Skip to content
MyFreud

Compulsive Lying: When It Stops Being a Choice

Most people tell almost no lies, and a small minority tell most of them. What separates ordinary dishonesty from compulsive lying, and what imaging found.

6 min read

Pop-art illustration of a man in close-up, brow furrowed, looking down and to one side.

Key takeaways

  • Lying is not evenly distributed. In studies of self-reported lying, most people report telling none on a given day, and roughly half of all reported lies come from around five percent of people.
  • That distribution is the reason compulsive lying is a meaningful category rather than a moral judgement. It describes a small group whose behaviour is statistically unlike everybody else's, not a difference of degree from the average person.
  • Pathological lying has been proposed as a distinct condition, with survey work putting self-identification at roughly eight to thirteen percent in one sample, but it appears in no diagnostic manual and there is no agreed threshold.
  • Brain imaging suggests dishonesty escalates through adaptation rather than through decision. The amygdala responds less to each successive self-serving lie, and the size of that drop predicts how much the next lie grows.
  • Most compulsive lying is not strategic and often is not even advantageous. The lies that puzzle people most are the pointless ones, and those are the ones least well explained by treating lying as a choice about payoff.

Compulsive lying describes lying that has stopped being a decision and started being a default, usually about small things, often with nothing to gain. It is not a diagnosis, and the most useful fact about it is statistical: lying is very unevenly distributed across people, so somebody who lies constantly is not a more extreme version of the average person but a genuinely unusual case.

That framing matters because the folk theory of lying is a theory of cost-benefit, and cost-benefit predicts almost none of what worries the families who go looking for this.

Most people barely lie

The assumption behind most advice about dishonesty is that everybody lies a bit and some people lie more. Prevalence research does not support the first half.

When people are asked to report the lies they told in the previous twenty-four hours, the most common answer is none. Around six in ten report zero, and the distribution has a long tail rather than a hump: a small minority report large numbers, and they account for a disproportionate share of the total. Work by [serota-2015-prolific-liars] found that about half of all reported lies came from roughly five percent of people.

Where reported lies actually come from Reported figures
~50% from about 1 person in 20
  • The most prolific 5% of people 50%
  • Everybody else combined 50%
~6 in 10 reported none at all
  • Told no lies yesterday 60%
  • Told at least one 40%

Share of self-reported lies by teller group, from Serota and Levine (2015). The category labels are ours; the split is theirs.

Read the second tab first. If most people told a steady trickle of small lies, compulsive lying would be a point on a continuum and the word would be doing little work. Because most people tell none, the heavy tellers are a separate population, and describing them as a category is a description of the data rather than a verdict on their character.

Is pathological lying a real condition?

Not officially, and not for want of anybody trying. Neither of the major diagnostic manuals lists it. Lying appears inside the criteria for other conditions, most explicitly as deceitfulness in antisocial personality disorder, but no manual treats chronic lying as a thing in its own right.

Where lying does appear formally, it appears as one feature among many, which is how personality itself is described in current models: as dimensions that vary by degree rather than types you either are or are not.

In 2020, [curtis-2020-pathological-lying] made the case that it should be, and ran a survey of 623 adults in which around thirteen percent identified themselves as pathological liars. Their respondents commonly dated the onset to adolescence. That is self-identification against a term with no fixed meaning, so the number is a starting point rather than a prevalence estimate, and the authors were clear about that.

What the proposal contributes is a set of thresholds to argue about: that the lying is persistent rather than situational, that it is out of proportion to any benefit, and that it causes distress or impairment to the person themselves. The last one is doing the most work, because it is what separates a condition from a complaint other people have about somebody.

The escalation finding

The most interesting piece of laboratory work here is about how lying grows. In one study, [garrett-2016-brain-dishonesty] put people in a scanner and gave them repeated opportunities to be dishonest for their own benefit. The dishonesty escalated with repetition, which was expected.

What was not expected was the neural pattern. Activity in the amygdala, which responds to emotionally significant events, dropped with each successive lie in a way consistent with adaptation. And the size of that drop on one trial predicted how much the lie would grow on the next.

The mechanism that suggests is worth stating carefully, because it is easy to overclaim. This was a laboratory task, with a specific kind of self-serving dishonesty, in a small sample. It does not establish that anybody’s real-world lying works this way. What it does is supply a plausible biological account of a pattern people describe constantly: that it got easier, and that nobody decided it should.

Telling the kinds apart

Most of the confusion in this area comes from one word covering behaviours that have almost nothing in common.

What it looks likeWhat is driving itDoes the person believe it
Everyday social liesSmall, brief, mostly to smooth an interactionPoliteness, avoiding a minor costNo
Defensive lyingReactive, follows a question, often about something already doneAvoiding a specific consequenceNo
Compulsive lyingFrequent, often unprompted, often no gain attachedHabit and self-presentationSometimes, partly
Deception as manipulationSustained, coordinated, aimed at a personControl or extractionNo, it is the point

That last row is a different phenomenon wearing similar clothes, and it is usually the one people actually need help with when they come looking. Our guide to emotional manipulation covers it, and the practical difference is that compulsive lying is chaotic and mostly harms the teller, while manipulation is organised and aimed.

