University of Maastricht: "The influence of promotion focus and prevention focus on ego depletion and self-control"
Frank Beckers, Eveline Eerdekens, Sylvia Gaul, Mathilde van Haperen, Olga Maurer, Hans Moesen
According to the limited energy model of self-control, repeated acts of self-control deteriorate due to a loss of energy. When the resource of energy becomes depleted, the individual is left in a state of ego depletion. We examined whether the type of regulatory focus can have an influence on ego depletion. The type of regulatory focus that a person can have is a promotion focus or a prevention focus. People with a promotion focus are concentrated on accomplishing their goals and maximizing their profit. In contrast, people with a prevention focus are concentrated on avoiding mistakes and failures and minimizing costs. Participants engaged in a self-control task. After this task, participants’ performance on a second self-control task, a calculation task, was measured. At this point, the regulatory focus was manipulated. In order to induce the promotion focus, we told participants they had to answer more than 90 % of the problems correctly to win a 25 euros reward. Participants in the prevention focus group were told they had no chance of winning the reward if they answered more than 10 % of the problems incorrectly.
Receiving a promotion or prevention focus task framing resulted in no significant differences in performance on the calculation task. However, an analysis of a regulatory focus list, taken at the beginning of the experiment, revealed an influence of interpersonal differences in focus. We found that participants who scored relatively high on the promotion items of the questionnaire solved a significantly larger amount of sums in the calculation task but also made significantly more mistakes than participants who scored relatively low on the promotion items.
The influence of promotion focus and prevention focus on ego depletion and self-control
Have you ever tried to stop smoking or to lose weight, or even both at the same time?
Then you know what it feels like to come home after a stressful day when you inhibited your impulses, your emotions and needs and you do not have the strength to say no to a cigarette or a piece of chocolate anymore. It is clear that saying NO to an attractive offer involves a lot of self-control. Self-control is the suppression of an immediate impulse or response to attain a goal over a longer period of time. Failures in self-control such as drug abuse or aggression are common in our society. Recent findings suggest that a reason for the loss of self-control is the depletion of energy. According to the limited energy model of Baumeister, Bratslavsky, Muraven, & Tice (1998), self-control is drawn from a limited resource, which can easily become depleted through self-control efforts like non-smoking or dieting. An act of self-control may deplete the energy required to exercise self-control in a different unrelated domain. For example, trying not to smoke if you are addicted to cigarettes requires a lot of self-control and depletes the available energy. So, if you try to say no to a cigarette, there is less energy left to successfully inhibit your urge to eat a desirable pizza. When the resource of energy becomes depleted, the individual is left in a state of ego-depletion. Consequently, self-control will be more likely to break down and fail (Wallace & Baumeister, 2002). This could be compared to a muscle that becomes tired after an act of effort.
In one of the first experiments, carried out about ego-depletion, (Baumeister et al., 1998), hungry participants were assigned to three conditions. In the first condition participants were allowed to eat 2 or 3 radishes, while a big plate with chocolate sweets stood beside (an act of self-control). This group had to control themselves by not eating from the tempting sweets. The second group was allowed to eat 2 or 3 chocolate candies and the third (control) group skipped the food part and went directly to the unsolvable puzzle task that followed. Trying to solve the unsolvable puzzle task involved self-control. The experiment indicated that participants of the first (radish) group in contrast to participants of the second and third group were less persistent on the unsolvable puzzle task which was the second self-control task for the first group. This study implies that self-control performance is dependent on the resource of energy. There is depletion of energy by the first group but not by the two others, which explains why those groups had more energy left on the unsolvable puzzle task and could go on trying longer. This implicates that there is a limited resource of self-control that becomes depleted by using it. But leads depletion invariably to loss of self-control? In several other studies evidence is found that other factors like motivation or expectation may overcome effects of ego-depletion or even have a positive effect on self-control.
