Philosophy › Section B · Critical thinking, reading, and writing › Topic B.1

The brain as an inference machine

Before you can reason well, it helps to know how your mind reasons when you are not watching. The brain draws conclusions all the time, mostly without telling you, and its usual successes explain its typical mistakes.

Why begin with the brain?

Philosophy asks you to reason carefully, so it pays to know how reasoning goes wrong. A striking lesson from psychology and cognitive science is that most errors in thinking are not signs of low intelligence. They are side effects of a brain that works very well most of the time. An inference is the mental step from some information to a conclusion. We picture it as slow and deliberate, like a detective weighing clues, but the brain also infers constantly and without awareness: recognising a friend across a street, or hearing speech as words rather than noise.

A caution about brain talk

Talk of what "the brain" does needs care, because brain and mind are not obviously the same thing. How conscious experience relates to neural activity is the mind-body problem, which the topics The mind as self: physicalism and Descartes, mind-body dualism, and behaviorism take up. A second caution matters more here. Psychology describes how people actually think; critical thinking asks how we ought to think. Facts about our habits can guide better strategies, but do not settle what good reasoning is. A bias being common does not make it acceptable.

Representations

Philosophers and cognitive scientists call the basic units of thought representations: mental states that carry information about something. Seeing a cup, remembering last summer and imagining a dragon all involve representing, and what is represented can be absent, past or fictional. Some representations are unconscious, carrying information that guides behaviour without your awareness. Because representations can be accurate or inaccurate, the key question is how the brain builds them.

Perception is inference, not photography

A tempting picture is that the eye is a camera and the brain stores the photograph. It fails, because perception routinely adds what is not there. Each eye has a blind spot where the optic nerve leaves the retina, yet you see no hole, because the brain fills it in. In the nineteenth century Hermann von Helmholtz described perception as unconscious inference: the brain treats sensory signals as clues and settles on their most likely cause. A family of current theories, often called predictive processing, holds that the brain constantly predicts its input and corrects itself by the errors. These theories are influential but still debated.

Worked example: the hollow mask

Look at the inside of a hollow face mask from a few metres away. Most people see an ordinary face bulging towards them. Why? 1. The light reaching the eye is ambiguous: the same shading fits a hollow face lit from one side and a convex face lit from the other. 2. Experience overwhelmingly favours convex faces, since nearly every face we see sticks out. 3. The brain settles on the more probable interpretation and overrides depth cues pointing the other way. 4. Lesson: what you experience is the brain's best guess, and a strong expectation can beat the evidence in front of you.

Emotion is part of reason

Much of the history of philosophy treated emotion as the enemy of reason. The neuroscientist Antonio Damasio challenged this in Descartes' Error (1994). He described patients with damage to parts of the frontal lobes who could still discuss choices in the abstract, yet decided disastrously in their own lives. His somatic marker hypothesis proposes that bodily feelings tag options as good or bad and steer us among them, so that someone without feelings is less rational, not more. The hypothesis is influential but contested, and some of its supporting experiments have been reinterpreted. The safer conclusion: emotion and reasoning are intertwined.

Homeostasis and allostasis

Emotions are tied to the body's self-regulation. Homeostasis keeps internal states, such as temperature and blood sugar, near a stable level by feedback: a drop is detected and corrected. Some of this reaches awareness as feeling, as when low blood sugar is felt as hunger. Allostasis, a term introduced by Peter Sterling and Joseph Eyer in 1988, is regulation by anticipation: the body prepares for needs before they arise, as when your heart rate climbs before a race begins. The psychologist Lisa Feldman Barrett argues that the brain constructs emotions from such predictions, drawing on past experience and learned concepts; hers is one of several competing theories.

Fast thinking, slow thinking and attention

As the psychology topic Problem solving explains, Daniel Kahneman popularised a contrast between fast, automatic System 1 and slow, effortful System 2, which he called a useful simplification rather than two brain parts. Effortful thought needs attention, and attention is narrow. In a 1999 study by Daniel Simons and Christopher Chabris, many people counting basketball passes in a video failed to notice a person in a gorilla suit walking through the scene. Hard thinking feels costly, but be wary of the story that it drains a fuel tank: the theory that self-control uses up a limited resource, called ego depletion, has failed several large multi-laboratory replications.

When intuition can be trusted

Heuristics, rules of thumb for quick inference, are usually good enough, and practice can make demanding tasks fluent: Mihaly Csikszentmihalyi called the absorbed state of a skilled person meeting a matched challenge flow. In 2009 Kahneman and the psychologist Gary Klein, long on opposite sides of the debate, agreed that expert intuition deserves trust when the environment is regular enough to learn and the expert has had long practice with clear feedback, as a chess master has. A newcomer to philosophy has had no such training, so first reactions to a puzzle are a starting point, not a verdict.

Worked example: answering an easier question

A bat and a ball cost $1.10 together, and the bat costs $1.00 more than the ball. What does the ball cost? 1. The answer that springs to mind is 10 cents. 2. Check it: then the bat costs $1.10 and the total is $1.20, not $1.10. 3. Solve it: if the ball costs \(b\) dollars, the bat costs \(b + 1\), so \(2b + 1 = 1.10\) and \(b = 0.05\). 4. Verify: a 5-cent ball and a $1.05 bat total $1.10 and differ by $1.00. 5. Diagnosis: the mind swapped the real question for an easier one, splitting $1.10 into $1 and the rest. Kahneman and Shane Frederick called this attribute substitution. A cognitive bias is a systematic error of this kind, produced by a normally useful shortcut.

Common mistakes to avoid

First, treating errors in thinking as signs of stupidity: they come from shortcuts that usually work, and they affect able people too. Second, believing perception simply records the world: it infers, fills gaps and can be overruled by expectation. Third, treating "the brain does it" as if it solved the mind-body problem. Fourth, sliding from how we do think to how we should think. Fifth, assuming emotion is always the enemy of reason. Sixth, repeating famous findings without asking whether they survived replication.

Key takeaways

Most of the brain's inferences are fast and unconscious, built from representations that can be accurate or not. Perception is unconscious inference (Helmholtz): the brain fills gaps and can let expectation beat evidence, as the hollow mask shows. Emotion and reason are intertwined (Damasio); the body regulates by feedback (homeostasis) and anticipation (allostasis). Attention is narrow, and intuition deserves trust only after long practice with clear feedback. Biases are systematic errors from useful shortcuts, such as substituting an easier question. Psychology describes how we think; critical thinking asks how we should.

Flashcards

Flashcard

Inference

The mental step from information to a conclusion

It can be deliberate or fast and unconscious.

Test your understanding

Question 1 of 5

Which statement best captures Helmholtz's idea of unconscious inference?

Question 2 of 5

A bat and a ball cost $1.10 together, and the bat costs $1.00 more than the ball. What does the ball cost?

Question 3 of 5

According to Kahneman and Klein, when does an expert's intuition deserve trust?

Question 4 of 5

A student argues: "Most people favour evidence that suits them, so that is the right way to weigh evidence." What is the mistake?

Question 5 of 5

Your heart rate climbs while you wait at the starting line, before the race begins. Which process does this best illustrate?

← Previous topicNext topic →