Psychology World — home
Blog

Signal detection theory: hits, misses, false alarms and the test that sorts any case

Signal detection theory separates sensitivity from bias. Learn the four outcomes, liberal vs conservative criteria, and everyday examples.

Cognition10 min read
By Psychology World

Signal detection theory says a yes or no answer reflects two parts at once. One part is how well you separate a signal from background noise. The other part is where you set your answer line. A faint tone, a face in a lineup and a disease test all share the same four outcomes: hit, miss, false alarm and correct rejection.

This article defines the theory in plain words. It shows the four outcomes in a table, explains sensitivity and bias with everyday cases, and separates the theory from the absolute threshold. It ends with a study workflow so the four terms stick.

What the theory says

You face a weak stimulus in noise. The stimulus can be a tone in a quiet room. The noise can be blood flow, breathing, room hum and neural activity. You must answer yes or no. Was something there.

The theory migrated from radar and communications work to psychology in the 1950s as a model of such decisions (EBSCO (2024). Signal detection theory.). The key shift is simple. The observer is not a passive receiver. The observer makes a decision under uncertainty. The decision adds a cognitive part to the sensory part. That is why the same ear can answer differently on two trials with the same sound. Read how sensation and perception differ before you continue, because the theory builds on that distinction.

Noise never leaves. A signal adds to it. An absent trial still gives some activity. A present trial gives signal plus noise. The two distributions overlap. Overlap is why errors are normal. They do not prove guessing. They reflect real subjective signals that vary from trial to trial (Wixted (2026). Signal Detection Theory.).

Fechner proposed noisy measurement in 1860. Engineers formalized detection for radar in 1953 and 1954. Tanner and Swets applied the model to visual detection in 1954. Green and Swets published the psychophysics book in 1966 (Wixted (2026). Signal Detection Theory.). The history matches the detection account (Detection theory.). The modern form keeps their core: separate what you can discriminate from how you decide.

The four outcomes

Each trial has two facts. Was a signal there. What did the observer say. Two facts give four outcomes. The names are fixed. Use them exactly. Teaching pages list them the same way (Hanover College (n.d.). Signal Detection Theory Introduction.) (Keiser University (2023). Signal Detection Theory: What It Is, Why It Matters, and Application.).

The four outcomes of a yes or no detection decision.
SignalObserver says yesObserver says no
PresentHit: the signal is there and you say yesMiss: the signal is there and you say no
AbsentFalse alarm: no signal is there and you say yesCorrect rejection: no signal is there and you say no

A hit and a correct rejection are correct answers. A miss and a false alarm are errors. A miss says no when yes was right. A false alarm says yes when no was right. This table is the whole vocabulary. Keep it in front of you.

Use this test for any example. Ask two questions in order. First, was a signal there. Second, what did the observer say. Then read the cell. A radiologist sees cancer on a scan and reports it. Signal present, answer yes. That is a hit. The same doctor reports cancer on a healthy scan. Signal absent, answer yes. That is a false alarm. An eyewitness picks the true culprit from a lineup. Signal present, answer yes. That is a hit. The witness picks an innocent filler. Signal absent for that person, answer yes. That is a false alarm (Detection theory.).

The test works for memory too. When a face from the study list appears and you call it old, target present with answer old is a hit. When a new face appears and you call it old, distractor only with answer old is a false alarm. The labels change from yes to old, but the logic stays the same (Wixted (2026). Signal Detection Theory.).

Sensitivity and bias

Two observers can hear equally well and score differently. The theory explains why. It separates sensitivity from bias.

Sensitivity is how far apart signal plus noise stands from noise alone. Strong signal, quiet room and full attention raise sensitivity. Weak signal, loud background and fatigue lower it. Sensitivity is a property of the evidence. You cannot move it by will on one trial.

Bias is where you set your yes line. Psychologists also call it criterion. A liberal line needs little evidence. You say yes often. You gain hits and you pay false alarms. A conservative line needs strong evidence. You say yes rarely. You cut false alarms and you pay misses. The decision adds top-down processing to raw input, because expectations and goals move the line.

A wartime sentry shows a liberal line. The sentry reports faint sounds because a miss costs lives. A peacetime sentry with the same ears sets a stricter line. Experience, expectations and physiological state move the threshold (Detection theory.). A doctor shows the same trade. When a miss means a lost patient and a false alarm means one more test, the doctor leans liberal. When a false alarm means a risky action and a miss means watchful waiting, the doctor leans conservative. The ears or eyes did not change. The payoff changed.

Instructions, signal frequency and payoff move the line in the lab. Tell people most trials contain a signal and yes answers rise. Tell people most trials are catch trials with no signal and no answers rise. Pay hits and punish false alarms and the line moves liberal. Punish false alarms and it moves conservative. The sound stays the same while hits and false alarms move together (EBSCO (2024). Signal detection theory.). That is why percent correct misleads. A worked example shows the point. A balanced observer and a liberal observer can share the same discriminability score while accuracy differs, because their lines sit in different places (Wixted (2026). Signal Detection Theory.). Accuracy differed. Ability did not.

Keep the pair straight. Sensitivity answers how clear the evidence is. Bias answers how much evidence you demand before you say yes.

Signal detection and the absolute threshold

Students mix the theory with the absolute threshold. The table keeps them apart. The absolute threshold is the intensity heard on half the trials (Keiser University (2023). Signal Detection Theory: What It Is, Why It Matters, and Application.). The theory rejects a fixed line and measures decision under noise (EBSCO (2024). Signal detection theory.).

