While political tensions certainly seem to be at an all-time high, it’s also true that we’ve long been a nation divided. Even if not recently, at some point many of us have found ourselves trying to convince someone with different beliefs to come over to our side. If I had to guess, it probably didn’t work all that well.
Our failure wasn’t likely due to lack of effort. Political scientists have spent decades on the sticky question of how to get people to change their beliefs. They’ve run experiments in labs and by knocking on doors. They’ve sent mailers and flooded people with rational arguments. And, you guessed it, they rarely saw success.
In one particularly nauseating experiment, scientists provided research participants with blatant misinformation via a fake news article. Shortly after, they provided them with a correction with actual data to dispute the initially wrong information. Frustratingly, the correction often didn’t do much to counteract the misinformation. In fact, certain groups of people – ones particularly tied to an identity – became more tied to the misinformation once given the correction. They felt more certain it was true.
It would be easy to read about this study and conclude that the zealots who became more attached to their beliefs are these crazy people on the political fringe. But the study actually suggests something about all of us. Maybe it’s not so extreme, but these kinds of phenomena happen readily for us outside of the political arena.
If I start out with the belief that I am terrible at tennis and then someone tells me that I’m wrong (they claim to know that I’m excellent at tennis), then I’m likely to not only not believe them, but to attach even more strongly to my belief that I’m a bad tennis player. Now I feel I have to prove or justify something, even if it’s not something I particularly like.
The researchers in the political study called it the “backfire effect,” and it seems clear to me that it cuts across situations and types of beliefs. When a belief is embedded enough, challenging it – even with supposedly solid data – only serves to make us cling more tightly to it. It’s not because we necessarily want to grasp the belief; it’s because by that point it’s so interconnected to our predictions of the world that our minds really won’t let us release it.
The Beliefs That Shape Our Lives
Our beliefs about politics or about our aptitude for tennis may or may not cut to our core, depending on who we are. But we all possess deeply held and felt beliefs about ourselves, others, and the nature of the world. Beliefs like, “I’m not worthy of other people’s care,” or “I can’t trust myself,” or “People are only out for themselves,” are the kinds that shape how we live our lives day by day, moment by moment. That’s why our beliefs end up shaping how and who we are.
We’re all familiar with thoughts and emotions – and we can go a step further to see beliefs as enduring ideas that we carry forward to make meaning out of our experience. Beliefs form through experience, whether that’s direct experiences we have or through the experiences of others that we observe up close or even far away. We form beliefs as humans precisely because they allow us to make predictions about the world.
The Brain as a Prediction Machine
You may have heard the idea before that our brains are ultimately just prediction machines. This isn’t metaphorical or abstract. According to what’s called the Bayesian brain hypothesis, the mind is essentially a series of predictions – guesses about everything from how something will look and smell and sound to how someone will respond to what will happen next in a situation.
The theory, which has been well supported, says that it would be far too inefficient for the brain to have to build its models of what’s happening around us in real time from the infinite pieces of data in our field of being. Instead, it takes in just enough information to compare it to all of our past experiences and then makes predictions. That mug will be hot. That bird will be fast. That person will be critical.
The Power of Prediction Errors
Then, our sensory organs will perceive information from the environment and help us update as needed. If we pick up the mug and it’s indeed hot, we confirm our prediction and it strengthens it for next time. If we pick it up and it’s cold, our brain registers surprise, or a prediction error.
This is all going to become very important in a moment when we get further into understanding how we can actively work on changing beliefs. For now though, just simmer in this for a moment. What’s being suggested here is that we don’t see things as they are – at least not initially – but rather as we expect them to be. These predictions aren’t corrected through rationality or being told they are wrong, but only through our direct experience. When our sensory experience shows us that our prediction was wrong, it often registers as a jolt. In that moment, our belief and indeed who we are changes.
This process is happening all of the time and at multiple levels – from the micro level (realizing we’ve been mishearing a song lyric) to the macro level (realizing we couldn’t trust our betraying partner after all). Our prediction errors are the basis of human learning. We have experiences, we develop predictions, we experience prediction errors, we update our predictions – and around and around we go.
Our beliefs operate by these principles. We hold our beliefs about ourselves and the world firmly in the world not until someone comes and argues us out of them, but until we experience prediction errors that allows them to update.
Why Arguments (and Affirmations) Often Fail
The concept of prediction errors – and how crucial they are in updating our beliefs – is powerful because it helps us understand why certain things do and do not work when we want to change. As I’ve said, simply arguing with someone about their belief doesn’t form a prediction error, so something vital is missing in the change process.
Now think about this as me ‘arguing’ with myself. I’ve been told by self-help gurus or the popular culture to remind myself that I am good or talented or unstoppable. Assuming I don’t historically believe those things, my brain perceives this is just the same way as if some political canvasser was coming to knock on my door and tell me that I’ve got it all wrong about my position on gay marriage. My brain isn’t distinguishing that it’s me arguing. It just experiences me as trying to talk me out of something.
I’m not going to be convinced of an alternative without some kind of prediction error. Show me the proof, our brains are essentially saying. Less words, more action.
I’m not trying to totally knock the power of alternate perspectives or personal affirmations here. But if we are going to use affirmations, we have to understand the nuance. Just slapping some healthier-seeming words on ourselves won’t do much because our minds want to do battle with them. That battle can actually create more tension and distress.
