ADVERTISEMENT
  About the SA Blog Network













The Moral Universe

The Moral Universe


Dialogues on the psychology of right and wrong
The Moral Universe HomeAboutContact

Psychological studies are not about you

The views expressed are those of the author and are not necessarily those of Scientific American.


Email   PrintPrint



I have some bad news that, I hope, will turn out to be good news.  Psychological studies are not about you.  They make few if any predictions about how you should live your life, how to tell if you’re an introvert, or anything else about you as an individual.

I’m not trying to pull a Carly Simon on readers, most of whom probably understand this already.  My point is that a consumer of psychology, as it’s offered up in blogs, newspapers, and popular books, could be more than forgiven for coming away with the conclusion that our studies are about them.  This is because oftentimes, popular psychology is explicitly pitched this way: as information about you—the individual reader—how your mind works, and even what you should do to make it work better.

Take a recent blog post titled “10 Simple Things You Can Do Today That Will Make You Happier, Backed By Science.”  The author makes some perfectly good suggestions for increasing well-being: get some exercise, spend time with family, act generously, and so forth.  What’s more, she cites some great research.  For instance, the study she mentions by Lara Aknin and colleagues, on positive “feedback loops” between generosity and happiness, is part of a robust and fascinating research program.

Perfectly good suggestions paired with great research sounds like a recipe for informative and useful science writing, so why am I so grumpy about it?  Because when put together in the wrong way, these ingredients can mislead people about psychology and its purpose, or worse, produce personal disappointment.

A useful way to describe this tension is by way of analogy to baseball.  Batting average, the percentage of times that a player gets a hit for each “at bat” (or trip to the plate) is maybe the most famous measure of baseball performance.  Players covet averages of .300 (30% hits) or above, and players batting above .300 make much more money than those below .300.  As such, two batters with averages of .320 and .290 are worlds apart.  Crucially, though, these two batters would be virtually indistinguishable if you only saw each of them at the plate once.  This is because a .030 difference translates to a difference of about 20 hits per season, or only one extra hit for every 30 or so at bats.

Like batting averages, psychological studies tell us about behavior over lots of cases, but say little about any one case. Photo credit: AJW on Tour via Flickr

Findings from psychological studies are a bit like batting averages.  Except—and this is critical—you’re not the batter.  You’re the at bat.  This is because psychology is a science of populations.  A typical study might include 200 people, dividing them into groups (say, people told to act generously versus those told to act selfishly), and demonstrate a statistically significant edge in happiness for one over the other.  Like a batting average, though, even strong differences across groups tell us virtually nothing about how generosity or selfishness would affect the happiness of any one person.

Suggesting the opposite—that science provides readers with magic bullets that will change their lives today—is harmful in at least two ways.  First, it misrepresents psychology’s goals.  Experimental psychologists do not, by and large, claim their studies reveal much about any one individual.  Suggesting that we do can produce a false sense that psychology is over-promising and under-delivering.  Second, this type of writing can make people feel like outliers when they are not.  If a reader believes science has proven that generosity will make them happy, and it doesn’t, this person might wrongly feel like a counter-normative Grinch.

The good news is that this type of writing is totally unnecessary, because averages provide just as powerful (or more powerful) a message when divorced from any one person.  Knowing players’ batting average—though it doesn’t tell you about any one at bat—is crucial for predicting how a team will perform over a season.  Similarly, psychological studies, without telling us about any one person, can tell us about how changes in behavior (again, think generosity) might affect the well-being of whole populations.  Such insights can inform policy on a broad scale; in fact, the newly created White House Social and Behavioral Science Team will use psychology in just this way.  Most sciences—including psychology—are much better suited to these broad applications than to telling any one person about their life.  That said, it probably wouldn’t hurt to get some exercise anyways.

 

Jamil Zaki About the Author: Jamil Zaki is an assistant professor of psychology at Stanford University, studying the cognitive and neural bases of social cognition and behavior. Follow on Twitter @jazzmule.

The views expressed are those of the author and are not necessarily those of Scientific American.





Rights & Permissions

Comments 4 Comments

Add Comment
  1. 1. ahhthesimplelife 5:24 pm 09/6/2013

    Hello Jamil, I have a different perspective on the blog post that you refer to. In fact, I have a similar post on my own blog “Happiness How-To’s” at http://www.ahhthesimplelife.com/happiness-tos/ . As a matter of fact, the Huffington Post has many similar articles at http://www.huffingtonpost.com/news/happiness/

    I believe that there is a flaw in your logic. To use an analogy – there’s a body of scientific research that suggests a plant-based diet is better for one’s health. According to your logic, I should ignore this dietary advice since I did not participate in any of the studies on which that advice is based. Of course, ideally, a person’s diet should be tailored to his particular health profile.

