Modus operandi and affect in Sweden: the Swedish version of the Regulatory Mode Questionnaire

Regulatory Mode Theory: locomotion and assessment orientations

In the classic theory of control (Carver & Scheier, 1990), assessment and locomotion are functionally interdependent parts of a single self-regulatory process. In the first step, through the assessment function, an individual compares the different mental goal representations and chooses the favorite one, then he/she compares the current state with the desired end-state. In the second step, through the locomotion function, the individual tries to reduce the current-end states discrepancy by starting and maintaining movement toward the target goal. Conversely, in the frame of Regulatory Mode Theory (Higgins, Kruglanski & Pierro, 2003), locomotion and assessment are independent functions, which can receive different emphasis. In this context, locomotion is the aspect of self-regulation concerned with “movement from state to state and with committing the psychological resources that will initiate and maintain goal-related movement in a straightforward and direct manner, without undue distractions or delays” (Kruglanski et al., 2000, p. 794) and assessment is “the comparative aspect of self-regulation concerned with critically evaluating entities or states, such as goals or means, in relation to alternatives in order to judge relative quality” (Kruglanski et al., 2000, p. 794).

There are two important assumptions in Regulatory Mode Theory: (1) temperament and socialization lead individuals to develop different degrees of concern with movement or progress (i.e., emphasizing locomotion) and different degrees of concern for standards and for critically evaluating alternatives (i.e., emphasizing assessment) and (2) the reasons for preferring locomotion or assessment are unrelated (Kruglanski et al., 2000). In other words, locomotion and assessment are independent and orthogonal biopsychosocial modes of operation, thus, some individuals can be high on one mode and low on the other mode, low on both modes, or high on both modes (cf. Garcia & Lindskär, 2016). The RMQ is indeed an instrument that helps to measure the chronic preference for one regulatory mode orientation over the other.

Applications of the RMQ

Previous studies have shown the high practical relevance and the wide potential applicability of the RMQ. Since its development, several studies have shown that the two regulatory modes are related with many important aspects, ranging from organizational behavior (e.g., preference for leadership style; Benjamin & Flynn, 2006; time management; Amato et al., 2014), to entrepreneurial aptitude (Amato et al., 2016), to cognition (e.g., cognitive biases, counterfactual thinking; Pierro et al., 2008), and economic decision-making (e.g., inter-temporal choices; Mannetti et al., 2009). Particularly relevant to the present study is the strong link between self-regulation and both health and well-being. Specifically, there is evidence that a prevalence of locomotion orientation is positively related to affective well-being (i.e., frequent experience of positive affect and infrequent experience of negative affect), while a prevalence of assessment orientation is associated with maladaptive psychological functioning and even psychopathological symptoms (Garcia et al., 2016b). For instance, locomotion is both indirectly and directly positively related to affective well-being. For example, people who are high in locomotion do not engage in social comparison and have therefore a tendency to be less materialistic, which in turn leads them to experience less negative affect (Giacomantonio, Mannetti & Pierro, 2013). Indeed, engaging in social comparison and self-evaluation, something a person does when in assessment mode, leads to materialistic thinking, which in turn leads to high levels of negative affect (Giacomantonio, Mannetti & Pierro, 2013). That being said, affective well-being is more that the absence of negative emotions (cf. Cloninger, 2004). In this context, people who are high in locomotion report having a sense of life purpose and this lead them to be more satisfied with their life (Vazeou-Nieuwenhuis, Orehek & Scheier, 2017). In addition, locomotion is directly related to higher levels of vitality (Kruglanski et al., 2000), higher levels of optimism (Kruglanski et al., 2000), happiness (Smith et al., 1993) and last but not the least, positive affect (Kruglanski et al., 2000; Garcia et al., 2015c; Garcia et al., 2016b). Individuals with a high assessment orientation, on the other hand, are more likely to present depressive symptoms (Hong, Tan & Chang, 2004), they are also more susceptible to suffer from borderline disorder (Bornovalova et al., 2008) and obsesive-compulsive disorder (Shalev & Sulkowski, 2009), and report high levels of negative affect (Kruglanski et al., 2000; Garcia et al., 2015c; Garcia et al., 2016b).

The present study

In the present study, based on Regulatory Mode Theory, we test the unidimensionality of the Swedish version of the RMQ scales (locomotion and assessment). We also test the reliability of the Swedish RMQ and the predictive validity of the regulatory mode scales by investigating the relationship between regulatory modes (locomotion and assessment) and affective well-being (positive and negative affect). The affectivity dimensions to test the RMQ scales’ predictive validity were found to be appropriate for different reasons.

