Sexual Communication
What's the secret to good sex? How about better sex? Great sex? Mindblowing sex?
Choose an adjective, and you'll find reports saying that it's not what you think.
Apparently, what you think is that it's all about technique. These reports hasten to inform you that the "secret" to good sex, or better sex, or whatever, depends on things that start outside the bedroom, such as knowing yourself, communicating with your partner, and expressing affection.
I feel that we've heard this advice often enough that when I noticed a new study focusing on communication and some of those other, often-discussed variables, I felt curious. What could this study tell us that sex therapists haven't been saying in interviews and podcasts and blog posts for many years now?
As it turns out, not much. It's a terribly weak study. All the same, it provides an especially good illustration of how our our lives have been transformed by statistics. We now have historically unprecedented knowledge about human sexuality, thanks to survey data, statistical techniques, and large samples. We've also experienced misunderstandings about sex created by an excessively statistical view of how people work. In short, for better and for worse, we've been "statisfied".
Statistics and sexual openness
Since the mid-20th century, statistics have given us increasingly detailed glimpses into peoples' sexual thoughts, feelings, and behaviors. Merely knowing what others do has reduced taboos on conversations about sex with partners, with young people, and with medical professionals. The numbers have helped destigmatize – and in some cases, decriminalize – certain behaviors. What were formerly considered perversions are now, in some cases, treated as preferences.
A simple example of what I have in mind are the Kinsey Reports of 1948 and 1953. By simply noting the percentages of people who've had homosexual feelings or experiences, for example, and by presenting homosexuality and heterosexuality as two ends of a continuous spectrum, these reports helped destigmatize homosexuality and spur the American Psychological Association to stop classifying it as a form of mental illness in 1973 (as per the DSM-III). More broadly, the work of Kinsey, and, later, Masters and Johnson, contributed to a growing openness about discussing sex, in part by quantifiying the amazing variety of sexual fantasies, responses and behavioral experiences people have. Whatever your views on the most pressing LGBTQ issues (including which acronym should be used), the fact that you can encounter discussion of these issues virtually anywhere is ultimately due, in part, to to statistical data that helped us appreciate that not all people, in the privacy of their bedrooms, have satisfactory sex, on a regular basis, missionary-style, with their heterosexual spouse. The stats suggest that this scenario might not even be the most common one.
Sexual openness and science
Presumably, those "secret to better sex" reports continue to emphasize openness, because some people struggle with being open. Scientists do too, in sense that even now, new studies are addressing topics that weren't covered in earlier, less open times. Here's a shocking example:
When sex was a relatively taboo subject, medical researchers studied genitalia primarily in connection with reproduction. The penis received a lot of attention because of its importance to reproduction (and to the researchers themselves, since, traditionally, most of them had one). The clitoris, being connected to pleasure merely, received far less attention. Hence, while knowledge of penile anatomy is quite advanced, it wasn't until October 2022 we learned that the clitoris has more than 10,000 nerve fibers. Why is that important? Because the estimated number previously taught to medical school students was considerably lower. Why the mistake? Because the previous estimate was derived from research on cows. Seriously. Research on cows discussed in a 1976 book. Experts have already begun commenting on important surgical implications of the underestimate.
This brings me to the new study. The focus of the study is not microanatomical but rather on verbal communication, along with other contributors to sexual and relationship satisfaction. The fact that any study could explore sexual satisfaction per se, apart from its role in reproduction, is a relatively new development in mainstream research. Again, we can thank the pioneering sex researchers for their statistics and other contributions to cultural shifts that made studies like this possible.
The new study
The study in question, conducted by Drs. Margaret Bennett-Brown (Texas Tech) and Amanda Denes (University of Connecticut), was published in the journal Communication Research earlier this month.
The rationale for the study was straightforward and clear: We know about the benefits of communication before and after sex, but studies haven't yet focused on communication during sexual interactions.
I was surprised when I read that. Experts have long advised us to be open with our partners about boundaries, needs, preferences, fantasies, etc. Doesn't that mean being open at all times?
A quick Google search reveals that the experts (marriage and family counselors, sex therapists, advice columnists, etc.) have been a step ahead of the researchers. Countless news reports, editorials, websites, blogposts, podcasts, etc. stress the importance of communication, and in many cases this includes talking with one's partner in the midst of sex. It's only the researchers who've restricted themselves to the before and after.
