Printer Friendly

The effect of self-control resource on risk preference.

In their daily lives, people plan their actions, such as buying lottery tickets, investing money in a project, or buying a house, with the objective of receiving a preferred benefit. This always involves a calculation of risk, because there are different costs and reward benefits in these activities. When facing these scenarios, people can exert self-control to manage the risk of their behaviors, thereby allowing them to make appropriate decisions. Thus, individuals' self-control can affect their risk preference in risky choice decision making.

The choice between a certain option (e.g., 100% chance of receiving US$65) and an uncertain option (e.g., 50% chance of receiving US$168) is defined as risky choice (Chen & He, 2011). Risk preference refers to the attitude people have toward risk, and is a matter of preferences. Kahneman, Slovic, and Tversky (1982) and Berns, Laibson, and Loewenstein (2007) investigated the role of self-control in risky choice, and found that people with greater self-control preferred to take a risk (high risk preference), whereas those with less self-control preferred not to take a risk. However, the mechanism of how an individual's self-control (high or low level, enhanced or depleted) can affect risk preference is still unclear.

Self-control is considered to be a personal process involving a set of skills, and as such, difficult to manipulate in a laboratory environment (Baumeister, 2002). To successfully manipulate self-control, Baumeister, Bratslavsky, Muraven, and Tice (1998) proposed the strength model of self-control. They defined self-control as a type of limited self-control resource (SCR), like muscular energy, in that the strength of self-control is determined by SCR. Hagger, Wood, Stiff, and Chatzisarantis (2010) in their meta-analysis, ascertained that an individual's SCR is depleted by overuse, resulting in ego depletion and poor performance in consecutive self-control tasks. Results of Chinese studies conducted by Sun (2008) and Dou (2013) indicated that Chinese participants also showed SCR depletion after finishing an SCR manipulation task. However, in an exploratory experiment on SCR and ego depletion, I found that individuals' SCR levels vary naturally and that finishing an SCR manipulation task affected individual participants' SCR level differently. Specifically, some participants showed depleted SCR and others showed enhanced SCR. Thus, I theorized that individuals' risk preference would be influenced by their SCR level at the onset of the manipulation task and also by the variance in individual SCR levels.

Individual SCR Level Differences and Risk Preference

Hofmann, Vohs, and Baumeister (2012) pointed out that individuals' SCR levels might be influenced by a wide range of daily activities and temptations, such as the enticement of icecream when one is on a diet. If an individual's SCR is, indeed, a limited psychological resource, it seems likely that SCR level will differ from one individual to another and also within each individual at different points in time. Moreover, an individual's SCR level may fluctuate according to the time of day (Baumeister & Heatherton, 1996). Thus, when two alternatives are equally attractive in risky choice, individuals whose SCR level is high may be better able to exert self-control to handle risk and, thus, would exhibit a stronger preference than others for choosing the higher risk alternative. Thus, I proposed the following hypothesis:

Hypothesis 1: SCR level will be associated with risk preference such that individuals with a high SCR level will exhibit a higher risk preference whereas individuals with a low SCR level will exhibit a lower risk preference.

Individual Variation in SCR Level and Risk Preference

Bruyneel, Dewitte, Franses, and Dekimpe (2009), Freeman and Muraven (2010), and Price and Yates (2010) found by using monetary scenarios, the Choice Dilemma Questionnaire (CDQ), the Balloon Analog Risk Task (BART), and mathematical computation tasks of varying difficulty, that, compared with a nondepleted SCR group, depleted SCR participants showed a higher risk preference. In contrast, Unger and Stahlberg (2011) reported that depleted SCR participants preferred low-risk options in an investment scenario. These conflicting conclusions can possibly be attributed to differences in the risky-choice tasks used in the studies. For example, as the emphasis in an investment scenario decision is on the benefit to the company, participants are likely to be risk averse. Furthermore, facing the risky choice of a financial loss or gain in an investment scenario may activate a participant's consideration of his/her future career and income in making a decision, unlike in the other tasks. On the other hand, Vohs et al. (2008) found that too many choices affected SCR. As the CDQ and BART tasks include many choices, this leads to SCR variance. Therefore, it was important to develop a risky-choice task that was equivalent to CDQ and BART but without these faults.

In addition, Muraven (2008) and Converse and DeShon (2009) reported that not all their participants exhibited SCR depletion after completing SCR manipulation tasks. Their findings indicated that SCR is a limited resource relative to the duration of a SCR manipulation task. When differences in the variation patterns in an individual's SCR level are considered, the possibility arises that participants with a higher primary SCR level than others will exhibit SCR depletion, and those with an initially lower SCR level will not exhibit SCR depletion upon completing an SCR posttest in a dual-task paradigm. Thus, I proposed the following hypothesis:

Hypothesis 2: Depleted SCR individuals will have a stronger risk preference than nondepleted SCR individuals.

