Skip to navigationSkip to content
REUTERS/Michaela Rehle
Integrating cognitive science insights into anti-sexual harassment trainings could make them more effective at teaching and engaging employees.

Anti-sexual harassment trainings should teach people how their brains work

Elizabeth Weingarten
By Elizabeth Weingarten

Senior fellow, New America's Better Life Lab

If you’d asked me in college how I would respond to my (future) boss if he rubbed my thigh in a bar and inquired about my sex life, I would have answered unwaveringly: I’d push his hand away, walk out of the bar, and report him to human resources the next day. Easy. Next question?  

But when I entered the workforce a few years later, I discovered reality is much murkier. What actually happened: his hand stayed on my thigh, I stuttered a guarded response to his question and took months to report him to HR (which initially did nothing to address my concerns). Far from just shaking it off, I felt uneasy around him and refused his future invitations to go out for drinks, which seemed to have a direct impact on the kind of work and feedback he gave me, and my “talent” in his eyes. For months, I felt lonely and depressed and worthless, like my promising career had just gone careening into a ditch. When this guy signaled that I wasn’t good at my job, it didn’t occur to me that maybe he was wrong. Especially because he was, as HR told me, one of the organization’s highest performers—and there just wasn’t much it could do beyond a slap on the wrist.    

Why the disconnect between how I thought I would act and my actual behavior? One partial explanation is affective forecasting error, a pattern of distorted or irrational thinking (also known as a cognitive bias) which deludes us into thinking we know how we’ll react to a high-stress situation like sexual harassment when really we have no idea.

In a new report on sexual harassment authored by me and my New America colleagues at the Better Life Lab, we suggest that integrating cognitive science insights like this one into anti-sexual harassment trainings could make them more effective at teaching and engaging employees on what is a particularly challenging subject matter. It’s an idea that’s somewhat revolutionary for trainings, which so often are run as check-the-box solutions to legal compliance.

To be sure, adjusting sexual harassment trainings in this way won’t replace necessary structural workplace changes: Involved organizational leadership, transparent reporting systems, clear and accessible policies, gender-equal pay structures, and gender parity across all levels of the institution—all of these things are essential pieces of an inclusive workplace. Integrating cognitive insights into training needs to be part of a holistic strategy to shift social and cultural norms that normalize harassment, and to dismantle the policy structures and systems that allow it to thrive.

Nor is our report meant to advocate for the sort of unconscious bias training that has become a common part of diversity efforts, which some research now suggests can be ineffective or even harmful. (According to Harvard research, when trainings are compulsory, participants can respond with “anger and resistance,” or with more hostility towards other groups afterward).

Rather, this approach is about “using cognitive science to engage people in learning, and combining it with data and neuroscience,” says Rory Gerberg, a consultant at the diversity and inclusion firm Readyset who runs sexual harassment and D&I trainings that incorporate cognitive science insights (and also happens to be a friend).

Principles like affective forecasting error engage participants because they can help explain empirical data or research findings, explains Gerberg. For instance: Why is it that so few people (only about 30 percent, according to the EEOC ) report sexual harassment when it happens? Or why is it that an unequal power dynamic so often exists between a harasser and victim? And why might we let certain people off the hook for harassment, but not others?

Where and how to plug in cognitive insights

One crucial part of any training is to deepen participants’ understanding of the nature of the problem—especially why it continues to be so pervasive. Some experts do this through exercises that show participants gaps in their own knowledge.

For instance, gender stereotypes can contribute to harassing behavior or “gender policing.”: The latter occurs when women ( or men) don’t act in accordance with their gender stereotypes—for instance, if a woman is in a male-dominated field like mining—and they are harassed and disparaged not for sexual purposes, but in order to drive them from that field and make them question their belonging. In order to shed light on how pervasive and blinding these stereotypes can be, Eden King, an associate professor of  industrial-organizational psychology at Rice University, asks her students to solve a common riddle.

The riddle sets up a story about a father and his son who get into a car crash that kills the father. The son is rushed to the hospital. But just as he’s going into surgery, the surgeon yells, ”I can’t operate on this boy, he’s my son!” How is this possible? Of course, the surgeon is the boy’s mother –a reveal that both often surprises participants (especially if they couldn’t get the riddle) and also demonstrates the pervasive nature of a problematic way of thinking.

