A false alarm in psychology is when a person reports detecting something that isn’t actually there. The term comes from signal detection theory, a framework psychologists use to understand how people make decisions under uncertainty. The American Psychological Association defines it as “an incorrect observation by the participant that a signal is present in a trial when in fact it is absent.” While the concept originated in laboratory research, it has practical implications in everything from medical screening to criminal justice.
How Signal Detection Theory Works
Signal detection theory breaks every detection task into four possible outcomes. Imagine you’re a radiologist scanning a mammogram, a security guard watching a monitor, or simply a participant in a psychology experiment listening for a faint tone. In each case, a signal is either present or absent, and you either say “yes, I detect it” or “no, I don’t.” That creates a simple two-by-two grid:
- Hit: The signal is there, and you correctly say yes.
- Miss: The signal is there, but you say no.
- False alarm: The signal is not there, but you say yes.
- Correct rejection: The signal is not there, and you correctly say no.
Hits and correct rejections are the two ways to be right. Misses and false alarms are the two ways to be wrong. False alarms and misses carry costs, while correct detections and correct rejections provide benefits. The balance between these outcomes reveals a lot about how a person perceives the world and makes decisions.
Why False Alarms Happen
Your brain is never working in silence. Even when no external signal is present, your sensory systems generate background neural activity, often called “noise.” This noise can occasionally feel like a real signal, especially when you’re primed to detect one. Think of lying in bed at night, convinced you heard a sound downstairs. Your auditory system produced a random spike of activity, and your brain interpreted it as something meaningful.
The boundary between “I detect something” and “I don’t” is set by an internal decision criterion. When that criterion is low (a liberal or lenient threshold), you’re more willing to say “yes,” which catches more real signals but also lets more noise through as false alarms. When the criterion is high (a conservative or strict threshold), you say “yes” less often, reducing false alarms but increasing the chance of missing a real signal. Research on sensory detection shows that when target evidence is weak, people naturally adopt a lower threshold to avoid misses, and that lower criterion means short sequences of random noise can accidentally trigger a detection response.
Sensitivity vs. Response Bias
Psychologists separate two distinct factors that shape someone’s false alarm rate. The first is sensitivity: how well a person can actually tell the difference between signal and noise. Someone with high sensitivity has a large gap between their “signal present” and “signal absent” mental distributions, making it easier to tell the two apart. Sensitivity is measured with a statistic called d-prime, calculated from the relationship between a person’s hit rate and false alarm rate. Larger values mean sharper discrimination.
The second factor is response bias: a person’s general tendency to say “yes” or “no” regardless of what they actually perceive. Two people with identical sensitivity can have very different false alarm rates simply because one is more willing to say “yes.” In one classic demonstration, a participant named Anita was more willing than a participant named Bob to say she recognized a stimulus, which gave her both a higher hit rate and a higher false alarm rate. Her ability to tell signal from noise wasn’t worse; she just had a more liberal threshold for responding.
False Alarms and Type I Errors
If you’ve encountered statistics, you may recognize the false alarm as a close relative of the Type I error. A Type I error occurs when a researcher rejects a true null hypothesis, concluding that an effect exists when it doesn’t. The logic is identical: something is declared present (a signal, a treatment effect, a difference between groups) when nothing is actually there. The probability of committing a Type I error is represented by alpha, which researchers typically set at 0.05, or 5%. Signal detection theory and statistical hypothesis testing are parallel frameworks for the same fundamental problem of distinguishing real patterns from noise.
False Alarms in Anxiety and Threat Detection
The false alarm concept helps explain why anxiety disorders can feel so overwhelming. People with anxiety tend to have lower thresholds for detecting threat, meaning they respond to harmless cues as if danger is present. Lab-based fear-conditioning studies have repeatedly linked exaggerated reactivity to safe stimuli with clinical anxiety.
A longitudinal study tracking college students through their first year found that heightened false alarms during a threat-conditioning task predicted later increases in generalized anxiety disorder and social anxiety disorder symptoms. Interestingly, the two disorders showed distinct false alarm patterns. Risk for generalized anxiety was predicted by increased anxiety toward safe stimuli that resembled a learned danger cue, a kind of overgeneralization. Social anxiety, by contrast, was predicted by elevated anxiety toward a broad range of safe stimuli, including ones that bore little resemblance to the original threat. In both cases, the core problem was the same: the brain’s threat detection system was generating too many false alarms, flagging safety as danger.
Medical Screening: A High-Stakes Example
Mammography is one of the clearest real-world illustrations of how false alarms play out. In the United States, about 10% of mammograms lead to a woman being called back for further testing. Of those callbacks, only about 7% ultimately lead to a cancer diagnosis. An analysis of 3.5 million mammograms found that roughly 345,000 were false positives, compared to about 3.2 million true negatives. Over a decade of annual screening, more than half of women will experience at least one false-positive result, and many will undergo a biopsy as part of follow-up testing.
These false alarms carry real psychological and behavioral costs. Research from the National Cancer Institute found that some women avoid future breast cancer screening after receiving a false-positive mammogram result. The screening system deliberately uses a liberal detection threshold (catching as many true cancers as possible) because the cost of a miss, an undetected cancer, is considered worse than the cost of a false alarm. But that tradeoff means a large number of healthy women experience unnecessary anxiety, additional procedures, and sometimes lasting changes in their screening behavior.
Eyewitness Identification and Wrongful Conviction
False alarms also have serious consequences in forensic psychology. When a witness picks an innocent person out of a police lineup, that is a false alarm: the witness reports detecting the perpetrator when the perpetrator isn’t the person they selected. Nationwide, mistaken eyewitness identifications have played a role in 75% of convictions later overturned by DNA evidence.
Many factors that increase lineup false alarms are outside the justice system’s control, including lighting at the crime scene, how long the witness saw the perpetrator, and cross-race identification difficulty. But lineup procedures themselves also matter. A large study by the American Judicature Society compared two formats: simultaneous lineups (all photos shown at once) and sequential lineups (photos shown one at a time). Simultaneous lineups produced an 18.1% rate of filler identifications (picking someone known to be innocent), while sequential lineups reduced that to 12.2%. That 5.9 percentage point difference was statistically significant and consistent with decades of laboratory research showing that sequential lineups reduce mistaken identifications without substantially reducing accurate ones.
Even with improved procedures, though, witnesses still identified an innocent filler 12.2% of the time. The false alarm rate in eyewitness identification can never be eliminated entirely, because the same noise and bias dynamics from signal detection theory apply whenever a person tries to match a memory to a face.
The Core Tradeoff
The most important thing to understand about false alarms is that they exist on a continuum with misses. Reducing one almost always increases the other. A smoke detector set to maximum sensitivity will catch every real fire but will also go off when you burn toast. Turning down the sensitivity eliminates the toast alerts but risks missing a smoldering wire in the wall. Every detection system, whether it’s a medical test, a human brain scanning for threats, or a witness trying to identify a suspect, faces this same tradeoff. The “right” balance depends entirely on the relative costs of each type of error in that specific situation.