Does the Color of Feedback Affect Investment Decisions? - IGI Global

International Journal of Applied Behavioral Economics, 2(3), 15-26, July-September 2013 15

Does the Color of Feedback Affect Investment Decisions? Tal Shavit, The School of Business Administration, The College of Management Academic Studies, Israel Mosi Rosenboim, The School of Business Administration, The College of Management Academic Studies, Israel Chen Cohen, Department of Economics and Management, The Open University of Israel, Raanana, Israel

ABSTRACT This paper presents a multi-period experiment that extends a classic experiment on investment allocation preferences by adding colors to the feedback returned to participants. The results show that investors allocate the same proportion of their investment to the stock and the bond funds without regard to the colors. However, red feedback activates an avoidance motivation (vs. an approach motivation), and this reduces chasing past returns. The authors also found that the color of the feedback affected the time needed to make a decision. Financial institutions might use colored feedback to encourage approach or avoidance motivations in their clients. Keywords:

Allocation Task, Approach, Avoidance, Colors, Feedback

INTRODUCTION A great deal of research has been conducted over the past century focusing on the physics, physiology, and psychology of color. Yet little is currently known regarding the effect of color on psychological functioning (Fehrman & Fehrman, 2004; Whitfield & Wiltshire, 1990), and the results of research on this issue are inconsistent. Some have proposed that red enhances cognitive task performance, as compared with blue or green (Kwallek & Lewis, 1990), while others have shown exactly the opposite DOI: 10.4018/ijabe.2013070102

(Elliot, Maier, Moller, Friedman, & Meinhardt, 2007; Soldat, Sinclair, & Mark, 1997). Studies that examined the effect of red color (Elliot et al., 2007; Maier, Elliot, & Lichtenfeld, 2008) documented a link between red and avoidance motivation or behavior. Avoidance motivation is taking (or not taking) action to avoid something unpleasant. It is a defense mechanism by which a person avoids unpleasant situation or unpleasant feelings like loss. An opposite behavior is the approach motivation. Approach motivation is taking action because you desire something “good” to come into your life. In approach motivation, which is activated by the colors blue or green, behavior is directed by

Copyright © 2013, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

16 International Journal of Applied Behavioral Economics, 2(3), 15-26, July-September 2013

positive situation or positive feelings like gain (Elliot & Covington, 2001). Elliot et al. (2007) focused on the effect of color on performance in achievement contexts. They conducted six experiments to test the hypothesis that that red impairs performance in achievement contexts, and that it does so in non-conscious fashion. They wrote, “achievement contexts are situations in which competence is evaluated and both positive outcomes (success) and negative outcomes (failure) are possible” (p. 156). They proposed that in achievement contexts red is associated with the psychological danger of failure. Over time, people learn to link red and danger in many contexts with possible negative outcomes. The results of their experiments provided strong support for their hypothesis about the effect of red on performance. Specifically, they demonstrated that the perception of red prior to an achievement task, evokes an avoidance motivation more often than a perception of green. They suggested that red carries the meaning of danger, specifically, the psychological danger of failure. Experimental results obtained by Elliot, Maier, Binser, Friedman, and Pekrun (2009) indicated that a brief perception of red in an achievement context induces avoidance behavior, and does so without conscious awareness. They suggested that one source of the link between red and danger is the societal association between red and danger in situations where negative possibilities are salient, such as stop signs, fire alarms, and warning signals. Another source is biologically based predisposition across species who interpret red as a danger signal in competitive situations (e.g. Setchell & Wickings, 2005). Mehta and Zhu (2009) conducted a series of six studies, using various tasks in several different domains. They demonstrated that red (vs. blue) can activate an avoidance motivation (vs. an approach motivation), and subsequently enhance performance on detail-oriented cognitive tasks (vs. creative tasks). Although there are many studies on the effect of color on decision making, to the best of our knowledge, there is no research testing