So far, so good. Bennett-Brown and Denes are studying a topic that hasn't received scientific scrutiny. More data is often a good thing.
The specific purpose of the study was to test a model of communication during sexual activity. This model makes five predictions.
—The first three predictions are that people who are more affectionate, more sexually confident, or more sexually assertive will engage in more verbal communication during sexual activity with their partners. (As you'll see, "more verbal communication" simply means talking more.)
—The fourth prediction is that more verbal communication during sex will be associated with greater sexual satisfaction.
—The fifth prediction is that greater sexual satisfaction will be associated with greater relationship satisfaction.
Of course, you could test these predictions by just obtaining peoples' trust and chatting with them informally. But statistics yield two kinds of information that tend to be impossible to obtain through conversation.
1. Statistics can tease out relationships among variables.
The model predicts, for example, that people who are more affectionate will talk more during sex, and more sexual communication fosters stronger relationships. This makes sense, but perhaps communication during sex has little or nothing to do with what's happening. Maybe being affectionate directly strengthens a relationship, regardless of how much or little talk occurs during sex. Statistical analyses allow these possibilities to be distinguished. If I were a mental health professional, I'd want the details. When couples are dissatisfied with their sex lives, should I be prepared to devote some time to discussing bedroom talk, or should I just focus on things like expressions of affection?
2. Statistics quantifies the strength of relationships among variables.
The model predicts an association between affectionateness and sexual communication, for instance. But nobody, including the researchers, expects that association to be perfect. Some people who are otherwise affectionate might not like to talk during sex. Some people who aren't very affectionate might talk a lot. These exceptions don't invalidate the model. They model only predicts that among a large group of people, more affectionate people tend to talk more during sex, and more talk leads to better sex. Statistics can describe the strength of these associations. Here again, if I were a mental health professional who supports couples, I'd want to know the details. The stronger the statistical relationship between these variables, the more I'd want to track the extent of affection and communication within each couple.
Study methods
Participants consisted of 165 individuals who reported currently being in a sexual relationship.
Now we arrive at the first big, honking limitation of the study: Extremely narrow demographics.
Specifically, all participants were undergraduates in a heterosexual relationship. 71% self-identified as white. Although the age range was 18 to 42, the standard deviation was quite small (2.09, around a mean of 19.32 years), which strongly hints that the sample more or less comprised a bunch of students in the age range 18 to 22, plus one 42 year old.
We can still learn something from this study, but we can't make generalizations to everyone on the basis of what we learn about young white heterosexuals. And, there's an additional problem around the ambiguity of the term "relationship".
Prior studies focused on couples in committed romantic relationships (marriages, long-term partnerships, etc.). For this particular study, undergraduates only have to report "being currently in a sexual relationship with someone of a different sex."
That covers a lot of ground. Literally the only thing you can be sure of is that these relationships were fairly new, because, except for that 42 year old, the number of years an undergraduate could've been in a sexual relationship is relatively small. Meanwhile, some of the participants may have been in a monogamous relationship with the person they intend to marry, while others might've merely been hooking up with someone, non-exclusively, on a regular basis. Lots of other things fall under the heading of "relationships" too. In short, the study might tell us something about sexual communication and satisfaction, but it's not positioned to say much about relationships, because we don't know what kinds were represented in the sample.
The study is already heading south. Once you get to the measures, you discover it's gone straight to hell, methodologically speaking.
The key variable is verbal communication during sex with one's partner. Let's pause for a moment. What would you want to know about this variable?
If I were the researcher, I'd want to know things like how often each member of the couple talks during sex, how much they say, what they say, and how their partner feels about what's said.
What the researchers did was to present four statements for participants to either agree or disagree with (e.g.,"I say such-and-such during sex."). Each participant thus received a score ranging from 0 to 4 indicating how much they talk while having sex.
There are at least three problems with this approach. (Warning: some graphic language ahead.)
1. Supposedly, this measure tells you how often a person talks during sex, but it's not a very sensitive measure. Someone who tends to say one or two words every time will get the same score as a person who's constantly talking.
2. At least one of the items includes examples that are, at minimum, distracting. Here, verbatim, is that item:
“I say words during sexual activity (e.g., Oh God!, Yes!, Right there!, Pussy, Cock)”.
I'm sorry, but I can't read that with a straight face. It sounds like the porn version of the original Batman. ("RIGHT THERE! BANG! ZOWIE!")
