Stroop task, phonemic task and tiredness levels and significant differences in performance under the two conditions. These results support the notion that subsequent performance in self control tasks will be hindered and priming participants will have an effect on the performance in following tasks, however, other explanations for these effects are also discussed. Introduction The human ability for self control has been much more widespread than any other animal; this suggests that the evolutionary pressures that guided the selection of traits that make up human nature must have found self control to be a powerful and imperative skill for survival, with many functions (Baumeister, 2005, as cited in Gailliot et al., 2007). Self control is a central function of the self and an important key to success in life. Self control has attracted increasing attention from psychologists because understanding it will allow us to understand the nature and function of the self and self control also has many practical implications (Baumeister, Vohs, & Tice, 2007). A lot of recent studies have focused on the strength model (Baumeister, Bratslavsky, Muraven, & Tice, 1998). This model proposes that there is a limited store of self control resources, acts of self control cause short term impairments (ego depletion) in subsequent acts of self control (even on unrelated tasks) (Baumeister, Vohs, & Tice, 2007). For example, after resisting the temptation to eat freshly baked cookies, participants in one study quit sooner on a subsequent task requiring effortful persistence, compared with participants who did not have to resist eating the cookies (Baumeister, Bratslavsky, Muraven, & Tice, 1998).
Recent investigations have shown that the state of ego depletion can be counteracted by replenishing the resource through rest and relaxation (Tyler & Burns, 2008, as cited in Hagger, Wood, Stiff, & Chatzisarantis, 2010), vicarious restoration (Egan, Hirt, & Karpen, 2012) or by consuming a glucose product (Gailliot et al., 2007). Inadequate self control has been linked to risky and impulsive behaviour (Sultan, Joireman, & Sprott, 2011; Unger & Stahlberg, 2011), and the promotion of unethical behaviour (Gino, Schweitzer, Mead, & Ariely, 2011). The aims and rationale of the current study is to test the effects of ego depletion on self control and whether beliefs about self control will have make a difference to subsequent performance in self control tasks. Participants carried out two tasks in which they had to exert a degree of self control in order to successfully complete them. Shortly after, they were given a short ego depletion task. Participants were then instructed to carry out the same tasks, with different components, under different conditions (limited or unlimited). The subjective tiredness of the participants was recorded before and after the ego depletion task. The hypotheses are to see whether ego depletion will increase responses times on the Stroop test, reduce the number of words produced in the phonemic fluency test and increase subjective feelings of tiredness and whether priming participants to believe that self control resources are infinite will protect them against ego depletion. Method Participants
The performance and subjective feelings of tiredness of psychology students, aged between 18 and 37, was recorded (N=139, 110 female). The number of participants that were assigned to each experimental condition was uneven, in the limited prime condition (N=75, 60 female) and in the unlimited prime condition (N=64, 50 female). Participants were chosen using the opportunity sampling method. Design
The first part of the study is a within subjects design comparing phase one task performance to phase two performance. The second part of the study is a between subjects design comparing task performance between participants primed to believe self control was unlimited and those primed to believe it is limited (phase two only). The dependent variable for hypothesis 1 was the time taken to read the list (in seconds), the independent variable was the time before and after the ego depletion task (also in seconds). The dependent variable for hypothesis 2 was the number of words produced and the independent variable was the number of words before and after the depletion task. The dependent variable for hypothesis 3 was the rating of tiredness and the independent variable was the tiredness before and after the depletion task. For hypothesis 4 the two conditions, ‘limited’ and ‘unlimited’ were the independent variables and the between subjects effects on the three tasks were the dependent variables. Apparatus/Materials
A stopwatch program on our mobile phone devices was used to record the time taken to read the lists for the Stroop task and a tiredness rating questionnaire was used which used a 25 point likert scale. The paragraph of text which was copied out as the depleting task was also supplied. Procedure
The procedure was based on two parts. In the first phase the participants completed two tasks: the Stroop task and phonemic fluency task and rated their subjective tiredness. The first self control task was the colour-conflict Stroop test, where the words were presented as ‘RED’, ‘YELLOW, ‘GREEN’, & ‘BLUE’, but were printed in an incongruent colour. The time taken to read this list was recorded. The second task was a measure of phonemic fluency, where participants were given a letter and had to write down as many words as possible in one minute. Participants completed this for three different letters (F A S). In addition to these cognitive tasks subjective ratings of tiredness were assessed (which use a 25 point likert scale). In the second phase of the study scripts were read out to the participants that either primed them to believe that self control resources were limited or unlimited (see appendix). Participants were then instructed to copy it our but exclude the letter ‘E’ and spaces between words, this acted as the ego depletion task as it is a natural learnt responses to include the letter ‘E’ and spaces when writing so these natural responses need to be controlled.
