Self-Regulatory Strategy Use,Efficacy,and Strategy-Situation-Fit in Self-Control Conflicts of Initiation,Persistence,and Inhibition

Method

This study aimed at generating a list of the strategies that people use in their daily lives to help themselves succeed in self-control conflicts of inhibition, initiation, and persistence. This list of strategies was used as the basis for the Main Study in which participants were asked about their strategy use in daily life.

Results and Brief Discussion

In sum, participants reported 508 self-control conflicts. 210 (41.3%) of these were categorized as initiation conflicts, 129 (25.4%) as inhibition conflicts, and 105 (20.7%) as persistence conflicts. For 64 (12.6%) conflicts, participants indicated that they were neither conflicts of initiation, persistence, or inhibition but a different type of situation. ¹ Across conflicts, participants described 525 different strategies, with an average of 3.8 strategies (SD = 3.5) per conflict. 256 (48.8%) of these strategies referred to conflicts of initiation, 127 (24.2%) to conflicts of persistence, and 92 (17.5%) to conflicts of inhibition. Through the various iterations described above, the coding schema arrived at a list of 26 different strategies, most of which could be used for all three types of conflicts. These were: “Changing the environment,” “removing or reducing distractions or temptations,” “seeking social support,” “adopting a process focus,” “distracting oneself from the activity or temptation,” “anticipating self-reward,” “focusing on negative consequences,” “focusing on positive consequences,” “goal setting,” “monitoring one’s goal progress,” “planning/scheduling,” “putting oneself under (social, time, and performance) pressure,” “reasoning in favor of the target activity/against the temptation,” “reminding oneself of a commitment to oneself or others,” “self-affirmation,” “self-talk,” “thinking of the near finish,” and “suppressing the impulse to quit or to give in to the temptation/forcing oneself to initiate the aversive activity.” The strategies “changing the activity itself” and “task enrichment” were suitable only for conflicts of initiation and persistence. In addition, the strategies “substituting the temptation,” “indulging a little,” and “taking a break” were suitable for only one type of conflict and, in turn, not considered in the main study. The full list of strategies, and their descriptions from the coding scheme is displayed in Table 1S on the OSF.

The study shows that there is great variety in the ways that people strategically self-regulate during conflicts of inhibition, initiation, and persistence. In our perspective, the self-regulatory strategies are located on a useful level of abstraction, that was prompted by participants’ own responses and that seems to provide a good balance between parsimony and differentiation (see also Hennecke et al., 2019). It should therefore be useful for application in an experience sampling study where participants are asked to choose from a list those strategies they deployed in a recent self-control conflict.

Note that even though we applied a bottom-up approach, the strategies reported furthermore have their equivalents in the self-regulation literature. Not only was it possible to assign the strategies to the categories or stages proposed in the process model of self-regulation (Duckworth et al., 2016) (except for “emotion regulation” which can occur at various stages and in a variety of ways and was, due to a lack of specificity, not considered in the Main Study), but many of the more specific strategies, for example, “changing the environment”, “reducing distractions”, “distraction”, “anticipating self-reward”, or “goal setting”, have already been reported and studied elsewhere (e.g., Bandura, 1976; Ent et al., 2015; Goldfried & Merbaum, 1973; Grunschel et al., 2016; Klein et al., 1999; Kuhl, 1984; Locke & Latham, 2002; Mischel et al., 1972; Pintrich, 2000; Schunk, 1990; Zimmerman, 1990, 2000).

Method

Descriptive Statistics

Participants provided a total of 41,461 reports. They experienced a self-control conflict at 33.9% (n = 14,067) of measurements, with a mean number of conflicts per participant of M = 19.6. Conflicts of initiation (n = 7,310, 17.6%) were the most common type of self-control conflict, followed by conflicts of persistence (n = 3,832, 9.2%) and conflicts of inhibition (n = 2,925, 7.1%). Initiation conflicts were reported most frequently for studying (34.1%), commuting (11.6%), and work (11.0%). Persistence conflicts were reported most frequently during studying (50.8%), work (13.5%), and classes (10.7%). Temptations tended to occur primarily in the domains of food (43.3%), media (13.8%), and sleep (11.4%).