How a fixed threshold differs from a decision under noise.
QuestionAbsolute thresholdSignal detection theory
What it asksIs there a fixed weakest levelHow do you decide under noise
What it measuresIntensity heard on half the trialsSensitivity apart from bias
Role of noiseTreated as errorTreated as part of the model
Role of decisionIgnoredCentral with four outcomes
What moves answersStimulus strength onlyStrength plus expectations, motivation, fatigue and payoff

Threshold work in the 1800s assumed a door-like line. Below it, never heard. Above it, always heard. Measurement failed that idea. The same weak sound was heard once and missed next. Researchers kept the 50 percent point as a definition, but they saw contamination. Fear of missing, fear of looking foolish and payoff moved reports. The theory gave a method to measure sensitivity apart from that bias (EBSCO (2024). Signal detection theory.). It does not replace the threshold. It explains why the threshold wobbles.

A faint tone in a silent room shows both ideas. Threshold language asks what volume reaches half yes. Theory language asks, at that volume, how many hits, misses, false alarms and correct rejections occur, and where the observer set the line. Use threshold for weakest level. Use the theory for decisions with noise.

Where you meet it

The faint tone is the lab case. Headphones, warning light and catch trials with no sound measure hits against false alarms (EBSCO (2024). Signal detection theory.). The clinic is the applied case. A test result must separate disease from healthy variation (Wixted (2026). Signal Detection Theory.). A liberal test catches more cases and alarms more healthy people. A conservative test alarms fewer healthy people and misses more cases. Neither accuracy alone tells which test is better. You need to know the costs (Lynn and Barrett (2014). Utilizing Signal Detection Theory.).

The lineup is the legal case. The witness must separate a studied face from new faces. You score a hit when you say old to an old face. You score a false alarm when you say old to a new face. Confidence tracks signal strength in a graded way, which fits a continuous evidence model better than a fixed line (Wixted (2026). Signal Detection Theory.).

The classroom is the study case. A recognition test mixes old words with new distractors. You score a hit when you say yes to a target. You score a false alarm when you say yes to a distractor. Longer study raises sensitivity. Payoff and base rate move bias. Low base rate with rare targets pushes conservative answers. High payoff for hits pushes liberal answers. The utility tutorial makes the same point with threat faces, jury choices and travel speed: payoffs, base rates and similarity set the best line, not accuracy alone (Lynn and Barrett (2014). Utilizing Signal Detection Theory.).

Everyday listening fits too. A car clunk in traffic asks the same two questions. Did the engine make the sound. Did you act as if it did. A new car owner dismisses the faint sound. An old car owner with repair history books the service. Same signal strength, different line (Basics of Signal Detection Theory.).

How to keep the four terms apart when you study

State the test out loud before you drill. Was a signal there. What did the observer say. Then name the cell. Do five fresh examples from memory. Then sort five new ones. Short repeated sessions beat one long reread for terminology.

Look the four terms up in the Psychology World dictionary and save them to one list. Use the list for Flashcard rounds, Question rounds once the list holds four or more words, and Typing to test recall. Our guide on how to memorise psychology terms explains why this type of practice test is effective. Our article on spaced repetition explains how to put time between the sessions.

Psychology World is a psychology dictionary with study games. You save the terms that you confuse to one list, and then you drill that list. The Question game is a multiple-choice test, and it needs a list of at least four words. In the Typing game, you type the answer yourself.

Frequently asked questions

What is signal detection theory in simple terms

It is a model of yes or no decisions with noisy evidence. It separates how clear the signal is from how much proof you demand before you say yes. Clear signal with strict line and weak signal with lax line can give the same accuracy for different reasons.

Can you give a simple example of signal detection

Hear a faint tone with headphones while a light warns you. Some trials have a tone, some have only noise. You score a hit when you say yes with a tone. You score a miss when you say no with a tone. You score a false alarm when you say yes with no tone. You score a correct rejection when you say no with no tone.

Who developed signal detection theory

Fechner proposed noisy measurement in 1860. Peterson, Birdsall and Fox formalized detectability for radar in 1954. Tanner and Swets applied the model to visual detection in 1954. Green and Swets published the psychophysics book in 1966 that carried the ideas into psychology (Wixted (2026). Signal Detection Theory.).

Why does signal detection theory matter for psychology

It stops researchers from reading accuracy as pure ability. Accuracy mixes sensitivity with bias. The theory measures the two apart, so instruction, base rate and payoff effects stop looking like changes in the senses. It also guides real decisions in clinics, lineups and labs.

Sources

  1. 1.Wixted (2026). Signal Detection Theory. Open Encyclopedia of Cognitive Science.
  2. 2.Hanover College (n.d.). Signal Detection Theory Introduction.
  3. 3.Hanover ISLE (n.d.). Basics of Signal Detection Theory.
  4. 4.Keiser University (2023). Signal Detection Theory: What It Is, Why It Matters, and Application.
  5. 5.EBSCO (2024). Signal detection theory.
  6. 6.Lynn and Barrett (2014). Utilizing Signal Detection Theory. Psychological Science.
  7. 7.Detection theory. Wikipedia.

Look the terms up yourself

18,000+ psychology and psychotherapy terms in English and German, on iPhone, iPad, Mac and Apple Vision.

See how Psychology World works
All posts