Cognitive bias: when our brain plays tricks on us

A cognitive bias, such as confirmation bias, is a systematic distortion of judgment: the brain produces a predictable, reproducible error that moves us away from rational thinking. It is not a simple one-off mistake due to fatigue or a lack of information. It is a structural mechanism, rooted in the very way our cognition works.

The distinction from an ordinary error is fundamental. A one-off error can be corrected as soon as better data is available. A cognitive bias, by contrast, persists even when the data is available, because it operates upstream of conscious reasoning.

These distortions are closely linked to heuristics: quick, intuitive mental rules that the brain uses to process information without mobilizing its full analytical power. Heuristics are useful, even indispensable. But they produce cognitive biases when they are applied outside the context where they are effective.

According to the psychologist Daniel Kahneman[2], our brain works in two distinct modes. System 1 is fast, automatic and intuitive: it handles most of our everyday decisions by relying on beliefs, reflexes and associations. System 2 is slow, analytical and deliberate: it allows rigorous reasoning, but consumes a great deal of cognitive energy and is rarely called upon. Cognitive biases thrive in system 1.

For mental health professionals, understanding these mechanisms is particularly valuable. Identifying a patient's cognitive biases means better understanding their automatic thoughts and their judgments about themselves or about the world. It is also one of the areas of cognitive behavioral therapy: an anxious person may overestimate a danger, then test this prediction against the facts during an exposure. C2Care allows the practitioner to conduct this type of exposure in virtual reality, within a controlled setting.

The answer is precise. Cognitive biases were formalized in the 1970s by two Israeli-American psychologists: Amos Tversky and Daniel Kahneman. Their work showed experimentally that human beings are not rational agents: their judgments systematically deviate from the predictions of logic and probability.

This discovery shook up economics as much as psychology. It gave rise to behavioral economics, which incorporates human cognitive limits into the modeling of economic decisions.

In 2002, Daniel Kahneman received the Nobel Prize in Economics for all of this work (Amos Tversky had died in 1996 and the prize is not awarded posthumously). Since the 1970s, the field has expanded considerably: neuroscientists, philosophers and researchers in cognitive psychology have identified hundreds of additional biases, constantly refining the taxonomy of human judgment.

The short answer: there is no official list, and counts range from a few dozen to several hundred. But this figure varies depending on the classifications used.

Some researchers group several closely related phenomena under a single bias, while others distinguish them finely. This is why the numbers vary from one list to another. The cognitive bias codex, a visual reference tool widely circulated in academic and professional circles, offers an overview structured by functional categories.

What matters more than the exact number is the breakdown into families: six main categories make it possible to understand how each bias affects a specific cognitive process, from perception to personality.

There is no single classification. A common teaching presentation distinguishes six families, according to the cognitive function involved.

Sensorimotor and perceptual biases

These biases affect the way the brain interprets sensory information. Visual illusions are the most immediate example: two segments of identical length look different depending on their graphic context. The distortion does not come from the eye, but from the brain's processing. Note: a perceptual illusion is not strictly synonymous with a cognitive bias. The former is tied to the sensory system, the latter to information processing and judgment. But the two often interact.

Attentional biases

Our attention is selective by nature. The brain cannot process every available stimulus at once: it chooses, and that choice introduces distortions. The cocktail party effect illustrates this selectivity: in a noisy room, you pick up your own name spoken at the other end of the hall. Inattentional blindness is its flip side: focused on a task, you fail to see what is happening right in front of you. These attentional biases directly influence what you perceive as important or trivial.

Memory biases

Memory does not work like a faithful recording. It reconstructs, selects, distorts. The primacy effect refers to the tendency to remember the first pieces of information received better; the recency effect, the last ones. The mere exposure effect leads us to prefer what we have already encountered, even without any objective reason. Base-rate neglect skews our probability estimates by making us over-weight salient memories.

This type of bias can also matter in clinical settings, when a patient reports their memories of a situation.

Judgment biases

These biases alter the way we evaluate a situation or form an opinion. Among the best documented: the ambiguity effect (a preference for options whose outcome is known), the attribution bias (the tendency to explain other people's behavior by their personality rather than by the context), the self-serving bias (taking credit for successes and blaming failures on circumstances), the status quo bias (a preference for the current state of affairs), the egocentric bias and hindsight bias (the feeling of having always known what was going to happen). These distortions of judgment have a direct impact on decision-making, both professional and personal.

Reasoning biases

Here, it is logic itself that goes off the rails. Confirmation bias leads us to seek out and retain only the information that supports a pre-existing belief. Erroneous probability assessment leads to overestimating rare but striking events. The representativeness bias leads us to judge the probability of an event by its resemblance to a mental prototype. Finally, mental availability makes us estimate the frequency of a phenomenon by how easily examples come to mind. These reasoning biases are at the heart of collective errors of judgment.

Personality-related biases

Some biases are amplified or modulated by an individual's personality traits and deep-seated beliefs. A strong need for internal consistency will reinforce confirmation bias. A tendency toward mistrust will color interpretation biases in interpersonal relationships. These personality-related biases affect self-perception and the way one interprets other people's intentions, which makes them particularly relevant in a therapeutic context.

The anchoring bias is one of the most studied biases in cognitive psychology and behavioral economics. Its mechanism is simple: the first piece of information received on a subject serves as a mental anchor, and all subsequent estimates gravitate around it, even when that information is arbitrary or irrelevant.

Tversky and Kahneman demonstrated it experimentally as early as the 1970s[1]. In their studies, participants had to estimate a number after seeing a randomly drawn value. Their final estimate remained systematically influenced by that initial number, even when they knew it was random.

In everyday life, this bias is omnipresent:

• In commercial negotiation: the first price announced steers the whole discussion, whether it is too high or too low.
• In price estimation: an item shown at 500 euros and then marked down to 300 seems like a good deal, regardless of its real value.
• In medical decision-making: a first diagnosis can influence the interpretation of subsequent tests, even among experienced professionals.

This is precisely where the strength of the anchoring bias lies: it affects even experts in their own field. Knowledge does not automatically protect against this mechanism. Slowing down one's thinking and checking one's estimates against independent references remain the most effective strategies for limiting its impact.

A heuristic is not a bias. It is a fast, intuitive rule of reasoning, a kind of mental shortcut that the brain uses to make decisions without mobilizing its full analytical power. These shortcuts have real value: they allow us to act quickly in complex environments, to handle a flow of information that system 2 could not absorb in real time.

Let us take a concrete example. An emergency physician who instantly recognizes the signs of a stroke relies on a heuristic refined by experience. This rapid recognition saves lives. It is not a bias, it is encoded expertise.

The problem arises when a heuristic is applied outside the context in which it is effective. According to Kahneman's model[2], system 1 generates fast responses suited to the majority of everyday situations. But in new, complex or high-stakes contexts, these same shortcuts produce systematic errors, that is to say cognitive biases.

Heuristics and biases are therefore two sides of the same mechanism: useful in one context, problematic in another. Understanding this distinction helps you better identify when to slow down and engage system 2, rather than letting intuition decide alone.

Examples of cognitive biases in everyday life

You probably come across them every day:

• Confirmation bias: you mostly read the articles that support your opinion.
• Availability bias: after a news report on a crash, flying seems more dangerous to you, because the example comes easily to mind.
• Hindsight bias: once the outcome is known, you feel you always knew how it would end.
• Self-serving bias: a success comes from your qualities, a failure comes from the circumstances.
• All-or-nothing thinking: a mistake in a file becomes "I'm useless."

The Linda problem

This is one of Tversky and Kahneman's best-known experiments[3]. Linda is described as a bright young woman, committed to fighting discrimination, and people are then asked which is more probable: that she is a bank teller, or that she is a bank teller and a feminist activist. Many choose the second answer, even though a combination of two conditions cannot be more probable than one of them alone. The description "resembles" an activist, and that resemblance wins out over logic.

What to remember about cognitive biases

A cognitive bias is a systematic and predictable error of judgment, often linked to mental shortcuts that are useful in other respects. There is no official list, and no one escapes them, not even experts. You cannot eliminate them, but you can slow down on important decisions, look for what contradicts your first idea, and ask for an outside perspective.