By Jason Hreha

Article

Could Six Beliefs Explain Many Cognitive Biases?

By

Published 5 min read

You walk out of a presentation convinced that everyone noticed the moment your voice cracked. You assume they could tell how nervous you felt. You also suspect that most reasonable people agreed with your recommendation.

Psychologists have different names for these judgments. The spotlight effect. The illusion of transparency. The false consensus effect.

Three labels. But perhaps not three separate mental machines.

In a 2023 paper, psychologists Aileen Oeberst and Roland Imhoff propose a more economical explanation. Seventeen familiar biases, they argue, may grow from six prior beliefs combined with one broad tendency: we process new information in ways that fit what we already believe.

It is an appealing theory. It may also help clean up the crowded world of cognitive biases. But we need to be precise about what the paper says. Oeberst and Imhoff are proposing a framework for researchers to test. They have not discovered six master biases that explain every error the mind makes.

The basic recipe: a belief plus friendly evidence

The paper calls the shared process belief-consistent information processing. That is a mouthful, but the idea is familiar. We do not meet every piece of evidence as a blank slate. We already believe things about ourselves, our groups, other people, and the way the world works. Those beliefs can affect which evidence we find, notice, remember, and accept.

A belief does not have to be obviously irrational. It does not even have to be false. “My experience is a reasonable reference” is often a useful assumption. Your own experience is the part of the world you know most directly. Trouble begins when that reasonable starting point quietly becomes the standard for judging what everyone else saw, knew, or believed.

Belief-consistent processing can happen in several ways:

  • You search for evidence that supports your view.
  • You notice friendly evidence faster.
  • You interpret an ambiguous fact in your favor.
  • You remember supporting examples and lose track of the awkward ones.
  • You explain away a contradiction as a strange exception.
  • You behave in a way that draws out the response you expected.

None of this requires a tiny person inside your head deliberately hiding the bad evidence. The processing can be quick and unreflective. That is part of what makes the proposal interesting: several effects that look different on the surface may share the same machinery underneath.

One belief, five different biases

Return to the presentation.

You felt your heart pounding, heard the shake in your voice, and knew exactly which slide made you lose your place. Your experience was full of these details. The audience's experience was not. They were looking at the chart, thinking about lunch, or wondering when the meeting would end.

If you use your own experience as the reference for theirs, several familiar effects can follow.

The task changes. The proposed starting point does not: my experience is a reasonable reference. Oeberst and Imhoff's claim is not that these five effects are identical. It is that they may be different results of the same belief meeting different questions.

The six proposed beliefs

The paper connects 17 biases with six beliefs. Some of the names overlap with one another, and some have their own competing explanations. This is the map the authors propose:

Proposed beliefBiases the paper connects to it
1. My experience is a reasonable reference.Spotlight effect; illusion of transparency; illusory transparency of intention; false consensus; social projection
2. I make correct assessments of the world.Bias blind spot; hostile media bias
3. I am good.Better-than-average effect; self-serving bias
4. My group is a reasonable reference.Ethnocentric bias; in-group projection
5. My group and its members are good.In-group bias or partisan bias; ultimate attribution error; linguistic intergroup bias; intergroup sensitivity effect
6. People's attributes, rather than context, shape outcomes.Fundamental attribution error or correspondence bias; outcome bias

This is not a list of the “six core cognitive biases.” The beliefs are proposed starting assumptions. The biases are the patterns that may appear when those assumptions shape how someone handles information.

Why believing you are right creates its own problems

The second proposed belief sounds almost comically obvious: I make correct assessments of the world. Of course you think your assessments are correct. If you thought one was wrong, you would change it.

That creates a problem when you judge the judgment itself.

Suppose two people read the same political news story. One calls it fair. The other calls it hostile to her side. When you believe your own view is accurate, disagreement can become evidence that the article, the other reader, or both are biased. Your own interpretation escapes inspection because it is serving as the measuring stick.

That logic can produce the bias blind spot: other people's biases are easier to see than your own. It can also contribute to the hostile media effect, in which opposing partisans judge the same coverage as unfair to their side.

Again, one proposed belief. Different tasks. Different labels.

Why this theory is useful

Psychology has a label problem. Researchers often find an interesting pattern, give it a name, and build a small literature around it. Another group finds a related pattern and gives that one a different name. Soon the field has a cabinet full of labeled drawers, but little agreement about how many mechanisms are hiding inside.

Names are not useless. A good label helps us recognize a recurring pattern. But a label is the beginning of an explanation, not the end of one. Saying “spotlight effect” tells you what happened. It does not necessarily tell you why.

A shared-process theory forces stronger questions. If several biases depend on the same prior belief, people who hold that belief more strongly should show more of those patterns. A change that weakens belief-consistent processing should affect several related biases, not just one. And if two biases that look related respond differently, perhaps the shared explanation is wrong or incomplete.

That is how a useful framework earns its keep. It does not merely reorganize a list. It creates predictions that could fail.

What the 2023 paper does not prove

Here is the part that will get lost if this idea turns into an infographic.

The 2023 paper is a theoretical synthesis. It brings existing findings together and proposes a common framework. The authors work through the first two beliefs in the most detail, then extend the logic to the other four. That is serious conceptual work. It is not the same as a program of experiments showing that all 17 effects reduce to six beliefs and one process.

Other mechanisms may contribute to the same bias. A belief may be weak, strong, deliberate, or barely noticed. The situation may amplify it or shut it down. Two people can arrive at the same judgment by different routes. A neat map does not make that mess disappear.

A 2025 follow-up by Oeberst, Dorothee Mischkowski, and Imhoff connects the framework with coherence-based reasoning: the tendency to build an internally consistent story from beliefs and evidence. That may help explain how a prior belief reshapes the meaning of new information. But the follow-up is also theoretical. It sharpens the questions; it does not settle them.

So, could six beliefs explain many cognitive biases? Yes, possibly. Do we know that they do? No.

A better way to use the idea

The tempting response is to memorize six beliefs instead of 200 bias labels. That would be an improvement in storage, perhaps, but not necessarily in judgment.

A more useful response is to work backward from a specific decision. What were you trying to estimate, explain, remember, or choose? What did you believe before the evidence arrived? Then look at what happened to the evidence on the way through.

  • What did you search for?
  • What caught your attention?
  • Which examples came to mind first?
  • How did you explain the evidence that did not fit?
  • What result would have made you change your mind?

These are questions suggested by the framework, not a proven debiasing treatment. Their value is that they move you past naming a bias and toward examining the judgment that produced it.

Before adding another label, ask a harder question: what did the person already believe, and what happened to the evidence?

Sources