In a study conduced by Tenbült, Martijn and Merckelbach (2001) participants had to watch a horrible movie. The ego depletion group was told to suppress their emotions while watching the movie. The other group did not have to suppress their emotions. After watching the movie one group was told by the experimenter that: “after the suppression of a negative experience, people tend to think that their performance on a physical task would decrease. This however is not the case”. The participants of this group, although depleted, showed better performance on a physical test than the participants of the group with no depletion that did not receive this instruction. They even performed better than the group who did not suppress their emotions while watching the movie. According to these findings, ego depletion seems not to occur as soon as expectations of people are changed. In other words, it seems that the way people perceive acts of self-control can have a crucial influence on self-control performance.
Another study by Muraven and Slesstra (2003) investigated the effect of motivation on self-control. In one experiment, some participants were told that the experimental procedures would have a great benefit for developing a better treatment for the Alzheimer’s disease, so they thought their effort would be important. The rest of the participants were not told about the potential benefits of the experimental procedures. During the depletion phase, the participants had to work either on a cognition task or on a memory task. To measure the self-control performance, the participants were then introduced to work on an unsolvable puzzle task. The results show that participants who were motivated through the mention that the purpose of the study was to provide scientific evidence about the Alzheimer’s disease, performed better on the puzzle task than participants who were not depleted as well as participants who were depleted but did not get this motivation. This provides evidence that self-control becomes not weaker after ego depletion if there is sufficient motivation available.
It becomes clear that ego depletion is not inevitably a consequence of repeated self-control. Factors like motivation and implicit expectations have a great influence on people who are exposed to repeated self-control. This evidence gives reason to search for more factors that might have influence on self-control, to understand why some people can offer enough self-control to stop smoking and dieting at the same time. Another factor that could have an influence on ego depletion is the type of regulatory focus a person has. If the type of regulatory focus a person has is a promotion focus, then he or she is concerned with obtaining the presence of positive outcome from success and avoiding the negative outcome from failure. It is a focus on advancement, accomplishment, and aspirations. However, if the type of regulatory focus a person has is a prevention focus, then she or he is concerned with avoiding negative outcomes from success and avoiding the presence of negative outcomes from failure. It is a focus on safety, responsibilities, and obligations (Higgins, Shah, & Friedman, 1997). People with a prevention focus are concentrated on avoiding mistakes and minimizing costs. People with a promotion focus are concentrated on maximizing their profit. It seems that the type of focus one has, can have a great influence on different domains. For example, in a study about the effect of focus on emotions by Higgins, Shah and Friedman (1997) it was found that there is a relationship between the strength of the regulatory focus and the strength of emotions. Participants exposed to a promotion focus experienced stronger feelings of happiness and satisfaction when attaining a goal and stronger feelings of disappointment when failing to attain a goal. Contrary, participants exposed to a prevention focus experienced calm and relaxed feelings when attaining a goal and felt tense and uneasy when failing to attain a goal. Another study by Förster, Higgins and Bianco (2003) revealed that participants with a promotion focus are faster but less accurate on a proofreading task, while participants with a prevention focus were more accurate but not so fast. Based on the previous findings discussed, the question arises if the type of regulatory focus can have an influence on ego depletion. Referring to the example given at the beginning of the introduction the current question is: Is someone who stops smoking because of avoiding lung cancer (prevention focus) more successful than someone who stops smoking with the goal of winning the “Ironman” on Hawaii (promotion focus)?
In the present study, the influence of the regulatory focus on self-control performance is going to be tested in four conditions. In order to investigate trait regulatory focus, all participants fill out a questionnaire to determine their type of regulatory focus and then a task that involves self-control or a task that involves no self-control. For the first three depletion groups, ego depletion is induced through a manipulated Stroop task. The control group perform on an easy Stroop task which requires no self-control. In order to investigate state induced regulatory focus, the regulatory focus of the participants in the first two conditions is manipulated by the instructions given by the researcher. The participants in the remaining two conditions receive no manipulation. These instructions concern the second task, a self-control task in the form a of a calculation task. The question is whether participants who are exposed to a promotion focus perform different on this calculation task compared to participants who are exposed to a prevention focus.
