Lead
Your three-year-old tells their first unmistakable lie. Cookies are missing from the kitchen counter. There are crumbs on their face. "I didn't eat any," they say.
For a moment you almost smile. Then a small worry takes its place: My child has started lying.
It is worth pausing on how to read that moment. A lie at this age is not a signal that discipline has failed. It is a behavior that psychological researchers have documented as a cognitive milestone — appearing on the same developmental curve as other markers of emerging mental capacity. From the perspective of developmental psychology, what you are watching is less "misbehavior" and more "the acquisition of the ability to model another person's mind."
This article traces when and why children begin to lie, how the quality of lying changes with age, and what the research — primarily Talwar and Lee's body of work — suggests about how to respond. The goal is to give you language for observing what is happening before the binary of scold-or-not-scold.
When Does Lying Begin?
The standard experimental paradigm for studying children's lying is the temptation resistance task, commonly called the peeping task. An experimenter places a sound-making toy behind the child, instructs the child not to turn around and look, then leaves the room. On returning, the experimenter asks: "Did you peek?"
Lewis, Stanger, and Sullivan used this paradigm with three-year-olds in 1989 and found that the majority of children who had succumbed to temptation and turned around either denied having peeked or refused to answer when asked [1]. Three years old — when verbal exchange is just becoming reliably stable — and children were already selecting statements that diverged from the facts.
Evans and Lee (2013) extended the paradigm to children ages two to three and found that while two-year-olds who peeked were more likely to confess honestly, the rate of denial climbed as children approached age three [2]. They interpreted this shift not as a moral regression but as the product of developing executive function: a set of cognitive control processes — inhibition, working memory, and mental flexibility — that enable goal-directed behavior and self-regulation — the cognitive control capacities required to produce and sustain a lie [2].
Lying does not appear one day as a bad habit acquired from somewhere. It emerges around age three in most children, tracking the same developmental curve as inhibitory control and cognitive flexibility.
Lying and Theory of Mind Move Together
Why does executive function connect to lying? The relevant concept here is theory of mind: the ability to attribute mental states — beliefs, desires, intentions — to other people and to understand that their mental states may differ from one's own — the ability to represent that other people have mental states (beliefs, desires, intentions) that differ from your own.
Wimmer and Perner introduced the classic test in 1983: the false-belief task, sometimes called the Maxi task [3]. Maxi puts chocolate in a green cupboard and goes outside to play. While Maxi is away, his mother moves the chocolate to the blue cupboard. Where will Maxi look when he returns? Most children younger than four or five answer "the blue cupboard" — where they know the chocolate is — rather than the green cupboard where Maxi left it [3]. Understanding that another person holds a belief that differs from reality, and acting on that understanding, consolidates slowly around ages four to five.
Talwar and Lee connected theory of mind to lying in a systematic program of research. In a 2008 study of children ages three to eight, they found that the ability to deny having peeked (in the peeping task) was associated with first-order false-belief understanding and inhibitory control — while the ability to maintain that lie without contradicting oneself was associated with second-order false-belief understanding: the ability to reason about what one person believes another person believes, requiring two nested mental-state representations: grasping not just "what does the other person believe?" but "what does the other person believe I know?" [4]
To construct a lie, in other words, the child must operate on the premise that the listener's mental world contains different information than theirs. Three-year-olds who deny having peeked frequently contradict themselves immediately afterward. A common pattern: the child says "I didn't peek" and then promptly names the toy they saw. Polak and Harris documented this pattern in detail: three- to five-year-old deniers regularly left behind inconsistent follow-up statements that gave them away [5]. The quality of the lie tracks the sophistication of the theory of mind behind it.
The Lies of a Three-Year-Old, a Five-Year-Old, and an Eight-Year-Old Are Not the Same Thing
Summarizing Talwar and Lee's work [4,6]:
- Age 3: Denial is possible, but subsequent statements contradict it. Facial expressions and behavior leak the truth easily.