Participants were then instructed to complete the same tasks as in phase one, this time a different list for the Stroop task was used, the order of which was counterbalanced and different letter sets were used for the phonemic fluency task (P L W). Subjective levels of tiredness were also recorded.
Results Hypothesis 1, 2 and 3
To examine the effect that ego depletion has on performance and subjective feelings of tiredness histograms were used to assess the distribution, all data sets were found to be normally distributed (see appendix). Skewness statistics were used to check the distribution statistically (also see appendix). The mean was calculated to assess the central tendency and the standard deviation was also measured to determine the spread of the data. Table 1: Mean and standard deviation statistics for each strategy before and after depletion
Mean (±SD)
Phase 1
Before depletion
Phase 2
After depletion
Stroop (seconds)
Fluency
Tiredness
70.66 (±12.36)
40.29 (±8.21)
12.66 (±4.50)
61.95 (±10.36)
39.35 (7.95)
15.27 (±4.52) Figure 1: Bar graph showing the means before and after Depletion for each strategy image The bar graph of means in Figure 1 shows that after depletion the mean decreases for the Stroop Task and Fluency Task and as expected the subjective feelings of tiredness increase. Statistical analysis for Hypothesis 1, 2 and 3
The design type is within subjects so in order to analyse the effect of the ego depletion manipulation on task performance and tiredness a related samples t-test must be carried out. A related samples t-test showed that ego depletion manipulation led to a significant increase in Stroop performance, t (138) = 10.94, p < .001 (one-tailed). Stroop reaction times were significantly higher before the ego depletion task (M = 70.66, SD = 12.36) as opposed to after the ego depletion task (M = 61.95, SD = 10.36). A related samples t-test also showed that ego depletion led to a significant decrease in the number of words produced in the phonemic fluency task, t (138) = 2.07, p = .02 (one-tailed). The number of words produced was significantly higher before the ego depletion task (M = 40.29, SD = 8.21) as opposed to (M = 39.35, SD = 7.95). A related samples t-test showed that the ego depletion task led to an increase in the level of subjective feelings of tiredness, t (138) = -7.19, p < .001(one-tailed). The reported level of tiredness was significantly lower before the ego depletion manipulation (M = 12.66, SD = 4.50) as opposed to (M = 15.27, SD = 4.52). Hypothesis 4
To test the effects of priming the participants to believe that self control resources were either limited or unlimited skewness statistics were conducted to check the distribution statistically (see appendix). The mean was calculated to assess the central tendency and the standard deviation was also measured to determine the spread of the data. Table 2: Mean and standard deviation statistics for each task under each condition
Mean (±SD)
Limited condition
Unlimited condition
Stroop
Fluency
Tiredness
62.70 (±11.20)
37.93 (±8.30)
16.08 (±4.26)
61.08 (±9.30)
41.02 (7.23)
14.31(±4.66)
Figure 2: Bar graph showing the means for each task for each condition image This bar graph shows that the mean for the Stroop task and the score given for subjective feelings of tiredness were higher in the limited condition whereas the mean for the fluency task was higher in the unlimited condition. Statistical analysis for Hypothesis 4
An unrelated samples t-test was used to analyse the effects that beliefs about self control had on ego depletion because the design type was between subjects. Subjects were either tested under the limited or unlimited condition. An unrelated samples t-test showed that participants in condition one (limited resource) had insignificantly greater response times on the Stroop (M = 62.70, SD = 11.20) than participants in condition two (unlimited resource; M = 61.08, SD = 9.30), t (137) = .919, p = .180 (one-tailed). An unrelated samples t-test showed that participants in condition one (limited resource) produced a significantly lower number of words in the phonemic fluency task (M = 37.93, SD = 8.30) than participants in condition two (unlimited resource; M = 41.02, SD = 7.23), t (137) = -2.315, p = .011 (one-tailed). An unrelated samples t-test showed that participants in condition one (limited resource) had significantly higher subjective feelings of tiredness (M = 16.08, SD = 4.26) than participants in condition two (unlimited resource; M = 14.31, SD = 4.66), t (137) = 2.34, p = .011 (one-tailed). Discussion The first part of the study was to see whether ego depletion will hinder performance on self control tasks and the second part was to see whether priming participants to believe that self control