Does self-control improve with practice? Evidence from a six-week training program.

TLDR: There was no effect of training on any measure of self-control and the implication is that training self- control through repeated practice does not result in generalized improvements inSelf-control.

Abstract: Can self-control be improved through practice? Several studies have found that repeated practice of tasks involving self-control improves performance on other tasks relevant to self-control. However, in many of these studies, improvements after training could be attributable to methodological factors (e.g., passive control conditions). Moreover, the extent to which the effects of training transfer to real-life settings is not yet clear. In the present research, participants (N = 174) completed a 6-week training program of either cognitive or behavioral self-control tasks. We then tested the effects of practice on a range of measures of self-control, including lab-based and real-world tasks. Training was compared with both active and no-contact control conditions. Despite high levels of adherence to the training tasks, there was no effect of training on any measure of self-control. Trained participants did not, for example, show reduced ego depletion effects, become better at overcoming their habits, or report exerting more self-control in everyday life. Moderation analyses found no evidence that training was effective only among particular groups of participants. Bayesian analyses suggested that the data were more consistent with a null effect of training on self-control than with previous estimates of the effect of practice. The implication is that training self-control through repeated practice does not result in generalized improvements in self-control. (PsycINFO Database Record

Figures

Table 2 Key Outcome Measures

Figure 3 Structure of the Study.

Figure 2 Types and Measures of Training Transfer.

Table 1 Characteristics of Published Studies on Self-Control Training

Figure 1 Flow of Participants Through the Study.

Full text

This is a repository copy of Does self-control improve with practice? Evidence from a
six-week training program.
White Rose Research Online URL for this paper:
http://eprints.whiterose.ac.uk/107549/
Version: Accepted Version
Article:
Miles, E., Sheeran, P., Baird, H. et al. (3 more authors) (2016) Does self-control improve
with practice? Evidence from a six-week training program. Journal of Experimental
Psychology, 145 (8). pp. 1075-1091. ISSN 0096-3445
https://doi.org/10.1037/xge0000185
This article may not exactly replicate the final version published in the APA journal. It is not
the copy of record.
eprints@whiterose.ac.uk
https://eprints.whiterose.ac.uk/
Reuse
Unless indicated otherwise, fulltext items are protected by copyright with all rights reserved. The copyright
exception in section 29 of the Copyright, Designs and Patents Act 1988 allows the making of a single copy
solely for the purpose of non-commercial research or private study within the limits of fair dealing. The
publisher or other rights-holder may allow further reproduction and re-use of this version - refer to the White
Rose Research Online record for this item. Where records identify the publisher as the copyright holder,
users can verify any specific terms of use on the publisher’s website.
Takedown
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by
emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request.

1
Running head: EFFECTS OF SELF-CONTROL TRAINING
Does Self-Control Improve With Practice? Evidence from a 6-Week Training Program
Eleanor Miles, Paschal Sheeran, Harriet Baird, Ian Macdonald, Thomas L. Webb, and Peter R.
Harris
Word count: 11,056 (excluding references and tables)
Acknowledgements
This project was made possible through the support of a grant from the John Templeton
Foundation. The opinions expressed here are those of the authors and do not necessarily reflect
the views of the John Templeton Foundation. We thank Zoltan Dienes for guidance on Bayesian
statistics.

2
Abstract
Can self-control be improved through practice? Several studies have found that repeated
practice of tasks involving self-control improves performance on other tasks relevant to self-
control. However, in many of these studies, improvements after training could be attributable to
methodological factors (e.g., passive control conditions). Moreover, the extent to which the
effects of training transfer to real-life settings is not yet clear. In the present research, participants
(N = 174) completed a 6-week training program of either cognitive or behavioral self-control
tasks. We then tested the effects of practice on a range of measures of self-control, including lab-
based and real-world tasks. Training was compared to both active and no-contact control
conditions. Despite high levels of adherence to the training tasks, there was no effect of training
on any measure of self-control. Trained participants did not, for example, show reduced ego
depletion effects, become better at overcoming their habits, or report exerting more self-control
in everyday life. Moderation analyses found no evidence that training was effective only among
particular groups of participants. Bayesian analyses suggested that the data was more consistent
with a null effect of training on self-control than with previous estimates of the effect of practice.
The implication is that training self-control through repeated practice does not result in
generalized improvements in self-control.
Keywords: self-control, self-regulation, intervention, ego depletion, self-control training