There may be an explanation of advocates of the ego depletion theory like Baumeister who will say that the one who performs better on a self-control task will have used more energy for solving it than normal performers. If that would be the case the better performers would perform worse on a following self-control task than the normal performers, because their energy level previous to the second self-control task was lower than the one of the normal performers. The possible result that the better performers on the first self-control task would additionally perform better on the second self-control task would indicate that the ego depletion theory of Baumeister would not be concordant with the possible result stated above. To test this, a geometric tangram puzzle that is unsolvable is presented.
Design and Participants
The 80 participants were second year students of Maastricht University who are participating in the course 2.5A at the faculty of Psychology. We did not inform the participants about the content of our experiment because it would influence the results of our study. They received the information after the experiment was done. The participants were randomly assigned to one of the following 4 conditions: Depletion/Promotion Focus, Depletion/Prevention Focus, Depletion/No Focus, No Depletion/No Focus.
The experimenter welcomed the participant and explained that the experiment was about concentration and that he/she had to do a few tasks, like completing questionnaires and do some concentration tests. After this explanation, the participant was asked to sign a form of consent. The next step was the completion of two questionnaires, one concerning the mood of the participant and the other for determining the regulatory focus (prevention or promotion focus) people have.
In order to manipulate depletion, the second part of our experiment was a self-control task, which was a variation of the Stroop task for the 3 Depletion groups (Wallace & Baumeister, 2002). The participants got a list of 135 words containing colors (red, green, yellow, blue) that were written in another color than the meaning of the word. The subjects were asked to name the color each word was printed in. Only if the word was printed in red, they had to ignore the ink color and read the meaning of the word aloud. This required self-control, because the participants had to override their impulses to read the words instead of telling the ink color. The task became more difficult with the additional instruction of reading the word when the ink was red, because the pattern of overriding the impulse had to be overridden in this case. The participants were asked to complete the task as quickly as possible. To increase the pressure we measured the time it took every participant to complete the task and we pretended to mark every correct answer of the participant. This data was not registered because it was not relevant for our study.
The No Depletion condition got a list with the same amount of words containing colors that were written in the same color. They were asked to read the list aloud. In contrast with the other three groups, this group was not depleted because the participants did not need to override their impulses to read the text, since the color of the text matched the meaning of the text. Again, we asked the participants to solve the task as quickly as possible. To increase the pressure we measured the time it took every participant to complete the task and we pretended to mark every correct answer of the participant, but this data was not registered. When this task was completed the participants were asked to indicate their mood on 5 point scale, ranging from very bad to very good.
The next task was a self-control task in the form of a calculation task. The participants saw the calculation problems on a computer screen and they had to compute the calculation by typing the correct answer using the keyboard. When they were finished with one problem, the next problem appeared on the screen. It was not clear for the participant how many problems there were. In addition, the participants had to put on headphones through which they could hear a voice naming random digits during this task. This was the distracter of the calculation task. The task involved self-control because the participants had to ignore the voice naming random digits. In this phase of the experiment we manipulated the focus of the participants. Before participants started the task, they were exposed to different instructions. All the groups were given the instruction to solve as many calculation problems as they could. The promotion focus group additionally got the information that they had a chance of winning 25 Euros if they answered more than 90% of the problems correctly. The prevention focus group got the information that they had no chance of winning 25 Euros if they answered more than 10% of the problems incorrectly. The control groups only got the instruction to solve as many problems as they could, but after they finished the, task they were told that there would be a lottery with the best 10% of the participants, so they had a chance to win 25 Euros. We measured how many problems the participants in each group solved in 8 minutes.