First of all, the affectivity dimension range from pleasant/unpleasant engagement (for example: enthusiastic and active for positive affect, anger and fear for negative affect), to unpleasant/pleasant disengagement (for example: sad and bored for positive affect, calm and serene for negative affect) (Watson, Clark & Tellegen, 1988; Watson & Clark, 1994). Thus, as locomotion and assessment (Kruglanski et al., 2000), positive affect and negative affect are best conceptualized as two independent orthogonal dimensions that are part of a complex adaptive system and not as ends of one single dimension (Garcia, 2011; Garcia, in press-a). These two affectivity dimensions reflect approaching behavior (positive affect) and avoidant behavior (negative affect) (Garcia, 2011; Garcia, in press-a). Thus, we expected locomotion to predict positive affect or approaching behavior and assessment to predict negative affect or avoidant behavior (cf. with Kruglanski et al., 2000, who also suggested a positive correlation between locomotion and positive affect because locomotion or forward movement contributes to a sense of progress and a positive correlation between assessment and negative affect because high assessment involves ruminating self-evaluation that may highlight the discrepancies between one’s actual self and a desired self).

Secondly, according to Regulatory Mode Theory, temperament and socialization lead individuals to develop different degrees of locomotion and assessment. Indeed, Kruglanski and colleagues (2000) found a positive correlation between locomotion and extraversion (r = .38; p < .001), a negative correlation between locomotion and neuroticism (r = −.20; p < .001), a positive correlation between assessment and neuroticism (r = .41; p < .001), and a negative correlation between assessment and extraversion (r = −.03; nonsignificant). Notably, positive and negative affect are almost identical with these personality traits (i.e., extraversion and neuroticism, respectively) (Larsen & Ketelaar, 1991; see also Garcia et al., 2015a; Garcia et al., 2015b; Rapp Ricciardi et al., 2014; Garcia & Moradi, 2013; Garcia, 2012; Garcia et al., 2010). In addition, compared to extraversion and neuroticism, the two affectivity dimensions involve more mood and social traits (Gunderson et al., 1999). Therefore, these affectivity dimension are not only seen as temperamental dispositions or signal sensitivity systems, but also as being complementary to extraversion and neuroticism (Tellegen, 1993). We argue that this supports our expectation of a positive relationship between locomotion and positive affect and between assessment and negative affect. In addition, this suggests a negative or non-significant relationship between locomotion and negative affect and between assessment and positive affect. Hence, we expected a stronger relationship between locomotion and positive affect than between assessment and positive affect. Conversely, we expected a stronger relationship between assessment and negative affect than between locomotion and negative affect.

Ethical statement

After consulting with the Network for Empowerment and Well-Being’s Review Board we arrived at the conclusion that the design of the present study (e.g., all participants’ data were anonymous and will not be used for commercial or other non-scientific purposes) required only informed consent from the participants.

Participants and procedure

A total of 655 students (357 university students and 298 high school students) in the West of Sweden (males = 408 females = 242, and five participants who didn’t report their gender, age_mean = 21.93 ± 6.51) participated in the study. All participants were informed that taking part of the study was voluntary and anonymous. They were presented with paper-pencil Swedish versions of the RMQ and the Positive Affect Negative Affect Schedule.

Measures

Preliminary descriptive analyses

We conducted Preliminary analyses to test some statistical issues, such as the presence of outliers and collinearity between variables. We first standardized the scores of each variable and tested if any cases had larger standardized scores than ±3.29, as recommended by Tabachnick & Fidell (2007). The analysis detected no cases of univariate outliers. The variables had skewness between −.10 to .64 and kurtosis between .70 and .12, thus, indicating that the data did not violate the assumption of normality (see Tabachnick & Fidell, 2007, p. 80). Moreover, we examined the distributions of the items and linearity by scatter plots. The results showed the absence of any signs of violation of normality and linearity assumptions. Finally, all the significant correlations among the variables were not so high as to threaten statistical multicollinearity and singularity. Indeed, the correlation coefficients range between .08 and .53, thus not constituting a cause for multicollianiarity and singularity issues (see, Tabachnick & Fidell, 2007, p. 88). Table 2 show the correlations between variables in the presents study, along Cronbach’s alphas and both means and standard deviations for each variable.

Locomotion

The first confirmatory analysis was conducted using the 12 items to estimate locomotion as one factor. The analysis showed that chi-square value was significant (Chi² = 268.79, df = 54, p < .001), the goodness of fit index was .93 and the root mean square error of approximation for the default model was .08. The chi-square statistic is heavily influenced by sample size (Kline, 2010), with larger samples (in this study N = 650) leading to larger value and therefore, a larger likelihood of being significant. However, other values (the goodness of fit index and the root mean square error of approximation) indicated that the hypothesized model fits the data well. All the regression weights/loadings between the factor (i.e., locomotion) and its 12 items were significant at p < .001 (ranging from .32 to .72) with the exception of two items that had low loadings (.29 for the reversed item 13: “When I finish one project, I often wait awhile before getting started on a new one.” and .15 for item 16: “When I decide to do something, I can’t wait to get started”.) which were also significant at p < .001 (see Fig. 1 and Table 3). Nevertheless, the results mirror those in the original article (Kruglanski et al., 2000) in which the fit of the one-factor model for both the locomotion and assessment scales were higher than .90 for goodness of fit index, RMRs around .10, and the overall alpha for locomotion was .82.

