The Influence of Risk-Taking Personality on

Behavior in Romantic Relationships

Ericka N. Dennis

Abstract

An experiment was conducted to investigate the correlation between risk-taking personality and risk-taking behaviors in romantic relationships.  The Personality and Romantic Relationships survey was administered to 85 students (38 males, 46 females, one did not state) at a small liberal arts college in the Midwest, ranging from first-year students to seniors.  The survey consisted of three sections: part one was original, part two was adapted from Zuckerman�s �Sensation Seeking Scale,� and part three was adapted from a quiz in Cosmopolitan magazine.  The main hypothesis was that an individual�s scores on parts one and two of the Personality and Romantic Relationships survey would be related to their score on the Cosmopolitan �Do You Take Risks in Love?� survey.  There were little to no significant findings for each of the tests run on the data.  Therefore, none of the hypotheses were supported due to a lack of significant findings, and it was inconclusive whether or not individuals with risk-taking personalities also tended to take risks in romantic relationships.  One practical application of this research could be to orient information toward potential high risk students for sexually transmitted diseases.

            In the last few years, there has been a rising number of people participating in risk-taking behavior.  This behavior ranges from thrill-seeking activities such as skydiving, to extreme sports such as snowboarding, to high risk social behaviors such as binge drinking.  One might wonder what motivates people to take such risks.  Another question posed could be, �What type of personality characteristics do risk-taking people possess?�  In particular, the researcher wanted to know whether risk-taking behavior was also manifested in romantic relationships.

Risk-taking behavior is an issue because of how quickly it has been popularized in recent years, and also because accidents happen from time to time.  People who take part in risky activities are always taking a chance, even if it is small, on getting injured or even losing their life.  For example, in the United States there were 13 serious injuries and two fatalities due to hang gliding accidents in 2003 (United States Hang Gliding Association, 2003).  Furthermore, there were 25 fatalities as a result of skydiving accidents in 2003 (United States Parachute Association, 2003).

The researcher�s focus in studying risk-taking personality was to find whether it influenced behavior in romantic relationships, and if so, in what ways.  Are individuals with risk-taking personalities also risk-takers in romance?  Do people who enjoy the thrill of a risky physical activity also enjoy the thrill of a risky move in their dating life?  The experiment conducted was aimed at finding out how much of an effect that risk-taking personality has on behavior in romantic relationships.

            A large body of research has suggested that risk-taking personality does have an effect on one�s behavior in romantic relationships.  The following studies suggested that an individual who has a risk-taking personality may be more likely to also take risks in the romantic arena of their lives.  Some of the studies discussed risk-taking personality in general, and some discussed how it seems to be evident in romantic relationships. 

When an individual experiences the universal emotion of fear, the fight-or-flight response prepares the body for what may happen.  The perception of risk triggers several physiological changes.  A flow of adrenalin, noradrenalin, and growth hormone are produced.  As the brain recognizes danger, one�s heart begins to beat up to three times faster, blood pressure increases, breathing rate becomes faster, and the mouth may become dry (Llewellyn, 2003).  For individuals with risk-taking personalities, this natural protection system is often ignored as the individual pursues the thrill of a risky activity.  What causes people to go for the �fight� part of the fight-or-flight response and engage in risk-taking behavior?

Llewellyn (2003) described several areas of theory to explain risk-taking behavior.  According to Freud and psychoanalytic theory of the early 1900�s, it was abnormal for an individual to overcome natural fears, and risk-taking behavior was seen as an indicator of a diseased mind.  Psychoanalysts could not understand why people would willingly participate in activities that would risk their lives.  They therefore came to the conclusion that risky behavior should be classified as a sign of suicidal tendencies, a death wish (which Freud called Thanatos), or �repressed feelings of masculine inadequacy� (Llewellyn, 2003).  It was proposed that individuals who took part in risky activities were illogical and possibly pathological.  This theory is no longer widely used to explain risk-taking behavior.  Criticism of the psychoanalytic theory in regards to risk-taking behavior is that there is no evidence to support the idea that risk-taking individuals are pathological.  Studies that have explored the mental health of risk-taking individuals have been inconclusive or contradictory, and some have even found that risk-taking behaviors lead to an increase in self-esteem.  Even animal behavior contradicts psychoanalytic theory regarding risk-taking behavior, as animals �have been shown to take risks for social reasons; the underlying theory being that by taking risks that a less able animal would have to avoid they demonstrate the superiority of their genes and become more attractive as a mate� (Llewellyn, 2003).  Studies have found this occurrence in humans too, with �people who engage in high risk sports (e.g. parachuting) seen to be more sexually attractive� (Llewellyn, 2003).