After completing the calculation task, participants rated their mood again, and were asked to complete the last task. Participants had to put down pieces of paper (that were shaped like triangles, rectangles or other symmetric forms) on a template, so that the pieces exactly fitted the template. The puzzle was unsolvable; it was not possible to get the exact form of the template with the pieces they got. The participants were asked to solve the puzzle and were told that they could quit this task whenever they wanted. We measured the time each participant persisted in trying to solve the puzzle.
Finally, we asked the participants to indicate their mood on a scale and gave them a last questionnaire with questions about their age, gender and what they thought our experiment was about. (See Table 1 for an overview of the experimental procedure.)
Table 1 Graphical representation of the experimental procedure
|Conditions||1: Questionnaire: Regulatory Focus (Promotion or Prevention Focus)||2: Depletion Task (Easy or Difficult Stroop Task)||3: Mood||4: Self-Control Task: Mathematical Task (Different Instructions-Different Focus)||5: Mood||6: Self-Control Task: Unsolvable Geometric Puzzle (Energy Level)||7: Mood||8: Questionnaire (Age, Gender, What Experiment Was About)|
|Condition 1 (experimental group): Depletion/Promotion Focus||Questionnaire||Difficult Stroop Task||Indicate Mood on Scale||Chance of Winning 25 Euros if more than 90% of the Problems Correct||Indicate Mood on Scale||Tangram Puzzle||Indicate Mood on Scale||Questionnaire|
|Condition 2 (experimental group): Depletion/Prevention Focus||Questionnaire||Difficult Stroop Task||Indicate Mood on Scale||No Chance of Winning 25 Euros if more than 10% of the Problems Incorrect||Indicate Mood on Scale||Tangram Puzzle||Indicate Mood on Scale||Questionnaire|
|Condition 3 (control group): Depletion/No Focus||Questionnaire||Difficult Stroop Task||Indicate Mood on Scale||No Focus||Indicate Mood on Scale||Tangram Puzzle||Indicate Mood on Scale||Questionnaire|
|Condition 4 (control group): No Depletion/No Focus||Questionnaire||Easy Stroop Task||Indicate Mood on Scale||No Focus||Indicate Mood on Scale||Tangram Puzzle||Indicate Mood on Scale||Questionnaire|
In order to determine a participant’s regulatory focus we used a questionnaire with half of the questions considering a promotion focus and half of the questions considering a prevention focus. The participants had to rate each question on a 5 point scale varying from totally disagree (1) to totally agree (5).
In order to deplete participants we used a Stroop task as well as a mathematical task. There were two versions of the Stroop task, a difficult and an easy version. Both Stroop tasks were presented on a piece of paper (size A4).
We used a Compaq desktop computer with Windows XP to run a Delphi programmed mathematical task. Participants used the keyboard to type the right answer, and the mouse to verify the answer. If participants did not enter an answer or entered a wrong answer the program still continued. The program stopped after 8 minutes. A total of 120 sums were programmed in the task. During the mathematical task, the participants wore headphones which were playing a voice naming random digits. We used a Winamp-player in Windows XP to play these interfering audio-files. The last task was an unsolvable geometric puzzle. A classical tangram puzzle, consisting of six geometric figures was used.
Mood was measured 4 times during the experiment by the item: “My mood at this moment is:”
The questionnaire items were rated on 5 point scale ranging from very bad (1) to very good (5).
In order to measure the participants’ self-control performance we used a calculation task that required the participants to calculate as many problems as possible within 8 minutes. The computer program counted the number of solved problems and the percentage of correctly solved problems.
Unsolvable puzzleThe participant was told to try to recreate a geometric figure that exactly fitted the template. We measured the time (in seconds) they spent on trying to solve the puzzle.