- Around age 5: As first-order false-belief understanding consolidates, children gain some ability to manage follow-up statements so they do not contradict the initial denial.
- Ages 7–8: With the development of second-order false-belief understanding, children can maintain a contextually consistent lie across a conversation.
There is also a qualitative shift in motivation. Talwar, Murphy, and Lee (2007) studied "white lie" telling in children ages three to eleven [6]. Children were given an unwanted present and observed for whether they concealed their true reaction. Most children across ages used a white lie — and the rate increased with age. Even without prompting from a parent, 68% of children produced a white lie; when a parent prompted, the rate rose to 86% [6]. The social skill of softening the truth to avoid hurting someone grows quietly inside the same category of behavior we label "lying."
Treating all lies as a single category subject to a single response loses important information. A self-protective lie, a playful lie, and a consideration-for-others lie are meaningfully different even when they use the same words.
One Step Before the Scolding
None of this means welcoming deception. It means that before responding, there is room for one brief question: What was this child trying to figure out when they chose this lie?
- They hid a fact to avoid being scolded: they predicted a parent's emotional response. Theory of mind is operating.
- They thought they wouldn't be caught: they modeled what the parent knows and doesn't know. Theory of mind again.
- They said "it wasn't me" because they knew it would upset a younger sibling: they simulated a sequence of social consequences.
None of these capacities is undesirable. The problem is not the faculty — it is what the child was trying to protect. Asking about the context that produced the lie — rather than prosecuting the fact of the lie — returns more information and typically a more honest account.
A consistent finding in the empirical literature: threats of punishment do not increase honest disclosure. In Talwar and colleagues' research, a simple request — "I'd be happy if you told me the truth" — was more effective than a warning of punishment at eliciting truthful reports from children [7]. This is not a prescription to never impose consequences. It is a separate, additional observation: the threat of punishment also functions as pressure to become a more skilled liar.
Logging a first lie in Memori with a date attached is a minor act of developmental record-keeping. Looking back at it a year later, the three-year-old's transparent contradiction and the four-year-old's tidier version are different events. The record tells you that your child is moving through something, not stuck in something.
Summary
Children's lying is not a failure of parenting. It emerges on the same developmental schedule as theory of mind and executive function [1,2,3,4]. Three-year-old lies contradict themselves and leak through expression; five-year-old lies are more consistent; by the early school years, socially motivated white lies are also in the repertoire [4,6].
One second's worth of observation — what was this child modeling when they chose this lie? — shifts the lie from a problem behavior to a developmental clue. The date your child told their first lie is probably worth remembering.
References
- Lewis M, Stanger C, Sullivan MW. Deception in 3-year-olds. Dev Psychol. 1989;25(3):439–443. doi:10.1037/0012-1649.25.3.439.
- Evans AD, Lee K. Emergence of lying in very young children. Dev Psychol. 2013;49(10):1958–1963. doi:10.1037/a0031409. PMID: 23294150.
- Wimmer H, Perner J. Beliefs about beliefs: representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition. 1983;13(1):103–128. doi:10.1016/0010-0277(83)90004-5. PMID: 6681741.
- Talwar V, Lee K. Social and cognitive correlates of children's lying behavior. Child Dev. 2008;79(4):866–881. doi:10.1111/j.1467-8624.2008.01164.x. PMID: 18717895.
- Polak A, Harris PL. Deception by young children following noncompliance. Dev Psychol. 1999;35(2):561–568. doi:10.1037/0012-1649.35.2.561. PMID: 10082026.
- Talwar V, Murphy SM, Lee K. White lie-telling in children for politeness purposes. Int J Behav Dev. 2007;31(1):1–11. doi:10.1177/0165025406073530. PMID: 18997880.
- Talwar V, Arruda C, Yachison S. The effects of punishment and appeals for honesty on children's truth-telling behavior. J Exp Child Psychol. 2015;130:209–217. doi:10.1016/j.jecp.2014.09.011.