According to evolutionary theory, risk-taking behavior has been passed to humans through evolution.  During the more treacherous time periods of the past, humans had to take risks daily in order to survive.  Living without the commodities of modern times, early man had to take risks to find food, shelter from inclement weather conditions, and other necessities.  Through the process of natural selection, individuals who possessed genes related to risk-taking behavior would pass them down, therefore causing risk-taking behavior to be more common than non-risk-taking behavior.  The implication of this theory for the present day is that �the way in which we are socialized effects the way in which our genetic heritage is expressed, and perhaps people go rock climbing� because they are attempting to express the primal instinct to take risks in a modern world� (Llewellyn, 2003).

According to contemporary theory, personality traits play an important role in distinguishing between people who enjoy taking risks and those who would rather play it safe.  Individuals react differently to the same situations, and these differences are caused by variations in personality traits from person to person.  �There are a small number of broad traits which account for the majority of people's personalities, and the two broad traits of Extraversion vs. Introversion and Emotional stability vs. Neuroticism are clearly the most important� (Llewellyn, 2003).  Regarding other dimensions of personality, Eysenck proposed one more dimension, Psychoticism vs. Humaneness, while Costa and McCrae proposed three more dimensions: Openness, Conscientiousness, and Agreeableness.  It has been suggested that personality traits are about 50% genetically determined, which relates to the evolutionary theory that much of our behavior today is influenced by our genetic history.  Other variance in personality traits comes from socialization, especially parental upbringing (Llewellyn, 2003).

In applying personality traits to risk-taking behavior, one must look at whether the dimensions of personality help to predict risk-taking behavior.  A considerable amount of research supports the idea that individuals who take risks tend to be high in the Sensation Seeking trait, a constituent of the Psychoticism vs. Humaneness trait.  The Sensation Seeking theory was first developed in the 1950�s by Marvin Zuckerman.  While performing sensory deprivation experiments, he hypothesized that the individuals who had volunteered for the experiments shared several of the same personality traits.  They seemed to be �especially venturesome and inquisitive, eager to have new and exciting experiences even if they did contain a degree of social or physical risk� (Llewellyn, 2003).  A number of studies have suggested that people who participate in high risk activities tend to be high in the Sensation Seeking trait, which supports the hypothesis that people take risks because they want to have new and exciting experiences.  Twin studies have shown that about 60% of the Sensation Seeking personality trait is genetically determined (Llewellyn, 2003).  In regards to gender differences, males tend to be higher in the Sensation Seeking trait than females.  Regarding age differences, the Sensation Seeking trait tends to decrease with age.  These differences can somewhat account for why many individuals who take potentially deadly risks are young men.  Nonetheless, there are many women who exhibit the Sensation Seeking trait, evidenced by a rising number of women taking part in such risky activities as extreme sports and binge drinking. 

According to Zuckerman�s work on the Sensation Seeking trait, there are four sub-dimensions to the trait: (a) �Thrill and Adventure Seeking� which is the desire to take physical risks such as high risk sports, (b) �Experience Seeking� which is the desire for new and unique experiences, (c) �Disinhibition� which is the willingness to partake in risky social behaviors such as unprotected sex and binge drinking, and (d) �Boredom Susceptibility� which is an aversion for monotony.  Zuckerman�s four sub-dimensions support the idea that various forms of risk-taking behavior may all be connected with a general �risk taking personality� (Llewellyn, 2003).  On risk-taking personality in regards to romantic relationships, Zuckerman says, �Many of life�s decisions involve a balance between anticipated reward and risk. The married man or woman who is charmed by an attractive stranger weighs the threat to his or her valued marital relationship and the possibilities of contracting a sexually transmitted disease against the possibility of a thrilling sexual encounter� (2000).