the influence of color on financial decision making or assets allocation. This is surprising since it is common in financial markets to use different colors for gain and loss. Usually red is used for losing assets, and green for gaining assets. Based on the hypothesis that red can activate an avoidance motivation and green can activate approach motivation, it is interesting to test how the different colors change the behavior of decision makers in an investment context. We suggest that beside the effect of losses or gains, decision makers are also affected by the color in which the losses or the gains are presented. Specifically, a return on the decision maker’s investment which is presented in red will activate avoidance behavior. Avoidance behavior in the context of investment means that the decision maker might take less risky action in his portfolio to avoid the unpleasant feelings of loss or to avoid any potential regret from loss (Bar-Hillel & Neter, 1996; Loomes & Sugden, 1982). On the other hand, a return on the decision maker’s investment which is presented in green will activate approach motivation. The approach motivation in the context of investment means that decision makers are more active in their portfolio and possibly will take more risk. The investment task is different form learning tasks or other assignments that were used in previous studies on the effect of colors on decision making. As mentioned above, Elliot et al. (2007) found that red impairs performance in achievement contexts. We suggest that investment task is an achievement task, because it is a task in which competence is evaluated, and both positive outcomes and negative outcomes are possible. To test the effect of colors on the decision makers’ investment behavior, we conducted a multi-period experiment using the same assets, bond fund (BF) and stock fund (SF) as in Thaler, Tversky, Kahneman, and Schwartz (1997). We explore how the color of feedback on investments’ return might affect investment strategy. We conducted four experimental treatments with different subjects in each treatment. Each subject in each group allocated funds between the assets for 120 experimental periods. Differ-



ent colors were used to present the feedback in each treatment. In three treatments, the returns after each period were shown in single color: red, green and a neutral color (black). There were no other differences between the treatments, which enabled us to isolate the effect of the colors. Specifically, in the green treatment both losses and gains were shown in green; and in the red treatment, both losses and gains were red. This made it possible to compare the behavior of the decision makers when exposed to different colors, and neutralize the effect of loss or gain. If the colors have no effect, we would expect to find the same results in all the treatments. We also conducted a forth treatment in which the loss were shown in red and gains in green, as in real life situations. The experimental results reveal that the decision makers tended to invest according to an extreme version of the diversification 1/n heuristic (Benartzi & Thaler, 2001) throughout the entire experiment (investing close to 50% in each of the funds) without any relation to the color of the feedback. However, we did find that the use of red induces avoidance behavior, in the sense that subjects tended to make fewer changes in their allocation when the feedback was presented in red, compared to treatments when the feedback was presented in green or black. The remainder of this paper is organized as follows: The next section presents the theoretical hypotheses based on previous literature. It is followed by a section presenting the experimental procedures, and another that presents the results. The final section summarizes and concludes.

HYPOTHESES First, we refer to the way investors allocate their fund between the assets. Some findings show that people use different heuristics for making decisions on different occasions (Kahneman & Frederick, 2002; Tversky & Kahneman, 1974). According to the 1/n heuristic, when facing a decision about building a portfolio, investors divide the money equally among the alternatives suggested. Benartzi and Thaler (2001) showed

that the composition of a pension (i.e. the division between stocks and bonds) was greatly influenced by the investment options offered to employees. For example, researchers gave questionnaires to employees of the University of California, and asked them to choose how to allocate monthly deposits in their pension fund between two different funds. When the researchers offered the choice of a BF and an SF, the average allocation to shares was 54%. When employees were asked to choose between an SF and a “balanced” fund (which invests 50% in BF and 50% in SF), the average allocation to shares was 73%. When asked to choose between a BF and a “balanced” fund, the average allocation to the SF fell to 35%. In addition, researchers examined the diversity bias on the basis of real data obtained about the savings of 1.56 million people, and found a clear correlation between the investment options offered to employees and their allocation among stocks and bonds in the pension fund. Harry M. Markowitz (quoted in Zweig, 1998) told researchers: I should have computed the historical covariance of the asset classes and drawn an efficient frontier. Instead I visualized my grief if the stock market went way up and I wasn’t in it, or if it went way down and I was completely in it. My intention was to minimize my future regret, so I split my (pension scheme) contributions 50/50 between bonds and equities. (p. 114) Hypothesis 1: Feedback color effect: H1: Subjects will allocate their fund equally between assets following the 1/n heuristic without any relation to the feedback color. Next, we refer to the way investors are affected by the feedback. Investors tend to rely on small samples of past experiences to decide upon their action, meaning that they chase past returns (Benzion, Erev, Haruvy, & Shavit, 2010; Chevalier & Ellison, 1997; Erev & Barron, 2005; Hendricks, Patel, & Zeckhauser, 1993; Ippolito, 1992; Johnson, Tellis, & Macinnes, 2005; Sirri & Tufano, 1998). Choi, Laibson,



