Psychology’s Greatest Misses (Part 1/3)

A tour of your favorite wrong ideas.

In the early 20th century, many physicists perceived a crisis: anomalies such as blackbody radiation, the Michelson–Morley null result, and observations of discrete atomic spectra conflicted with classical predictions, and new frameworks—such as Einstein’s special relativity and the emerging quantum theory—were beginning to replace them.

In some sense, it was less of a “crisis” and more how science is supposed to work. Many empirical observations didn’t fit existing theories, so new ideas were necessary to build a more accurate and complete picture of the physical universe.

One could be forgiven, therefore, for assuming that the replication crisis across social sciences, including psychology, is a similar beast. A person could reasonably mistake what is going on now as something of a Kuhnian revolution, a transition to a new, more productive theoretical framework.

The replication crisis is, however, something else altogether. It is more akin to the financial crisis of 2008, when everyone discovered that the models they trusted were illusions built on bad assumptions.

Replication is a cornerstone of science. If a field’s findings systematically don’t replicate, then it’s not similar to the situation in physics, in which new theories were needed to account for new data. Instead, it’s not clear that that field was doing science in the first place. The solution to the replication crisis isn’t new theories—it’s new procedure. That is the crisis faced by psychology.

We at Living Fossils have been critical of the field of psychology because we believe in the promise of a science of the mind to improve the world. When this promise is squandered, whether in the form of bad therapy or poor science, we mourn the loss.

In this series, we discuss some common hypotheses educated laypeople believe from the field of psychology, particularly social and clinical psychology. All of them are either downright wrong or completely misunderstood, adding weight to the possibility that psychology may have, so far, done more to confuse than clarify.

The 3-Cueing System – RK

Steve Pinker’s book, The Language Instinct, was on my mind the first time I learned about the 3-Cueing System: a new trend in teaching reading. One part of the idea was that children learn to read the same way they learn to talk, so you should teach them the same way: expose them to a lot of written words—just like they are exposed to spoken words—let them guess at the word, and correct them when they are wrong.

Having read The Language Instinct, this idea struck me as, well, dum. Humans have a spoken language instinct, sculpted by evolution for many thousands of generations. Learning spoken language is an evolved natural competence. But because written words came about yesterday, evolution hasn’t had nearly enough time to develop a written language learning module.

Writing is more like math. Evolution didn’t give us a specialized system to learn it, so we have to do it the hard way, recruiting our general learning abilities, which takes a while and is painful.

There’s a great podcast I cannot recommend highly enough called Sold a Story that goes through the sad, sordid tale of good-intentioned people insisting on teaching young children to learn written language in a way that didn’t work. The podcast recounts countless children’s lives impacted by this horrible decision on the part of adults, all of whom should have known better.

Long story short: If you’re a parent, use phonics.

Attachment Styles – JZ

In infants, attachment as a mechanism—an evolved system that keeps young organisms close to caregivers—is one of the better-supported ideas in developmental psychology.¹ It appears in many mammals and birds, too, where proximity means protection.

Human infants can be categorized by their reactions to a caregiver leaving and returning (known as the “Strange Situation”) as anxious, avoidant, disorganized, or secure. The popular narrative is that people carry these “attachment styles” into adulthood—particularly into the arena of romance—like so much emotional baggage. The evidence suggests otherwise: early attachment explains only about 15% of the variance in later attachment, and roughly one-third of adults change attachment categories within a few years of their last measurement. Early patterns bias later ones, in other words, but don’t determine them.

Why not? Because far more important than how a person attached to their caregiver as an infant is who they are attaching to now. This is another example of the situation—the person or thing being attached to—mattering more than the unique person doing the attaching. Since attachment is largely a relationship-specific orientation, you can be “secure” with one partner and “anxious” with another, depending on who the partner is and how they make you feel.

To me, the way to make “attachment style” genuinely useful is to treat it as a description of strategies, not traits.² In familiar situations, people tend to reuse what worked in the past. So yes, if your mother was unavailable and you coped by escalating your bids for attention, you might do the same with a similarly distant partner—or boss. Attachment isn’t a trait you have as much as a pattern you might be inclined to repeat when the relationship context resembles one you’ve known before. Since relationships vary, so do the strategies people use.

Grief, Stages of – JZ

I’m not sure how movies, TV shows, and books would even grapple with grief if they didn’t have the five stages to lean on. These stages—first outlined in Elisabeth Kübler-Ross’s On Death and Dying (1969)—are:

1. Denial: “I must be dreaming.”

2. Anger: “Why did this happen to me?”

3. Bargaining: “I’d give anything to have her back for a day.”

4. Depression: “What’s the point of trudging on?”

5. Acceptance: “I miss him, but I know he’d want me to be happy.”

It’s one thing when pop culture runs with an idea. It’s quite another when, despite repeated criticisms from bereavement researchers, nearly half of clinicians still regard the following statement as either definitely or probably true: “The process of grief can be expected to progress through a predictable series of stages, starting with denial and ending with acceptance.”³

As a recent review put it:

Major concerns [with using the stages-of-grief theory] include the absence of sound empirical evidence, conceptual clarity, or explanatory potential. It lacks practical utility for the design or allocation of treatment services, and it does not help identification of those at risk or with complications in the grieving process. Most disturbingly, the expectation that bereaved persons will, even should, go through stages of grieving can be harmful to those who do not. Following such lines of reasoning, we argue that stage theory should be discarded by all concerned (including bereaved persons themselves); at best, it should be relegated to the realms of history.

