Site hosted by Angelfire.com: Build your free website today!

Running head:  POSITIVE AFFECT AND RISK-TAKING BEHAVIOR

 

 

 

 

 

 

 

Risky business:

Positive affect and risk-taking behavior

Elizabeth Cherson

Bryn Mawr College


Abstract

The purpose of this study was to explore the relationship between positive affect and risk-taking behavior.  The hypothesis, participants in the in the positive affect condition will be more willing partake in risk-taking behavior and will therefore pull more tickets than other participants, was investigated by the following procedure:  after watching a movie clip, approximately 5 minutes long, participants filled out a short online survey which inquired, among other things, as to their emotional response to the movie clip.  The participants then partook in risk-taking behavior by pulling a string of tickets out of an opaque box, similar to the game “Skunked” from the TV show “Let’s Make a Deal.”  The number of tickets pulled was the measure of the level of risk in which the participants were willing to engage.  The study found that amount of tickets pulled was not correlated with the participant’s mood, nor was it correlated with the participant’s assigned experimental condition, despite the fact that the movie clips did induce the desired emotions from the participants as desired.  These findings do not agree with previous studies.  The reasons why this study did not report similar findings to the findings of previous studies are discussed within the paper.
Risky business:

Positive affect and risk-taking behavior

            When someone is in a good mood, most times, others can tell.  The person smiles more; is more likely to stop and help a stranger; he or she might even whistle a happy tune.  However, there are other effects of being in a “good mood.”  Such effects have become the focus of more and more psychological studies since the 1970’s.  In fact, the number of studies conducted since that time has more than tripled.  Psychology, as a field, is no longer sole interested in “bad” behavior and “bad” moods; we are now only beginning to scratch the surface of “good” behavior and moods, and how they influence our cognitive behavior, attitudes and the probability of enacting one behavior over another.  Risk-taking behavior is one such behavior whose likelihood is affected by a positive frame of mind.  Studies have been conducted (Isen A., & Geva, N. 1907,  Isen, A., & Patrick, R. 1983, Isen, A. 2002, Nygren, T. 1998, and Roberts, K., Dimsdale, J, East, P., & Friedman, L. 1998) to determine exactly what kind of effect a good mood has on risk-taking behavior.  Positive affect is not so easily pinned down, though.  There are many aspects of a situation that can determine how positive affect will interact with risk-taking behavior and more specifically, how much a person with positive affect with risk:  how the positive affect is brought about, what kind of risk is involved, and what level of risk is being broached (Isen & Patrick, 1983).  If the risk were high and with concrete consequences, a subject would most likely choose to gamble little or nothing, in an attempt to maintain his or her positive state (Isen, 2002).  We must not think that people in a positive state all trip through the daisies with blithe smiles all day.  In reality, the positive affect leads people to think in a more complex manner that involves “elaboration and evaluation of outcomes, rather than just …” (Isen, 2002, p. 533) a blind, mad dash for whatever will maintain their good feelings and mood.  Isen also writes that positive affect may even nurture people to think about potentially negative information if it is important information.

            Studies have also found that people who experience strong emotion, such as a strong positive affect, have an increased likelihood of risk-taking behavior (Roberts et al., 1998).  But at what level would the risks that such emotionally volatile people take lie?  The Roberts et al. study surveyed youths in the San Diego area about their likelihood to participate in health-risk behavior like fast driving and unprotected sex with more than one partner.  Such behavior is considered by many to be extremely risky, with a very concrete consequence if one were to lose one’s “bet.”

            The study at hand did not measure behavior quite as risky or significant as health-risk behavior, but the hypotheses were the same:  people experiencing a positive affect are more likely to partake in higher risk behavior.  More specifically, participants in the study’s positive affect condition will pull a significantly higher number of tickets.

Method

Participants

The participants of this study included 24 males and 23 females.  They were recruited, not randomly selected, from the population of Bryn Mawr, Haverford and Lycoming Colleges.  The method of recruiting was to ask students if they would like to participate in a psychological study for the chance to win a 100-dollar prize.  They were not told the nature of the research.  After recruiting, the participants were randomly assigned to one of the experimental conditions.  Twelve were assigned to the negative condition, 16 were assigned to the neutral condition, and 18 were assigned to the positive condition.  The ages of the participants ranged from 18 to 23 years of age, with the mean age falling at 19.59 years. 