Madrian, and Metrick (2009) showed that investors who experience higher returns on their retirement savings increase those savings more than investors who experience less rewarding outcomes. In three laboratory experiments and one quasi-field experiment, Benzion et al. (2010) showed that adaptive investors who look to the past to adjust expectations about future returns will shun diversified funds. That is, an adaptive reaction to feedback implies underdiversification when the investor gets complete feedback on the performance of the diversified fund as well as its components in a given period Chasing past returns is also closely related to market overreaction (Chopra, Lakonishok, & Ritter, 1992; De Bondt & Thaler; 1985, 1990; Nosic & Weber, 2009; Offerman & Sonnemans, 2004) where demand overshoots in response to positive recent returns. As mentioned above, some studies have documented a link between red and avoidance motivation (Elliot et al., 2007; Maier et al., 2008; Mehta & Zhu, 2009). Specifically, they demonstrate that red (vs. blue) can activate an avoidance motivation (vs. an approach motivation), meaning that individuals will avoid a potentially unpleasant experience. Elliot et al. (2009) suggest that avoidance behavior may take on many different forms, including withdrawal, freezing, inhibition, and escape (Atkinson, 1964; Faneslaw, 1991; Lewin, 1935). These forms of behavior indicate that decision makers tend to take less action when the feedback is red. As a result, we expect that decision makers will make fewer changes in their portfolio when the avoidance motivation is activated. Hypothesis 2: Effect of red: H2: Feedback displayed in red can activate an avoidance motivation (vs. an approach motivation) and, as a result, decision makers will make fewer changes in their allocations each period, relative to the previous period. According to Mehta and Zhu (2009), “Faster reaction time to approach-related (or avoidance-related) anagrams would imply a stronger activation of an approach (or avoid-

ance) motivation” (p. 1227). This means that when red or green are used in the feedback screen, the reaction time will be shorter than if a neutral color is used, because green and red lead to activation and avoidance approaches, respectively. Hypothesis 3: Reaction time effect: H3: The reaction time to red or green feedback is shorter than that for neutral color feedback.

THE EXPERIMENTS The main goal of the experiments presented below is to clarify the relationship between the predictions presented in our hypotheses and human behavior.

Participants The participants in the experiments were 80 (45 male and 35 female) Israeli undergraduates, who received payment based on their decisions in the experiment. The mean age of participants was 25 years old with a range of 21-30 years.

Design and Procedure The experiments were conducted in a computer laboratory at a major university in Israel. Participants were randomly assigned to one of four between-subjects experimental investment feedback treatments: (a) Black (neutral) condition: the feedback on returns from assets was black; (b) Red condition: the feedback on returns from assets was red; (c) Green condition: the feedback on returns from assets was green, and (d) Mixed condition: the feedback on losses was red and feedback on gains was green (as in real life situations). The decision-making task was presented as an investment problem. The participants were asked to allocate 100 tokens between two assets in each period. The basic task in the experiment was repeated for 120 periods. Subjects received feedback on their investment after each period, with the feedback received by each group presented in different colors. After each period, participants observed



their earnings from all assets. Participants could not allocate any money accumulated in previous period. They were asked to allocate 100 tokens in each period, independent of the outcome of their allocation in any of the previous periods. This means that participants could not hold the assets after they lost or decide to sell gaining or losing assets (as in Thaler et al., 1997; Gneezy & Potters, 1997, Barron & Erev, 2003). Using this procedure, there is no disposition effect, which suggests that investors tend to sell winning transactions too soon and hold losing transactions too long (Chui, 2001; Shafran, Benzion, & Shavit, 2009; Shefrin & Statman, 1985; Weber & Camerer, 1998), because participants could not hold the assets or sell them. Even though each period was a separate decision, and participants could not hold the assets after they lost or decide to sell gaining or losing assets, we still expect (as mentioned above) that the return in each period will affect the allocation decision in the next period. The two assets were SF (stock funds) and BF (bond funds) with the same distributions as in Thaler et al. (1997), however, we did not use the terms “stocks” and “bonds” to prevent a labeling effect. The participants were offered: (1) Asset A, which was a SF with normal distribution, an expected return of 1% and standard deviation of 3.54% and (2) Asset B, which was a BF with normal distribution, an expected return of 0.25% and standard deviation of 0.177%. The return on this asset was limited to zero to prevent a negative return. The returns on each asset for each period were selected randomly by the computer from the possible distributions. The subjects were told that the tokens they earned would be converted to money at the end of the experiment, and that the final payment would be relative to their earnings in the experiment. The payment was 10% of the total earnings in New Israeli Shekel (NIS) plus a show up fee of NIS 20. The exchange rate at the time of the experiment was approximately NIS 3.7 to USD 1. Note that one might argue that this payment mechanism could be problematic, since it may influence the choices that participants make at

