Bayesian updating is the way you revise a prior belief after new evidence comes in. In Intro to Cognitive Science, it models how people and systems combine prior knowledge with incoming data.
Bayesian updating is the process of changing a belief after you get new evidence, using a prior probability and a likelihood to produce a new posterior probability. In Intro to Cognitive Science, it is one of the main ways researchers model how minds learn from experience instead of starting from scratch every time.
The basic idea is simple: you do not just ask, “What does the new evidence say?” You also ask, “What did I already believe before seeing it?” A Bayesian model starts with a prior belief, weighs the strength of the new evidence, and then updates to a posterior belief. That update can be small if the new data is weak, or larger if the evidence is strong and consistent.
This makes Bayesian updating useful for cognition because human thinking is rarely neutral. Your brain is always working with expectations, memory, and past experience. When you hear a word in a sentence, see an object in dim light, or judge whether someone is lying, your mind uses context to predict what is probably true and then adjusts if the evidence pushes back.
Cognitive science uses this framework in perception, learning, and decision-making. For example, if you already expect a blurry shape to be a bicycle, you are more likely to interpret the visual input that way unless something in the scene strongly suggests otherwise. That is Bayesian reasoning at work, with prior knowledge shaping interpretation.
A big reason this term shows up in the course is that it connects psychology and computation. Researchers use Bayesian models to describe both human behavior and machine systems that update beliefs over time. It also helps explain why people can be smart and biased at the same time, since an overly strong prior can make someone ignore helpful new evidence.
Bayesian updating shows how cognitive science treats the mind as an information processor that is constantly revising itself. That makes it a bridge term between psychology, computer science, and neuroscience, because it gives you a formal way to talk about learning from evidence.
It also helps explain common patterns in real behavior. If someone clings too tightly to an initial belief, they may under-update and keep making the same mistake. If they trust every new cue too much, they may overreact and become inconsistent. Bayesian updating gives you a way to describe both the helpful and unhelpful sides of that process.
In Intro to Cognitive Science, you will see this idea when the class talks about perception, inference, and decision-making. It gives you a lens for questions like why two people can look at the same situation and come away with different judgments, or why prior knowledge can make learning faster in some cases and more biased in others.
It also matters because it shows what makes cognitive science different from a plain description of behavior. Instead of just saying people have beliefs, it asks how beliefs change, what evidence counts, and when the update is rational versus distorted.
Keep studying Intro to Cognitive Science Unit 1
Visual cheatsheet
view galleryPrior Probability
The prior probability is the belief you start with before new evidence arrives. Bayesian updating begins there, so the prior sets the starting point for any revision. In cognition, priors often come from experience, memory, or expectations, which means two people can update the same evidence differently if their starting beliefs are different.
Likelihood
Likelihood is the part of Bayes' theorem that measures how well the evidence fits a hypothesis. In Bayesian updating, it tells you how strongly the new data should move your belief. A strong likelihood can outweigh a weak prior, but weak or ambiguous evidence may not change much.
Posterior Probability
The posterior probability is the updated belief after you combine the prior and the evidence. It is the outcome of Bayesian updating, not a separate starting point. In cognitive science, the posterior is what researchers compare to observed behavior when they test whether people are updating in a Bayesian-like way.
Behavioral Experiments
Behavioral experiments are one way cognitive scientists test whether people update beliefs in a rational or biased way. Researchers can change the evidence participants receive and see how their judgments shift over time. Bayesian updating gives a model for predicting those shifts and for spotting when people rely too much on prior expectations.
A quiz question or short answer might give you a scenario where someone starts with a belief, gets new evidence, and changes their judgment. Your job is to trace the update step by step: identify the prior, describe the evidence, and explain the new posterior. In a problem set, you may compare two people with different priors and show why they end up with different conclusions after seeing the same clue.
In a discussion or essay, use the term to explain perception, learning, or bias instead of just saying someone “changed their mind.” The stronger answer connects the belief update to expectation and evidence, showing whether the person updated too little, too much, or in a biased direction.
Confirmation bias is the tendency to favor evidence that supports what you already believe. Bayesian updating is a broader process for revising beliefs using evidence, and it can be rational or distorted depending on the priors and how evidence is weighed. In cognitive science, confirmation bias is often discussed as a failure to update well, not as the update process itself.
Bayesian updating is the process of revising a belief after new evidence arrives.
In cognitive science, it models how the mind combines prior knowledge with incoming information.
The update depends on the prior probability and the strength of the evidence, or likelihood.
A posterior probability is the result of the update, not the starting point.
The term shows up in perception, learning, decision-making, and studies of bias.
It is the process of changing a belief after new evidence comes in, using prior knowledge as a starting point. In Intro to Cognitive Science, it is used to explain how people make inferences, learn from experience, and adjust judgments over time.
You begin with a prior belief, compare it with new evidence, and then arrive at a posterior belief. The update is bigger when the evidence strongly matches or challenges the hypothesis, and smaller when the evidence is weak or unclear.
No. Bayesian updating is the general process of revising beliefs with evidence, while confirmation bias is the tendency to prefer evidence that supports what you already think. Confirmation bias can make updating less accurate because it distorts which evidence gets weight.
It explains why your brain does not treat every new input equally. Prior expectations shape how you interpret sights, sounds, and social clues, and then the new data pushes your belief one way or another. That makes it useful for modeling real-world perception and learning.