3
Does Self-Control Improve With Practice? Evidence from a 6-Week Training Program
Self-control, or the ability to control thoughts, behaviors, and feelings, seems to be
important for success in most areas of life (De Ridder, Lensvelt-Mulders, Finkenauer, Stok, &
Baumeister, 2012). During the past 15 years, much of the research into self-control has been
inspired by the strength model, which draws the analogy between self-control and a physical
muscle (for a review, see Muraven & Baumeister, 2000). This model proposes that, just as using
a muscle leads to temporary fatigue, exerting self-control leads to temporary reductions in self-
control performance; a phenomenon that has been termed ‘ego depletion’ (Baumeister,
Bratslavsky, Muraven, & Tice, 1998). However, the strength model also suggests that if self-
control is repeatedly exerted over time (interspersed with periods of rest), then the opposite
effect should occur. In other words, just as a muscle grows stronger with exercise, so self-control
should improve over time with practice (Muraven, 2010a; Muraven, Baumeister, & Tice, 1999).
The ego depletion effect has been the subject of hundreds of empirical tests, extensive
analysis of mediating and moderating factors, and much lively theoretical debate (e.g., Hagger,
Wood, Stiff, & Chatzisarantis, 2010; Inzlicht & Schmeichel, 2012; Carter, Kofler, Forster, &
McCullough, 2015). However, the strength model’s predictions about the long-term effects of
exerting self-control have received less attention. If practicing self-control improves subsequent
self-control, as has been suggested by several studies (e.g., Muraven, 2010a; Muraven et al.,
1999), then self-control training could benefit people facing everyday self-regulatory struggles
such as controlling emotions, breaking bad habits, and overcoming impulses. Yet, there are many
things that we do not yet know about the effects of training self-control through practice (see
Berkman, in press; Inzlicht & Berkman, 2015; Inzlicht, Schmeichel, & Macrae, 2014, for
discussion). For example, how reliable are training effects? Can we be sure that the observed

4
improvements are the result of practicing self-control? Perhaps most importantly from an applied
perspective, does training self-control indeed influence real-life outcomes that depend on self-
control? The present research sought to address these unanswered questions.
What Do We Already Know About the Effectiveness of Self-Control Training?
Table 1 provides an overview of prior studies of self-control training. Studies
investigating the effect of self-control training typically ask one group of participants to perform
a task requiring self-control over a period of weeks, while another group performs either no task,
or a task that does not require self-control. The performance of the two groups is then compared
on a subsequent self-control task, usually different to the one that was trained. A wide range of
tasks has been used to train self-control, such as using one's non-dominant hand, developing and
executing a personalized study or exercise program, completing the Stroop task, squeezing a
handgrip, or performing a logical reasoning task. The effect of training has also been assessed
using a wide range of tasks, such as tolerating pain, inhibiting aggressive inclinations or
behavior, ignoring a distracting video while performing a visual tracking task, solving anagrams,
or holding a handgrip. The common feature of all of these tasks is that they are believed to
require self-control.
How effective are these interventions in improving self-control performance? Hagger et
al. (2010) meta-analyzed the findings of studies that measured the effects of training on ego
depletion (i.e., performance on the second of two sequential tasks involving self-control). Across
9 tests, taken from 7 published papers, they observed that training significantly reduced ego
depletion, with an overall effect size of d
+
= 1.07. This suggests that practicing self-control is an
extremely effective intervention. To put this effect size in context, it places the effect of self-
control training on ego depletion at roughly the 95
th
percentile in terms of both the average effect

Frequently asked questions

Q1. What contributions have the authors mentioned in the paper "Running head: effects of self-control training does self-control improve with practice? evidence from a 6-week training program" ?