the very end. If they already have earnings, they may become much more risk-averse towards the end. If they are losing, there is an implied incentive for them to take excessive risks toward the end of the experiment. However, this payment mechanism was used in other studies such as Thaler et al. (1997), Gneezy and Potters (1997), Barron and Erev (2003) and Benzion et al. (2010), and we followed these studies. Moreover, our study is a between-subjects design, meaning that we compare the results of the four treatments with the same payment mechanism. If there is a bias in the payment mechanism its relevance and effect will be similar in all treatments. The instructions (See the appendix for the instructions) spelled out the experimental procedure, in a simple, non-technical manner but without giving any information about the actual payoff distribution for each asset. After distributing the written instructions, we gave participants time to review the experimental procedure carefully and ask questions, to be certain that they understood. The reaction times were measured for each subject. The first reaction time was the time that each participant spent on the decision how to allocate the endowment. The second reaction time was the time each participant spent looking at the feedback screen.

RESULTS The Single-Color Treatments First, we analyzed the effect of red and green colors in the single color treatments, by comparing them to the neutral color (black) treatment. The average (and STDV) allocations to SF were 53.67 (21.56), 50.38 (19.3) and 52.11 (25.78) for the neutral, green and red treatments, respectively. There was no significant difference between the average allocation between the red treatment and the green treatment (one-tailed Mann-Whitney Test, Z = 0.33, p = 0.38) or between the neutral treatment and the other treatments (one-tailed Mann-Whitney Test, Z = 0.62, p = 0.27 and Z = 0.22, p = 0.42 for green



and red treatments, respectively). In all of the conditions, the average allocation to the risky assets is not significantly different from 50 (one-tailed T-Test, t(19) = 0.76, p = 0.23; t(19) = 0.09, p = 0.47 and t(19) = 0.37, p = 0.36 for black, green and red treatments respectively). The results in all the treatment are consistent with Hypothesis 1 which suggests that subjects will allocate their tokens equally between assets without any relation to the feedback color. In Table 1, we present the average proportion of periods in which subjects did not change their allocation to the risky asset from one period to another. We also show the average proportion of times each subject changed the allocation to and from the previous return, out of the periods that subjects changed the allocation (and compare them using Wilcoxon Signed Ranks Test). We distinguish between changes toward a previous return and away from a previous return. Change toward previous return is (1) if the previous return is positive, and the allocation to the risky asset increases from the previous allocation, or (2) if the previous return is negative, and the allocation to the risky asset decreases from the previous allocation. Change away from previous return is (1) if the previous return is positive, and the allocation to the risky asset decreases from the previous allocation or (2) if the previous return is negative, and the allocation to the risky asset increases from the previous allocation. The proportion of no change is higher for the red treatment than for the green treatment (one-tailed Mann-Whitney Test, Z = 1.88, p =

0.03) or for the neutral treatment (one-tailed Mann-Whitney Test, Z = 1.88, p = 0.03). No difference was found between the green and the neutral treatments (one-tailed Mann-Whitney Test, Z = 0.20, p = 0.42). These results are consistent with the avoidance motivation since subjects in the red treatment did not choose to change their allocation more frequently, consistent with Hypothesis 2 and the avoidance approach. This means that participants receiving red feedback tended to freeze and inhibit behavior more than in the other treatments. From the total periods with changes, the proportion of changes towards previous return is lower for the neutral treatment than the green treatment (one-tailed Mann-Whitney Test, Z = 1.95, p = 0.03) and the red treatment (one-tailed Mann-Whitney Test, Z = 1.49, p = 0.07). No difference was found between the green and the red treatments (one-tailed Mann-Whitney Test, Z = 0.14, p = 0.45). Only in the neutral condition no difference was found between the changes to and away from previous return. In the color treatments, when deciding to react, subjects tended to change their allocation more to the previous return. These results indicate that colors encourage result-chasing behavior while neutral feedback encourages more random response to the feedback. Next, we present the average time each subjects spent on the decision how much to allocate and the time spent on looking at the feedback after each period (Table 2). The reaction (decision) time in the neutral treatment is longer than the reaction time in