Initiation conflicts were reported most frequently during the pursuit of academic goals (53.9%), housework goals (14.9%), and health goals (14.5%). Persistence conflicts were reported most frequently during the pursuit of academic goals (73.0%), financial goals (9.9%), and social goals (9.8%). Inhibition conflicts were reported most frequently during the pursuit of health goals (54.5%), academic goals (24.9%), and social goals (8.1%).

Strategy Choice

Main effects

Next, we wanted to examine how often strategies were selected and whether strategies that were experienced as generally more effective were also selected more often. Figure 3 and Table 1 show a large variability in strategy selection: Whereas “reappraisal” was selected in only 2.0% of conflicts participants attempted to regulate, “focusing on positive consequences” was selected in almost a third of those instances.OPEN IN VIEWER

Figure 3.

Bar graphs showing the predicted likelihood of using a self-regulatory strategy as a function of conflict type. Note. Bars with different letters differ significantly from each other after Šidák correction. * Significant after Benjamini-Hochberg correction (omnibus and heterogeneity test as well as mean efficacy) or after Šidák correction (conflict type differences within a given self-regulatory strategy).

OPEN IN VIEWER

Table 1.

Summary of Main Results in the Main Study.

Strategy	Stage	Choice, %	Freq. Rank	Efficacy	Efficacy Rank	Efficacy: Type Moderation	Choice: Type Moderation	Flexibility
Focusing on positive consequences	Cognitive	32.4	1	0.27	5	No	Yes
Suppression	Response	22.2	2	0.27	4	Yes	Yes	No
Reminding oneself of a commitment to oneself or to others	Cognitive	21.9	3	0.11	11	No	Yes
Focusing on negative consequences	Cognitive	21.4	4	−0.05	16	No	No
Thinking of the near finish	Cognitive	21.0	5	0.28	3	No	Yes
Task enrichment	Situation	20.3	6	0.33	1	Yes	Yes	No
Monitoring one’s goal progress	Cognitive	12.2	7	0.26	7	No	Yes
Planning/scheduling	Cognitive	11.4	8	−0.10	19	No	Yes
Putting oneself under pressure	Cognitive	10.3	9	−0.21	20	No	Yes
Goal setting	Cognitive	10.0	10	0.11	13	No	Yes
Distracting oneself from the activity or temptation	Attention	8.5	11	−0.25	22	Yes	Yes	Yes
Anticipating self-reward	Cognitive	7.9	12	0.26	6	No	Yes
Reasoning in favor of target activity or against the temptation	Cognitive	7.9	13	−0.22	21	No	Yes
Self-affirmation	Cognitive	7.6	14	0.16	9	Yes	Yes	No
Taking a substance	Situation	7.3	15	−0.06	17	No	Yes
Changing the environment	Situation	7.3	16	0.09	14	Yes	Yes	Yes
Seeking social support	Situation	7.0	17	0.11	12	No	Yes
Changing the activity itself	Situation	5.8	18	−0.07	18	No	No
Self-talk	Cognitive	5.3	19	−0.02	15	No	Yes
Removing or reducing distractions or temptations	Situation	5.2	20	0.14	10	No	Yes
Adopting a process focus	Attention	5.0	21	0.29	2	No	Yes
Reappraisal	Cognitive	2.0	22	0.18	8	No	No

Note. Freq = frequency; situation = situation selection or modification strategies; attention = attentional deployment strategies; cognitive = cognitive change strategies; response = response modulation; suppression = suppressing the impulse to quit or to give on to the temptation/forcing oneself to initiate the aversive activity.

Regarding general strategy efficacy, we found that “focusing on positive consequences,” “suppression,” “thinking of the near finish,” and “task enrichment” were, according to participants’ reports, the four most adaptive strategies in our sample. Table 1 shows that these four strategies were not only experienced as highly effective but that they were also highly popular, in that they were selected most, second most, fifth most often, and sixth most often, respectively. To test the hypothesis that strategies that were experienced as more effective were also selected more often, we computed a meta regression where we predicted the general likelihood of selecting a particular self-regulatory strategy (mean in Figure 3, i.e., the values in the choice and efficacy columns in Table 1, respectively) by the general subjective efficacy coefficient of each strategy (mean in Figure 2). This yielded a positive and significant coefficient, b = 0.17, SE = 0.07, p = .018, 95% CI [0.03, 0.30], r = .39, indicating that strategies that were perceived as more effective were selected more likely by the participants (and vice versa).

