The Perception Trap: Why the Extremist You Fear Probably Doesn't Exist
- NVC Rising Platform Desk
- May 8
- 6 min read
You're not imagining the divide. But you might be imagining who's on the other side of it.
Here's a number that stopped me cold: when researchers asked Democrats to estimate what percentage of Republicans believe that immigration should be reduced to zero, they guessed around 50%. The actual number is closer to 16%. Republicans made the same kind of error in reverse, consistently imagining far-left positions as majority Democratic views when they're held by a fraction of that party.
This is the finding at the center of the More in Common "Perception Gap" research: on average, Americans imagine nearly twice as many extremists on the other side as actually exist. What we think we're dealing with, and who we're actually dealing with, are two very different populations.
That gap has a name in Nonviolent Communication. Marshall Rosenberg called it an enemy image.
The Enemy You Built in Your Mind
An enemy image isn't a mistake you make. It's something that forms gradually, from a thousand small inputs — a news segment, a social media post, a dinner table argument that didn't go well. Over time, the image hardens. The other side stops being made up of individuals with specific views and starts becoming a type. A threat. A category.
The enemy image is self-reinforcing. Once you have one, you interpret everything through it. Ambiguous behavior becomes evidence. Neutral statements become coded attacks. A complex person becomes proof of what you already believed.
NVC doesn't dismiss the pain that creates enemy images. It doesn't say your concerns aren't real. What it does say is this: an enemy image is always partly a construction. And you can test it.
Why Your News Diet Is Making This Worse
Before you test it, you should know what's reinforcing it.
The More in Common data showed something that doesn't follow common sense: people who consume news "most of the time" are nearly three times more distorted in their perceptions than people who consume news only occasionally. And for Democrats specifically, having a postgraduate degree made the perception gap worse, not better. The most informed, most educated consumers ended up with the least accurate picture of who the other side actually is.
This is what a tribal information environment does. It doesn't just give you wrong information — it gives you the cognitive tools to build more elaborate and convincing wrong pictures of the people who disagree with you.
At the same time, Pew Research data from 2022 found that 62% of Republicans and 54% of Democrats held "very unfavorable" views of the opposing party, up from just 20-26% in 2002. And research published in PNAS Nexus in 2024 found that people's personal hostility toward the other party is regularly lower than they believe their own ingroup expects of them. People are performing contempt they don't fully feel, because they think that's what belonging to their side requires.
The divide is real. But a significant part of what's sustaining it is a kind of collective hallucination about who the other side is and what they believe.
The Observation/Judgment Problem
This is where NVC's framework becomes something more than feel-good advice.
Most of us aren't having contact with actual people across the political divide. We're having contact with the images we've built of them. And those images are built almost entirely from evaluations, not observations.
In NVC, the distinction is precise. An observation is what you could capture on a video recording: specific behavior, specific words, in a specific context. A judgment is the interpretation layered over that — the story you tell about what it means, who they are, what they want.
"He said the border should be controlled" is an observation. "He's a racist who doesn't care about human beings" is a judgment.
"She said she thinks we should defund the police" is an observation. "She's a radical who wants society to collapse" is a judgment.
We almost never fight about observations. We fight about the judgments we've stacked on top of them. And we almost never check whether the judgments are accurate.
The NVC practice here isn't soft. It's a discipline of reality-testing. It asks: what did this person actually say or do, stripped of my interpretation? And when I look at just that — just the observable behavior — what do I actually know?
Often, much less than we thought.
If this distinction resonates — and you want to practice it with others navigating the same questions — the NVC Learning Community is where that work happens. Live sessions, real conversations, a community of people learning to see more clearly.
What Happens When You Test the Image
In 2016, researchers David Broockman and Joshua Kalla published a study in Science that showed something remarkable. A single ten-minute conversation using active perspective-taking reduced prejudice measurably, and the effect held for at least three months. The effect was comparable in size to the fourteen-year national decline in anti-gay prejudice — from a single conversation.
This is what contact with an actual person does to an enemy image. Not argument. Not facts. Contact with someone's real fears, their actual story, what they need and why they're scared. The image can't survive intact when the person beneath it shows up.
NVC's contribution to this isn't only empathy as a warm feeling. It's a structural practice for making contact where contact had been blocked. You start with observation, not judgment. You ask about needs, not just positions. You look for what's actually driving the behavior, not what the behavior seems to confirm about your existing beliefs.
A Limit Worth Naming
NVC isn't an instruction to absorb hostility with equanimity. It is not asking you to seek understanding from people who are actively threatening you, or to treat every political position as equally deserving of empathic inquiry.
The practice is most useful for people who want to re-enter a conversation that has hardened into mutual enemy-imaging. For people who have drawn a boundary for safety reasons, NVC doesn't override that boundary. Safety comes first.
What it does offer is this: for the many people who are exhausted by the contempt, who know the divide is real but feel the war is somehow larger than the actual disagreement underneath it, NVC provides a practical place to start. Not by summoning more warmth. By practicing more precision.
One Move You Can Try This Week
Before the next political conversation you have — or before you post a reaction to something you read — try this.
Write down what you're reacting to. Then separate it into two columns: what actually happened (the specific words, the specific action), and what I'm making it mean (the interpretation, the character judgment, the story about who they are).
Look at the first column only. What do you actually know, from what was actually said or done?
Often you'll find the enemy image you were fighting is built more on the second column than the first. That doesn't mean your concerns aren't valid. It means you now have something real to work with, instead of the ghost of a threat that may have grown much larger than the person who created it.
The divide in this country is real. The anger on both sides is real. The unmet needs — for safety, dignity, belonging, and being heard — are real on every side of it.
But the extremist you've been arguing with in your head may not be quite who you think they are.
That's worth checking.
FAQ
What is the perception gap in politics?
The perception gap is the measurable difference between how extreme you believe the other side's views are and how extreme those views actually are. More in Common research found that Americans overestimate the prevalence of extreme positions by roughly 100% — imagining nearly twice as many extremists on the other side as actually exist.
What is an enemy image in NVC?
In Nonviolent Communication, an enemy image is a fixed mental representation of another person or group that filters everything they do through a negative frame. It builds gradually from accumulated inputs and becomes self-confirming. NVC treats it not as a moral failing but as a partly constructed portrait that can be examined and tested.
Does talking to the other side actually reduce polarization?
Research by Broockman and Kalla found that a single ten-minute conversation using active perspective-taking reduced measurable prejudice, with effects holding for at least three months. The mechanism isn't warmth or agreement — it's contact with a real person's actual fears and needs, which the enemy image can't survive intact.
Why does more information make polarization worse?
In a tribal information environment, more information produces more elaborate wrong pictures of the other side, not more accurate ones. More in Common data showed that heavy news consumers had nearly three times the perception distortion of occasional consumers, and education amplified this when the underlying information diet was polarized.
How do I apply the observation/judgment distinction to political conversations?
When you notice a strong reaction to something political, try separating what actually happened — the specific words, the specific action, what a camera would capture — from what you're making it mean: the story, the character judgment, the conclusions you're drawing. Look at the first column alone, and ask what you actually know.
Comments