The Cruelest Version of You Isn't Fake. It's the One With No Witnesses.
You've said things in a comment section you would never say to a stranger's face. Not because you're secretly cruel. Because something changes the instant your name disappears and the other person becomes a username instead of a face — and the version of you that shows up in that gap isn't a costume. It's not fake at all. That's the part nobody wants to hear.
The comfortable explanation is that the internet reveals "who people really are underneath." That's close, but it's not quite right, and the difference matters. The internet doesn't reveal a hidden self. It removes the machinery that normally keeps a real, pre-existing part of you in check — and once you understand which machinery, exactly, gets switched off, you stop being surprised by what climbs out.
The six switches Suler found, and why none of them require malice
In 2004, psychologist John Suler published "The Online Disinhibition Effect" in CyberPsychology & Behavior, naming six specific mechanisms that lower behavioral restraint online, none of which require the person to be a bad actor to begin with. Dissociative anonymity — no name attached, no accountability trail. Invisibility — no face, no tone of voice, no wince to register and correct against. Asynchronicity — you post and leave; the blowback, if it comes, lands in an inbox hours later, disconnected from the moment of writing. Solipsistic introjection — you fill in the other person's voice and reactions from imagination, which means you're arguing, in part, with a character your own mind wrote. Dissociative imagination — the online self gets filed as a separate character, a persona you can log out of, not "really" you. And minimization of authority — no uniform, no title, no visible hierarchy signaling that this person outranks you and deserves deference.
Individually, each of these is a minor permission slip. Stacked together, which is exactly what a comment thread does, they add up to something closer to a temporary suspension of the entire social-consequence system that governs face-to-face behavior. Suler's point, and it's the one that survives twenty years of internet escalation intact, is that none of these mechanisms manufacture a new impulse. They just clear the runway for whatever was already sitting there, waiting for permission.
Benign and toxic disinhibition run through the exact same door
The part of Suler's framework that gets flattened into "the internet makes people mean" is that he explicitly split disinhibition into two directions, and both use the identical mechanism. Benign disinhibition is a stranger opening up to an online support group about something they'd never say to a friend's face, or a shy person finding genuine self-expression somewhere they don't feel watched. Toxic disinhibition is the exact same anonymity, invisibility, and distance producing cruelty, threats, and what Suler called "blind catharsis" — venting without any of the personal growth that makes catharsis meaningful. Same six switches. Opposite outputs. The determining factor isn't the technology. It's what was already loaded in the chamber before the switches got flipped.
This is why platform design matters more than platform users are usually given credit for, and also why blaming "anonymous trolls" as a separate species of person misses the mechanism. A 2025 refinement of Suler's model, the Multidimensional Measure of Online Disinhibition, found the effect correlates most strongly with existing emotion-regulation difficulty — people already struggling to modulate anger or frustration face-to-face are the ones for whom the six switches do the most damage, because there was more pressure behind the door to begin with. The anonymity didn't install the rage. It just found the one room in the house with no lock.
The numbers are not abstract. Recent tracking puts roughly 34–40% of online harassment cases tied specifically to anonymous or impersonated accounts, and cyberbullying-victimization research shows the effect isn't contained to obvious targets — a 2025 survey found 42% of LGBTQ+ youth reporting digital harassment, a rate that tracks disinhibition research showing perpetrators frequently escalate once a victim responds, because the asynchronous, faceless exchange gives no natural braking point the way a real-time, face-to-face reaction would. The same faceless distance that fuels workplace ostracism runs on a related principle — remove the immediate human cost of an action, and the action gets easier to repeat, regardless of the platform it happens on.
The turn: the cruelty was never invisible. You were.
Here's the reframe that actually changes how you read your own worst online moments, and other people's: the internet didn't make you crueler than you are. It made you crueler than you're normally allowed to be, which is a very different claim, and a more useful one. The restraint you show face-to-face isn't fake goodness laid over hidden rage — it's a real regulatory system, built out of eye contact, tone, and immediate consequence, and it's simply absent online. What you say in that gap is real. It's just real without the machinery that usually filters it, the same way a person might say something true and cruel that alcohol removed the brake on, without alcohol having invented the sentiment from nothing.
That cuts both directions, and it's the part that actually matters: if the switches can be flipped on toward toxic disinhibition, the same six mechanisms can be deliberately used toward the benign kind — the honest disclosure you'd never manage in person, the support you give a stranger precisely because there's no face to perform for. The question worth asking before you post isn't "would I say this to their face." It's "what does it tell me that I only can say it because their face isn't there" — because that answer is the one that's actually true about you.
Follow The Shadow Self Media on TikTok · Instagram · Facebook