Moderation: Conflict Type

Figure 3 also shows how the selection of self-regulatory strategies was moderated by conflict type. Here, we focus on the five self-regulatory strategies, whose subjective efficacy was significantly moderated by conflict type. The full results can be found in the online on the OSF project page.

Regarding “changing the environment,” we found that this strategy was only perceived as effective in regulating inhibition conflicts (Figure 2). Figure 3 shows that this pattern is reflected in participants’ choice as well, as participants were more likely to select this strategy when confronted with inhibition than with initiation or persistence conflicts.

Similarly, “distracting oneself from the activity or temptation” was not only perceived as more effective during conflicts of inhibition compared to conflicts of initiation or persistence, but participants also selected it more often, as evidenced in Figure 3. In a similar vein, “suppression” was significantly less selected for conflicts of persistence than for other conflicts (Figure 3), which was also the only conflict type that “suppression” was not effective for in regulating them (Figure 2).

Regarding “self-affirmation,” we found that it was subjectively more effective for regulating conflicts of initiation and inhibition but not of persistence. However, Figure 3 shows that participants were only more likely choose this strategy, on average, when they were confronted with conflicts of initiation but not of inhibition compared to persistence conflicts. A similar disconnect was found for “task enrichment,” which was selected more often for conflicts of persistence compared to initiation (Figure 3), although Figure 2 shows that it was only effective for conflicts of initiation but not persistence.

Taken together, we found partial evidence for the notion that participants selectively choose strategies for the types of conflicts they also perceive them to be more effective for, particularly for initiation conflicts but to a lesser extent for inhibition conflicts. However, as Figure 3 indicates, these results were often heterogenous, indicating the presence of moderators.

Strategy-Situation Fit and Regulatory Flexibility

Finally, we examined whether individuals who favored strategies that were reported to be more effective for specific conflict types reported greater self-regulatory success in general, constituting our operationalization of strategy-situation fit or self-regulatory flexibility. We focused on the five strategies where we found conflict type differences in their efficacy. Given that Mplus does not have an option for applying error correction, we set the alpha = .025 for the following analyses given that there were two comparisons of interest per strategy.

For “changing the environment,” we found evidence for strategy-situation fit: Not only was this strategy experienced as more effective in regulating inhibition conflicts, but participants who selected this strategy more often when confronted with inhibition conflicts reported higher self-regulatory success on average than participants who chose it less often, but only compared to conflicts of persistence, b = 2.10, SE = 1.26, p = .021, 95% CI [0.09, 5.04], but not of initiation, b = 1.07, SE = 1.17, p = .144, 95% CI [−0.92, 3.75]. A similar pattern was found for “distracting from the activity or temptation.” Participants who selected this strategy more likely when confronted with conflicts of inhibition also reported higher subjective self-regulatory success compared to when confronted with conflicts of initiation, b = 1.52, SE = 0.84, p = .022, 95% CI [0.05, 3.32] or of persistence, b = 1.69, SE = 1.03, p = .029, 95% CI [−0.07, 4.09], although the latter coefficient was not significant when applying Šidák correction. These results demonstrate the importance of strategy-situation fit of these two strategies.

However, we did not find evidence for strategy-situation fit for the other three strategies “task enrichment,” “self-affirmation,” and “suppression,” as the person-specific slopes of the relationship between strategy use and conflict type were not significantly associated with subjective self-regulatory success on the between-person level. Although “task enrichment” was more effective for conflicts of initiation than of persistence, participants who selected “task enrichment” more likely when confronted with conflicts of initiation compared to persistence did not report significantly higher levels of subjective self-regulatory success, b = 0.81, SE = 0.52, p = .055, 95% CI [−0.20, 1.83]. Moreover, although “self-affirmation” was less effective for persistence conflicts but selecting it more likely in these situations was not significantly associated with higher self-regulatory success in general, b = 1.04, SE = 1.48, p = .220, 95% CI [−1.72, 4.16] and b = 3.54, SE = 3.45, p = .077, 95% CI [−1.56, 12.11], for initiation and inhibition conflicts, respectively. The same pattern was found for “suppression,” with b = −0.12, SE = 0.52, p = .411, 95% CI [−1.12, 0.91] and b = 2.77, SE = 4.07, p = .185, 95% CI [−3.90, 12.43], for initiation and inhibition compared to persistence conflicts. Thus, endorsing these three strategies in situations, in which they demonstrated comparatively higher self-regulatory success, was not significantly associated with higher levels of self-regulatory success in general.