Assessment

The second confirmatory analysis was conducted using the 12 items to estimate assessment as one factor. The analysis showed that chi-square value was significant (Chi² = 362.20, df = 54, p < .001), the goodness of fit index was .91 and the root mean square error of approximation for the default model was .09. However, based on other important statistical indexes, such as the goodness of fit index and the root mean square error of approximation, we assumed that the model fits the data well. All the regression weights/loadings between the factor (i.e., assessment) and its items were significant at p < .001 (ranging from .26 to .66) with the exception of reversed item 10 (“I don’t spend much time thinking about ways others could improve themselves”.) which had low loading .07 (ns, p = .115) (see Fig. 2). Hence, item 10 was removed from the analyses. After this, the analysis showed that chi-square value was significant (Chi² = 336.88, df = 44, p < .001), the goodness of fit index was .91 and the root mean square error of approximation for the model was .10 (see Fig. 3 and Table 3). All the regression weights/loadings between the factor (i.e., assessment) and its 11 items were significant at p < .001 with ranging from .26 to .66. Thus, the model with just 11 items fits the data well.

Locomotion, assessment, and affective well-being

In the third SEM, we conducted a path analysis using locomotion (comprising 12 items) and assessment (comprising 11 items) as the independent variables and positive affect and negative affect as the dependent variables. The analysis showed that chi-square value was significant (Chi² = 27.85, df = 1, p < .001). However, as for the other two models previously presented, the goodness of fit index (.98), comparative fit index (.93), incremental fit index (.93), and normed fit index (.93) indicated that the hypothesized model fits the data well. All the regression weights between the independent variables (locomotion, assessment) and the dependent variables (positive affect and negative affect) were significant at p < .001. Standardized parameter estimates of correlation and direct effects appear in Fig. 4 and both standardized and unstandardized coefficients are listed in Table 4.

Additionally, we tested assessment and locomotion’s equalities of regression paths by comparing the regression weights between locomotion-positive affect to assessment-postive affect and between locomotion-negative affect and assessment-negative affect. The result showed that the effect of locomotion on positive affect was stronger than the effect of assessment on positive affect (Critical Ratios for Differences between Parameters/Z statistic = −15.16, p < .001), while the effect of assessment on negative affect was stronger than the effect of locomotion on negative affect (Critical Ratios for Differences between Parameters/Z statistic = 10.73, p < .001). See Table 3.

Limitations and suggestions for future research

To provide a valid instrument to use in the Swedish context is useful at the theoretical, research, and practical levels. At a theoretical level, the results presented here support the notion of two independent constructs of self-regulation that lead to different approach (positive affect) and prevention (negative affect) behaviors (cf. Garcia, 2011; Garcia, in press-a). That being said, at this point a fully biopsychosocial theory of self-regulation needs to be developed using holistic models, such as, Cloninger’s ternary model of human personality (Cloninger, 2004). This model disentangles personality in biological (harm avoidance, novelty seeking, reward dependence, and persistence), psychological (self-directedness and cooperativeness), and social (self-transcendence) aspects of personality (Cloninger, 2004; Garcia et al., 2017c; Garcia et al., 2017a; Garcia et al., 2017b; Garcia et al., 2017d). Hence, we suggest that future research should find the Cloninger personality model useful to test if self-regulatory mode theory is biopsychosocial in nature. As far as we understand the theory, self-regulation should be biopsychosocial; but no other work has addressed the concept holistically. At the research level, the Swedish RMQ could be used in the fields of motivation, health, personality, and organizational settings.

At the practical level, the RMQ could be a useful tool of evaluation of current social challenges in Sweden. For instance, Swedish students report grades that are among the lowest of the other 32 OECD countries (PISA, Sverigesradio, 2013; SNAE, 2009; SNAE, 2012) along deteriorations in psychological health (Fleming et al., 2013; Ghofranipour et al., 2013). At the same time, many Swedish high schools decrease the number of physical education hours available for their students (Sverigesradio, 2013)—a type of activity that is well-documented for positive effects on psychological health and well-being among adolescents and young adults (e.g., Archer & Garcia, 2014). Interestingly, in a sample of Swedish adolescents, locomotion predicted frequently exercising, which in turn predicted higher grades, while assessment predicted higher grades but also lower levels of well-being (Garcia et al., 2015c). These findings suggest that interventions need to be holistic or biopsychosocial, that is, targeting all aspects of self-regulatory mode in order to target both academic achievement and well-being (cf. Garcia et al., 2016a). We suggest that the RMQ could be used as a tool in this endeavor. In addition, since regulatory mode can be situationally induced (Avnet & Higgins, 2003) and situations may influence the preference of one mode over the other (Kruglanski et al., 2010), then, educators and other practitioners might design person-centered interventions that enhance or promote locomotion orientation through movement or exercise.