More importantly, this item focuses narrowly on individual words and phrases. I can imagine some literal-minded participant disagreeing with the statement because they don't say those particular words. Or because they think to themselves: I don't "say words" during sexual activity. I talk. In sentences....
3. This measure doesn't address the content of what people say or how the partner reacts. Here again, some of the examples are problematic. Here's the only other item the researchers provided:
“I talk with my partner during sexual activity (e.g., about what you like, about your day, about what you want to do next)”.
That sudden, unexpected appearance of the phrase "about your day" is distracting at minimum. Talking about one's day during sex seems different from talking sexually during sex. I wouldn't fold these kinds of talk into the same item.
At least as important, we need to know how the partner feels about what's said. When you think of an older couple talking about their day in the midst of leisurely foreplay, it sounds intimate and sweet. You imagine they have a good relationship. But if a person keeps rehashing office politics even after their partner is quite aroused, well, you might expect some tension in the relationship.
(Sure enough, a 2022 meta-analysis found that the way people perceive the quality of sexual communication with their partner has a greater impact on satisfaction than the frequency of that communication.)
In sum, the approach to measuring sexual communication is so deeply flawed, the data are not going to be readily interpretable.
Affectionateness, sexual confidence, sexual assertiveness, and relationship satisfaction were measured by asking people directly about them, typically via ratings on a 7-point scale. More on these variables shortly.
Study findings
The results of the study mostly supported the predictions. The more affectionate and sexually confident people claimed to be, the more they claimed to talk during sex with their partners. The more people talked during sex, the greater their sexual satisfaction. And, greater sexual satisfaction predicted greater relationship satisfaction.
Along with the methodological limitations I've mentioned already, a separate issue is the focus on only one member of a couple. Besides overlooking that other person, this makes the study especially susceptible to something called omitted-variable bias.
Crudely speaking, this bias occurs when some variable that wasn't measured turns out to be responsible for the results. Here's a simple example: The researchers showed that the more a person talks during sex, the more they enjoy sex with their partner. Well, maybe people who talk more during sex feel more at ease with their partner. So, it's not the talk per se that makes the sex better, but rather the sense of comfort they have around the other person.
Omitted-variable bias can undermine the findings even more deeply than the way I just described. Imagine for a moment (this is a bit of a caricature) a guy who's loud, confident, cheerful, talkative, and not very self-aware. You've met this guy before. You've might heard the booming laughter before you saw him.
A guy like this may report a high level of sexual confidence because he's confident about everything. Since he's a big talker, he'll confirm that he talks during sex too. And, because he's a relentlessly positive person, he'll tell you that everything between him and his wife is great. Great sex, great relationship.
What this guy says might seem to confirm some of the researchers' hypotheses. But notice that he's not talking during sex because he's sexually confident. He's talking during sex because he talks all the time. And, he's not saying the sex is great because he talks a lot during sex. He's saying it's great because he thinks almost everything in life is great. Ditto for the relationship.
In short, overall self-esteem, general talkativeness, and positivity of outlook are some of the omitted variables that could easily account for the findings.
Two elephants in the room (i.e., the bedroom)
1. The researchers never did clarify why the the mere act of talking during sex might improve one's own sexual satisfaction, much less satisfaction with the relationship. The researchers focused on quantity rather than quality of talk, under the assumption that more is better, and they included any sort of talk, even if it wasn't erotic. I've already described some reasons why this doesn't work well as a hypothesis.
You might be tempted to say: It doesn't matter whether it's a reasonable hypothesis or not, because it's borne out by the results. My response would be that (a) the methodology is too weak to yield trustworthy data, (b) even if you could trust the results, the effects are small, and (c) even if the effects were large, they might be explained by omitted variables.
(I'm tempted to add that data have a way of turning out consistent with researchers' personal and theoretical biases, but I have no reason to suppose that's an issue here.)
2. Sexual satisfaction was measured with a single question "How would you describe your sexual relationship with your partner?" Participants provided what was ultimately recorded on a 7-point scale.
There are so many dimensions on which one might judge the physical, emotional, and perhaps even spiritual experiences of sex with one's partner. It might be uniformly splendid or uniformly awful. Perhaps it's somewhere in between, and one has mixed feelings (this aspect is great, that aspect is just so-so…) Reducing the complexity of the experience to a single number is not very informative. Recall that the sample consisted of anyone who reported being in a sexual relationship. How good you consider the sex in such a relationship depends a lot on your motives for being there. If it's a hookup, then having a great time, physically speaking, might be the main criterion, along with some influence from variables like how safe and respected you feel. If it's sex with the person you've been married to for three decades, the priorities may have shifted.