Individual Differences in Self-Regulatory Success in Three Types of Self-Control Conflicts

In our study, person differences alone predicted 11.6% of variance in subjective self-regulatory success during self-control conflicts, attesting to important individual differences. In addition, person × conflict type interactions explained an additional 34.3% of variance in subjective self-regulatory success. This means that to understand and predict when people will succeed in managing their self-control conflicts, it is important to take individual differences but also differences between conflict types into account: Whereas some people appear to be better at regulating themselves when confronted with one kind of self-control conflict, other people appear to be better at regulating themselves when confronted with another type of self-control conflict.

This finding is line with an attempt to develop a scale that assesses individual differences in global self-control separately for the three types of self-control conflicts (Hoyle & Davisson, 2016). Even though the authors report relatively strong correlations between the three subscales, they also report separation between them in terms of association with other personality traits. Neuroticism, for example, is more strongly associated with self-control by inhibition than with self-control by initiation. This corresponds to possible predictions based on cybernetic Big Five theory which distinguishes between two independent meta-traits that can reflect covariation between Big Five domains on a superordinate level (DeYoung, 2006; 2010). Stability, the first meta-trait, captures the covariation between (low) neuroticism, agreeableness, and conscientiousness and reflects a tendency towards resisting temptations that would disturb current goal pursuit. Plasticity, the second meta-trait, captures the covariation between openness and extraversion and reflects a tendency to pursue new goals and be flexible with regard to entertaining new perspectives on current goals and strategies. Given conceptual overlaps, we would expect that the stability-related traits (low) neuroticism, agreeableness, and conscientiousness correlate more strongly with self-control during conflicts of inhibition whereas the flexibility-related traits extraversion and openness correlate more strongly with self-control during conflicts of initiation (see Hoyle & Davisson, 2016, for the same argument).

The idea that people’s abilities to resolve the three types of conflict are relatively independent of each other, is also in line with the view that distinct aversive and appetitive motivational systems underlie affective and behavioral response tendencies and dimensions of personality (Carver, Sutton, & Scheier, 2000; Carver & White, 1994; Davidson, 1998; Gable, Reis, & Elliot, 2000; Gray, 1994; Harmon-Jones & Allen, 1997; Lucas, Diener, Grob, Suh, & Shao, 2000; Tellegen, 1985). On the one hand, the behavioral activation (or sometimes approach or facilitation) system (BAS) is responsible for approach and (most types of) positive affect (Carver & White, 1994; Fowles, 1993; Gray, 1994). It is sensitive to rewards and activity in this system causes a person to begin or increase movement towards goals (Carver & White, 1994). On the other hand, the behavioral inhibition system (or sometimes withdrawal) system (BIS) is responsible for inhibition in response to threats and for withdrawal behavior and anxiety (Carver & White, 1994; Davidson, 1992; Fowles, 1993; Gray, 1972, 1994). It is sensitive to punishment and it inhibits behavior that may lead to painful outcomes. Support for the relative independence of these two systems comes from research that has shown their independent neural bases (Gray, 1987a, 1987b; Quay, 1993; Sutton & Davidson, 1997). Moreover, personality differences in healthy adults as well as psychopathologies in clinical populations can be classified according to whether they are accounted for by BAS and BIS sensitivity (e.g., Fowles, 1993; Quay, 1988, 1993). Importantly, the distinction between the BAS and the BIS corresponds with the distinction of initiation/persistence and inhibition: Whereas the BAS should be mostly relevant for initiation and persistence of goal-directed behavior, the BIS should be more relevant to the inhibition of goal-inconsistent behavior. In our data, we also found that self-regulatory success in, on the one hand, initiation and persistence conflicts and, on the other hand, inhibition conflicts were relatively independent, as evidenced by a small association of r = .12. Thus, individuals differ with regard to how well they can control themselves during different types of conflict (Hoyle & Davisson, 2016).