Table 1. The average (STDV) proportion of periods each subject did not change his or her allocation or changed the allocation to the risky asset Out of periods with change Condition

No change

Toward return

Away from return

Wilcoxon Signed Ranks Test Z (sig’)

Neutral

0.35 (0.25)

0.53 (0.18)

0.47 (0.18)

0.56 (0.57)

All green

0.32 (0.23)

0.61 (0.13)

0.39(0.13)

3.06 (0.00)

All red

0.48 (0.23)

0.58 (0.15)

0.42 (0.15)

2.11 (0.04)

* STDV in brackets



Table 2. Average time in seconds (STDV) for decision and feedback Condition

Decision

Feedback

Neutral

5.67 (1.83)

2.70 (1.19)

All green

4.73 (1.22)

2.33 (1.37)

All red

4.43 (1.49)

2.15 (1.24)

* STDV in brackets

both the green treatment (one-tailed MannWhitney Test: Z = 1.92, p = 0.03) and in the red treatment (one-tailed Mann-Whitney Test: Z = 2.22, p = 0.01). No difference in reaction time was found between the red and the green treatments (one-tailed Mann-Whitney Test: Z = 0.95, p = 0.18). The time spent on feedback was longer in the neutral treatment than in the red treatment (one-tailed Mann-Whitney Test: Z= 1.79, p= 0.04) but not significantly different from the green treatment (one-tailed MannWhitney Test: Z= 1.19, p= 0.12). No significant difference in the time spent on feedback was found between the red and the green treatments (one-tailed Mann-Whitney Test: Z= 0.20, p= 0.42). The results regarding the reaction time support Hypothesis 3, because subjects spent more time on the decision how to allocate in the neutral treatment than in the red or green treatments. The findings regarding the time spent on the feedback support Hypothesis 3 only for the avoidance approach suggesting that the reaction time to red feedback is shorter than that of neutral color feedback.

the mixed treatment is 56.55 (STDV = 17.67), which is not significantly different (one-tailed Mann-Whitney Test: Z= 1.19, p= 0.12) from the green treatment which is 50.38 (STDV = 19.3). The proportion of no change in the mixed treatment is 0.41 (STDV = 0.24) and slightly higher (one-tailed Mann-Whitney Test, Z = 1.26, p = 0.105) than the proportion of no change in the green treatment which is 0.32 (STDV = 0.23). The decision time and the feedback time were 4.49 seconds (STDV = 1.13) and 2.17 seconds (STDV = 0.94), respectively. There are no significant differences between the reaction times in the mixed and the green treatments (onetailed Mann-Whitney Test: Z= 0.49, p= 0.31 and Z= 0.05, p= 0.48 for the decision time and the feedback time, respectively). The existence of red in some of the periods (only when there is a loss in the SF) in the mixed treatment had a slight effect on the decision makers, slightly increasing the rate of no change.

The Mixed-Color Treatment

This study shows that the decision makers tended to allocate their money between risky and riskless assets according to an extreme version of the diversification 1/n heuristic (Benartzi & Thaler, 2001). They invest on average close to 50% of their fund to each of the assets without regard to the color of the feedback. However, the use of red induces avoidance behavior, in the sense that subjects tended to make fewer changes in their allocation when presented with red feedback than when the feedback was presented in green or black. We even found that the existence of red only when there is a loss (as in real life situation), has a slight effect

In the mixed-color treatment losses were presented in red, and gains in green consistent with the practice in financial markets. When there was a gain in the SF, the feedback was presented in green, but the BF is always in green because there is no possibility of loss. When there is a loss in the SF, both colors appear in the feedback screen, SF in red and BF in green. The question is how the presence of red in the loss domain affects the subject’s behavior, and if this behavior differs from the green treatment. The average allocation to SF in

