Love Drugs
The Chemical Future of Relationships
Brian D. Earp and Julian Savulescu




FOR ALMOST A DECADE, the hormone oxytocin has been relentlessly hyped as a “one-ingredient recipe for a utopian society.” In the words of one of our favorite science writers, Ed Yong, this “molecular high-five, which is released when we hug, tweet, dance, and orgasm, has been linked to trust, cooperation, empathy and a laundry list of other virtues.” One online magazine has called oxytocin “the most amazing molecule in the world.” Other popular accounts have taken an alliterative turn, calling oxytocin a “cuddle chemical,” “hug hormone,” or “moral molecule.” And yes, some have even called it a “love drug” (no alliteration there, but closer to our hearts).

On Amazon you can buy a nasal spray called OxyLuv. The spray, which purports to improve your sex life while decreasing your anxiety, has earned three out of five stars for satisfaction—across an array of customer experiences. On the one hand, a Brian Williams (not, we presume, the erstwhile host of NBC Nightly News) claimed that the spray made him more “willing to engage,” but only when he poured it into his morning coffee. On the other hand, a self-declared researcher from the University of Washington wrote that he had analyzed the spray with two separate tests and found it to be “negative for oxytocin” and “consistent with tap water.”

Still, a range of products touts some version of oxytocin, including Oxiboost, Liquid Trust, and Attrakt Oxytocin Pheromone Spray Cologne for Him, which promises to “create lasting attraction” and “enhance existing relationships.” In other words, Amazon is peddling real-life love drugs, although, most likely, not very effective ones.

Nevertheless, these products are attempting to capitalize on research into actual oxytocin, a hormone that is involved in human attachment, as we have seen. But oxytocin doesn’t have all those mystical properties that have been attributed to it. In fact, it provides a very good illustration of just how complicated and messy designing any new real-life love drug will be.

Science time

Let’s start with a little science. Oxytocin is what’s called a neuropeptide, a protein-like molecule that neurons use to communicate with each other. It’s present in a lot of animal species, from reptiles—in the form of something called vasotocin, an evolutionary precursor—to mammals, including us primates. And it is intimately associated with social, sexual, and reproductive behaviors, such as mating and childbirth.

The part of the brain that produces oxytocin (OT, as we’ll say from now on) is called the hypothalamus, and it releases OT when it’s called for various jobs. For example, OT sometimes acts as a hormone, controlling functions like labor contractions, the release of breast milk, and cervical dilation during childbirth. But OT isn’t just for mothers and birthing. It also acts as a neurotransmitter in everyone when it’s released into the central nervous system, where it influences a whole slew of interpersonal judgments, motivations, and behaviors. It may not be a “moral molecule,” but it is a multi-talented molecule to be sure.

Much of what we know about OT comes from studies of voles. Voles are small rodents that, for our purposes anyway, come in two main varieties. Depending on the species, they follow either a monogamous or a polygamous mating strategy, and the difference appears to turn on the activity of OT, vasopressin, and dopamine and how their respective receptors are patterned in the brain.

The basic idea is that oxytocin and vasopressin help register and store information about social identity (so you know who your partner is) and then dopamine associates that information with some level of reward (so you feel motivated to spend time with your partner). Roughly speaking, the denser the cluster of OT and vasopressin receptors in the brain’s reward centers, the higher the social reward and the more exclusive the partner preference.

Prairie voles have a “high density” cluster and they are usually socially monogamous. Meadow voles have a “low density” cluster and they are usually polygamous. Critical studies have manipulated OT and vasopressin levels directly in the monogamous (prairie) species. Infusing oxytocin into the brains of the females and vasopressin into the brains of the males facilitated a pair bond even in the absence of mating. Normally the voles have to have sex for this to happen. In other studies, infusing an oxytocin blocker into their brains had the opposite effect: the voles didn’t form a partner preference even after mating (whereas normally they would).

To be clear, no one knows if human attachment relies on the exact same hormonal machinery as that seen in voles. But several researchers have argued that natural selection would be under pressure to conserve a biochemical system that is so central to the survival and reproduction of other mammals.

Neuroimaging studies in humans seem to support this. Mothers who observe a photograph of their own young child, compared to another young child they have known for about the same amount of time, but did not give birth to, show heightened activation in brain areas rich in OT, vasopressin, and dopamine receptors. This effect is amplified for mothers with a secure attachment style compared to those who are insecurely attached. And it correlates with OT levels in the blood. Similar activation patterns happen in adults exposed to an image of their romantic partner—with whom they report being “intensely in love”—compared to an image of a platonic friend.

Although none of these findings demonstrates that elevated OT levels in humans directly induce greater feelings of love or attachment—whether in the mother-child or romantic partner case—they do point to a clear correlation between OT and attachment that seems to mirror what has been observed in other species.

Correlation does not entail causation, as everyone knows. So recent experiments have gone even further and have attempted to investigate a potential causal relationship between OT and various phenomena associated with love and attachment.

Here is how these experiments basically work. Using a simple nasal spray to deliver OT into the brain through one nostril, these studies were placebo-controlled and double-blind. This means that neither the scientists nor the study participants knew who got the real OT versus a dummy spray until after the results were in. (This kind of design is usually considered the gold standard for assessing causation. On the other hand, it’s also been called a “golden calf” because scientists are sometimes so in awe of it they forget to properly examine the actual study materials, statistics, or results.)

These foundational studies have delivered a range of important, if tentative discoveries. Emphasis on tentative. In just the last year or so while writing this book, we have seen evidence that some of the early findings on OT might not replicate. As one of the main researchers behind the famous vole experiments wrote in a recent review, “Intranasal OT studies are generally underpowered,” meaning they have too few participants for the statistics to make sense; and, “There is a high probability that most of the published intranasal OT findings do not represent true effects.” Accordingly, “the remarkable reports that intranasal OT influences a large number of human social behaviors should be viewed with healthy skepticism.”

This a stunning admission for a subfield of science that has long been at the vanguard of the media-hype parade. Science, in contrast to the fast clip of that parade, moves slowly, and several major research fields have been grappling with how to improve their methods. This has become a big issue in light of the “replication crisis” that is now rippling through science and medicine. Brian has written a lot about this perceived crisis in a separate line of work. But in a nutshell, the normal ways of doing science over the past many decades—from initial data collection all the way through to peer review—have likely been churning out a lot of false alarms. Not all false alarms, but more than you’d hope. And it isn’t just OT research; in fact, it’s research in many different subfields of psychology, economics, biology, and other disciplines as well. What this means is that you should take what we are about to say with respect to specific OT findings with a decent-sized grain of salt. We think we have the bigger trends about right, but only time will tell which of the particular results we mention will stand up to independent replication.