Strategy Choice

Generally, we find that people deploy a large variety of strategies across these three types of self-control conflicts. Our data furthermore showed large differences in the popularity of different self-regulatory strategies. Some strategies were chosen rarely (e.g., “reappraisal”), others frequently (e.g., “focusing on the positive consequences”). When looking at people’s use of the different self-regulatory strategies, the relative contribution of the three sources of variance—person, conflict type, person × conflict type—also varied greatly. This implies that the extent to which each source of variance contributed to the prediction of strategy use, depends on which strategy is considered. If the contribution of person variance component is large, it implies that differences between people, for example, personal preferences, play a large role in predicting whether a strategy is used or not: Some people use a given strategy frequently, others rarely. This is the case, for example, for “focusing on negative consequences” (25.6% variance explained), “taking a substance” (19.8%), and “goal setting” (20.3%). In these examples, but also more generally, the portion of variance that could be attributed to person differences was quite large for strategy choice. It would be interesting to investigate further which person characteristics can explain this large amount of variance between people. One idea is that people take different approaches to goal pursuit and that this is reflected in their strategy choice. On the one hand, there may be the self-controlled way, which is goal-driven and focused and potentially indicated by larger scores on traits like self-control or the conscientiousness facet self-discipline. These persons may prefer strategies that could be described as goal-driven and focused, for example, strategies like “goal setting,” “monitoring progress,” “planning/scheduling,” “putting oneself under pressure,” “reminding themselves of a commitment,” or “suppression.” Others may take a “lighter” approach to goal pursuit, for example, by trying to make it more pleasant or even easier through “seeking social support,” “task enrichment,” or “anticipating self-reward.” To examine this, it might be worthwhile to correlate personality traits like conscientiousness or trait self-control but also individual differences in the quality of motivation (e.g., in controlled versus. autonomous motivation, e.g., Deci & Ryan, 1985; Sheldon & Elliot, 1999) with strategy use to account for the large individual differences between people. It may furthermore be interesting to take a person-centered approach to identify distinct profiles of people who prefer certain groups of strategies over others and link these profiles to these and other individual traits.

Regarding the variance component attributable to the type of self-control conflict only the strategy “distracting oneself from the temptation/activity” stood out, with 10.3% of variance explained by conflict type alone. This implies that whether this strategy was used depended largely on the type of conflict that was experiences. As another analysis showed “distraction” was used most often for conflicts of inhibition.

Lastly, if the contribution of the interaction of person × conflict type variance component is large, as is the case for “changing the environment” (12.2%), “focusing on the positive consequences” (11.4%), and “distraction from the activity/temptation” (11.2%), it implies that whether the strategy is used depends on both the person who is experiencing the self-control conflict and the conflict type in interaction. Or in other words: In these cases, some people prefer using it for one type of conflict, other people prefer using it for other types of conflict. This source of variance is highly interesting, in our perspective, given that our data also show that the subjective efficacy of some strategies—including “changing the environment”—is moderated by conflict type. In a situation where a strategy is clearly more effective for one type of conflict, but people vary in the types of conflict they prefer to use it for, it would make sense to design interventions that teach people who use the strategy for the “wrong” type of conflict to rather use if for the “right” type of conflict. For example, people who use “changing the environment” for conflicts of initiation and persistence, where it does not go along with higher subjective self-regulatory success, could benefit from learning that it is, in fact, only effective for conflicts of inhibition.

Strategy Efficacy

In terms of the subjective efficacy of other strategies, our analyses of main effects showed that 8 (out of 22) strategies were generally, that is, across the three conflict types, experienced as adaptive and three strategies as maladaptive. Considering conflict type as a moderator, however, revealed more nuance in the relations between strategy use and subjective self-regulatory success. In fact, some strategies, were only experienced as adaptive for certain types of self-control conflicts: “Changing the environment” was only reported as effective for conflicts of inhibition. “Task enrichment” was only reported as effective for conflicts of initiation. “Distracting oneself from the temptation/activity” had no significant effect during conflicts of inhibition and was, in fact, subjectively maladaptive if deployed during disliked activities participants wanted to initiate or persist in. This finding confirms our hypothesis insofar as we had predicted that the subjective effectiveness of distraction would be moderated by conflict type and that it would be negative for conflicts of initiation and persistence (see Hennecke et al., 2019). However, we had also predicted that, in line with previous findings (Mischel et al., 1972), distraction should be perceived as effective during conflicts of inhibition and this part of the hypothesis was not confirmed. Finally, “self-affirmation” and “suppression” had no perceivable beneficial effects during conflicts of persistence but were subjectively effective for conflicts of initiation and conflicts of inhibition.