In three experiments conducted by Beisswanger, Stone, Hupp, and Allgaier (2003), it was investigated whether individuals make decisions about romantic relationships differently when the situation pertained to themselves versus when it pertained to others.  The subjects used in the first experiment were 104 male and 97 female undergraduate students from a large public university.  They were each given a questionnaire with 11 scenarios to respond to, all dealing with relationships.  The results of this experiment found that individuals were �more likely to make decisions or give advice to friends encouraging risk-taking behavior than they were to take the same risks when deciding for themselves� (2003).  The subjects used in the second experiment were 66 male and 116 female undergraduate students from the same university.  They were each given a questionnaire with 12 scenarios, some from the first experiment and some new ones.  The results of this experiment suggested that the difference found in the first experiment is only valid for less important and non-permanent decisions, and does not apply when the decisions could have significant consequences.  The subjects used in the third experiment were 72 male and 92 female undergraduate students from the same university.  They were each given a questionnaire with the same 12 scenarios from the second experiment, along with possible consequences of decisions made in each scenario.  The results of this experiment showed that �the self-other differences were not eliminated when information on the potential consequences was explicitly presented to participants� (2003).  Nevertheless, an assessment of the reasons that the participants gave for their decisions revealed that they were more likely to regard the potential negative outcomes as important when the decision pertained to themselves than when it pertained to others.

In a study conducted by Weisskirch and Murphy (2004), it was determined that individuals who have many casual and close friends, or who frequently use the Internet for entertainment purposes, or who prefer punk, heavy metal, or reggae music also tend to be high in the Sensation Seeking trait.  For the study, 62 male and 75 female students from a public university in California completed Arnett�s Inventory of Sensation Seeking.  This questionnaire was created in 1992 and is a more modern version of Zuckerman�s Sensation Seeking Scale.  The subjects also answered questions regarding their Internet usage, music preference, and close relationships.  The results indicated that �there are subtypes of sensation seekers. Some may need physical thrills, and others may satisfy their need for stimulation through being with friends, listening to music, or using the Internet� (2004).  In addition, social psychology indicates that we choose risk-takers for leadership positions, which further supports the results of the study by Weisskirch and Murphy.

Eysenck and Zuckerman (1978) found that there is a correlation between the dimensions of personality measured by Eysenck�s Personality Questionnaire and Zuckerman�s Sensation Seeking Scale.  For their study, the participants were 219 American male and female undergraduate students and 879 male and female English twins.  All participants took both the EPQ and the Sensation Seeking Scale (Form V).  The results suggested that �sensation-seeking was positively correlated with the traits of extraversion and psychoticism as measured by the EPQ� and that �there was no relationship between sensation-seeking and the trait dimension of neuroticism� (1978).

Research conducted by Huth-Bocks (1996) explored the relationship between personality traits and sensation seeking, along with gender differences in risk-taking activity such as promiscuous sex and substance abuse.  For the study, the Sensation- Seeking Scale (Form V) was given to 22 male and 25 female undergraduate students, followed by a personality inventory using McCrae and Costa�s Big Five personality dimensions (Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism).  It was determined that the Extraversion trait predicted sensation-seeking in females, and that low Agreeableness and low Conscientiousness predicted sensation-seeking for both genders.  In addition, males were found to be considerably more interested in taking risks than were females.  These results suggest what type of person is most likely to take part in risky activities in a college student population, and can be used to improve education and prevention programs.

