Cover Story

Love Stories: Adventures in the Study of Attraction



Commitment, openness, appreciation, creativity, patience — these qualities underlie the strongest romantic relationships, but they could just as well apply to the scientists who study relationships. Love, desire, and romance are far from simple phenomena, but this hasn’t deterred psychological scientists from diving in to explore the full range and complexity of these fundamental aspects of the human experience.

In a nod to Valentine’s Day, the Observer asked APS Fellows Lisa Diamond (University of Utah), Eli Finkel (Northwestern University), Nickola Overall (University of Auckland), and Harry Reis (University of Rochester) and psychological scientists Jessica Maxwell (Florida State University) and Meredith Chivers (Queen’s University) about the discoveries, challenges, and new directions in the study of love, desire, dating, and commitment.

How did you originally become interested in your line of research?

Samantha Joel: In the relationships unit of my intro to social psychology class in undergrad, our instructor introduced the Investment Model, and why it is that people wind up in unhappy relationships. The model puts forth two reasons — low investment and high alternatives. Are those the only reasons? How do they work? What other reasons could be out there? I became obsessed.

I switched my major to psychology, transferred to a university with labs that were studying relationships (University of Toronto), and I never looked back.

Jessica Maxwell: I first learned about sex and relationship research in my undergraduate studies, after taking a human sexuality course, and a course on intimate relationships. I always knew I was passionate about this area of research, but it wasn’t until a few years into my graduate studies that I really began pursuing this line of inquiry.

Eli Finkel: I always wanted to know how relationships work — why we find some people sexier than others, why some marriages succeed while others fail, etc. The major development for me came in a social psychology course at Northwestern where I discovered that it was possible to make a living by asking and answering questions like those.

Meredith Chivers: My interest in sexual psychophysiology emerged when working in a clinical sexuality research lab that focused on male sexuality. The large gaps in our knowledge of women’s sexual response intrigued me, and when I started grad school at Northwestern University, I jumped at the opportunity to be trained in sexual psychophysiology at the Kinsey Institute.

Harry Reis: I have always been fascinated by relationships. As soon as I discovered that you could study them empirically, I was hooked. It was an easy, no-brainer decision.

Lisa Diamond: When I was applying to graduate school, I was extremely interested in studying LGBT youth (this was the early 1990s, and it was actually a relatively new topic at the time). I was interested in studying the role of romantic relationships in the early development of lesbian and bisexual women, and that led me to a broader interest in love, attachment, and close relationships.

Have you made any discoveries that were unexpected?

Jessica Maxwell: I am working on a project right now where I’ve found that sex after conflict (“make-up sex”) is less satisfying than sex on days without conflict, which goes against lay notions that make-up sex is something that is really hot and passionate.

Another finding that may be unexpected to some is that believing sex takes work is associated with higher relationship and sexual satisfaction. It doesn’t always sound sexy to say your sex life takes effort and work, but my research shows it’s a beneficial belief.

Nickola Overall: Many people, including psychological scientists, may believe that conflict is bad for relationships, and that the best way to maintain relationships is to soften conflict with expressions of love and forgiveness. However, our research has shown that anger and hostility can sometimes produce increases in relationship well-being because these types of behaviors directly target problems, motivate change, and convey commitment and investment in the relationship. Although conflict can be tough and difficult to manage, it can also offer the opportunity for relationships to grow and become more secure.

Eli Finkel: I’d always assumed that it was possible, in principle, to develop algorithms that could use some sort of self-report data to match people who are more compatible than chance, but it looks like that’s impossible (Finkel et al., 2012; Joel, Eastwick, & Finkel, 2017). I’d always assumed that the expectations we bring to our marriages have increased systematically across the centuries, but it turns out that our expectations are actually decreasing in major ways (Finkel, 2017).

Meredith Chivers: Over the past two decades, we have discovered that cisgender women who are sexually attracted to men have unique patterns of sexual response. These women show significant sexual response to visual and narrative stimuli that depict women, although they do not report feeling sexual attraction to women, or report a history of sexual interactions with women. This is unexpected because women are showing an appetitive response — sexual arousal and desire — to sexual cues that have no incentive history associated with them.

We have observed this pattern using a number of methodologies, including genital responses measured using plethysmography and thermography, self-reported sexual arousal, neural responses assessed using fMRI and EEG, visual attention, and other cognitive measures, and it has been replicated in my lab and by others.