Together, these results highlight the importance of not repeating, in self-control research, the “fallacy of uniform efficacy” which was previously committed in research on emotion regulation (Bonanno & Burton, 2013) and shows that generalizing results about strategy efficacy from studies looking at only one type of self-control (e.g., persistence conflicts: Hennecke et al., 2019; inhibition conflicts: Friese & Hofmann, 2016; Lopez et al., 2021; Milyavskaya et al., 2021; Williamson & Wilkowski, 2020) to other self-control conflicts or self-control more generally may lead to wrong conclusions. More generally, these results highlight the complexity of self-regulation in the real world.

Which type of self-control conflict is experienced might only be one example for a situational variable that moderates subjective strategy efficacy. As argued above and elsewhere (Hennecke & Bürgler, 2020), additional variables that might interact with each other might complicate deriving clear conclusions about strategy efficacy: Individual trait and state differences (e.g., in self-control, working memory capacity, cognitive load), differences in goals (e.g., between approach and avoidance goals), or other demands of the self-control conflicts (e.g., how difficult it is to resolve, whether a disliked activity is boring, mentally effortful, or physically effortful) might all contribute to the momentary efficacy of a given strategy. However, given the large number of strategies and situational factors, finding ways to examine the complex interplay between strategies and situational factors represents an important challenge for future research.

The Association of Strategy Choice and Efficacy

The relatively high correlation between the subjective efficacy of strategies and their popularity shows that subjectively effective strategies were also, in general, used more frequently. Given that a correlation does not inform us about the causal direction of this association, we can only speculate about the underlying processes. First, people may choose the strategies they perceive to be more effective more often. This might reflect both inaccurate as well as accurate metacognitive knowledge about the efficacy of self-regulatory strategies for given self-control conflicts (Bürgler et al., 2021). Second, it may be the case that strategies that are more frequently used become more successful, for example, through practice. Third, a third variable, for example, a strategy’s ease of implementation, may affect both its efficacy and its popularity. The current research cannot tell us which of these explanations holds true.

Despite the overall positive correlation between the strategies’ popularity and their subjective efficacy, there are nevertheless a couple of strategies whose popularity diverges from their subjective efficacy. The strategy “focusing on negative consequences,” was popular overall (frequency rank 4) but also relatively ineffective overall (efficacy rank 16). “Self-affirmation” and “task enrichment” were most popular for the conflict types that they were not most effective for. In contrast, the strategy “adopting a process focus” was relatively effective (efficacy rank 8) but unpopular (frequency rank 21). These divergences might reflect misbeliefs about certain strategies. It could be a fruitful endeavor for applied research to correct such misbeliefs and inform people about strategies that are similarly applicable but more effective (e.g., “focusing on the positive consequences” instead). It might also be the case that some effective strategies are used rarely because they can only be implemented under certain conditions. “Adopting a process focus” might, for example, be a strategy that with its attention to how an activity is performed imposes cognitive load and is therefore not applicable when people need to focus on other aspects of the task at hand. The attention of a person studying for an exam might, for example, already be fully absorbed by the contents that need to be understood and memorized, leaving no room for simultaneously focusing on the process of studying itself. During other types of self-control conflicts, for example, when mental load is low, for example, during physical exercise, the strategy might be applicable, nevertheless.