CONCLUSION



on the decision maker by reducing changes in their allocation. The real-life implication of this study is that presenting feedback on an investment in red induces avoidance behavior and, as a result, participants change their allocations less frequently. Financial institutions such as banks, mutual funds or pension funds could adopt our findings, and use red if they want their clients to react according to an avoidance motivation or green if they want their clients to react according to an approach motivation. For example, pension fund managers prefer to induce an avoidance motivation in their clients since changing investments each period is not the best strategy for long-term investments. On the other hand, fund managers who profit from the clients’ transactions might want to encourage their clients to adopt an approach motivation, and make more changes. Our contribution to the existent literature is that we are the first who look at the effect of colors on financial decision making. However, our findings are limited to laboratory conditions but do indicate some different behaviors for different colors. Future research might examine our findings in real life situations, and with professional investors. It would also be interesting to examine the influence of other colors on investors’ behavior in the future. We leave this issue for future research to explore. Another technical implication of this study is the use of different colors when designing experiments. Elliot et al. (2009) suggest that red and green are often used in experiments as cues signaling loss/failure/bad and gain/ success/good, respectively (e.g., Dijksterhuis & Smith, 2002; Förster, Higgings & Idson, 1998; Guarnaschelli, McKelvey, & Palfrey, 2000; Rothermund, Wentur, & Bak, 2001; Trommershauser, Maloney, & Landy, 2003). Regarding experimental studies, Elliot et al. (2009) argue, “This use of red and green unwittingly confounds the effect of motivational framing with the effect of color” (p. 372). They also warn, “It is undoubtedly the case that color not only varies by hue but also by lightness and chroma, thereby adding further extraneous variance to the data. Red has a systematic influence

on behavior in achievement contexts indicates that red, and color more generally, should be used with great care in experimental designs, assessment procedures, and beyond” (p. 372). We show that the conclusions from experimental studies can be affected by the color of the feedback. For example, from the neutral treatment one would conclude that subjects are not affected by previous return, because there is no difference between change to or away from the previous return. However, when looking at the treatments using colors, one would conclude that subjects are chasing past returns, consistent with the literature.

REFERENCES Atkinson, J. W. (1964). An introduction to motivation. New York, NY: D. Van Nostrand Company, Inc.. Bar-Hillel, M., & Neter, E. (1996). Why are people reluctant to exchange lottery tickets? Journal of Personality and Social Psychology, 70, 17–27. doi:10.1037/0022-3514.70.1.17. Barron, G., & Erev, I. (2003). Small feedback-based decisions and their limited correspondence to description-based decisions. Journal of Behavioral Decision Making, 16, 215–233. doi:10.1002/bdm.443. Benartzi, S., & Thaler, R. E. (2001). Naïve diversification strategies in defined contribution saving plans. The American Economic Review, 91(1), 79–98. doi:10.1257/aer.91.1.79. Benzion, U., Erev, I., Haruvy, E., & Shavit, T. (2010). Adaptive behavior leads to under diversification. Journal of Economic Psychology, 31, 987–994. Chevalier, J., & Ellison, G. (1997). Risk taking by mutual funds as a response to incentives. The Journal of Political Economy, 105, 1167–1200. doi:10.1086/516389. Choi, J. D., Laibson, B., Madrian, B., & Metrick, A. (2007). (forthcoming). Reinforcement learning and savings behavior. The Journal of Finance. PMID:20352013. Chopra, N., Lakonishok, J., & Ritter, J. R. (1992). Measuring abnormal performance—do stocks overreact? Journal of Financial Economics, 31, 235–268. doi:10.1016/0304-405X(92)90005-I.