In a study conducted by Donaldson (1989), it was determined that likenesses in sensation-seeking are positively correlated with sexual satisfaction and contentment in long-term heterosexual relationships.  For the study, the Sensation Seeking Scale (Form IV) was given to 31 heterosexual couples, ages ranging from 17-45 years.  Donaldson hypothesized that �couples who were similar on the sensation-seeking dimension would report greater sexual satisfaction and contentment in their relationships than those who were less well-matched� (1989).  The results of the study supported that hypothesis, along with a related hypothesis that the sensation-seeking, sexual satisfaction, and contentment in the relationship were stronger for subjects who had been together for more than one year.  

Research conducted by Tobacyk and Thomas (1980) investigated the connection between masculinity, femininity, and the Sensation Seeking trait.  Tobacyk and Thomas hypothesized that masculinity would be directly related to sensation seeking, while the opposite would be true for femininity.  For the study, the Bem Sex-Role Inventory and the Sensation Seeking Scale were given to 129 female and 159 male undergraduate students.  The results indicated that for both genders, �masculinity was significantly and directly associated with sensation seeking� and that for males only, �femininity was significantly and inversely associated with sensation seeking.�  Tobacyk and Thomas also explored the correlation between masculinity and femininity according to the four Sensation Seeking sub-dimensions.  After obtaining masculinity and femininity scores of the participants, the researchers found that masculinity was more often associated with the Thrill and Adventure Seeking, Disinhibition, and Boredom Susceptibility sub-dimensions, while femininity was more often associated with the Experience Seeking sub-dimension.

Daderman, Meurling, and Hallman (2001) compared and contrasted the personality traits of non-socialized and socialized people who were sensation seekers.  They used a group of 47 male juvenile delinquents as the non-socialized subjects, and a group of 18 male Swedish air force pilot recruits as the socialized subjects.  The Sensation Seeking Scale, the Karolinska Scales of Personality, and the Eysenck Personality Questionnaire were administered to all subjects.  Results showed that juvenile delinquents and air force pilot recruits were both high in sensation seeking, but in separate sub-dimensions of the trait.  The juvenile delinquents were high in the traits of impulsiveness, somatic anxiety, and extraversion-sociability, while the air force pilot recruits were high in the thrill and adventure seeking trait and the disinhibition trait.

In a study conducted by Zuckerman and Kuhlman (2000), the Life Experiences Questionnaire and the Zuckerman-Kuhlman five-factor personality questionnaire was given to 260 undergraduate students.  The risky behaviors on the survey were put into six categories: smoking, drinking, drugs, sex, driving habits, and gambling; Zuckerman and Kuhlman hypothesized a connection with personality and those behaviors.  They found that for all six categories, behavior was related to �scales for impulsive sensation seeking, aggression, and sociability, but not to scales for neuroticism or activity� (2000).  They also found gender differences in risk-taking on the scale for impulsive sensation seeking.  Males scored significantly higher than females in impulsive sensation seeking, drug use, risky driving behavior, and gambling.  Zuckerman and Kuhlman also investigated possible biological causes of risk-taking personality; for example, the enzyme monoamine oxidase is low in individuals with high sensation seeking behavior.

Research conducted by Jones (2004) described how �sexual imposition, dyadic trust and sensation seeking� have an effect on risky sexual behavior in young women from metropolitan areas.  The researcher had 257 young urban women complete an Audio Computer-Assisted Self-Interview regarding their sexual behavior.  Results showed that the factors of sexual imposition, dyadic trust, and sensation seeking accounted for 18.3% of the subjects� risky sexual behavior.  Jones also found that �pressure to satisfy a male partner sexually was more common than physical coercion� (2004).  Unexpectedly, there was a negative correlation between dyadic trust and risky sexual behavior, suggesting that women had sexual relations with men that they did not trust or have an intimate connection with.  Finally, sensation seeking was found to have a positive correlation with the subjects� risky sexual behavior.