Strategy-Situation Fit and Individual Differences in Regulatory Flexibility

Our analyses focusing on the moderation of strategy efficacy by conflict type have supported the notion that some strategies are, at least in participants’ experience, more effective for some types of conflict than for others. What these analyses cannot show is whether individuals who use a particular strategy flexibly for the “right” types of conflicts actually benefit from this flexible usage, that is, by experiencing greater self-regulatory success overall. This ability to create strategy-situation-fit, should, however should be an important aspect of individual differences in regulatory flexibility, and in turn, of individual differences in people’s general self-regulatory competence. A similar idea was already proposed for emotion regulation, where three aspects of individual differences in regulatory flexibility were proposed: (1) context sensitivity, (2) strategy repertoire, and (3) feedback monitoring (Bonanno & Burton, 2013). In the self-control domain, there is, as of yet, only evidence that people benefit from a larger repertoire of self-regulatory strategies (Bürgler et al., 2021), from monitoring their use, the situation, and the efficacy of self-regulatory strategies for self-control (Bürgler et al., 2021), and from being more variable in their use of self-regulatory strategies (Wenzel et al., 2021). Evidence for the importance of context sensitivity, which has been defined as “the ability to perceive impinging demands and opportunities from the situation context […] and to determine the most appropriate regulatory strategy in response to those demands and opportunities” (Bonanno & Burton, 2013) was still lacking.

We believe that the current results provide preliminary evidence that this aspect of regulatory flexibility, the ability to select strategies that fit a given situation can, indeed, be an important component of individuals’ self-control: People who used the strategies “changing the environment” and “distraction” more often for conflicts of inhibition conflicts, for which they were found to be more effective than for the other conflict types (for which they were either maladaptive or neither adaptive nor maladaptive), reported higher self-regulatory success overall. These results suggest that regulatory flexibility may be an important aspect of individual differences in self-regulatory competence.

From a theoretical perspective, it would also be desirable to investigate associations of global individual differences in self-control and regulatory flexibility. So far, research has yielded unequivocal results: Feedback monitoring, on the one hand, has been found to correlate positively with trait self-control, implying that it may be one of the processes through which trait self-control “gets outside the skin” (Bürgler et al., 2021). Strategy repertoire, on the other hand, has been found to not be correlated with trait self-control, implying that strategy use may represent a different route to self-regulatory success independent of trait self-control (Bürgler et al., 2021). Given this lack of an association between trait self-control and strategy repertoire, strategy-situation fit may also not be correlated with global individual differences in self-control, given that a large repertoire should be a precondition for people’s ability to create strategy-situation fit (Bonanno & Burton, 2013).

Limitations and Future Directions

The two studies presented here represent a good start for studying the complexities of self-regulatory strategy use in daily life. The Pilot Study relied, with the DRM, on a bottom-up approach and was able to produce a list of self-regulatory strategies on which various coders could agree and that likely reflects the way lay people act when confronted with self-control conflicts in their daily lives. With its focus on within-person effects, this study took the notion seriously that hypotheses about intraindividual processes should be investigated as such, that is, in within-person analyses (Molenaar & Campbell, 2009). Despite these methodological advantages of the current work, both studies are not without limitations.

The potentially largest limitation of this work might have resulted directly from our ambition to capture the complexity of self-regulation “in the wild.” Even with our sample of 14,000 reported daily self-control conflicts, the evidence for the effectiveness of some strategies for some types of self-control conflicts should be considered preliminary. This is the case because some of the 22 different strategies were reported rarely for some type of conflicts. In turn, their efficacy is estimated based on relatively few observations. At the same time, we wonder what feasible replication studies could look like, given that larger samples may be difficult to attain for these types of studies that are burdensome for participants and therefore expensive for researchers who need to provide appropriate reimbursement. We do not think that, to reduce complexity, simply reducing the number of different strategies participants can choose from is advisable. Assessments with fewer strategies to choose from would either have to lump together different strategies that do not reliably correlate with each other and that do not perform equally well (e.g., “focusing on the positive consequences” and “focusing on the negative consequences” are both forms of reappraisal but only “focusing on the positive consequences” appears beneficial for self-control, see Hennecke et al., 2019; Wenzel et al., 2022) or refrain from assessing the full range of strategies that participants actually use “in the wild.” In our perspective, both options would not do the complexity of self-regulation in everyday life justice. And this complexity is probably even larger than our study suggests. In fact, self-control conflicts vary on many more dimensions than the distinction between initiation, persistence, and inhibition. Initiation and persistence conflicts, for example, may refer to activities with very different aversive demands, including boring activities, physically effortful activities, or cognitively effortful activities (Hennecke et al., 2019). Further complicating the issue, many of the strategies investigated can be implemented in different ways. For example, task enrichment during an aversive task can be achieved through music, TV, phone talk, eating, drinking, etc. It is possible that certain implementations of the same strategy are generally more effective than others, or it is possible that certain implementations can be used across different types of conflicts (or, more generally, situations) with similar success, while others, such as watching TV, might be more limited in where they can be implemented effectively. Even though this study already provides a rather detailed account of the different self-regulatory strategies people can use, it is possible that future studies would benefit from an even more nuanced differentiation of the self-regulatory strategies or “tactics” through which strategies can be implemented (McRae et al., 2012). We hope that future research does not shy away from investigating such complex strategy-by-situation-interactions even though progress may be incremental, and findings may require cross-validation.