Materials

The room used to conduct this study was a small computer lab and lounge area.  In this room was a computer on a desk with a chair, a small coffee table upon which sat the television used, and a couch upon which the participants sat while viewing the video clips.  There were three video clips used as the three conditions of the experiment. A five minute segment of “The Holocaust: In Memory of Millions” video was used as the negative condition.  The segment consisted of a woman talking about her experiences as a Holocaust survivor.  A five-minute segment of “The Architecture of Frank Lloyd Wright” was used as the neutral condition.  The segment consisted of a woman talking about the life and architecture of Frank Lloyd Wright.  A five-minute segment of “Finding Nemo” was used as the positive condition.  The segment consisted of Nemo’s father, Marlin, and a group of friendly sea turtles riding an ocean current.  The apparatus used to view the clips was a small television and VCR combination situated approximately 1 meter from the participant.  The short survey was an online survey, taken on the computer in the same room in which the video clips were viewed.  The questions to be filled out were:

Gender

Age

  1. Did you find the movie interesting?
  2. How much did you learn?
  3. How much did you enjoy the movie?
  4. How did you feel after the movie clip?
  5. Would you recommend this movie to a friend?
  6. Would you want to watch the whole movie?

A 1-7 scale followed all numbered questions.  The ticket box used for the risk-taking behavior measurement was a small cardboard box, approximately 8in x 4in x 3in, covered in a green tissue paper.  There was a small hole in the lower right hand corner of one of the box’s shorter ends, through which the participant could pull a strand of tickets.  The box prevented the participants from seeing the next ticket before it was pulled.  Inside the box was a large roll of tickets.  The experimenters placed a dog sticker on one ticket, approximately 65 tickets from the starting ticket, to act as a “skunk.”

Procedure

After giving informed consent, the participant was seated approximately 1 meter from the television/VCR monitor and was asked to view a short film clip.  The clip was the clip associated with the participant’s randomly assigned experimental condition, positive, negative or neutral.  The clip lasted approximately 5 minutes, during which, the experimenter left the room.  After the conclusion of the clip, the experimenter returned and the participant was asked to complete a short survey at the computer.  Once the online survey was submitted, the experimenter read the participant the following statement:  “We had mentioned that you had the chance to win 100 dollars, well, this is how it works:  somewhere on this roll of tickets, there is a ticket with a dog sticker on it.  You can pull as many tickets out of the box as you want, and for each ticket you pull, a chance to win the money will be entered for you.  There is a ticket somewhere on the roll with a dog sticker on it.  If you pull the ticket with the dog, you will lose all of your chances to win the money.  Go ahead and start pulling tickets.  Let me know when you feel that you’ve finished before you tear off the tickets.”  After the participant was satisfied with the number of tickets that he/she had pulled, the experimenter recorded the participant’s subject number, randomly assigned condition, and the number of tickets that he/she had pulled. The participant was then debriefed as to the nature of the study and that in reality only one chance to win the prize money would be entered for each participant.

Results

The variable “feel,” as measured by the forth question on the online survey, was gauged on a 1-7 scale, with 1 as very sad, 4 as neutral, and 7 as very happy.  I recoded this variable into a new variable, “feeling,” by recoding responses 1, 2 and 3 in the “feel” variable as a 1, or “sad,” in the new “feeling” variable.  I recoded a response of 4 in the “feel” variable as a 2, or “neutral,” in the “feeling” variable.  Responses of 5, 6 and 7 were recoded as a 3, or “positive,” in the “feeling” variable.

A univariate analysis of variance exposed three sub-hypotheses:  1. Factor A: Did participants differ in the number of tickets pulled when compared by movie condition group?  2.  Factor B:  Did participants differ in the number of tickets pulled when compared by how they felt about the movie clip they had watched?  3.  Interaction of A with B:  Does the effect of movie condition (on number of tickets pulled) depend on how the participant felt about the movie clip he/she watched?

This test showed that for hypothesis 1 (for factor A):  F(2, 39) = 0.222, p = 0.802, the number of tickets pulled did not differ significantly between the negative, neutral and positive movie condition groups.  It also showed that for hypothesis 2 (factor B):  F(2, 39) = 0.155, p = 0.857, the number of tickets pulled did not differ significantly according to the emotion the participant felt after the movie clip.  Finally the univariate analysis of variance revealed that for hypothesis 3 (interaction of factor A with factor B):  F(2,39) = 0.378, p = 0.688, there was no significant interaction between how the participant felt after viewing the movie clip and what condition movie clip the participant watched that acted on the number of tickets the participant pulled.

A one-way independent-samples ANOVA revealed that movie clip condition had a significant effect on the level and type of feeling felt after viewing the clip (F(2, 44) =11.140, p<0.001).  Post-hoc tests indicated that participants in the negative condition (M=2.83, SD=0.937) felt significantly sadder (less happy) than participants in the neutral condition (M=4.19, SD=1.047) and the positive condition (M=4.89, SD=1.410).  Participants in the neutral condition (M=4.19, SD=1.047) felt significantly sadder (less happy) than participants in the positive condition (M=4.89, SD=1.410) and but not significantly happier than participants in the negative condition (M=2.83, SD=0.937) (Tukey’s HSD,  p>0.05).   Participants in the positive condition (M=4.89, SD=1.410) felt significantly happier than participants in the negative condition (M=2.83, SD=0.937) but not participants in the neutral condition (M=4.19, SD=1.047) (Tukey’s HSD, p>0.05). 