Chui, P. M. W. (2001). An experimental study of disposition effect: Evidence from Macau. The Journal of Psychology and Markets, 2(4), 216–222. De Bondt, W. F. M., & Thaler, R. (1985). Does the stock market overreact? The Journal of Finance, 40, 793–805. doi:10.1111/j.1540-6261.1985.tb05004.x. De Bondt, W. F. M., & Thaler, R. (1990). Do security analysts overreact? The American Economic Review, 80, 52–57. Dijksterhuis, A., & Smith, P. K. (2002). Affective habituation: Subliminal exposure to extreme stimuli decreases their extremity. Emotion (Washington, D.C.), 2, 203–214. doi:10.1037/1528-3542.2.3.203 PMID:12899354. Elliot, A. J., & Covington, M. V. (2001). Approach and avoidance motivation. Educational Psychology Review, 13(2), 73–92. doi:10.1023/A:1009009018235. Elliot, A. J., Maier, M. A., Binser, M. J., Friedman, R., & Pekrun, R. (2009). The effect of red on avoidance behavior in achievement contexts. Personality and Social Psychology Bulletin, 35(3), 365–375. doi:10.1177/0146167208328330 PMID:19223458. Elliot, A. J., Maier, M. A., Moller, A. C., Friedman, R., & Meinhardt, J. (2007). Color and psychological functioning: The effect of red on performance attainment. Journal of Experimental Psychology. General, 136, 154–168. doi:10.1037/0096-3445.136.1.154 PMID:17324089. Erev, I., & Barron, G. (2001). On adaptation, maximization, and reinforcement learning among cognitive strategies. Psychological Review, 112(4), 912–931. doi:10.1037/0033-295X.112.4.912 PMID:16262473. Faneslaw, M. S. (1991). The midbrain periaqueductal gray as a coordinator of action in response to fear and anxiety. In A. Depaulis, & R. Bandler (Eds.), The midbrain periaqueductal grey matter: Functional, anatomical, and immunohistochemical organization (pp. 151–173). New York, NY: Plenum. doi:10.1007/978-1-4615-3302-3_10. Fehrman, K. R., & Fehrman, C. (2004). Color: The secret influence (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Förster, J., Higgins, E. T., & Idson, L. C. (1998). Approach and avoidance strength during goal attainment: Regulatory focus and the “goal looms larger effect. Journal of Personality and Social Psychology, 75, 1115–1131. doi:10.1037/0022-3514.75.5.1115 PMID:9866180.

Gneezy, U., & Potters, J. (1997). An experiment on risk taking and evaluation periods. The Quarterly Journal of Economics, 152(2), 631–645. doi:10.1162/003355397555217. Guarnaschelli, S., McKelvey, R. D., & Palfrey, T. R. (2000). An experimental study of jury decision rules. The American Political Science Review, 94, 407–423. doi:10.2307/2586020. Hendricks, D., Patel, J., & Zeckhauser, R. (1993). Hot hands in mutual funds: Short-run persistence in relative performance. The Journal of Finance, 48, 93–130. doi:10.1111/j.1540-6261.1993.tb04703.x. Ippolito, R. A. (1992). Consumer reaction to measures of poor quality: Evidence from the mutual fund industry. The Journal of Law & Economics, 35, 45–70. doi:10.1086/467244. Johnson, J., Tellis, G. J., & Macinnis, D. J. (2005). Losers, winners, and biased trades. The Journal of Consumer Research, 32(2), 324–329. doi:10.1086/432241. Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute substitution in intuitive judgment. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases. New York, NY: Cambridge University Press. doi:10.1017/ CBO9780511808098.004. Kwallek, N., & Lewis, C. M. (1990). Effects of environmental color on males and females: A red or white or green office. Applied Ergonomics, 21, 275–278. doi:10.1016/0003-6870(90)90197-6 PMID:15676782. Lewin, K. (1935). A dynamic theory of personality. New York, NY: McGraw-Hill. Loomes, G., & Sugden, R. (1982). Regret theory: An alternative theory of rational choice under uncertainty. The Economic Journal, 92, 805–824. doi:10.2307/2232669. Maier, M. A., Elliot, A. J., & Lichtenfeld, S. (2008). Mediation of the negative effect of red on intellectual performance. Personality and Social Psychology Bulletin, 34, 1530–1540. doi:10.1177/0146167208323104 PMID:18768745. Mehta, R., & Zhu, R. (2009). Blue or red? Exploring the effect of color on cognitive task performances. Science, 323, 1126–1229. doi:10.1126/science.1169144 PMID:19197022. Nosic, A., & Weber, M. (2009). Overreaction and asset allocation: An experimental analysis, working paper.



Offerman, T., & Sonnemans, J. (2004). What’s causing overreaction? An experimental investigation of recency and the hot hand effect. The Scandinavian Journal of Economics, 106, 533–553. doi:10.1111/j.0347-0520.2004.t01-1-00376.x.

Thaler, R. H., Tversky, A., Kahneman, D., & Schwartz, A. (1997). The effect of myopia and loss aversion on risk taking: An experimental test. The Quarterly Journal of Economics, 112, 647–661. doi:10.1162/003355397555226.

Rothermund, K., Wentura, D., & Bak, P. M. (2001). Automatic attention to stimuli signaling chances and dangers: Moderating effects of positive and negative goal and action contexts. Cognition and Emotion, 15, 231–248.