The current data include 80 participants. The conditions depletion/no focus and no depletion/no focus were compared to examine whether the Stroop task did induce depletion. One-way ANOVA revealed no significant effect of the depletion task on the percentage of correctly solved problems in the calculation task, F(1,38)= .92; p = .34. This indicates that we cannot find comparable differences between the depletion and no depletion conditions. In other words, this suggests that the difficult Stroop task in comparison with the easy Stroop task did not induce (enough) depletion. Thus, our manipulation was not successful.
We analyzed the effect of the depletion task on the time spent on the puzzle in the same way. The results did not show any significant outcome either, F(1,38) = .04; p = .85. These findings might imply that our depletion task, the Stroop task, did not deplete the subjects and therefore reveals no results.
The participants in the first condition were exposed to a promotion focus. The participants in the second condition were exposed to a prevention focus. These two conditions were compared in the analysis using One-way ANOVA, with percentage of correctly solved calculations as dependent measure. There were no different outcomes in the calculation task for the two foci, F(1,78)= .72; p = .85. In other words, the focus manipulation did not have an effect on self-control performance.
Again, we analyzed the effects of the focus manipulation in the same way on the time spent on the puzzle. No significant effects were found F(1,38) = .04; p = .92. This means that the two conditions did not differ in their effect on performance in the puzzle task.
In order to look at the influence of the trait regulatory focus, we rated the participants´ sort of focus with a mean median split. So, a participant got a high or low score on the promotion focus, and a high or low score on the prevention focus. Thus, we had two independent conditions in which each participant was rated high or low. One-way ANOVA showed a significant result for participants with a high promotion focus in contrast to participants with a low promotion focus. People in the high promotion condition solved more calculation problems than participants with a low promotion focus F(1,78) = 4.57; p = .04. However, the low promotion people solved more problems correctly F(1,76) = 4.35; p = .04. The mean scores can be observed in Table 2. There were no results found when we compared the high prevention and the low prevention conditions, as seen in Table 3 (all F`s <1).
Table 2: Means in promotion condition. Standard deviations are given in parentheses
|High Promotion Focus||Low Promotion Focus|
|Percentage correct solved problems in %||72,04 (1.965)||78,07 (2.119)|
|Number of solved problems||36,00 (7.528)||32,65 (6.312)|
Table 3: Means in prevention condition. Standard deviations are given in parentheses
|High Prevention Focus||Low Prevention Focus|
|Percentage correct solved problems in %||75,19 (2.069)||74,92 (2.017)|
|Number of solved problems||25,10 (6.443)||26,46 (7.646)|
The time spent on the puzzle task did not show any significant differences between the four conditions, F(1,38) = 0,83; p = .37. So, there is no effect of the regulatory focus people have on their performance in the unsolvable puzzle task.
We measured the mood of the participants at four times during the experiment. The mood changes were analyzed by a One-way ANOVA with the time as independent variable. These four analyses reveal no differences in mood between the groups at each time (all F’s < 1); but we found that the mood of the participants got significantly worse after the calculation task, as we can see in Figure 1.
Figure 1: Means of mood scores at the four times. The X-axis represents the time and the Y-axis represents the mean score of mood in the 4 groups.
In the present study we tried to investigate the influence of promotion focus and prevention focus on ego depletion and self-control. Unfortunately, no significant effects were found. Whether participants received a different type of focus or whether they received no focus, did not elicit a significant difference in performance on our main variable, the percentage of correctly solved sums on the self-control task (the calculation task). The absence of any significant difference between groups may be explained by several reasons.