In spite of its theoretical and practical relevance, the present study has some limitations. We didn’t include other measures of individual differences (e.g., Big Five) in order to test the incremental validity of regulatory mode in the prediction of affective well-being. Even if previous studies have shown that the relationship between regulatory mode and, for example, psychological vitality or energy, remain unchanged while controlling for the Big Five personality factors (Kruglanski et al., 2000), it would be useful to address this issue in cross-cultural studies (cf. Delle Fave & Bassi, 2009, who suggest cross-cultural studies as important to disentangle happiness and related constructs). It is also worth noticing that our sample was unbalanced with respect to gender. The higher presence of males, compared to females, could lead to a selection bias. Despite that earlier studies, using samples from other cultures, have not indicated any significant gender differences (Kruglanski et al., 2000), it could be interesting to investigate the presence/absence of this difference in a Swedish sample and its effect on the factor structure of the RMQ. After all, gender differences seem to depend on the level of egalitarianism within the culture—although counter-intuitive, gender differences in, for example, personality traits, self-esteem, subjective well-being, and depression are larger in relatively high gender egalitarian cultures, such as Sweden (Schmitt et al., 2016). Another concern is that our sample only comprised high school and university students. Hence, further studies should include a more heterogeneous sample to address differences in, for example, gender, education level and also age. Moreover, although most of the studies using the RMQ disregard of the lie scale, as we did, and some have also applied only one of the scales (e.g., locomotion; Pierro et al., 2013), it might be useful to consider the presence of social desirability in future studies. Also, longitudinal designs would provide insights on the Swedish RMQ’s test-retest validity and perhaps also a better understanding of the causal relationships between regulatory mode and affectivity.

Finally, we urge person-oriented approaches for the investigation and development of Regulatory Mode Theory, especially since peoples’ tendency to prefer either locomotion and assessment is probably influenced by unrelated biological (e.g., temperament), psychological and social resources and conditions (Kruglanski et al., 2000). In other words, locomotion and assessment are independent and orthogonal biopsychosocial modes of operation, thus, some individuals can be high on one mode and low on the other mode, low on both modes, or high on both modes (cf. Garcia & Lindskär, 2016). Person-oriented statistical approaches are methods that focus on internal patterns, in contrast to variable-oriented approaches which focus on individual differences (Lundh, 2015). For instance, from a person-oriented framework these two modes of operation (locomotion and assessment) within the individual can be seen as interwoven components with whole-system properties (Bergman & Wångby, 2014). The outlook of the individual as a whole-system unit is then best studied by analyzing patterns of information (Bergman & Wångby, 2014; Bergman, Magnusson & El-Khouri, 2003; Bergman & Wångby, 2014; Bergman & Magnusson, 1997; see also Cloninger, Svrakic & Svrakic, 1997, who explain nonlinear dynamics in complex adaptive systems). There is, indeed, an increasing amount of person-oriented methods that can be used (Garcia, in press-a; Garcia, in press-b; Garcia & Lindskär, 2016; Garcia, MacDonald & Archer, 2015; Bergman & Lundh, 2015; Valsiner, 2015; Molenaar, 2015; Laursen, 2015; Asendorpf, 2015; Von Eye & Wiedermann, 2015; Aunola et al., 2015; Baker, 2015; MacDonald & Kormi-Nouri, 2013) as well as statistical software aimed for these endeavors (ROPstat; Vargha, Torma & Bergman, 2015).

Conclusion and final remarks

One way to improve the applicability—and consequently the usefulness—of a measure is to verify its reliability and validity in different cultural contexts. The present study tested and confirmed the one-dimensionality structure and the reliability of the two scales of the RMQ (i.e., locomotion and assessment) in the Swedish context. Moreover, the present study confirmed the discriminant validity of self-regulation modes on affective well-being dimensions. This is important because the two affectivity dimensions are also measures of anxiety, which involves high negative affect, and depression, which involves both high negative affect and low positive affect (Clark & Watson, 1991). We suggest that the Swedish version of the RMQ, with only minor modifications, is a useful instrument to tap individual differences in locomotion and assessment. In sum, the present study contributes to the validation of the RMQ in the Swedish culture, adds support to the theoretical framework of self-regulatory mode, and suggests that the regulatory mode system could be a psychological complex adaptive system, that along with the biological affectivity system, forms a larger biopsychosocial system of self-regulation.