Rosenbloom (2003) examined the connection of risk evaluation with personality and with risk-taking behavior.  For the study, 55 females and 20 males from a university in Israel were given the Sensation Seeking Scale (Form V), along with the Inventory of Risk Evaluation and the Inventory of Risk Taking.  A positive correlation was found between risk-taking behavior and the sensation-seeking trait, while a negative correlation was found between risk evaluation and the sensation-seeking trait.  It was determined that individuals who were high in sensation-seeking tended to be higher in risk-taking behavior, while individuals who were low in sensation-seeking were higher in risk evaluation. Therefore, low risk-taking behavior might be attributed to the individual�s evaluation of risk.

Research findings from a study done by Littig and Branch (1993) suggested that differences are not necessarily found in every aspect of personality when it comes to sensation seeking.  The subjects used in the study were 20 male and 20 female undergraduate students, all African American, from a large public university in Washington, D.C.  They were given the Sensation Seeking Scale (Form V), and then shown a series of ten Japanese ideographs.  Each subject saw the ten ideographs at different speeds.  After viewing the ideographs, the subjects filled out rating scales regarding how much they liked each ideograph.  Results showed that the subjects generally preferred faster speeds of viewing, whether they were high or low in sensation seeking.

In a study conducted by Aron, Aron, Heyman, Norman, and McKenna (2000), the researchers hypothesized that the quality of long-term romantic relationships can be increased when the couple participates together in �novel and arousing activities� rather than �mundane activities.�  For the first part of their study, Aron and colleagues administered the Dyadic Adjustment Scale and the Idealistic Distortion Scale to their participants, which included 90 females and 22 males who were involved in a serious relationship.  For the second part of the study, the same two surveys were administered to 42 females and 38 males.  The researchers then conducted a series of three experiments in a controlled laboratory setting, with 28 couples as their participants, ages ranging from 17 to 44.  Some of the couples were assigned an exciting activity to participate in together, some were assigned a pleasant activity, and some were assigned a mundane activity.  The couples were observed and videotaped during the activities, and their reactions and communication with each other was evaluated by the researchers.  Their findings indicated that �the initial exhilaration in the early relationship years may be due to the novelty and arousal of forming the relationship; but when, over time, this novelty and arousal inevitably decline, couples might instead maintain a high level of experienced relationship quality by engaging together in novel and arousing activities, so that the positive effect from such activities becomes associated with the relationship� (2000).

The central hypothesis of the current research on risk-taking personality and romantic relationships is that an individual�s scores on parts one and two of the Personality and Romantic Relationships survey would be related to their score on the Cosmopolitan �Do You Take Risks in Love?� survey.  Related hypotheses are that if an individual identified himself or herself as having a self-concept of 6 or 7 on a Likert scale, or was male, or a first-year student or sophomore in college, then he or she would be more likely to perform behaviors such as: approaching an attractive person at a bar, cheating on a partner, lying to a partner, going home with a stranger, etc.  The researcher expected to find a significant positive correlation between risk-taking personality and risk-taking behavior in romantic relationships.  The rationale for this expectation was that every individual�s personality type has an effect on the decisions that they make in all areas of their lives, including romantic relationships. 

In the current study, a risk-taking person was operationalized as having a majority of responses of 5, 6, or 7 on the Likert scale questions on part one of the Personality and Romantic Relationships survey.  In addition, the individual was considered a risk-taking person if the majority of their responses on part two of the survey were the sensation-seeking response.  Risk-taking behavior refers to �any behavior that has a significant degree of uncertainty about the losses associated with its outcome� (Rosenbloom, 2003).  Risk-taking behavior in romantic relationships refers to a range of behaviors such as: inviting a new romantic interest to a wedding, approaching an attractive stranger in a social setting, being rejected after making romantic advances toward a stranger, cheating on a partner, lying to a partner, being the first to say �I love you� in a relationship, going home with a stranger, etc.

Method

Participants

            The participants (N=85) were acquired from four academic classes and the Student Government Association at a small Midwestern liberal arts college.  The classes included Earth Science, Introduction to Psychology, Experimental Psychology, and the First-year Seminar.  These classes were chosen in order to acquire participants from a variety of ages and majors.  There were more first-year students than sophomores, juniors, and seniors (36 first-year students, 10 sophomores, 14 juniors, and 24 seniors [one did not state]).  There were 38 male and 46 female participants (one did not state).  It is assumed that the participants came from a variety of majors and socioeconomic statuses, but as for race and ethnicity, the majority of the participants were Caucasian.  Also, the majority of the participants were in the age range of 18-23; however, there were a very small number of participants over the age of 23, who were non-traditional students.

