How can we scientifically measure love?

Question

I want to understand the psychology of love, and I found the Triangular Theory of Love on Wikipedia from this question, which I think is very believable. However, there is a big section on mixed support where it only says a little bit about THAT love was measured, not HOW:

"Sternberg measured his theory on couples who were roughly the same age (mean age of 28) and whose relationship duration was roughly the same (4 to 5 years)."

"The two other most obvious problems with Sternberg's theory of love are as follows. The first is a question of the separate nature of the levels of love. The second is a question of the measures that have previously been used to assess the three levels of love."

Can we scientifically measure love (without just asking the lovers)? How?

Answer 1

One way to measure love is to look at behaviors that people engage in to express love.

Chapman (1995) theorized that there were five broad classes of behaviors that people would engage in to express love: (1) words of affirmation, (2) spending quality time, (3) giving gifts, (4) acts of service, and (5) physical touch.

Goff, Goddard, Pointer, and Jackson (2007) developed a survey instrument to measure expressions of love. They created a series of questions that were designed to measure one (and only one) of the different behaviors that Chapman laid out. These involved asking questions about how a lover does things like: "giving presents," "offers encouragement," "spends time with me," "holds my hand," and "does yard work."

The goal of this work was mainly to see if the Chapman classes of expressions of love matched the kinds of behaviors that people wanted in a lover. The researchers administered the survey to a few hundred people, and in general found that the Chapman classes were a good set of high-level descriptions. People who tended to want one kind of behavior in a lover, e.g., receiving words of affirmation, also wanted other kinds of behavior from that class, e.g., receiving compliments.

References

Chapman, G.D. (1995). The five love languages: how to express heartfelt commitment to your mate. Chicago, IL: Northfield Publ.

Goff, B. G., Goddard, H. W., Pointer, L., & Jackson, G. B. (2007). Measures of expressions of love. Psychological Reports, 101, 357-360.

Answer 2

To add some general theoretical background...

Answering this is very complicated because the answer depends on how you define emotions, whether you see emotions as latent or emergent, whether you recognize high heterogeneity (behavior, cognition, physiology) within emotion categories, whether you view emotions as circumscribed in the brain or emerging from distributed neural networks, whether you view love as basic (like fear, disgust) or complex/higher order, and so on…

Emotion measurement is tricky business, and while our means of emotion measurement (e.g., facial coding [FACS], behavioral coding [SPAFF], psychophysiology [impedance cardiography, EKG, etc.]), have served us well in a lot of ways, they are also very limited.

From a constructivist perspective (a la Lisa Feldman Barrett or James Russell), self-report is the gold standard. On this view, emotions are just the labels that we use to organize and communicate our situated affective experiences. There is not a one-to-one mapping of behavior, cognition, and physiology to emotion categories like love. That is, love is a heterogeneous construct. It is highly variable in how it presents itself.

From a basic emotions perspective (a la Ekman, Izard, Panksepp, Keltner, etc.), we can use the methods suggested by Josh and Arnon without much concern. We can infer from patterns of behavior, cognition, facial expression, and physiology that the person is experiencing love. We should also be able to do this reliably.

So while we can certainly measure love "scientifically," how we do it depends on your theoretical orientation.

Answer 3

This BBC documentary reviews a number of methods for measuring love that have enjoyed some success. To summarize:

• Dr. Angela Rowe of the University of Bristol presents subjects with unfavourably distorted, unindistorted, and favourably distorted photographs of their love partners, and asks them to identify the undistorted one. Subjects in love tend to select photographs distorted in favour of their partner, while others select undistorted or unfavourably distorted photos.
• Bartels & Zeki (2000) identify differences in fMRI images of the brain while showing subjects in the early stages of their relationships photos of their love partners as compared to photos of controls. When viewing their love partners, areas of the brain associated with pleasure light up while areas associated with critical thinking are inhibited.
• Harker & Keltner (2001) claim they can predict romantic love and relationship success from college yearbook photos.

Here are some others of interest:

• Marazziti et al (1999) found that there are neurochemical markers (density of the platelet 5-HT transporter) for romantic love that are indistinguishable from severe OCD (they won an Ig Nobel prize for this).
• Masuda (2003) does a meta-analysis of several love scales, from Zick Rubin's original Love Scale (1970), to the popular Passionate Love Scale (PLS) (1986), to John Alan Lee's Color Theory of love (1973).