Introduction
Being in a close romantic relationship is a significant lifestyle change in comparison with being single. Apart from the self-evident impacts on loneliness and happiness, it may influence how much a person values another human being compared to all others or even themselves. This perception of a close one as more valuable could logically extend to risky behavior and resulting anxiety. Life is full of potential risks and dangers, and with interpersonal relationships being a significant part of life, it may be worthwhile to examine the connections between the two. Ghassemi, Bernecker, and Brandstätter (2020) propose that when an individual’s loved one engages in risky behavior, it produces more anxiety than if it was the individual facing that risk themselves, and the closer their relationship, the larger the discrepancy. The purpose of the study is to examine how the degree of interpersonal closeness can shape risk perception, and how much interpersonal relationships can impact emotional response to risk, as well as other relevant contributing factors.
The authors see a gap in the literature, where extant risk studies may have neglected the interpersonal part of risk-taking, affective response to it, and the self-other discrepancies. They focus on these aspects of risk studies to contribute to an under-researched topic. The present study is firmly rooted in risk studies and their core tenets. The risk as feelings hypothesis contributes to the core understanding of the affective response to the risk that the paper is based on. Another significant theory that contributes to the research is the interdependence theory, which describes how romantic dyads experience different feelings jointly. Many other contributing studies tangentially hint at the interpersonal aspect of risk perception. The cited works include both very recent and reasonably old studies, which provide sufficient rationale for the present study as a theoretical basis and as an indication of a gap.
The exact hypothesis of the paper is as follows: “individuals are more anxious when a significant other intends to engage in risk than when they intend to engage in the same risk themselves.” Several main variables, which appear in the experimental studies, can be derived from this hypothesis. The first dependent variable is the amount of anxiety experienced as an effective response to risk, the second dependent variable is the propensity to enact risky plans, and the independent variable is who faces that risk (the person, their significant other, or an acquaintance). There were several other control variables and study-specific secondary hypotheses, which, however, all revolved around the primary one and were designed to supplement it and examine it more thoroughly.
Method
The study followed an experimental design, with six experimental studies carried out to explore the primary hypothesis and their own smaller secondary hypotheses. The sixth study is not reported in the main study due to an error in data manipulation. Studies 1-4 follow very similar patterns set in Study 1, while Study 5 is performed differently to produce more grounded results. The main exploratory method was presenting participants with scenarios designed to elicit an emotional response, so there was no intervention, nor a treatment.
Study 1 had a sample of 166 individuals in committed romantic relationships, acquired from the university where the authors work, as well as research-related online resources. Study 2 had a sample of 193, presumably acquired by similar means. Study 3 used a German panel provider to get a sample of 166, 16 of which were excluded. Study 4 used 194 students recruited from the university’s subject pool. Study 5 was focused explicitly on couples, which were recruited from the university and car dealerships, 75 couples in total. Interestingly, Studies 1-4 had a predominately female sample, possibly due to the nature of the topic. All participants were promised a small compensation in the form of money or a mail-order coupon, while the students were promised partial course credit.
The experiments in the study were primarily focused on presenting the participants with hypothetical scenarios and asking them to explain their emotional reactions. As such, there are not many ethical considerations that could be violated by this study. A possible exception could be Study 5, which had the participants go on a reasonably lengthy car ride along a less-familiar route. For a research paper that treated car rides as inherently risky, subjecting the participants to that exact risk for data is ironic. That said, the participants gave informed consent, were both responsible adults and experienced drivers, and the study received the approval of the Institutional Review Board.
Results
The results of each study were statistically processed using regression models in R. As predicted, the results of Study 1 indicated that the participants regarded the partner’s risky behavior as more anxiety-inducing than their own. The risk posed to distant acquaintances was creating similar levels of anxiety to own risk. Additionally, women exhibited more anxiety than men in, presumably, all three scenarios. The propensity to engage in risky behavior followed similar trends, with participants recommending their significant other not face as much risk as they. Moreover, the conscious analysis of risk significantly predicted anxiety but did not differentiate among the three scenarios. Also consistent with the predictions, the strength of the relationship influenced the anxiety felt towards the partner engaging in risky behavior.
With Study 1 being the baseline, other studies added other factors to confound the experiment and more effectively prod at the hypothesis. Study 2 added scenarios where the participants faced risk together with their partners, which significantly decreased the anxiety of risk to self, but the anxiety remained much higher for the partner. Study 3 had the participants name several possible outcomes of the risky scenarios, which were more severe for the partner and acquaintance rather than self. Study 4 differentiated the closeness of the relationship and controlled for the social desirability of answering a certain way. The results indicated that anxiety increased with the closeness of the relationship and that answering a certain way was not caused by social desirability to any significant degree. Study 5, although noticeably different than the other studies, produced similar results, with anxiety over the car ride being higher when the partner was driving, and the perceived control of the situation being higher for self than the partner.
Discussion
All studies point to the existence of a discrepancy between the self and the other in affective reaction to risk. Namely, people experience much more anxiety over perceived risk to their partner, or an otherwise close person, than themselves or some distant acquaintances. Despite the differences in overall anxiety between genders, that discrepancy was universal. The authors mention that the higher severity of cognitively analyzed risks for the partner is more likely an effect of anxiety rather than the cause. They also mention the cognitive bias towards rewards for risk when thinking about own behavior, which may overshadow the adverse effects, but not appear when thinking about someone else. The limitations of the study include non-representative samples with an overabundance of women, simplistic measures, and self-report assessment. Future research should go deeper into the self-other discrepancy and prod at its causes, effects, implications, and deeper underlying mechanics of interpersonal relationships regarding the affective reactions to risk.
References
Ghassemi, M., Bernecker, K., & Brandstätter, V. (2020). “Take care, honey!”: People are more anxious about their significant others’ risk behavior than about their own. Journal of Experimental Social Psychology, 86, 103879.