Why It's Good to Discipline Children For Lying: Speculations on Theory of Mind

Psychologists define theory of mind as “…the ability to attribute mental states…to ourselves and others.” It is the capacity to know that other people have their own subjective experiences, thoughts, and so on.

Theory of mind (ToM) develops in children between 4 and 6 years of age. You can observe the development of ToM by playing hide and seek with a toddler, and watch how they get better at the game as they grow older. Toddlers seem to think that if they can’t see you, you can’t see them, so they hide in really stupid places like this:

Children realize as they get older that they need to hide where you can’t see them, so they pick better hiding spots. Eventually, they’re as good or better than you at hide and seek. Hide and seek shows, then, that a toddler can’t comprehend that you see and know things differently than they do, but as young children they obtain this capacity. By the time you’re struggling to find your kid in your own house, they know that you experience life from a different perspective, and with different beliefs about the world.

Another way in which children demonstrate ToM is through lying. Psychologist Dr. Kang Lee at the University of Toronto has demonstrated that around 2 years old, toddlers start telling white lies, and they get better at telling more complex lies as they enter early childhood. In order to lie, a child must know that another person knows differently about a given situation than they do. For example, if a child lied to me about whether he ate my cookies, he would have to know that I do not know for sure whether he in fact ate my cookies. Lying is ToM in action; a lying child makes use of their knowledge that I know differently than they do.

National Geographic did a video in 2017 on Dr. Lee’s work on ToM, titled: “Why Science Says It’s Good for Kids to Lie.” I first watched this video in an Introductory Psychology course several years ago. The clickbait-y title calls into question why we punish kids for lying, and whether we should stop punishing children for lying because lying is an indicator of healthy cognitive development. My professor at the time argued (in a fallacious fashion) that lying is evidence of normal cognitive development, so therefore we shouldn’t punish children for telling lies. Fallacious argumentation aside, my professor’s comments, along with the title of the video, made me think of an explanation for why disciplining a lying child is also evidence of ToM development. More precisely, it is good for children to understand that they are being punished for lying, because it is evidence that they’ve actually completed their development of ToM. Allow me to explain.

My parents, and most parents I know, justify punishment for lying on the basis of reciprocity; if I don’t lie to you, and you don’t lie to me, we both benefit from our shared agreement to be honest. Reciprocity as the justification for discipline is fairly common. You’ve probably walked by the toy aisle in Walmart and witnessed a child throwing a temper tantrum because their parent refuses to buy them a new toy. Parents know that if they don’t deal with this selfishness now, their child may throw more temper tantrums in the future. Parents discipline their child so sooner or later the child will understand that they can’t get their own way all the time, and that they wouldn’t appreciate it if someone else mistreated them for such selfish reasons.

I think that if a child understands why they’re being disciplined for lying (reciprocity), then they’re demonstrating a higher order of ToM than the kind required just to tell a lie. Think of it like this: If I lie to you, I have to know that you know differently than I do; i.e., I have ToM. If I know that you are capable of lying to me, then I must know that you know differently than I do, and that you know that I know differently than you do; i.e., if I know that you’re capable of lying to me, then I know that you have ToM as well. If I want to make a reciprocal agreement not to lie to you, I have to know that you’re capable of lying to me in the first place – otherwise, I may as well go try and make a reciprocal agreement with an inanimate object.

So, lying demonstrates that children have some ToM; the part that says that they know that you know differently. But I think children demonstrate complete ToM by coming to understand the reciprocal agreement against lying. By complying with the agreement, the child shows that they know that you know that they know differently. Notice that I have split ToM into two sub-types: (i) Knowing that you know differently (what we commonly call ToM); and (ii) knowing that others know that you know differently (what I think is a higher order sub-type of ToM). By splitting ToM into these two sub-types, I can further split the development of ToM into three stages:

Stage 1 - The First Person:The child is capable of knowing only in the first person. They are aware of the existence of their own mental states, but not of the existence of the mental states of others.

Stage 2 - The Second Person:The child is capable of knowing in the first and second person. They are aware of the existence of their own mental states, and the mental states of others.

Stage 3 - The Third Person: The child is capable of knowing in the first and second person, but now they can form a third person point of view about themselves and others. They are aware of the existence of their own mental states, and the mental states of others. In this stage, the child is also aware that others have knowledge of mental states that are not their own.

I’ll clarify further as to why I’ve chosen these names for each stage.

Consider hide and seek again. Imagine that I am a toddler. If you and I were playing hide and seek, I’d probably hide somewhere stupid – maybe behind a curtain – because I don’t understand that you can’t see me. My line of reasoning for choosing my hiding spot would be something like this: “If I can’t see, we can’t see.” ‘I’ and ‘we’ are, respectively, the singular and plural first person pronouns. The toddler that sucks at hide and seek is in stage 1 of their ToM development. At this stage, a toddler can’t lie because they have no cognitive representation of the ‘other’ as someone with different mental states; that comes in stage 2.

Now imagine that I’m 5 years old. I can hide in a really good hiding place now, because I know that you see different things than I do, so you therefore know differently than I do about my location. In this way, hide and seek is really just a game of lying about location. My reasoning for why I chose this hiding spot is like this: “If I go where you can’t see me, then you won’t find me.” At this point, I have a cognitive representation of you as the ‘other’, so I can label you using the second person pronoun ‘you’. If I were to steal your cookies and lie to you about it, my reasoning would be similar: “I know that you don’t know that I have your cookies.” So, the second person stage is really just what Dr. Lee and most psychologists would describe as ToM.

A couple of years later, I’m 7 years old and you’re getting annoyed at my lying. You start disciplining me, I keep lying, but finally I stop lying because it hits me that if I lie to you, there’s nothing stopping you from lying to me. And, I realize that when you lie to me, it really drives me nuts. I know that you’re lying about that bag of chips you said were all gone – I saw you stuff them behind the pillow! By now, I can abstract both you and I out into the third person perspective, and imagine that you can lie to me. It’s like I can imagine us as characters in a story, where (suppose for the sake of clarity that you’re a woman named Sally) I, Nick, lie to you about your stolen cookies, but you, Sally, lie to me about the hidden chips! My reasoning would be like this: “He lies to her about the cookies, but she lies to him about the chips!” It’s as if in realizing that you can lie to me, I have abstracted us into a story where we are characters that are both lying to one another, so I am narrating this story in the third person by shifting between the perspectives of both characters. I can finally understand why you don’t like it when I lie to you, because you know that others can lie to me and I won’t like it – that’s reciprocity! Now, whether I think I can get away with breaking the reciprocal rule is another question, but at least I understand why the rule exists, and therefore understand why you’re disciplining me. I now know that you too can see the world from the second person perspective, so you also have this thing Dr. Lee calls ‘theory of mind’.

I think we should expand the concept of ToM into these three stages to avoid using the label to describe only the second person stage. They’re at least conceptually distinct stages, and I have given some reason for thinking that the third person stage develops after, and not at the same time as, the second person stage. Who knows, maybe disorders that present with ToM deficits (such as schizophrenia) would be better understood by breaking the deficits down into these three stages.

I’d love to know what you think of the stages. Do they make sense? Is this as far as the ToM concept can be broken down? Can you think of any empirical uses for these three stages? Let me know in the comments below, and I’ll try to get back to you.

Thank you for reading. I hope you’ve enjoyed the ramblings of this undergraduate.

Nicholas Murray