Even if it did make sense to reduce judgements of sexual satisfaction to a single number, doing so introduces the possibility of what psychologists call a halo effect. Decades of studies predict that if you think someone is a wonderful person but the sex is just ok, you might slightly overestimate the sex because you admire the person. Likewise, if you're having great sex with an awful person, your rating of the sex, though still high, might diminish a bit in light of their awfulness.
Sex and communication
Although we can't learn much from the study I reviewed here, others are informative. Here are a few examples.
—Lots of studies link better communication to better sex (in many senses of the word "better"). Both talking and listening are important.
—Some studies focus specifically on the benefits of communicating needs, preferences, and boundaries, or reflecting with one's partner on what was great or not-so-great about prior experiences. (Positivity and tact are recommended for the backward-looking conversations.)
—Some studies show how communication can build affection and trust between partners, meanwhile reducing conflicts that might influence the emotional atmosphere of sexual interactions.
—Some studies, building on Carol Dweck's work, detail the benefits of communication that reflects a growth mindset. If you assume that the quality of sex with your partner is determined by your individual identities and preferences – and is therefore fixed – it will be harder to foster improvement than if you have a growth mindset, where you assume that improvements are possible but may take time and effort, and that obstacles along the way are opportunities for growth.
Conclusion: Sex and statistics
As I mentioned, we can thank statistics for playing a small role in documenting the extraordinary variety of our sexual thoughts, feelings, and activities. Prior to the mid-20th century, such granular information about these mostly private experiences was unimaginable. Statistics provide certain kinds of information that can't be attained through other means, including personal experience. The novelist Georges Simenon claimed to have slept with 10,000 women; Wilt Chamberlain claimed twice that many, but I wouldn't call this great data. Men exaggerate. Even if these particular men were telling the truth, their samples would be limited to women willing to sleep with them. Those are large samples, though they pale in comparison to the millions of sex surveys participants since the 1950s – and, in any case, size ultimately doesn't matter much. Sleeping with people wouldn't teach you anything unless you've been observant and reflective.
Statistical data depicts clearly, if somewhat superficially, the desires and behaviors of large numbers of people, and the data can help increase our tolerance for desires and behaviors that differ from our own. In the 1999 film American Beauty (spoiler alert), Lester Burnham is shot by Frank Fitts after Colonel Fitts realizes that he's sexually attracted to Lester and kisses him. Projecting the blame for his homosexual impulses onto Lester is a purely Freudian scenario, and one that might've been averted if Fitts had known that according to Kinsey's 1948 report, roughly half of men acknowledge having reacted sexually to other men at least once in their lives. More broadly, public acceptance of homosexuality and other alternative sexual orientations and behaviors has been promoted by numbers such as that. In a word, statistics have made us more tolerant.
Statistics make us more vigilant too with respect to behaviors that can't be tolerated. For instance, roughly 1 to 5% of adult males are estimated to have pedophilic tendencies, a range that may seem small percentage-wise but translates into disturbingly large absolute numbers.
At the same time, we sometimes overestimate how much statistics can tell us about human sexuality, even when our interest is purely numerical. When experts get specific about how many people are predisposed to rape, child molestation, and other forms of sexual deviance, we need to be skeptical, because at least some people deny those tendencies on surveys and/or successfully repress them (e.g, not all pedophiles engage in molestation), while others act out but don't get caught.
The study I reviewed here illustrates a different sort of overestimation. In trying to reveal something about intimate experiences, the study got tripped up by some fairly run-of-the-mill challenges, such as reducing complex experiences to a single number, or using statistics to compensate for weak methodology. (The researchers' approach to statistical analysis was fine; it's most everything else that was problematic.) A study needs to begin with a reasonable hypothesis and a representative sample. You need to know your sample well enough to be able to say who they represent. And, you need to use measures that capture the experiences you wish to study. If they don't, the numbers you feed to the statistical analyses won't yield credible findings. Garbage in, garbage out.
Bottom line
Communication with your partner is a good thing. Communicating with your partner about sex is a good thing. Communicating with your partner sexually during sex is probably a good thing too, but, in the words of a brand-new guru (ChatGPT), "Every individual is unique and what works for one person may not work for another." Pretty sage advice from a celibate being.
Thanks for reading!