This discovery raises questions about how sexual orientations manifest, how sexual attractions develop, how sexual cues acquire their emotional salience, and, most intriguing to me, why this pattern of response is found with cisgender, heterosexual women but not with queer women (including trans women) or men.

Lisa Diamond: Probably the most unexpected discovery, and the one that really changed my own thinking, was the fact that sexual orientation doesn’t necessarily “orient” one’s capacity for romantic love. In my research, I found that lesbian women sometimes fell in love with their close male friends, even when they weren’t attracted to them, and heterosexual women sometimes fell in love with female friends. I soon found that this has been true throughout human history, and it led me to investigating the biobehavioral independence of sexual desire and romantic attachment.

What are some of the biggest practical challenges you face in conducting your research?

Samantha Joel: Perhaps the biggest challenge is that relationships are really messy (like many interesting human phenomena), and it can be hard to extract the signal from the noise. How do you know you’re capturing these processes the way they really unfold, and that your conclusions are going to withstand the test of time?

If we want to produce nuanced, robust, and generalizable findings, we need to pool our resources more and coordinate our efforts. We need more team science.

Nickola Overall: My primary methodological aim is to assess how naturally-occurring emotional and behavioral dynamics shape the course of people’s lives and relationships. This means getting couples to record their experiences repeatedly across daily life (experience or daily sampling studies), video-recording couples discussing relationship problems or trying to support each other (behavioral observation studies), and following couples across months or years to assess how these daily and behavioral dynamics predict changes in personal and relationship health and well-being (longitudinal designs).

Large dyadic longitudinal studies like these take years to collect, are hugely expensive, and — given the consequences of relationships for health and well-being — can produce ethical dilemmas regarding intervening with distressed couples. These complexities and challenges step up further when assessing families (couples and children), when targeting specific populations (e.g., low SES, violent couples, minority couples, etc.), and when examining processes that are harmful (e.g., aggression during conflict, poor parenting during family interactions).

Eli Finkel: At present, I’m working to launch a study of relationship dynamics in family businesses. We’re recruiting four people linked to each business: two siblings who are actively involved in running the business, and each sibling’s significant other. Doing so allows us to leverage Dave Kenny’s social relations model (e.g., Kenny & La Voie, 1984) to answer lots of cool questions, but it requires time-consuming data collection.

Let’s say we want to include a three-item measure of liking. For a participant to report on how much she likes each of the other three people, and how much each of those people likes her, she completes 18 items. Ideally, we would also assess perceptions of others’ liking of one another (e.g., A’s perception how much B likes C), which requires dozens of additional items. And we might want to know about perceptions of others’ liking for one another (e.g., A’s perception of how much C thinks B likes her).

But what if I also wanted to measure — with similar appreciation of interdependence — evaluations of competence, views about who works well together, etc.? Such a survey rapidly becomes prohibitively time-consuming, especially if we want to study atypical research samples (e.g., senior executives).

Meredith Chivers: Sexuality research is still associated with discomfort and taboo for people outside the area. When preparing ethics and grant applications, we need to take extra care to present our work as professionally as possible, and frame the work as scientific. I have, for example, had an ethics review board question the scientific merit of federal grant-funded research on women’s sexual response, for no specific reason other than doubting the benefits of deeper knowledge about women’s sexual arousal.

Harry Reis: Recruiting couples is much harder than recruiting individuals. If we had as much access to couples as we do to individuals, my lab would be much more productive!

Lisa Diamond: One of the biggest difficulties for the entire field of relationship research involves recruiting truly diverse samples. As Benjamin Karney has passionately argued, relationship researchers spend far too much time studying white, middle-class couples, well-functioning couples, and the knowledge we generate from this research doesn’t necessarily generalize to couples from more diverse backgrounds, and especially couples under economic stress. In terms of studying sexual-minority individuals (and couples), it can be difficult to recruit individuals who are more closeted, and yet it’s really important to make sure that we are not just studying the most openly-identified LGBT individuals.

Are there challenges to this work that people may not realize?

Samantha Joel: Compared with some other fields, it’s harder for relationship researchers to openly share our data — particularly couples data — because of the risk of romantic partners finding the data and discovering each other’s responses. For my field to get on board with open data sharing, we need infrastructure in place for sharing data that more fully protects the confidentiality of the participants. I think we’ll get there, but we’re not there yet.