I designed two studies to test these hypotheses. In Study 1, I tested the first hypothesis by investigating the relationship between an individual's SCR level and his/her risk preference via a choice of two options in risky-choice scenarios. In Study 2, I tested Hypothesis 2 using a modified dual-task paradigm. I then analyzed overall risk preference and the effect of variation in individuals' SCR level on their risk preference.

Study 1

Method

Participants. I recruited 150 students at Zhejiang University (76 women) as voluntary participants ([M.sub.age] = 19.68, age range 18-26 years). I contacted the students via a poster at the university. All participants reported their ethnicity as Han Chinese, and gave informed consent before taking part.

Procedure. Each participant responded to risky-choice scenarios in a laboratory setting. They were asked to suppose that the risky-choice scenarios were happening in daily life. Participants completed an SCR test and then chose the option they most preferred in each of four randomly sequenced risky choices. When they had finished, participants were thanked and given a notebook as a gift with a value of CNY5 (approximately US$0.60).

Measures

Of the four items in the SCR test (Finkel et al., 2006), two were positively keyed ("energetic" and "lively") and two were negatively keyed and reverse-scored ("exhausted" and "drained"). Participants reported their current SCR level using a 7-point scale ranging from 1 (totally disagree) to 7 (totally agree). The total score on the four items was considered to be the participant's SCR.

In the second part, each risky-choice scenario contained a certain option (A) and an uncertain option (B). To exclude the effect of too many choices on SCR variance (Vohs et al., 2008), a different group of 93 Zhejiang University students were presented with 10 risky-choice scenarios in a preexperiment and, of these, four paired risky-choice options were selected for use in Study 1. As the preexperiment participants showed a close to 50% chance of choosing Option A, the attractiveness of Options A and B was equal. The four risky-choice scenarios were: Scenario 1, Option A: Individual is sure to receive CNY100; Option B: A 50% chance of receiving CNY290. Scenario 2, Option A: Individual is sure to receive CNY200; Option B: A 75% chance of receiving CNY420. Scenario 3, Option A: Individual is sure to receive CNY40; Option B: A 50% chance of receiving CNY95. Scenario 4, Option A: Individual is sure to receive CNY10; Option B: A 50% chance of receiving CNY25. Participants scored 1 point each time they chose Option B in a scenario. A higher score indicated a higher risk preference and vice versa.

In the four scenarios, there was a 50% chance of choosing Option A or B, so the risk preference of 93 participants was in accordance with normal distribution (from 0 to 4), and the theoretical mean risk preference was 2. Because the results of Study 1 and Study 2 indicated that the average risk preference of participants was close to 2, further analysis of the relationship between SCR and risk preference was supported.

Data Analysis

In terms of the measurement of risk preference, the scores of the four risky-choice scenarios received a high validity rating compared with the theoretical expected value of 2, M = 2.09, SD = 1.31, t(149) = 1.19, p = .421. The scores for the SCR test showed high reliability in measuring SCR (Cronbach's [alpha] = .86). The regression analysis indicated that the individual participants' SCR level could predict their risk preference effectively (regression equation: risk preference = .87 + 0.22 x SCR), F(1, 149) = 7.43, p = .007. Thus, Hypothesis 1 was supported.

Study 2

Method

Participants. In this experiment 55 students at Zhejiang University voluntarily participated. They were also recruited via a poster at the university. As three participants did not complete the SCR manipulation task and were thus excluded, I analyzed 52 participants' data (16 women, [M.sub.age] = 21.92, SD = 1.99, age range 19-27 years). All participants reported their ethnicity as Han Chinese, and gave informed consent before taking part.

Procedure. To manipulate SCR, I used a dual-task paradigm (Hagger et al., 2010), in which each participant completed the experiment in a laboratory setting. In step 1, each participant completed an SCR pretest ([SCR.sub.1]). In step 2, each participant completed a control cognitive processing task. In step 3, each participant completed an SCR posttest ([SCR.sub.2]), and in step 4, each participant responded to four risky-choice scenarios and completed the Chinese version of the Positive and Negative Affect Scales (PANAS; Chen & He, 2011) to measure risk preference and emotion, respectively.

Measures and Materials

I tested SCR by ASCR ([SCR.sub.2]--[SCR.sub.1]). The four items used for measuring [SCR.sub.1] and [SCR.sub.2] were the same as for Study 1 and were randomly sequenced.