resources are infinite will protect them against ego depletion. The results of the related samples t-test showed that there was a significant increase in Stroop performance, the responses times decreased, so hypothesis 1 was rejected. However, there was a significant reduction of the number of words produced in the phonemic fluency task and a higher level of subjective feelings of tiredness was reported, therefore, hypothesis 2 and 3 were both accepted. Hypotheses 4 was accepted because participants under the unlimited condition performed better on the Stroop task and the phonemic fluency task and reported less tiredness. These findings support the notion that engaging in an activity that uses self control depletes limited resources and makes subsequent tasks involving self control difficult. These findings seemed to support a study conducted by Baumeister, Vohs and Tice (2007), where the act of resisting the temptation to eat chocolates and eat unappetising radishes made participants more likely to perform poorly in solving puzzles and lack persistence, as compared to people who had not exerted self-control. One of the problems with the methodology of this study is that the use of a stopwatch for the measurement of the responses times for the Stroop task allows for human; a way to eliminate this inaccuracy would be to have more than one person recording the time so then we can look at the interobserver reliability as an indication of the accuracy of our scores. The role of glucose in self control has been proposed by Gailliot and Baumeister (2007), so an interesting extension of this study would be to include a blood glucose test as another dependent variable and test whether the blood glucose levels will decrease as a result of self control tasks, supporting or opposing the view that glucose is the limited energy resource responsible for failures in self control (Gailliot et al., 2007). We could also introduce gender as a factor of this study and analyse the effect, if any, gender has on self control. In summary, findings support the notion that engaging in an ego depletion task will facilitate the decline of performance in subsequent tasks. By building on this knowledge, psychology may be able to improve the mental health and well-being of many people, especially those people who struggle with self control, such as drug and alcohol abuse, overeating and risky sexual behaviour. References 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, 74(5), 1252-1265 Baumeister, R.F., Vohs, K.D., & Tice, D.M. (2007). The strength model of self-control. Current Directions in Psychological Science, 16(6), 351-355 Egan, P.M., Hirt, E.R., & Karpen, S.C. (2012). Taking a fresh perspective: vicarious restoration as a means of recovering self-control. Journal of Experimental Social Psychology, 48(2), 457-465 Gailliot, M.T., & Baumeister, R.F.
(2007). The physiology of willpower: linking blood glucose to self-control. Personality and Social Psychology Review, 11(4), 303-327 Gailliot, M.T., Baumeister, R.F., Dewall, C.N., Maner, J.K., Plant, E.A., Tice, D.M, Brewer, L.E., & Schmeichel, B.J. (2007). Self-control relies on glucose as a limited energy source: willpower is more than a metaphor. Journal of Personality and Social Psychology, 92(2), 325-336 Gino, F., Schweitzer, M.E., Mead, N.L., & Ariely, D. (2011) Unable to resist temptation: how self-control depletion promotes unethical behaviour. Organisational Behaviour and Human Decision Processes, 115(2), 191-203 Hagger, M.S., Wood, C., Stiff, C., & Chatzisarantis, N.L.D. (2010). Ego depletion and the strength model of self-control: a meta-analysis. Psychological Bulletin, 136(4), 495-525
Sultan, A.J., Joireman, J., & Sprott, D.E. (2012). Building consumer self-control: the effect of self-control exercises on impulse buying urges. Marketing Letters, 23(1), 61-72
Unger, A., & Stahlberg, D. (2011). Ego-depletion and risk behaviour: too exhausted to take a risk. Social Psychology, 42(1),
28-38 Appendix Hypothesis 1, 2 and 3 Figure 3: Histogram showing the distribution of the response times on the Stroop task (seconds) before the depletion task image Figure 4: Histogram showing the distribution of the response times on the Stroop task (seconds) after the depletion task image
Figure 5: Histogram showing the distribution of the number of words produced in the phonemic fluency task before the depletion task image
Figure 6: Histogram showing the distribution of the number of words produced in the phonemic fluency task after the depletion task image Figure 7: Histogram showing the distribution of the subjective feelings of tiredness before the depletion task image Figure 8: Histogram showing the distribution of the subjective feelings of tiredness after the depletion task image The histograms in Figures 3 - 8 shows that all data sets are normally distributed. Table 3: Skewness statistics Skewness