Trommershauser, J., Maloney, L. T., & Landy, M. S. (2003). Statistic decision theory and the selection of rapid, goal-directed movements. Journal of the Optical Society of America, 20, 1419–1433. doi:10.1364/ JOSAA.20.001419 PMID:12868646.

Setchell, J. M., & Wickings, E. J. (2005). Dominance, status signals, and coloration in male mandrills (Mandrillus sphinx). Ethology, 111, 25–50. doi:10.1111/j.1439-0310.2004.01054.x.

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124–1131. doi:10.1126/science.185.4157.1124 PMID:17835457.

Shafran, S., Benzion, U., & Shavit, T. (2009). Investors’ decision to trade stocks: An experimental study. Journal of Behavioral Finance, 10(2), 81–88. doi:10.1080/15427560902902150.

Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and Uncertainty, 5, 297–323. doi:10.1007/BF00122574.

Shefrin, H., & Statman, M. (1985). The disposition to sell winners too early and ride losers too long: Theory and evidence. The Journal of Finance, 40, 777–790. doi:10.1111/j.1540-6261.1985.tb05002.x.

Weber, M., & Camerer, C. F. (1998). The disposition effect in securities trading: An experimental analysis. Journal of Economic Behavior & Organization, 33(2), 167–184. doi:10.1016/S01672681(97)00089-9.

Sirri, E. R., & Tufano, P. (1998). Costly search and mutual fund flows. The Journal of Finance, 53, 1589–1622. doi:10.1111/0022-1082.00066. Soldat, A. S., Sinclair, R. C., & Mark, M. M. (1997). Color as an environmental processing cue: External affective cues can directly affect processing strategy without affecting mood. Social Cognition, 15, 55–71. doi:10.1521/soco.1997.15.1.55.

Whitfield, T. W., & Wiltshire, T. J. (1990). Color psychology: A critical review. Genetic, Social, and General Psychology Monographs, 116, 387–412. PMID:2289687. Zweig, J. (1988). How the big brains invest at TIAACREF. Money, 27(1), 114.

Tal Shavit is the head Associate Dean in the School of Business Administration in the Collage of Management Academic Studies, Israel. Tal’s current research interests are Behavioral finance, Socio-Economics, Experimental Economics and business Ethics. Tal has experience in economic and finance consulting and training of investing managers. Tal has teaching experience in Finance, Securities analysis and Behavioral Finance in various MBA and BA programs. Tal has Bs.c. and Ms.C degrees in Economics and Management from the Technion – The Israeli Technology Institution, and PhD in Economics from the Ben-Gurion University of the Negev. Mosi Rosenboim is a faculty member in Guilford Glazer School of Business and Management, Ben-Gurion University of the Negev and in the School of Business Administration in the Collage of Management, both in Israel.



Mosi’s current research interests are finance, behavioral finance, and regional economic policy. He sits on the Board of Directors of a public firm and has experience in economic and finance consulting. Mosi has teaching experience in finance, behavioral finance, investment, securities analysis, and financial management in various MBA and BA programs. Mosi has B.A., Master, and PhD degrees in Economics from the Ben-Gurion University of the Negev. Chen Cohen is a faculty member in the department of Economics in Ashkelon Academic College. Chen’s current research interests are Economic theory, Game Theory, Contest Theory, Auction Theory, Experimental Economics and Behavioral Fiancé. Chen has teaching experience in Economics and finance in various MBA and BA programs. Chen has B.A., Master, and PhD degrees in Economics from the Ben-Gurion University of the Negev.



APPENDIX Instructions for Experiments The experiment involves a series of investment decisions. For each financial period, you need to decide how to divide your initial cash between two assets, “Fund A,” and “Fund B.” In each period, you will be asked to decide how much to allocate the each fund, ranging from 0 (nothing assigned to the fund) to 100 (everything assigned to the fund). The total amount you assign to fund A and B should total 100. After each decision, you will receive information about the performance of Fund A, Fund B, and your total earnings. The information appears on a computer screen, which displays yield percentages obtained from Fund A, Fund B, and the portfolio. Then, you will be asked to decide on your allocation for the next period.

Payment for the Experiment Payment for the experiment will be 10% of your total earnings in all periods (in NIS) plus a show up fee of NIS 20. We will be happy to answer any questions at any stage. Thank you for your participation.