Control cognitive processing task. When each participant had finished the [SCR.sub.1] test in the laboratory, the experimenter explained that the purpose of the experiment was to "Explore factors that interfere with everyday problem solving, such as identifying an incorrectly placed letter in a series of sentences in English, and then develop treatments for learning disabilities such as attention deficit hyperactivity disorder." After they had read the introduction, participants were given a piece of A4 paper with 22 lines of meaningless English letters typed on it in 8-point font. The first 11 lines were printed in 50% black and the other 11 lines were printed in 35% black to better manipulate SCR (Tyler, 2008). The experimenter asked participants to find the letter "e" with the following rules: (1) Find all instances of the letter "e"; (2) Look at the next letter after the "e", and if this letter is one of the following four vowels--"a", "i", "o", or "u," circle the letter "e" (e.g., "[??]a", "[??]i", "[??]o", or "[??]u"). If the letter after the "e" is a consonant, draw a deletion line through the "e" (e.g., "[??]n"). This task contained the following vowel pairings: 40 "ea", 10 "ei", 20 "eo", and 10 "eu", and "e" was paired with consonants in 120 instances, This task took approximately 8 minutes to complete and proved to be effective in manipulating participants' SCR (Tyler, 2008). After participants finished the SCR manipulation task, each participant reported their SCR posttest score ([SCR.sub.2]).

Risk preference was tested by the four randomly sequenced risky-choice scenarios used in Study 1.

Emotion was tested by the 10-item PANAS, ranging from 1 (totally disagree) to 7 (totally agree). In this study, Cronbach's [alpha] for this scale was .89.

Data Analysis

The results of the SCR test showed that it had high reliability in measuring [SCR.sub.1] and [SCR.sub.2]. (Cronbach's [alpha] were .85 and .84, respectively). There was no significant gender difference in terms of age, [M.sub.men] = 22.08, SD = 2.13; [M.sub.women] = 21.56, SD = 1.63; t(50) = 0.87, p = .389; risk preference, [M.sub.men] = 2.36, SD = 1.36, [M.sub.women] = 1.81, SD = 1.38; t(50) = 1.34, p = .186; positive affect, [M.sub.men]= 11.91, SD = 4.15, [M.sub.women] = 21.31, SD = 3.50; t(50) = 1.87, p = .071; or negative affect, [M.sub.men] =13.89, SD = 5.21, [M.sub.women] = 12.00, SD = 4.35; t(50) = 1.26, p = .212.

The [SCR.sub.1] scores of the 52 participants were significantly higher than their [SCR.sub.2] scores, M[SCR.sub.1] = 19.15, SD = 3.83; M[SCR.sub.2] = 17.51, SD = 3.72; t(51) = 4.34, p < .001, and their risk preference was higher than the theoretical value of 2, but this difference was not significant (M = 2.19, SD = 1.37; t(51) = 1.01, p = .32). These results indicated an ego-depletion effect and that an increase in risk preference at a group mean level was nonsignificant.

Data were then classified into two groups by means of [DELTA]SCR ([SCR.sub.2] -[SCR.sub.1]): One group was labeled the depleted SCR group ([DELTA]SCR < 0, n = 30), and the second was the nondepleted SCR group ([DELTA]SCR [greater than or equal to] 0, n = 22). As shown in Table 1, there were no significant differences between the depleted and nondepleted SCR groups in terms of age, F(1, 51) = 3.14, p = .082; positive affect, F(1, 51) = 0.98, p = .326; negative affect, F(1, 51) = .005, p = .946; or [SCR.sub.2], F(1, 51) = 1.39, p = .244, but there was a significant difference in risk preference between the depleted and nondepleted groups, F(1, 51) = 4.70, p = .035, with depleted SCR participants being less likely to show a risk preference when presented with a risky choice. Thus, Hypothesis 2 was supported.

However, there was a significant difference between the depleted and nondepleted groups in terms of [SCR.sub.1], F(1, 51) = 7.12, p = .01. The depleted SCR group had significantly decreased SCR, t(29) = 7.64, p < .001 and the nondepleted SCR group had significantly increased SCR, t(21) = - 3.31, p = .003. This means that, when completing a self-control task, individuals with relatively higher SCR were prone to activating their SCR, whereas those with relatively lower SCR were prone to deplete their SCR.

Discussion

My approach in this study differed from the limited strength model of self-control, in that I supposed that (a) individuals' baseline level of SCR may affect their subsequent self-control behavior, (b) the effect of SCR depletion or nondepletion is determined by individuals' SCR level rather than group SCR level, and (c) individuals' SCR level and the variance in this level would have a separate impact on self-control behavior.