There can absolutely be a role for affirmations or mantras when we ground them in the principles of prediction error. When our self-talk is rooted in reminding ourselves of times that belief hasn’t proved true or in helping us pursue an opportunity for a prediction error, it can support real change.
If we are struggling with a belief that we are incapable at work, it can in fact help to spend some time reflecting on a time when you thought you were incapable at work and you ended up being very able to do what you needed to do. That’s referencing back to a time when a prediction error happened, and research tells us that this can strengthen our capability going forward.
Alternatively, if we believe we are incapable at work and we use an affirmation like, “Even if I struggle with this project at work, I will be kind to myself,” we can set ourselves up to be willing to work on the project. And being kind to ourselves after, regardless of outcome, might be a prediction error for our minds – especially if we are used to getting self-critical.
Therapy as a Home for Prediction Errors
I love thinking about what makes therapy actually helpful – in part because despite science proving that it is, it’s hard to say for sure why. This model of prediction error is a core part of how I think about therapy changing us.
Some therapy may use these concepts explicitly in terms of the content of the sessions.That would mean we might be talking about identifying some of our core beliefs and then creating opportunities for prediction errors in our daily lives. For example, we could think together about what it would be like to have a hard conversation with someone while soothing the part of ourselves that is jumping to conclusions. Those kinds of experiences could potentially help us see that we are more capable than we believed and that people may respond differently than we believed.
But to me, what’s more fundamental is that therapy can itself serve a source of prediction error. That’s getting at the process and nature of therapy more than the content.
“I couldn’t ever share this with anyone.” Wow, I just did.
“No one could ever tolerate my anger.” Oh look, my therapist seems to be staying with me.
“I will be abandoned if I ask for what I need.” She was glad to help me.
“It’s better to forget that ruptures happen than talk about.” Wow, it feels good that we cleared that up.
A good therapist is helping to facilitate prediction errors relationally throughout the process. Whether or not it’s discussed explicitly – and I think there is value in naming it – in therapy we are updating our beliefs through the specific relational process it entails.
Trauma Therapy and Prediction Errors
To take one last step further, we can think about the work of many forms of trauma therapy as embedded in the idea of prediction error as well.
Let’s take EMDR, one of the most widely discussed trauma treatments in many years and a treatment we offer at Galia. In EMDR, a therapist is asking a client to reprocess memories of difficult situations. Other therapies have been based on the idea that if we tell the story of trauma over and over, it will eventually lose it’s charge. That often doesn’t happen – and it’s really hard on the person telling it, so EMDR is different. In EMDR, we are revisiting memories (or even future states) that are living within us in order to, this time, experience them differently.
We are processing them in a specific way that introduces prediction errors. We are imagining a situation, for instance, while not experiencing the same habitual emotions. We are remembering while feeling deeply connected to the therapist or ourselves – which is often very different from how the original experience happened.
It’s been said that events become trauma not only because they were distressing, but because of the aloneness we felt in that distress. Trauma isn’t just about hardship, it’s about lack of supportive presence. When we revisit the memory, our minds expect to encounter the aloneness feeling again. When they don’t, we’ve created a prediction error.
That might seem small, but it’s incredibly profound. We are essentially creating a prediction error in our own history, allowing us to rewrite what happens after. Because of the way our brains store memories (they store the most recent version of the memory – like a file update), the rewritten memory can become the one that influences our subsequent feelings and behaviors.
The Surprising Power of Curiosity
There was another experiment conducted by political researchers that had startlingly different results than the one I mentioned earlier. In this field experiment, volunteers knocked on doors and asked people to talk about their perspectives on trans rights. They didn’t try in any way to debate the people at their doors; instead, they asked them only to recall a moment when they themselves felt judged or excluded. The volunteers simply listened.
Three months later, the randomized trial found that transphobia had significantly fallen among the people in these conversations. In fact, this decrease was greater than the average decrease in homophobia from 1998 to 2012. And interestingly, whether the canvasser was trans themselves didn’t seem to matter.
This study was somewhat of a landmark. It was politically meaningful, but I think it’s even more broadly meaningful. The way I see it, the respondents experienced a prediction error. They expected to have their beliefs challenged or their identity critiqued, and instead they were met with curiosity. They had the chance to experience – via remembering feeling judged – an error in the notion that this is an “other people” issue, which undercut their bias.
The Final Flex: Living With Uncertainty
This model is useful in understanding how to reshape beliefs, but there’s a final caveat that I want to mention. It’s that ultimately, even via prediction error, the goal isn’t necessarily to update our or others’ old beliefs with new ones that are just as rigid. Any rigid belief – whether that I’m completely terrible or that I’m completely wonderful – will get us into trouble and lead to suffering.
The real flex is letting our prediction errors show us that our maladaptive beliefs no longer hold, but then seeing if we can stay in the realm of possibility a little longer. The place where we don’t have to grasp around for a different belief, but that we can be exist in a space of not always being sure.
This might sound a little weird, but I’m personally convinced that getting more comfortable with uncertainty is actually the key to better mental health. To over an oversimplified example, instead of, “I won’t succeed,” turning into “I will succeed,” we can turn it into, “I may or may not succeed. I will live with either.”
It’s staying loose and free from any rigidly held belief. It’s letting our beliefs breathe and staying connected to ourselves through the process. Maybe we never thought we could be that flexible or that open. We might just prove ourselves otherwise.