In addition, both studies rely on self-reports which come with a couple of caveats. While the DRM has good ecological validity for assessing people’s activities and experiences in other areas of research (Kahneman et al., 2004), its validity for studying processes of self-control is unknown. We nevertheless hope that given its heavy reliance on internally produced structural reminders of the previous day, participants’ memories are more accurate than when asked to report their self-regulatory strategies more globally. Likewise, whereas experience sampling is by now a very common method for studying self-control (e.g., Converse et al., 2019; Hennecke et al., 2019; Hofmann et al., 2012; Milyavskaya et al., 2021; Williamson & Wilkowski, 2020), its validity for assessing self-control in daily life is not entirely clear either. To date, there has been no cross-validation of results from experience sampling through direct observation. In the case of our own assessments, this means that ultimately, we do not know for sure whether people’s self-reported strategy use and self-regulatory success would converge with objective assessments. One danger is that people who believe that a given strategy is successful, even if this is a misbelief, may report that they regulated themselves more successfully, even if they did not. A desire for consistency may cause such biased responding (Festinger, 1957).

Nevertheless, there are good reasons why we used these self-report measures. First, many strategies are intrapsychic strategies, which cannot easily be observed from the outside we studied, among them the five most popular strategies “focusing on positive consequences,” “suppression,” “reminding oneself of a commitment to oneself or to others,” “focusing on negative consequences,” and “thinking of the near finish.” Second, we were determined to study a wide variety of self-control conflicts, including the three types but also, within these types, conflicts that entailed various temptations and activities. Assessing self-regulatory success across a great variety of different self-control conflicts and many participants does not seem feasible with more objective measures. This is not to say that future studies should not extend the current work by, for example, zooming in on specific strategies that can be observed directly, for example, “task enrichment,” and in combination with objective indicators of success.

Another limitation of our assessment lies in the fact that we assessed strategy use with single items rather than more comprehensive scales in the main study. Even if single-item measures are not necessarily worse than multiple-item measures (Bergkvist & Rossitr, 2007) when constructs are quite narrow (Sackett & Larson, 1990), we cannot be sure that we have assessed strategy use with high reliability and with each item fully representing its underlying construct. Single-item measures are, however, very common in experience sampling studies, where otherwise the participant burden would be too high (Friese & Hofmann, 2016; Hofmann et al., 2012; Milyavskaya et al., 2021).

Both methods, the DRM and experience sampling, are furthermore restricted to assessing conscious experiences, here: the conscious deployment of strategies during consciously experienced self-control conflicts. Generally, people may not always be able to accurately report about their intrapsychic processes and it is not clear whether strategy deployment is an exception (Nisbett & Wilson, 1977). Accordingly, both studies did not assess and evaluate strategic attempts that work unconsciously, without being noticed and remembered by participants. Such unconscious regulation has been shown to be effective in helping people to self-control their behavior (Fishbach et al., 2003) but cannot be captured with our method. It is therefore restricted to the assessment of conscious self-regulation.

Finally, we only asked participants to name or choose strategies they used during self-control conflicts that they experienced. Hence, our method did not capture strategies that people use to select of modify situations in a way that prevents self-control conflicts from occurring in the first place (Duckworth et al., 2016; Ent et al., 2015). Some of the strategies we investigated were furthermore used rarely, at least for some type of conflicts, thereby reducing the power for our analyses. This is reflected in the sometimes relatively large confidence intervals around the effect sizes in Figure 2. It would require an even larger sample or a longer period of data collection to gather more data on the effectiveness of rarely used strategies.