I designed two studies to test the two hypotheses with respect to risk preference. The results supported my suppositions by demonstrating that (a) there was a relationship between the participants' SCR level and the level of their risk preference, (b) SCR is relatively unlimited when individuals exert self-control to regulate their behavior, and (c) in a dual-task paradigm individual variations in SCR had greater predictive power than did group variations.

These results may help clarify previous contradictory findings. That is, Bruyneel et al. (2009), Freeman and Muraven (2010), and Price and Yates (2010) reported that members of depleted SCR groups had a higher risk preference, whereas Unger and Stahlberg (2011) held the opposing view. I suggest that these researchers should have analyzed the variance in individuals' SCR level and compared the effect on risk preference of individuals' SCR depletion and nondepletion, as it was this procedure that helped me to clarify the data. In addition, future researchers may benefit from considering the SCR level of each participant before risky decisions are made. This would be helpful in interpreting Unger and Stahlberg's results, because the preliminary tasks of reading, comparing, and analyzing investment information may also have depleted the participants' SCR.

Theoretical and Practical Implications

There are far-reaching implications in the results, which support an unlimited resource theory of self-control. First, the results of the two studies indicate that each individual's SCR variance was determined by his/her primary SCR level, such that individuals with relatively high SCR may have had their SCR depleted, but those with relatively low SCR may have experienced SCR adaptation or activation. Indeed, Baumeister and Alquist (2009) ascertained that humans may activate implicit self-regulation resources to exert successful self-control when SCR is scarce. Therefore, I suggest that SCR is relatively unlimited.

Second, there are also implications in the findings of the two studies for the mechanism of how both the level and variance in individuals' SCR affect risky behavior. When the participants had not completed a task involving self-control (Study 1), their risky behavior was determined by their SCR level, whereby individuals with a higher SCR level than others had shown a higher preference (perceived greater control over risk) for risky behavior. However, when participants had just completed a task involving self-control (Study 2), their risky behavior was determined by the variance in their SCR level. Individuals with depleted SCR showed a higher preference for risky behavior, suggesting that they may have been unable to resist the enticement of more money. Thus, individuals had to compensate for the "loss," that is, negative state, by showing a higher risk preference.

Third, the results of Study 2 excluded the effect of emotion on risky behavior because negative and positive affect did not differ between depleted SCR and nondepleted SCR participants. This means that individuals' SCR does not vary with emotion (Hagger et al., 2010). Nevertheless, as Hagger et al. (2010) found that controlling emotion was an effective way of exhausting SCR, future researchers should compare emotional conditions for differences in SCR and, in order to investigate the effect on risk behavior of emotion as well as SCR variance, a control condition should be included.

In addition, the question remains whether or not there are cultural differences regarding SCR level and variance.

I suggest that there are no cultural differences. If self-control is a limited resource, each individual should have a different level of SCR at different times, and SCR should be depleted or nondepleted depending on the individual's exertion of self-control. In future, I will test both individual and cultural differences in SCR level and variance to further elucidate the mechanism of how they affect self-control behavior.

Limitation

A limitation in this study was that the nondepletion of SCR was passively activated. Active methods, such as drinking glucose (Gailliot et al., 2007), meditation (Friese, Messner, & Schaffner, 2012), and real monetary stimulus (Boucher & Kofos, 2012) have been shown to be highly effective in replenishing SCR. Future researchers should adopt these techniques to replenish and enhance individual SCR levels and also to investigate the effect of active SCR enhancement on individuals' risk behavior.

http://dx.doi.org/10.2224/sbp.2014.42.8.1335

References

Baumeister, R. F. (2002). Yielding to temptation: Self-control failure, impulsive purchasing, and consumer behavior. Journal of Consumer Research, 28, 670-676. http://doi.org/dphtf7

Baumeister, R. F., & Alquist, J. L. (2009). Is there a downside to good self-control? Self and Identity, 8, 115-130. http://doi.org/d65v3d

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, 1252-1265. http:// doi.org/bszcmm

Baumeister, R. F., & Heatherton, T. F. (1996). Self-regulation failure: An overview. Psychological Inquiry, 7, 1-15. http://doi.org/bbv42g

Berns, G. S., Laibson, D., & Loewenstein, G. (2007). Intertemporal choice--Toward an integrative framework. Trends in Cognitive Sciences, 11, 482-488. http://doi.org/dwmndk

Boucher, H. C., & Kofos, M. N. (2012). The idea of money counteracts ego depletion effects. Journal of Experimental Social Psychology, 48, 804-810. http://doi.org/s32