A one-way independent-samples ANOVA revealed that movie clip condition did not have a significant effect on mean number of tickets pulled (F(2, 44) =0.139, p=0.871).  A post-hoc test was not run, as effect was not significant.

A one-way independent-samples ANOVA revealed that feeling induced by movie clip did not have a significant effect on mean number of tickets pulled (F(2, 43) =0.091, p=0.913).  A post-hoc test was not run, as effect was not significant.

The frequencies show that the sample is biased 2 ways, while it is evenly balanced in 1 manner.  The number of males and females is almost exactly even, so in that manner, the sample is unbiased.  Yet, when it comes to age and movie clip condition, the sample is biased.  More participants were run in the positive condition than in the negative condition or the neutral condition.  The sample was also biased in that more participants were 19 or 20 years old than all other age groups combined.  This age majority does not reflect the Bi-College community as a whole.

A correlation test showed that movie clip condition and feeling induced by the movie clip were positively correlated (positive movie clip generally produced a feeling that was on the happy end of the survey question’s answering spectrum).  The correlation test also showered that there was no correlation between movie clip condition and number of tickets pulled, nor was there a correlation between feeling induced by movie clip and number of tickets pulled.

Discussion

            My hypothesis was that positive affect, as induced by the positive condition movie clip, would cause the participants who watched the positive condition movie clip to be more likely to take risks, thereby pulling more tickets than participants in the other conditions did.  This hypothesis was not supported.  Although the results showed that the movie clips did significantly induce the desired emotions, no significant relationship was found between the participant’s condition and the number of tickets pulled, nor between the induced emotion and the number of tickets pulled.  This is illustrated not only by the lack of correlation found between number of tickets pulled and the movie clip condition and the feeling induced by the movie clip, but also by the one-way independent-samples ANOVA tests that were conducted.  These tests also showed the lack of interaction of the factors movie clip condition and feeling induced by movie clip with the mean number of tickets pulled.

            I believe that the lack of consistency of this experiment’s results with the results found in other studies stems from the risk measurement factor of the experiment and how the emotions were induced.  Since the participants only stood to gain something which they did not have and not lose something which they did have, the risk factor may not have been as salient as we had hoped.  The over all axiom for positive affect’s interaction with risk-taking behavior says that positive affect will result in the participant being less likely to take a large risk, but more likely to take a small risk (Nygren, 1998).  From this, we can assume that in this experiment the risk was most likely a small one, since the participants would not be losing anything, only potentially gaining.  The lure of 100 dollars is quite strong for most college students, strong enough to coax them to participate in this study, but I believe that the threat of losing 100 dollars that they never had is not enough to make the risk that they took in pulling out more tickets a large and significant risk.  This behavior definitely would not equal the risk involved in walking across a fraying rope bridge over a 100-foot deep canyon.  Our results’ lack of consistency with prior research could also stem from weak emotions being induced by the movie clips.  It is possible that, while the movie clips did induce some emotion, the emotion was not strong enough to truly illustrate positive affect’s interaction with risk-taking behavior.

I believe that future researchers might wish to explore the venues of this research further.  This study leaves many unanswered questions, and many new possible experiments to be done.  Among them are, how would our results have differed if the participants had stood to lose something valuable, instead of standing to gain the 100 dollars?  Would they have been more cautious?  Would there have been a significant difference in the amount of risk taken if the results were again compared by movie clip condition?  This research has very real implications for every day life, such as, if the positive affect really does increase a person’s likelihood to take small risks, then should casinos show funny movies in the nickel slots areas, but show sad or dramatic movies in the high roller areas?  In addition, if a truly significant, marked effect is evidenced, we might be able to monitor a person with depression’s risk-taking behavior in order to better understand the depth and pervasiveness of his or her depression, and through such understanding, better help that person.


References

Isen A., & Geva, N. (1907).  The influence of positive affect on acceptable level of risk:  The person with a large canoe has a large worry.  Organizational Behavior and HumanProcesses, 39, 145-154.

Isen, A., & Patrick, R. (1983). The effect of positive feelings on risk taking:  when the chips are down.  Organizational Behavior and Human Performance, 31, 194-202.

Isen, A., (2002). A role for neuropsychology in understanding the facilitating influence of positive affect on social behavior and cognitive process.  Handbook of Positive Psychology, 38, 528-540.

Nygren, T., (1998). Reacting to perceived high- and low-risk win-lose opportunities in a risky decision- making task:  Is it framing or affect or both?  Motivation and Emotion, 22, 73-97.

Roberts, K., Dimsdale, J, East, P., & Friedman, L., (1998). Adolescent emotional response to music and its relationship to risk-taking behaviors.  Journal of Adolescent Health, 23, 49-54.