson

J. B.

Higgins

D. M.

(2002). Higher-order factors of the Big Five predict conformity: Are there neuroses of health? Personality and Individual Differences, 33, 533–552.

15.

Eid

Deiner

(1999). Intraindividual variability in affect: Reliability, validity and personality correlates. Journal of Personality and Social Psychology, 76, 662–676.

16.

Eyberg

S. M.

Nelson

M. M.

Boggs

S. R.

(2008). Evidence-based psychosocial treatments for children and adolescents with disruptive behavior. Journal of Clinical Child and Adolescent Psychology, 37, 215–237.

17.

Friedman

B. H.

(2007). An autonomic flexibility–neurovisceral integration model of anxiety and cardiac vagal tone. Biological Psychology, 74, 185–199.

18.

Granic

Hollenstein

(2003). Dynamic systems methods for models of developmental psychopathology. Development and Psychopathology, 15, 641–669.

19.

Granic

Hollenstein

(2006). A survey of dynamic systems methods for developmental psychopathology. In Cicchetti

Cohen

D. J.

(Eds.), Developmental psychopathology (2nd ed., pp. 889–930). New York, NY: Wiley.

20.

Granic

Hollenstein

Dishion

T. J.

Patterson

G. R.

(2003). Longitudinal analysis of flexibility and reorganization in early adolescence: A dynamic systems study of family interactions. Developmental Psychology, 39, 606–617.

21.

Granic

Lamey

A. V.

(2002). Combining dynamic systems and multivariate analyses to compare the mother–child interactions of externalizing subtypes. Journal of Abnormal Child Psychology, 30, 265–283.

22.

Granic

O’Hara

Pepler

Lewis

M. D.

(2007). A dynamic systems analysis of parent–child changes associated with successful “real-world” interventions for aggressive children. Journal of Abnormal Child Psychology, 35, 845–857.

23.

Gruber-Baldini

A. L.

Schaie

K. W.

Willis

S. L.

(1995). Similarity in married couples: A longitudinal study of mental abilities and rigidity–flexibility. Journal of Personality and Social Psychology, 69, 191–203.

24.

Haken

(1983). Synergetics: An introduction. Berlin, Germany: Springer-Verlag.

25.

Hartelman

P. A.

van der Maas

H. L. J.

Molenaar

P. C. M.

(1998). Detecting and modelling developmental transitions. British Journal of Developmental Psychology, 16, 97–122.

26.

Hollenstein

(2007). State space grids: Analyzing dynamics across development. International Journal of Behavioral Development, 31, 384–396.

27.

Hollenstein

(2011). Twenty years of dynamic systems approaches to development: Significant contributions, challenges, and future directions. Child Development Perspectives, 5, 256–259.

28.

Hollenstein

Granic

Stoolmiller

Snyder

(2004). Rigidity in parent–child interactions and the development of externalizing and internalizing behavior in early childhood. Journal of Abnormal Child Psychology, 32, 595–607.

29.

Hollenstein

Lewis

M. D.

(2006). A state space analysis of emotion and flexibility in parent–child interactions. Emotion, 6, 656–662.

30.

Johnson

S. L.

Nowak

(2002). Dynamical patterns in bipolar depression. Personality and Social Psychology Review, 6, 380–387.

31.

Kazdin

A. E.

(1997). Parent management training: Evidence, outcomes, and issues. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1349–1356.

32.

Kelso

J. S.

(1995). Dynamic patterns: The self-organization of brain and behavior. Cambridge, MA: MIT Press.

33.

Kuppens

Allen

N. B.

Sheeber

L. B.

(2010). Emotional inertia and psychological maladjustment. Psychological Science, 21, 984–991.

34.

Kuppens

Sheeber

L. B.

Yap

M. B. H.

Whittle

Simmons

J. G.

Allen

N. B.

(2012). Emotional inertia prospectively predicts the onset of depressive disorder in adolescence. Emotion, 12, 283–289.

35.

Lavictoire

Snyder

Stoolmiller

Hollenstein

(2012). Affective dynamics in triadic peer interactions in early childhood. Nonlinear Dynamics in Psychology and the Life Sciences, 16, 293–312.

36.

Leach

(1967). A critical study of the literature concerning rigidity. British Journal of Social and Clinical Psychology, 6, 11–22.

37.

Lewis

M. D.

(1997). Personality self-organization: Cascading constraints on cognition–emotion interaction. In Fogel

Lyra

M. C. D. P.

Valsiner

(Eds.), Dynamics and indeterminism in developmental and social processes (Vol. 294, pp. 193–216). Hillsdale, NJ: Lawrence Erlbaum Associates.

38.

Lewis

M. D.

(2000). The promise of dynamic systems approaches for an integrated account of human development. Child Development, 71, 36–43.

39.

Lewis

M. D.

(2005). Bridging emotion theory and neurobiology through dynamic systems modeling. Behavioral and Brain Sciences, 28, 169–245.

40.

Lewis

M. D.

(2011). Dynamic systems approaches: Hot enough? Cool enough? Child Development Perspectives, 5, 279–285.

41.

Lewis

M. D.

Ferrari

(2001). Cognitive-emotional self-organization in personality development and personal identity. In Bosma

H. A.

(Ed.), Identity and emotion: Development through self-organization (pp. 177–198). New York, NY: Cambridge University Press.

42.

Lewis

M. D.

Granic

(1999). Who put the self in self-organization? A clarification of terms and concepts for developmental psychopathology. Development and Psychopathology, 11, 365–374.