Testing Materials

            A survey entitled �Personality and Romantic Relationships� was used.  The survey consisted of three parts.  The first part included demographic information and original questions written by the researcher, regarding self-concept and several risk-taking behaviors.  The responses for these questions were on a 7-point Likert scale, with (1) being the lowest desire to participate in a risky activity, and (7) being the highest desire to participate in a risky activity.  The second part of the survey was comprised of 13 items adapted from Zuckerman�s Sensation-Seeking Scale (Zuckerman, 1979).  For each item, the respondent was to circle one of two choices which best described him or her.  The third part of the survey consisted of nine multiple-choice questions that were adapted from a survey in the May 2003 issue of Cosmopolitan magazine.  The instructions before each section of the survey were clear and concise.

Procedure

            The �Personality and Romantic Relationships� survey was administered to all participants at the beginning of a class period and also at the beginning of a Student Government Association meeting.  Before distributing the survey, the researcher briefly explained to the participants that the survey was being used as part of a study regarding personality and romantic relationships.  The researcher also read a statement of informed consent to all participants.  Upon completion of the survey, each participant handed the survey back to the researcher.

            For data entry, the Statistical Package for the Social Sciences (SPSS) program was used.  Initially, each response was entered into the database as it was.  After completion of the data entry, several responses had to be recoded.  (See appendix for a copy of the survey.)  In the demographics section of the survey, a �fifth-year senior� was coded as a senior.  In the second part of the survey, item numbers 13, 18, 19, and 22 were recoded so that a response of (A) was the sensation-seeking quality, and a response of (B) was the non-sensation-seeking quality.  In the third part of the survey, every question except the first one was recoded so that a response of (A) was the high-risk behavior, a response of (B) was the average-risk behavior, and a response of (C) was the non-risk behavior. 

            After recoding the necessary data, several tests were run on the data. In order to get a general idea of the results, the researcher ran descriptive statistics and crosstabs first.  To find whether a participant�s scores on parts one and two of the survey correlated positively with their scores on part three of the survey, a comparison of means was used.  Correlations were used to find the relationship between high self-concept and risk-taking behavior in romantic relationships, and between class (year in college) and risk-taking behavior in romantic relationships.  T-tests were run to determine the relationship between gender and risk-taking behavior in romantic relationships. 

Results

Hypothesis 1: Risk-taking personality and risky romantic behaviors

            Table 1 shows a comparison of means from each risk-taking activity in part one of the survey with the high-risk response (1), average-risk response (2), and non-risk response (3) from all nine questions in part three of the survey.  For each high-risk response on part three of the survey, the means for each of the risk-taking activities are all in the range of 5 to 7.  This means that participants who selected the high-risk response in part three, had also indicated that they have a high desire to try some or all of the risky activities in part one.  In addition, the means for the risk-taking activities were typically lower for the average-risk response and non-risk response.  However, it is inconclusive whether or not the researcher�s central hypothesis was supported.  The researcher was not able to find significance in the correlation between the means for part one and part three of the survey.  Furthermore, the researcher found no significant relationship between part two and part three of the survey.

Table 1.  Comparison of Means for Part One and Part Three of the Personality and Romantic Relationships Survey.

Hypothesis 2: Self-concept and risky romantic behaviors

Table 2 shows the correlation between self-concept and one measure of risk-taking behavior in part three of the survey (refer to question 29).  This was the only significant relationship found with self-concept and risk-taking behavior in romantic relationships.  The correlation was negative, meaning that participants with a high self-concept (6 or 7 on Likert scale) were more likely to choose the high-risk response on part three of the survey.  The relationship was significant at the .05 level (r = -.223, p = .042).