What this does not tell you

Nothing here supports diagnosing anybody. The prevalence work measures reported lies, which requires honest reporting about dishonesty, and the obvious problem with that has never been solved. The escalation study is a scanner task with a financial incentive, not a life. And the case for pathological lying as a category rests on a survey and a review rather than on a body of treatment trials.

It also does not establish that compulsive lying is outside somebody’s control. Adaptation makes something easier; it does not make it automatic. The reason to prefer the word compulsive over the word dishonest is that it points at the right intervention, not that it removes responsibility.

What helps

Treating the lying as the target rather than as evidence. Therapy for this tends to work on the moment before the lie: what the truth would cost, what the lie is protecting, and what happens when the protection is dropped once. That is a behavioural project, and it needs the person to want it.

Reducing the number of occasions. A lot of compulsive lying is maintained by environments that reward performance and punish ordinariness. Fewer situations that demand an impressive answer means fewer defaults to fire.

For the person on the other side: stop testing. Asking a question you already know the answer to produces one more lie and teaches nothing. Say what you know, say what it cost you, and stop there.

Keeping your own record. Not as ammunition, but because the most disorienting part of living around this is losing confidence in your own account of events. Writing down what happened, on the day it happened, protects that.

When to seek help

Speak to a doctor or a mental health professional if lying is costing you relationships, work or money and you have tried and failed to stop, or if you find you cannot always tell which version you have given somebody. If you are on the receiving end and it has left you doubting your own memory, that is worth raising for its own sake, separately from whatever is happening to them.

If you are having thoughts of harming yourself, treat that as urgent and contact your local emergency services or a crisis helpline.

How MyFreud can help

MyFreud is useful here for the record-keeping that this particular problem eats: a private daily note of what actually happened and how it felt, kept somewhere that does not get renegotiated later. For anyone working on their own lying, the same log is where the pattern becomes visible, which is usually the step before anything changes.

Download MyFreud and start today: App Store or Google Play.

Frequently asked questions

What is the difference between compulsive lying and pathological lying?

Neither term has an agreed clinical definition, which is worth knowing before anyone uses one about you. In common usage compulsive lying describes lying that has become habitual and largely automatic, often about small things and often with no gain attached. Pathological lying usually implies something more elaborate and more sustained, sometimes with the person appearing to lose track of what they have claimed. Researchers who have tried to formalise the second have proposed criteria around duration, distress and impairment, but no manual has adopted them, so in practice the two words get used interchangeably by everyone except the small number of people studying it.

Why do people lie about things that do not matter?

This is the pattern that convinces relatives something is genuinely wrong, and it is also the least well explained. Treating lying as a calculation about advantage predicts strategic lies and predicts nothing about inventing a detail nobody asked for. The more plausible accounts involve habit and self-presentation rather than gain: a story that runs slightly better, told often enough that the telling stops requiring a decision. What the escalation research adds is a reason the threshold might keep dropping over time even when nothing is being gained.

Can someone believe their own lies?

Sometimes, partially, and this is less exotic than it sounds. Memory reconstructs rather than replays, so a story repeated often enough acquires the qualities that usually mark a real memory: detail, fluency, and a sense of familiarity. That is a normal property of memory rather than a special property of liars. What it means practically is that asking somebody to admit a lie can genuinely fail to produce an admission, because the honest answer at that moment is not clean.

Is compulsive lying a symptom of a personality disorder?

It can be, and it frequently is not. Deceitfulness appears in the criteria for antisocial personality disorder, and lying features in descriptions of several other patterns, so it is one of the things a clinician assessing personality will ask about. But plenty of people who lie compulsively meet no criteria for any personality disorder, and the presence of lying on its own tells you very little. Our guide to [the personality disorder types](/personality/personality-disorder-types/) sets out what the categories actually contain.

How do you help someone who lies compulsively?

Start by dropping the interrogation, which is the intervention almost everybody tries first and which reliably makes things worse. Catching somebody out produces a new lie to cover the old one, and it also makes you the person they now manage rather than the person they might tell. What tends to work better is being specific about the effect rather than the offence, keeping your own account of events without arguing about theirs, and treating a therapy referral as the goal rather than a confession. Note that they have to want it: nobody has found a way to make this change from the outside.

References

  1. 1.Serota KB, Levine TR ( 2015). A few prolific liars: variation in the prevalence of lying. Journal of Language and Social Psychology. doi:10.1177/0261927X14528804
  2. 2.Curtis DA, Hart CL ( 2020). Pathological lying: theoretical and empirical support for a diagnostic entity. Psychiatric Research and Clinical Practice. doi:10.1176/appi.prcp.20190046
  3. 3.Garrett N, Lazzaro SC, Ariely D, Sharot T ( 2016). The brain adapts to dishonesty. Nature Neuroscience. doi:10.1038/nn.4426