43.

Lewis

M. D.

Granic

(Eds.). (2000). Emotion, development, and self-organization: Dynamic systems approaches to emotional development. New York, NY: Cambridge University Press.

44.

Lewis

M. D.

Granic

(2009). Bed timing: The “when-to” guide to helping your child sleep. Toronto, Canada: Collins Canada.

45.

Lewis

M. D.

Liu

(2011). Three time scales of neural self-organization underlying basic and non-basic emotion. Emotion Review, 3, 416–423.

46.

Lewis

M. D.

Todd

(2007). The self-regulating brain: Cortical-subcortical feedback and the development of intelligent action. Cognitive Development, 22, 406–430.

47.

Lewis

M. D.

Zimmerman

Hollenstein

Lamey

A. V.

(2004). Reorganization in coping behavior at 1½ years: Dynamic systems and normative change. Developmental Science, 7, 56–73.

48.

Lichtwarck-Aschoff

Hasselman

Cox

Peppler

Granic

(2012). A characteristic destabilization profile in parent–child interactions associated with treatment efficacy for aggressive children. Nonlinear Dynamics, Psychology, and Life Sciences, 16, 353–379.

49.

Lichtwarck-Aschoff

Kunnen

van Geert

(2009). Here we go again: A dynamic systems perspective on emotional rigidity across parent–adolescent conflicts. Developmental Psychology, 45, 1364–1375.

50.

Lichtwarck-Aschoff

van Geert

Bosma

Kunnen

(2008). Time and identity: A framework for research and theory formation. Developmental Review, 28, 370–400.

51.

Luchins

A. S.

Luchins

E. H.

(1959). Rigidity of behavior: A variational approach to the effect of Einstellung. Eugene, OR: University of Oregon Books.

52.

Lunkenheimer

E. S.

Hollenstein

Wang

Shields

A. M.

(2012). Flexibility and attractors in context: Family emotion socialization patterns and children’s emotion regulation in late childhood. Nonlinear Dynamics in Psychology and the Life Sciences, 16, 269–292.

53.

Lunkenheimer

E. S.

Olson

S. L.

Hollenstein

Sameroff

A. J.

Winter

(2011). Dyadic flexibility and positive affect in parent–child coregulation and the development of child behavior problems. Development and Psychopathology, 23, 577–591.

54.

Paulhus

D. L.

Martin

C. L.

(1988). Functional flexibility: A new conception of interpersonal flexibility. Journal of Personality and Social Psychology, 55, 88–101.

55.

Piaget

(1954). Construction of reality in the child. New York, NY: Basic Books.

56.

Rothbart

M. K.

(2007). Temperament, development, and personality. Current Directions in Psychological Science, 16, 207–212.

57.

Rottenberg

(2005). Mood and emotion in major depression. Current Directions in Psychological Science, 14, 167–170.

58.

Rottenberg

Gross

J. J.

Gotlib

I. H.

(2005). Emotion context insensitivity in major depressive disorder. Journal of Abnormal Psychology, 114, 627–639.

59.

Schultz

P. W.

Searleman

(2002). Rigidity of thought and behavior: 100 years of research. Genetic, Social, and General Psychology Monographs, 128, 165–207.

60.

Suls

Green

Hillis

(1998). Emotional reactivity to everyday problems, affective inertia, and neuroticism. Personality and Social Psychology Bulletin, 24, 127–136.

61.

Thelen

Smith

L. B.

(1994). Dynamic systems approach to the development of cognition and action. Cambridge, MA: MIT Press.

62.

Thompson

R. A.

(1990). Emotion and self-regulation. In Thompson

R. A.

(Ed.), Nebraska symposium on motivation, 1988: Socioemotional development. Current theory and research in motivation, Vol. 36 (pp. 367–467). Lincoln, NE: University of Nebraska Press.

63.

Tomkins

(1962). Affect imagery consciousness. New York, NY: Springer.

64.

van der Mass

H. L. J.

Hopkins

. (1998). Developmental transitions: So what’s new? British Journal of Developmental Psychology, 16, 1–13.

65.

van der Maas

H. L. J.

Molenaar

P. C. M

. (1992). Stagewise cognitive development: An application of catastrophe theory. Psychological Review, 99, 395–417.

66.

van Geert

. (1994). Dynamic systems of development: Change between complexity and chaos. New York, NY: Harvester Wheatsheaf.

67.

van Geert

. (1998). A dynamic systems model of basic developmental mechanisms: Piaget, Vygotsky, and beyond. Psychological Review, 105, 634–677.

68.

van Geert

. (2011). The contribution of complex dynamic systems to development. Child Development Perspectives, 5, 273–278.

69.

von Bertalanffy

. (1968). General system theory: Foundations, development, applications (rev. ed.). New York, NY: George Braziller.

70.

Westphal

Seivert

N. H.

Bonanno

G. A.

(2010). Expressive flexibility. Emotion, 10, 92–100.

71.

Wichers

Peeters

Geschwind

Jacobs

Simons

C. J. P.

Derom

van Os

. (2010). Unveiling patterns of affective responses in daily life may improve outcome prediction in depression: A momentary assessment study. Journal of Affective Disorders, 124, 191–195.

72.

Witherington

D. C.

(2011). Taking emergence seriously: The centrality of circular causality for dynamic systems approaches to development. Human Development, 54, 66–92.