    I think articles such as Belle Beth Cooper’s “10 Simple Things You Can Do Today That Will Make You Happier Backed By Science” and my “Happiness How-To’s” provide some valuable information to our readers. I doubt that anyone reading these articles expects that if they follow the tips they will magically become happy. However, trying some of them out may well put them on the path to a happier life.

    Link to this
  2. 2. stalder 9:34 pm 09/6/2013

    Psychological studies might be about you

    I strongly agree that popular-press authors go a little too far if they point fingers at an individual reader, or tell an individual reader how he or she would or should behave, based on research studies. Research studies are based on averages.

    But to say that “psychological studies are not about you” goes too far the other way, in my view. One comparison (I know it’s not perfectly equivalent) is to tell one who is eligible to vote that one should not worry about voting because most elections are not decided by one vote. From another angle, don’t worry about voting because studies that might predict dips in voter turnout are not about you but rather based on the average eligible voter.

    Perhaps more apt, it’s like telling a bystander within a large group of bystanders not to worry about having ignored a victim, or not to worry about trying harder to step up and help a victim, because diffusion of responsibility (in the “bystander effect”) only occurs on average. On average, an individual bystander is less likely to help the more bystanders there are, but not all of us succumb to this effect. Probably the best advice, by the way, if you are ever a victim with a crowd of onlookers is, if you’re able, to point to a particular individual in the crowd and request help – this strategy tends to slice through diffusion of responsibility (although this result is only on average, so this strategy might not always work). This strategy does not presume that the individual to whom you point would have succumbed to diffusion of responsibility, but this strategy does increase the chance that you receive the help you need.

    Even though research studies are based on averages, the results probably pertain to some of the readers, especially if the data are normally distributed. By comparison, in a bimodal distribution, for example, very few if any participants have necessarily scored the average.

    Dr. Zaki is not telling anyone not to vote or not to try to help a victim in need. In fact, Dr. Zaki sort of encouraged readers to exercise (presumably because exercise is helpful on average). I acknowledge that my above comparisons are not perfect, and I’m sorry if I am missing any crucial part of Dr. Zaki’s argument. However, I would encourage each individual reader of psychological studies to be open to the possibility that the research is indeed about him or her, even though it need not be. Voter turnout and helping behavior might increase. Such potential positive consequences of my suggestion (depending on the nature of the research) can be weighed against the potential negative consequences noted by Dr. Zaki, and then we can try to see where the net effect falls for the average reader.

    Link to this
  3. 3. Jamil Zaki in reply to Jamil Zaki 1:47 pm 09/7/2013

    stalder and ahhthesimplelife:

    You both make compelling points and I really appreciate having this in the discussion. I think it’s a tricky–but VERY important–job to translate findings from a scientific field to the public. One of the trickiest parts is how to describe the relevance of what we do to the public. Of course our studies are not *totally* irrelevant to individuals’ lives; averages, of course, are made of individuals. Plant-based diets, exercise, etc. are great examples of averages that I think we should take to heart when making personal choices. I also love the example of voting.

    That said, I do think it’s tricky to insert too many specifics about what people should expect from making a given choice, listening to a given study, and so forth. Let’s go back to the example of voting. You’re right: if I were to tell you “voting doesn’t matter,” that’d be wrong. But, if I were to tell you “if you vote, your candidate will win,” and that turns out to not be true, it could shake your faith in the voting system. Similarly, I’ve had *many* people tell me that they took advice from one study or another, and it didn’t work for them, and now they don’t believe that psychology can tell them anything.

    So I think the challenge here comes in striking a balance for readers: giving them advice that is useful (ahhthesimple life and Belle Beth Cooper both do this, as I pointed out), but not “overpromising” in a way that could lead to disappointment or loss of trust in a scientific discipline, especially if that loss of trust comes from a simple misunderstanding about how our field works.

    Again, thank you for the great thoughts!

    Link to this
  4. 4. belarius 12:49 am 09/8/2013

    As in most things, declarations about whether any psychology study is or is not “about” a person is a function of several factors. One of these is the style or “flavor” of the research (and psychology varies a great deal). Another is “effect size,” a statistical measure of how powerful an effect is.