Bruyneel, S. D., Dewitte, S., Franses, P. H., & Dekimpe, M. G. (2009). I felt low and my purse feels light: Depleting mood regulation attempts affect risk decision making. Journal of Behavioral Decision Making, 22, 153-170. http://doi.org/dw7dmf

Chen, H.-X., & He, G.-B. (2011). The effect of construal level on intertemporal choice and risky choice. Acta Psychologica Sinica, 43, 442-452. http://doi.org/s34

Converse, P. D., & DeShon, R. P. (2009). A tale of two tasks: Reversing the self-regulatory resource depletion effect. Journal of Applied Psychology, 94, 1318-1324. http://doi.org/c9ftpn

Dou, K. (2013). The study of how ego-depletion affects decision behavior: Evidence from behavior andERP. Unpublished master's thesis [In Chinese]. Guangzhou University, Guangzhou, China.

Finkel, E. J., Campbell, W. K., Brunell, A. B., Dalton, A. N., Scarbeck, S. J., & Chartrand, T. L. (2006). High-maintenance interaction: Inefficient social coordination impairs self-regulation. Journal of Personality and Social Psychology, 91, 456-475. http://doi.org/bhxmss

Freeman, N., & Muraven, M. (2010). Self-control depletion leads to increased risk taking. Social Psychological & Personality Science, 1, 175-181.

Friese, M., Messner, C., & Schaffner, Y. (2012). Mindfulness meditation counteracts self-control depletion. Consciousness and Cognition, 21, 1016-1022. http://doi.org/s33

Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., Plant, E. A., Tice, D. M.,... 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, 325-336. http://doi. org/bvj3t8

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, 495-525. http://doi.org/ btsrsr

Hofmann, W., Vohs, K. D., & Baumeister, R. F. (2012). What people desire, feel conflicted about, and try to resist in everyday life. Psychological Science, 23, 582-588. http://doi.org/s36

Kahneman, D., Slovic, P., & Tversky, A. (Eds.). (1982). Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press.

Muraven, M. (2008). Autonomous self-control is less depleting. Journal of Research in Personality, 42, 763-770. http://doi.org/cqq95s

Price, D. A., & Yates, G. C. R. (2010). Ego depletion effects on mathematics performance in primary school students: Why take the hard road? Educational Psychology, 30, 269-281. http://doi.org/ dw5hr3

Sun, Y.-J. (2008). The effect of ego depletion on the performance of athletes [In Chinese]. Unpublished doctoral dissertation, Beijing Sport University, Beijing, China.

Tyler, J. M. (2008). In the eyes of others: Monitoring for relational value cues. Human Communication Research, 34, 521-549. http://doi.org/fcgd9g

Unger, A., & Stahlberg, D. (2011). Ego-depletion and risk behavior: Too exhausted to take a risk. Social Psychology, 42, 28-38. http://doi.org/cg7qmf

Vohs, K. D., Baumeister, R. F., Schmeichel, B. J., Twenge, J. M., Nelson, N. M., & Tice, D. M. (2008). Making choices impairs subsequent self-control: A limited-resource account of decision making, self-regulation, and active initiative. Journal of Personality and Social Psychology, 94, 883-898. http://doi.org/fsmw86

XIANG-HUI YAN

Zhejiang University

Xiang-hui Yan, Department of Psychology and Behavioral Sciences, Zhejiang University. This research was supported by the National Natural Science Foundation of China (71271189).

Correspondence concerning this article should be addressed to: Xiang-hui Yan, Room 302, Department of Psychology and Behavioral Sciences, Xixi Campus, Zhejiang University, No. 148 Tianmushan Road, Hangzhou, Zhejiang 310028, People's Republic of China. Email: ynnuwhu@163.com
Table 1. Differences Between Depleted SCR and Nondepleted SCR Groups
in Terms of Age, Affect, [SCR.sub.1], [SCR.sub.2], and Risk Preference

                  Depleted SCR group   Nondepleted SCR group

                      M     SD               M      SD

Age                22.33   2.16           21.36    1.62
Positive affect    20.27   3.90           19.13    4.27
Negative affect    13.26   7.73           13.36    5.47
[SCR.sub.1]        20.30   2.13           17.00    3.62
[SCR.sub.2]        17.59   4.07           18.23    3.83
Risk preference    2.53    1.31           1.73     1.35
COPYRIGHT 2014 Scientific Journal Publishers, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Yan, Xiang-Hui
Publication:Social Behavior and Personality: An International Journal
Article Type:Report
Date:Sep 1, 2014
Words:4008
Previous Article:Modulation of facial processing by identity information in a face similarity evaluation task.
Next Article:The power of motivation-goal fit in predicting entrepreneurial persistence.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |