Confirmation Bias in AP Psychology

Confirmation bias is the filter, belief perseverance is what survives it

The neighbor term that causes the most trouble is belief perseverance, and the cleanest way to separate them is process versus outcome. Confirmation bias describes how you handle incoming information. Belief perseverance describes a belief that stays put even after the evidence says it's wrong.

The course links them directly. In Topic 4.2 (Attitude Formation and Attitude Change), someone showing belief perseverance may fall into confirmation bias, holding on to a belief no matter what the evidence for or against it shows. In Topic 2.2 (Thinking, Problem-Solving, Judgments, and Decision-Making), confirmation bias is one of the ways human judgment goes wrong. So it spans Unit 2: Cognition and Unit 4: Social Psychology and Personality, and it belongs to a bigger argument in both units: thinking is efficient but not neutral.

Here's a scenario. Maya is sure her new coworker is lazy. She notices every time he's late, reads his quiet days as slacking off, and forgets the weekend he covered her shift. That filtering is confirmation bias. Later, her manager shows her his performance numbers, which are the best on the team, and Maya still thinks he's lazy. That stubborn leftover belief is belief perseverance.

A quick check for scenario questions: if the stem focuses on how someone picks, reads, or recalls evidence, choose confirmation bias. If it focuses on someone keeping a belief after it's been shown false, choose belief perseverance.

Seeking, reading, and remembering: three places the bias shows up

A scenario doesn't have to show all three to count. Any one is enough.

• Seeking: choosing sources that agree with you, like only following news accounts that share your political views.
• Interpreting: reading unclear evidence as support, like deciding an ambiguous text from a friend means they're mad at you, because you already thought so.
• Remembering: recalling the hits and forgetting the misses, like a fan of a "lucky" jersey remembering the wins while wearing it but not the losses.

The interpreting piece ties confirmation bias to schemas from Topic 2.2. When new information gets squeezed into an existing schema without changing it, that's assimilation. Confirmation bias is one reason people assimilate instead of accommodating, which would mean revising the schema to fit the new evidence.

It also connects to stereotypes in Topic 4.2. Stereotypes can come from biased perceptions, cause them, or both. Confirmation bias is one way that loop runs: you notice group members who fit the stereotype and overlook the ones who don't, so the stereotype feels confirmed.

Look-alike answer choices and what each one protects

In a scenario-based question, it can help to distinguish confirmation bias from other errors in Topic 2.2. A useful question to ask yourself is what the person in the stem is holding on to.

If it's a belief, and they're filtering evidence to keep it, that's confirmation bias. If it's a problem-solving strategy that worked before, that's mental set. If they can't see a new use for an object, that's functional fixedness. If they're judging how common something is by how easily examples come to mind, like fearing plane crashes after seeing one on the news, that's the availability heuristic.

Two more terms are easy to confuse with it. Hindsight bias is the "I knew it all along" feeling after an outcome is known, so it looks backward at an event instead of filtering ongoing evidence. Cognitive dissonance is the discomfort of an action clashing with an attitude, which pushes a person to change one of them. Dissonance can lead someone to avoid uncomfortable evidence, but the term itself names the unease, not the evidence filtering.

Confirmation bias as the variable in a summarized study

In the Article Analysis question, the bias can be the thing being measured. One possible setup: a researcher measures confirmation bias and job performance in employees and finds r = -0.42. That's a moderate negative correlation, meaning more bias goes with lower performance. A top answer points out that a correlational design can't show that reducing the bias would improve performance, because a third variable (like stress or experience) could drive both.

Other studies operationalize the bias by counting behavior. Participants might choose articles from a pool of confirming and disconfirming sources, with the number of confirming articles as the dependent variable. Or they rate how relevant supporting versus contradicting evidence is on a 1-10 scale. Be ready to find a mean or median by hand, since no calculator is allowed. For confirming-article counts of 3, 5, 7, 8, 9, 10, 11, 12, 14, and 16, both the mean and the median are 9.5.

In the Evidence-Based question, confirmation bias can support a claim about why attitudes resist change. Naming the term won't earn credit by itself. Tie a specific result from a source to the definition, for example, "participants rated supporting evidence as more relevant than contradicting evidence, which shows they interpreted information in favor of their existing belief."