    Dr. Zaki’s research belongs to the category of “social neuroscience,” which merges the experimental tradition of social psychology with technologies that image the human brain (chiefly fMRI and EEG). The description he gives is completely accurate for most of the work done in that domain. Social psychologists, for the most part, study large groups of people and, in doing so, tease out subtle group effects; here “subtle” is loosely analogous to “a small effect size.” At the same time, human imaging methods are also averages: The spatial resolution of these techniques is far too fuzzy (in both time and space) to resolve actual neural circuitry. Instead, they identify average activity in millions of neurons at a time, on time scales much slower than neurons fire. Describing Dr. Zaki’s research in these terms is not an insult: It’s simply a factual description of the limits of certain research paradigms.

    However, not every psychological topic deals in small effect sizes, nor is every study performed by collecting only a few pieces of data from many hundreds of participants. For example, much of the work on the basic processes of memory sets unambiguous limits on how well the memory of almost everyone can be expected to work. A large body of work should convince us that (1) memories are somewhat plastic and very subjective, to such a degree that eyewitness testimony should be considered fairly weak evidence (2) “flashbulb memories” are no more accurate than other kinds of memory, despite seeming very clear to those who hold them and (3) the number of things we can simultaneously think about (our “working memory”) is actually quite constrained, but surprising feats can be achieved within these limits through the use of memorization strategies (or “mnemonics”). Unless you are a savant like Kim Peak, *these are statements about you,* as well as being about just about everyone you know. Psychologists can say this with some certainty because the effect sizes in question are unambiguous. Differences in batting average may be small between pro batters, but the difference between having them bat blindfolded vs. bat with earplugs is going to be a lot bigger than 20 hits a season, and the efficacy of the blindfold tells us a lot about what senses a batter need to hit the ball.

    Another important consideration is the explanatory power (i.e. the predicted effect sizes) associated with different groups. Some psychological categories of person (e.g. introverts) are diverse, diffuse, and difficult to generalize about. Other groups, however, are quite distinct from the general population. For example, although unmedicated paranoid schizophrenics are a diverse group, they are sufficiently distinct from the general population that (1) almost everyone will agree that they are ‘not normal’ and (2) a trained clinician can in many cases recognize their condition within just a few minutes. Obviously, such a group is still varied enough for there to be important exceptions, but these two groups (introverts vs. unmedicated paranoid schizophrenics) differ so dramatically in their explanatory power that Dr. Zaki’s consideration about the personal relevance of psychological research applies only to one of them. Two pro batters might be hard to distinguish, but even a casual observer would notice immediately if Dr. Zaki or I tried our hand against a major league pitcher.

    There is a temptation to think (1) that the psychology one is most familiar with is representative of psychology in general, and (2) that the way in which journalists write about psychology reflects how psychologists think about their own research. This temptation, incidentally, is the “availability heuristic,” which is well understood in psychology, and that absolutely applies to you. It also applies to myself, and I’m afraid it applies to Dr. Zaki as well. However, the one-sentence quip, as usual, short-changes the situation. Instead, whether psychological research applies to you depends on the specific research methods, its reported effect sizes, and on the ways in which you are atypical.

    Where does that leave our understanding of pop psychology? Dr. Zaki is absolutely right that most people who are trying to sell you a way of life are misrepresenting (and usually misunderstanding) the nature of the research they cite. In practice, it is difficult for a lay audience to read primary research, and journalists often treat molehills like mountains because misunderstanding the science nets them a flashier headline (see “confirmation bias,” which also applies to you). There is no simple solution. Psychologists must communicate more clearly about the relative predictive strength of their theories (even if that means they have to admit that some of their results amount to molehills), and both journalists and readers need to be more skeptical about the difference between “statistical significance” (i.e. “is the effect real?”) and “practical significance” (i.e. “how big is the effect size?”). Dr. Zaki’s warning is a step in the right direction, insofar as it helps people develop appropriate skepticism. That said, beware blanket generalizations of *any* kind.

    Link to this

Add a Comment
You must sign in or register as a ScientificAmerican.com member to submit a comment.

More from Scientific American

Scientific American Back To School

Back to School Sale!

12 Digital Issues + 4 Years of Archive Access just $19.99

Order Now >

X

Email this Article



This function is currently unavailable

X