In this Spotlight feature we focus on Marina Milyavskaya, PhD, Assistant Professor of Psychology at Carleton University. Dr. Milyavskaya’s research focuses on factors that predict successful goal attainment. These factors include trait differences between people as well as situational differences that influence whether individuals succeed or fail to attain goals in particular contexts.
Q: The mission of SOBC is largely tied to research related to self-control. In your recent paper in Motivation Science titled “The many faces of self-control: Tacit assumptions and recommendations to deal with them,” you presented a compelling argument that several implicit assumptions have guided this research for years that now need to be addressed explicitly to aid scientific progress (Milyavskaya, Berkman, & De Ridder, in press). In your view, what are the main perils of proceeding with research that leaves these assumptions tacit and untested?
Broadly speaking, maintaining these assumptions can prevent the field from developing in a coherent way and remaining relevant, as it increasingly can focus on a narrow set of questions constrained by these assumptions. More specifically, different assumptions come with different costs. For example, the definitional assumptions result in researchers frequently talking past each other. The literature on self-regulation and self-control is growing at a rapid pace, and if we use different terminologies it makes it very difficult to keep up with what’s being published in the field, and also to have a common conversation that allows us (as researchers) to agree on a concrete set of findings or recommendations. Another assumption, that self-control is a capacity, has led a lot of research efforts to be spent on examining how to improve this capacity, disregarding alternative potentially more promising avenues, such as focusing on motivation. Explicitly testing these assumptions and going beyond them can help develop and test novel questions, and hopefully uncover new approaches to improving self-regulation.
Q: One of the assumptions in much self-control research is that successful self-control is necessarily effortful. That assumption may be particularly difficult to challenge because of the general understanding in people’s everyday lives that self-control often entails an increase in mental effort in order to resist a tempting choice, such as choosing an unhealthy cheeseburger in the moment that the waitperson stops by to take orders. Nevertheless, there is evidence that effort may not be as closely intertwined with self-control success as people assume, including some of your own work showing that goal progress may be better predicted by perceived ease than by effort. If self-control can indeed be applied successfully in a way that does not always involve substantial effort, are there implications for developing effective low-effort interventions to change health behaviors?
Most recent research indeed suggests that most effective self-control is effortless (or at least less effortful than active resistance). For example, people who are considered to have ‘high trait self-control’ are not better at effortfully resisting temptations, but experience less frequent and less strong temptations. Effortless self-control is also automatic, and this automaticity can arise in numerous ways, including, for example, appraising the ‘temptation’ as less desirable or incompatible with broader goals. The ultimate effortless self-control is having good habits, as habits unfold automatically and effortlessly. This certainly has many implications for interventions. First, we need to figure out what can help automatize healthy behaviours. For example, research on implementation intentions has thoroughly examined this in the lab, but needs to be better translated into actual interventions (although some successful interventions do incorporate it). Additionally, changing the environment so that there are fewer temptations, changing appraisals to focus more on the benefits of the healthy options, shifting motivation, and enlisting the help of supportive others are all possible intervention targets that could in theory facilitate self-control without requiring effort. Much more research is needed on the best way to implement some of these ideas and especially to scale them in such a way that they are more broadly accessible. In some ways, the research on nudging tries to do just that, and may be a promising avenue.
Q: Two other assumptions that you highlight correspond to what have been termed the “jingle” and “jangle” fallacies, which are, respectively, the notions that the same term may be erroneously applied to different constructs and that different terms may be applied to largely overlapping constructs. The jingle assumption is especially important to address for SOBC research because its method depends on identifying potential target mechanisms that are sufficiently consistent to be compared across different studies. This consistency is necessary to draw conclusions about which particular mechanisms are linked to behavior change and which interventions influence those particular mechanisms. As you point out, this jingle assumption pertains to measures of state and trait versions self-control. For example, a measure of inhibition of a prepotent tendency in the lab (“state self-control”) may be quite unrelated to a temperament-level measure of impulsivity (“trait self-control”). How might researchers effectively disentangle state and trait self-control constructs to prevent erroneously considering them as isomorphic, and what refinements in terminology might be helpful in this regard?
I think there is increased recognition that lab measures and trait self-report measures can be quite distinct, as I have seen a few papers recently (in addition to my own) that have pointed to this issue. More research and theory is needed to understand exactly how they are related or dissociable, including the mechanisms by which traits can lead to better outcomes – presumably it’s by improving in-the-moment (state) self-control, but that is far from clear. So in many ways the research agenda of the SOBC in identifying target mechanisms is very much aligned with what the field needs to disentangle these constructs. In the meantime, researchers who discuss ‘self-control’ can be more specific in qualifying what they mean (e.g., using ‘trait self-control’, ‘effortful self-control’, or ‘effortless self-control’), to identify the precise aspect that they are studying. At the same time, the jangle fallacy risks cropping up even more if people are using drastically different terms. It’s certainly a tough balance to manage, and an issue that does not have a simple solution.
Q: Some research on self-control seems to make the assumption that one’s ability to achieve successful behavioral outcomes depends exclusively on one’s capacity for self-control. In contrast, recent research suggests that contextual factors also matter independently from self-control capacity. In particular, research shows that self-control is more likely to be engaged in some contexts relative to others, and different kinds of motivation are associated with different behavioral outcomes. How might the process of intervention development benefit from manipulating context, and possibly motivation, rather than capacity?
Changing capacity indeed does not seem to work. There are now numerous meta-analyses coming to this conclusion. So yes, changing to focus of interventions to change people’s contexts and help them improve or change motivation is critical for success. In my own research on goal pursuit I find that 85-90% of the variation in goal progress is among a person’s goals, and only 10-15% is across people. That is, it’s not the case that some people are better at self-regulation, but some goals are more likely to be attained than others, and this is likely influenced by context. Some research has examined the possibility of changing contexts by removing temptations and other obstacle in the environments, enlisting the help of supportive others, and setting concrete plans (implementation intentions); these have all been shown in the literature to be effective in improving self-regulatory outcomes (i.e., goal attainment). Additionally, motivation is consistently linked with better self-regulation, with some recent work suggesting that people with high trait self-control are more successful because they are more likely to pursue personally important and interesting goals (rather than pursuing goals to please others or because of guilt); such want-to goals are in turn more likely to be attained. Indeed, some of the newer models of self-control (including a really interesting one by Elliot Berkman and colleagues) proposes that self-control is a choice based on numerous inputs, including a person’s values and interests; by shifting the focus on those (i.e., why is my health important to me), we can help people make better self-regulatory choices.