Statistic
Standard Error
Stroop Phase 1
Fluency Phase 1
Tired Phase 1
Stroop Phase 2
Fluency Phase 2
Tired Phase 2
-.095
.082
-.361
.395
.060
-.395
.206
.206
.206
.206
.206
.206
The skewness statistic is not more than twice the standard error for these tasks in phases 1 and 2; therefore, all the data has been statistically proven to be normally distributed. Table 4: Results of the SPSS test for the related samples t-test for performance on the self control tasks and tiredness before and after the ego depletion
Paired Differences t df
Sig. (2-tailed)
Mean
Std. Deviation
Std. Error Mean
95% Confidence Interval of the Difference
Lower
Upper
Pair 1
Stroop Phase 1 - Stroop Phase 2
8.7129
9.3914
.7966
7.1379
10.2880
10.938
138
.000
Pair 2
Fluency Phase 1 - Fluency Phase 2
.942
5.369
.455
.042
1.843
2.070
138
.040
Pair 3
Tired Phase 1 - Tired Phase 2
-2.604
4.271
.362
-3.321
-1.888
-7.189
138
.000
Hypothesis 4 The scripts used to prime the participants to believe that self control resources were either limited or limited are as follows: Limited Script: “Recent research has shown that if you participate in self control tasks then you will find subsequent acts of self control more difficult. Think of a muscle; if you were to exert this muscle for an amount of time it would become fatigued and then become less efficient or even unable to perform a task. Only after sufficient rest and recuperation can the muscle act efficiently again. As a physical system the brain acts in the same way- applying mental effort to control oneself results in mental self control resources becoming depleted, so eventually self control becomes more difficult. I’m sure everyone here has had an experience were they have felt drained after exerting self control.”
Unlimited Script: “Recent research has shown that self control resources in humans are relatively limitless, and failures control oneself are usually the result of a conscious choice not to control behaviour. Unlike muscles which become fatigued and less efficient (or even unable to perform) after repeated performance of a task, the brain is able to maintain acts of self control. This is because muscles require a huge amount of oxygen to exert effort while the brain only requires a stable amount of oxygen whether self control is exerted or not, so it is essentially protected from fatigue induced by self control. I’m sure everyone here has had when they have had to control their behaviour for prolonged periods of time and not found it tiring.” Table 5: Skewness statistics for the Stroop and phonemic fluency task and tiredness for the two conditions ‘limited’ and ‘unlimited’
Skewness
Limited
Unlimited
Statistic
Standard Error
Statistic
Standard Error
Stroop
Fluency
Tiredness
.358
.070
-.058
.277
.277
.277
.347
.278
-.646
.299
.299
.299
The skewness statistic is not more than twice the standard error for these tasks in both conditions, apart from the tiredness scores in the unlimited condition; therefore, the majority of data has been statistically proven to be normally distributed. Table 6: Results of the SPSS test for the unrelated samples t-test for Stroop scores in unlimited compared to limited self control conditions (Phase 2)
Independent Samples Test Levene 's Test for Equality of Variances t-test for Equality of Means
F
Sig. t df
Sig. (2-tailed)
Mean Difference
Std. Error Difference
95% Confidence Interval of the Difference
Lower
Upper
Stroop
Equal variances assumed
2.446
.120
.919
137
.360
1.6208
1.7643
-1.8681
5.1096
Equal variances not assumed .932
136.912
.353
1.6208
1.7388
-1.8176
5.0591
Table 7: Results of the SPSS test for the unrelated samples t-test for phonemic fluency scores in the unlimited compared to limited self control conditions (Phase 2)
Independent Samples Test Levene 's Test for Equality of Variances t-test for Equality of Means
F
Sig. t df
Sig. (2-tailed)
Mean Difference
Std. Error Difference
95% Confidence Interval of the Difference
Lower
Upper
Fluency
Equal variances assumed
2.246
.136
-2.315
137
.022
-3.082
1.332
-5.716
-.449
Equal variances not assumed -2.340
136.932
.021
-3.082
1.317
-5.687
-.477 Table 8: Results of the SPSS test for the unrelated samples t-test for subjective feelings of tiredness ratings in the unlimited compared to limited self control conditions (Phase 2) Independent Samples Test Levene 's Test for Equality of Variances t-test for Equality of Means
F
Sig. t df
Sig. (2-tailed)
Mean Difference
Std. Error Difference
95% Confidence Interval of the Difference
Lower
Upper
Tired
Equal variances assumed
.560
.456
2.337
137
.021
1.767
.756
.272
3.263
Equal variances not assumed 2.320
129.048
.022
1.767
.762
.260
3.275