In this paper, a 6-week training program of either cognitive or behavioral self-control tasks was used to test the effects of practice on a range of measures of self control, including lab-based and real-world tasks. 

Q2. What is the main effect of self-control training on behavior?

As self-control influences the extent to which individuals can overcome habits and enact intentions (De Ridder et al., 2012; Neal, Wood, & Drolet, 2013), it is possible that self-control training could reduce habitual control, and increase intentional control, over behaviors. 

Q3. How many participants were excluded from the analyses of the effects of training on prejudice?

Sixteen participants who identified themselves as Asian were excluded from theanalysis of the effects of training on prejudice and four participants who reported that they were unable to eat chocolate were excluded from the analyses of the effects of training on chocolate consumption. 

Q4. What did the authors analyze for the effects of training on outcomes?

The authors also analyzed the effects of training on outcomes using condition as a four-levelvariable (behavioral training, cognitive training, active control, no-contact control), as opposed to collapsing across training and control conditions. 

Q5. What is the role of training self-control in improving well-being?

Greater self-control is associated with improvedwell-being (e.g., De Ridder et al., 2012) and training self-control might therefore be expected to enhance well-being. 

Q6. What are the main ideas of Inzlicht et al. (2014)?

Inzlicht et al. (2014) also offer a number of other promising suggestions for improving self-control that go beyond the conception of effortful inhibition, such as changing goal appraisals and responding to self-control failures with acceptance. 

Q7. How many participants were required to achieve 90% power?

Power analyses based on the effect size estimate of d+ = 1.07 from Hagger et al.’s (2010)meta-analysis of training effects on ego depletion indicated that 40 participants would be required, split between the training and control conditions, to achieve 90% power (two-tailed). 

Q8. What did the participants report that their tasks required more effort?

Training participants also reported that their tasks required more effort, but only when rated retrospectively, t(171) = 3.70, p < .001, d+ = 0.56, and not when rated immediately after each task, t(132) = 0.00, p = 1.00, d+ = 0.00. 

Q9. What is the main idea of training self-control?

Although the central idea of programs designed to train self-control involves improvingpeoples’ ability to inhibit a dominant response, self-control involves more than just the effortfulinhibition of impulses (Fujita, 2011) and, as such, there may be multiple ways to improve this skill (Inzlicht, Legault, & Teper, 2014). 

Q10. What is the Bayes factor for the experiment?

The Bayes factor was 0.23, which indicates that their data are more consistent with a null effect of training on ego depletion than with the experimental hypothesis. 

Q11. How did the MANOVA find no effect of condition on behavior?

A MANOVA on these four variables found no effect of condition on behavior, F(4, 168) = 0.53, Wilk's 】 = 0.99, p = .71, partial さ2 = .01. 

Q12. What are the main effects of training on self-control?

As such, the present research measured participants’ performance of two behaviors that have been shown to depend upon self-control resources, but have not yet been tested as training outcomes; eating chocolate and displaying prejudice. 

Q13. How did the authors test for differences in the magnitude of the correlations between conditions?

As correlations are not normally distributed, the authors tested for differences in the magnitude of these correlations between conditions using Mann-Whitney U tests. 

Q14. What did the authors do to determine whether the association between implicit attitudes and behavior was moderated by training?

The authors also conducted hierarchical regression analyses to examine whether the association between implicit attitudes and behavior was moderated by training. 

Q15. What is the main prediction of the strength model?

The present findings also have implicationsfor the other key prediction of the strength model – that exerting self-control temporarily reduces self-control performance. 

Q16. How was the data for each of the behaviors skewed?

Data for each of the behaviors was positively skewed, such that most participants performed the behaviors relatively infrequently.