In her trainings, Gerberg  relies on a different kind of reveal. She asks her participants to guess the percentage of sexual harassment that’s reported, and then shares the actual (very low) numbers. Piquing participants’ interest in this way, she says, means they’ll more actively engage with the material, and remember it better.

It also sheds light on a critical cognitive gap between our perceptions and reality. Although much of what motivates human behavior hinges on a simple question—is everyone else doing it?—we’re not great judges of what other people are actually doing, particularly if we can’t see them in action.

That’s particularly true when it comes to sexual harassment. Researchers have found, for instance, a significant gap between men’s perceptions of how many of their peers are engaging in harmful harassing behavior, which they estimate to be nearly everyone, and the reality, which is much smaller. This misperception may prevent men from intervening to stop bad behavior (rendering things like bystander intervention training less effective), or even make them more likely to engage in the behavior themselves because, they reason, everyone is doing it. For instance, someone may privately think it’s wrong to make sexist jokes, but mistakenly think that everyone else holds a different view—that sexist jokes are no big deal. Behavioral scientists call that phenomenon “pluralistic ignorance.” It’s a brain behavior that can keep people silent, willing to look the other way, and give rise to toxic and hostile workplace environments that tolerate or normalize sexual harassment. Explaining pluralistic ignorance is another way to help elucidate why sexual harassment happens, and to give people an opportunity to discuss what kinds of harassing or sexist behavior have become normal in their office. It’s this dialogue that can shift cultural norms, which can ultimately help drive behavior change.  

Power is another factor that can widen the gap between perception and reality, and another potential discussion point in a training that incorporates cognitive science explanations for why harassing behavior so often flows from an unequal power dynamic. Having power can, for instance, make individuals more likely to think others will be romantically interested in them, sometimes leading them to perceive romantic interest when it doesn’t exist. This is known as the overperception bias. Researchers also find that regardless of power and gender, many of us underestimate how hard it can be for the recipient of an unwanted romantic gesture to say “no” to a request, a fact that can be useful for any employee thinking about hitting on a colleague.

Learning about moral licensing could also help close the reality-perception gap by exploring our tendency to dismiss or excuse certain kinds of negative behaviors. We engage in moral licensing when there’s a disconnect between our behavior and the perception we have of ourselves as good, moral people. In this case, we’re forced to justify our behavior so that it matches with this image, lest we experience cognitive dissonance. One way we do this is by using a virtuous act  like running a 5K to offset a weekend of Netflix and chilling. At work, we may excuse or dismiss bad behavior (like sexual harassment) by remembering our past good behavior (like starting a women’s empowerment initiative).

Just as we morally license our own behavior, we can do the same for the behavior of people with whom we work. If you’ve seen your boss do good things—for instance, starting that women’s empowerment initiative—you might be more likely to forgive him for a few inappropriate comments at a meeting, or even for making a pass at you at a party. Could an understanding of this dynamic give you more confidence to call him out next time it happens?   

The way forward

Once we understand a little bit more about our own brains, we can journey into the minds of our colleagues through a tactic called perspective-taking, which some research suggests could also prevent harassment. The process is straightforward: it involves employees considering, some experts say most effectively through peer-to-peer dialogue, an unfamiliar perspective to them. The goal is for individuals to develop more empathy for people with different perspectives, and to translate what can be an abstract concept for some into something concrete and visceral.   

It’s one thing to engage participants in a sexual harassment training, and it’s another thing to actually change attitudes and behavior. “There is a long-standing debate within psychology and related fields, with no clear consensus, about what leads to a reduction in biased behavior,” Meg Bond, the Director of the Center for Women & Work at University of Massachusetts, Lowell, told me in an email. It’s a bit of a chicken and egg dilemma: Some camps say you need to change the way people think in order to nudge behavior, others believe behavior change will shift the way people think. For her part, Bond is “moving toward framing [ anti-harassment] interventions around changing behavior versus cognitions.”

For my part, learning about affective forecasting error gave me some solace, and helped me understand why my expectations and behavior didn’t line up. It allowed me to stop beating myself up and gave me tremendous empathy for women who don’t immediately act in the wake of harassment. There’s power, it turns out, in understanding the why of harassment—and not just the when and the how.  

Elizabeth Weingarten is a senior fellow in New America’s Better Life Lab.