Jessica Maxwell: A particular challenge that people may not think of is that sex doesn’t happen as frequently as other life events. The average couple has sex about once per week. If you are designing a daily survey where you track people every night, you will have to make a longer survey (e.g. 3 weeks) to capture multiple instances of sex.

Another challenge is that participants may have different definitions of what “sex” is. Couple members can even disagree as to whether they had sex the night before, which can pose problems for data analysis.

What do you see as the most exciting new directions for this work?

Jessica Maxwell: Being able to access a wide participant pool through online recruitment allows researchers to collect data that captures a wider range of sexual experiences and orientations, as well as ethnicities. I’m excited to see the insights gained from examining more diverse relationships.

I am also excited to start applying implicit measurement to the study of sexual relationships, to look at how our automatic attitudes about sex with our partner can be improved.

Regarding technology, I do some research on casual sex, which has become a lot more readily available with the rise of smartphone apps. As dating apps and online dating continue to become more normative, it will be interesting to see whether existing relationship phenomena change.

Nickola Overall: The time and expense of large dyadic samples, coupled with the growing need for replication across studies and contexts, has set the scene for international collaborations between relationship scientists in different labs across many countries. These collaborations increase the quality and replicability of relationship science and make for a broader, more representative view of relationship processes.

Eli Finkel: The easy answer here involves the emergence of “big data” and computational research methods, but I’m not particularly sanguine about those developments when they are applied to the relationships space (i.e., what happens once people have actually met). Thus far, nobody’s figured out how to use big data to track stuff like that.

At the moment, I’m more excited about the integration of relationship science with the broader marketplace of ideas. One idea I’m playing with these days, for example, is whether insights from relationship science can help to alleviate the extreme partisanship tearing apart many of our societies.

Harry Reis: We’ve now got the tools to examine couples’ behavior in its natural context. That’s so much more informative than surveys and lab observation. Two decades from now, we’ll know a lot more than we know now, and that should pave the way for better interventions and prevention programs.

Science Provides a Valentine Gift Giving Guide
Gifts to spouses and partners are a staple of Valentine’s Day. But results of a recent preregistered study published in Psychological Science suggest that our romantic gift-giving may provide the recipients some momentary elation at the expense of genuine satisfaction.

Adelle Yang of the National University of Singapore and Oleg Urminsky of the University of Chicago’s Booth School of Business theorized that people gravitate towards the gifts that they anticipate will elicit the most enthusiastic emotional responses, rather than those that the recipients themselves would prefer or would derive the most satisfaction from. The researchers tested this hypothesis in a series of studies involving both real and imaginary gift-giving decisions.

In one online study, 357 participants imagined they were either part of a gift-receiving couple or one of the couple’s gift-giving friends. They then saw pictures and descriptions of two similarly priced pairs of mugs – one set was personalized and the other had an ergonomic design. They rated how much they liked each option, which option they preferred, and predicted the emotional response and satisfaction that each option would elicit.

Regardless of whether they were giving or receiving the gift, participants anticipated that the personalized mugs would elicit a stronger emotional response than the ergonomic mugs would. Givers thought the couple would be equally satisfied with the two mug options and tended to prefer the personalized mugs, a preference driven by the emotional response they anticipated from the couple. Receivers, on the other hand, showed no preference for one option over the other.

In another online study, 295 participants in romantic relationships evaluated pairs of similarly priced Valentine gifts. The choice pairs included a dozen roses in bloom or 2 dozen roses about to bloom, a bouquet of fresh flowers or a bonsai plant, and a heart-shaped basket containing cookies or fruit. Again, givers were more likely than receivers to choose the option that they thought would elicit the strongest immediate reaction, such as the bouquet of fresh flowers, over the option that was likely to deliver more long-term satisfaction, such as the bonsai plant.

Findings from additional studies revealed that givers’ preference for gifts with a “wow” factor disappeared when they learned that they wouldn’t be able to see the recipient’s reaction.
When Yang and Urminsky asked people to think about gifts they had actually given or received, they found that people seem to derive the most enjoyment from receiving gifts, such as books and money, that givers often shy away from because they tend not to elicit strong emotional reactions.


Yang, A. X., & Urminsky, O. (2018). The smile-seeking hypothesis: How immediate affective reactions motivate and reward gift giving. Psychological Science, 29(8), 1221–1233.,

APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines.

Please login with your APS account to comment.