Firstly, it is possible that the induction of depletion was not strong enough. The Stroop variation task of Wallace and Baumeister (2002) may not have caused sufficient depletion in this research. Future studies may use an ego depletion task that is longer in time and more difficult to complete for the subject. Secondly, while participants were completing the calculation task, they got distracted by the presence of the experimenter and new participants in an unconformable way. New participants came and asked for several things like new appointments and if it was the correct room etc. This distraction may have caused irregularities in performance. Thirdly, one third of our participants were German. In comparison to the Dutch participants, for German participants it may have been less distracting that the numbers playing on the headphones were in Dutch. If we take for granted, that people count in their mother tongue, the distraction would have been less strong for the German speaking subjects. This could have influenced the performance on both the Stroop and the calculation task for German students. Future research should try to reach conformity in test situations for all participants. Fourthly, another factor influencing the experiment might be the reward. Most participants showed no interest in the 25 euros reward and were more concerned about the study-point they got for participating in the experiment. Unfortunately, a higher amount of money was not to our exposal due to the faculty funding. Also, our formulation that people make chance on winning 25 euros might have been not strong enough. To tell people that they win 25 euros if they reach the condition might have been more effective in order to correctly induce a certain kind of regulatory focus.
In addition, participants performed the experiment in one single room. However, for the puzzle task, participants were placed outside this room. Outside this room, other groups were performing their experiments as well. The presence of other groups and their participants might therefore have influenced the performance of participants on the puzzle task.
Another factor might be the different experimenters. This could have influenced objectivity. It was crucial for the experiment that the instructions for the participants given by the experimenter were constant over conditions. However, the way of expressing the instructions might have slightly differed between experimenters, and therefore influenced the results. Also, the gender of experimenters could have had an influence. Male experimenters might have had a more authorizing pose, which provoked an increased in performance of participants.
Finally, a few days after the experiment started, some participants heard from their fellow students that the puzzle task was unsolvable. These students might have stopped earlier, knowing that trying to solve the puzzle was pointless.
Although our main question was not answered in this study, we did find an interpersonal effect of regulatory focus. We found that people who scored relatively low on the promotion focus items of the first questionnaire (Higgins et al., 2001) solved a higher percentage of sums correctly on the counting task in comparison with people who scored high on the promotion focus items. Furthermore, our data suggests that people who scored higher on promotion focus solved in general a larger amount of sums. The explanation of this finding could be that people who tend to score lower on promotion focus items make fewer mistakes, because they are more concentrated on the prevention of mistakes. These findings are in line with the results of a study by Förster, Higgins and Bianco (2003) which revealed that participants with a promotion focus are faster but less accurate on a proofreading task while participants with a prevention focus were more accurate but not so fast. However, in our study no significant effects were found for the scores on the prevention items of the questionnaire. Therefore, it remains difficult to draw definite conclusions about the influence of prevention and promotion on task performance.
The current experiment was a first step in considering the effects of regulatory focus on ego depletion. Although no significant effects were found, future experiments may prove different. Future studies may take the above mentioned possible confounding factors into account. With few questions answered and more to come, researchers studying this field will have a great task to accomplish.
Baumeister, R.F., Bratslavsky, E., Muraven, M., & Tice, D.M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 5(74), 1252-1265.
Förster, J., Higgins, E.T., & Bianco, A.T. (2003). Speed/Accuracy decisions in task performance: Build-in trade-off or separate strategic concerns? Organizational behavior and human decision processes, 90, 148-164.
Higgins, E.T., Shah, J., & Friedman, R.S. (1997). Emotional responses to goal attainment: Strenght of gedulatory focus as moderator. Journal of personality and social psychology, 72, 515.
Higgins, E.T., Friedman, R.S., Harlow, R.E., Idson, L.C., Ayduk, O.N., & Taylor, A. (2001).
Achievement orientations from subjective histories of success: Promotion pride versus prevention pride. European Journal of Social Psychology, 31, 3-23.
Muraven, M., & Slesstra, E. (2003). Mechanisms of self-control failure: Motivation and Limited resources. Journal of Personality and Social Psychology, 29, 894-906.
Tenbült, P., Martijn, C., & Merckelbach, H. (2001). Getting a grip at self-control. Manuscript submitted for publication.
Wallace, H.M., & Baumeister, R.F. (2002). The effects of succes versus failure feedback on further self-control. Self and Identity, 1, 35-41.