Well said.

Power Posing – RK

In 2012, millions of people were instructed to go stand in a bathroom stall like Rocky.

Amy Cuddy’s TED talk on “power posing” became one of the most-watched talks in the history of the platform. The idea was disarmingly simple and intuitively appealing: hold an expansive, “high-power” posture for two minutes and your testosterone would rise, your cortisol would decrease, and you would become more willing to take beneficial risks. This change, in turn, could start a cascade towards a promotion, a better job, and so on. The 2010 paper on which the talk was based says that “…posing in displays of power caused advantaged and adaptive psychological, physiological, and behavioral changes, and … a person can, by assuming two simple 1-min poses, embody power and instantly become more powerful...” The inevitable book would become an international best-seller.

But short, punchy effects with small samples in social psychology—this was 2010—turned out to be exactly the kind of thing that didn’t age well once people started checking.

One attempted replication in 2015 failed to find the key hormonal and behavioral results, though they did find that people in the Rocky pose reported feeling more powerful. Shortly after that, the first author on the original paper—not the senior person, with the TED talk and best-seller—wrote, “I do not believe that ‘power pose’ effects are real.”⁴

Failures to replicate have continued, though maybe the risk-taking part will be vindicated.⁵

Generally, however, what started as a set of bold, exciting claims—it gets into your hormones!—has been reduced to the possibility that you can change how you feel, a bit, by making yourself big.

Yawn.

Stereotype threat – RK

When I was in graduate school, it seemed to me that half the field of social psychology was studying what was called “stereotype threat.” The idea was that you can cause people to perform worse on some task—math problems, or whatever—by reminding them, even subtly, that they are a member of a group that is stereotypically bad at that task.

Shortly after I got to Penn, perhaps around 2005 or so, two undergraduate students presented some research in which they tried to replicate and extend the work on stereotype threat. They were disappointed and puzzled that they could not replicate the effect.

I knew why. It’s not true. In contrast, one of the few areas of social psychological research that has held up is Lee Jussim and others’ work on what is called stereotype accuracy, which is just what it sounds like. Turns out people don’t suddenly become less smart if you remind them other people think they are, but people can use information about what category people are in to make (limited, true) guesses about their traits.

Stereotype threat was a popular topic of study for one of the usual reasons: the work was really a political message, a judgy message. Isn’t it so bad bad bad that people have stereotypes and oh my god clutch my pearls those stereotypes are undermining achievement on tests and other tasks.

Transference – JZ

The notion of transference—that clients unconsciously redirect feelings and expectations from important figures in their past onto their therapist—remains a cornerstone of many psychotherapies, particularly those descended from psychoanalysis (known as psychodynamic psychotherapies).

Given that much of what Freud theorized turned out to be wrong, what does the research have to say about transference? Essentially nothing.

Despite its well-established importance in psychoanalytic theory, there is a scarcity of empirical evidence on the relationship between a therapist’s transference interpretation (TI) and therapeutic outcome. The current scientific literature shows no consensus on the existence and nature of such an association. - source

In my view, the legitimacy of transference is undermined for many of the same reasons as attachment styles. Both assume that people carry fixed templates for relationships—pre-written scripts into which new partners, friends, or therapists are cast. Yet in my experience, clients usually experience me as a distinct person with whom they form a genuinely new relationship.⁶

Of course, a therapist might in some way resemble a parent, partner, or sibling, and that resemblance can sometimes prompt useful emotional processing. But isn’t this just ordinary perception at work? We interpret new people through the lens of familiar ones, just as we make sense of new buildings by comparing them to buildings we’ve seen before. Noticing who someone reminds us of doesn’t mean we’re re-enacting an old dynamic with them.

And even if we do re-enact—and resolve—an old dynamic, there is no reason to think we’ll walk away cured of our troubles. Some therapists treat “resolving the transference” as a panacea, overlooking the fact that because relationships come in many forms, a resolution in one does not generalize to the rest.

Further research is needed to say what, if anything, is going on here.

Why Do Bad Ideas Travel So Far?

Great question.

In the next post in this series, we’ll discuss more myths before attempting to answer it. Thanks for reading and stay tuned.

1. John Bowlby deserves credit for breaking the ice on this idea.

2. For another example of something that might be a strategy rather than trait, see Rob’s article on Depression.

3. To be fair, things have improved substantially over the past twenty years, and clinicians still possess a more accurate understanding than the general public.

4. For the technically inclined, see this p-curve analysis.

5. Reviewing the papers in a Special Issue on the topic, it is noted that there was “a reliable non-zero effect on felt power.” Here is the pdf, if you’re interested.

6. “Transference” is sometimes used colloquially to mean simply how a client treats their therapist. For example: “They have quite the transactional transference to me: they expect to come in, pay, and get their money’s worth of wisdom,” or “They treat me like a god—everything I say gets written down as profound.” To me, these are better understood as ordinary relational dynamics. They may be complex, but they do not justify a separate or special term—certainly not one with so much historical baggage.