New Queen’s study examines children’s lies

Study examines how children with and without autism interact with the truth

Queen’s psychology professor Elizabeth Kelley and PhD candidate Annie Li conducted the study.
Queen’s psychology professor Elizabeth Kelley and PhD candidate Annie Li conducted the study.

All children tell lies, but children with autism aren’t as good as covering them up. These findings were the result of a one of a kind Queen’s study, conducted by psychology professor Elizabeth Kelley and PhD candidate Annie Li, as well as Kang Lee from the University of Toronto.

This study is the first in Canada to test whether autistic children will lie.

“There’s been lots of studies that look at children’s understandings of other people’s lives but no studies that looked at whether children will tell a white lie on their own,” Kelley said. “This is interesting because there’s lots of research that shows children with autism don’t understand beliefs and feelings of other people, and one of the reasons we lie is to change other people’s beliefs or protect their feelings. We felt this was important thing to investigate.”

Nineteen children with autism, ages six to 12, and 30 typically developing children, ages six to 10, participated in the study that was conducted from 2007 to this September. The study looked at two kinds of lies.

“The experiment looked at anti-social lies which are lies that are used to cover our own mistakes or misdeeds, and it also looked at pro-social lies, which are white lies used to protect another person’s feelings,” Kelley said.

“In the anti-social lies test, we had kids play a guessing game where they had to guess what toy was hidden based on the noise we played for them. For example, we’d have a chicken clucking noise and the kids would guess a chicken toy was hidden,” she said.

Before the third and final guessing trial, someone knocked on the door and told the experimenter s/he had to leave to answer the phone. The experimenter told the children not to peek at the toy.

“There was no link between the sound we played and the toy in this trial. We played Christmas music before the experimenter left and had an Elmo toy, so the child would only know the toy if they did in fact peek,” she said.

When asked, the children who peeked would lie to cover it up, Kelley said. Hidden cameras showed that 15 out of 30 typically developing children who were tested peeked at the toy and seven of these children lied and said they saw the wrong toy.

Kelley said this is very different than the children with autism. While 15 out of 19 children with autism peeked, 14 lied and said they did not peek and only one of the 15 said they saw the wrong toy.

“In other words, children with autism are able to lie, but aren’t as good as covering up their lies,” she said.

Kelley and Li received unexpected results in the pro-social lies test.

“We played another game with the children, where we told them they’d get a great prize, the experimenter left and left prizes for children, and the prize was just a bar of soap,” Kelley said.

“When the experimenter came back in they asked the children if they liked the prize. We expected the children with autism to say they didn’t like the prize, but 83 per cent of the children with autism said they liked the prize,” she said, adding that 68 per cent of typically developing kids said they liked the prize.

Still, Kelley said the motivation for the children’s lies in this study remain unclear.

“We are interested in pulling apart why these kids lied,” Kelley said. “It’s hard to figure out why they’re lying. Is it learned behaviour or do they understand what they’re doing?”

A lot of kids couldn’t answer the question as to why they told the white lie, Kelley said.

“We got a lot of responses like ‘I don’t know’ or ‘that’s what you’re supposed to say’. They really fell apart when you asked, and it’s hard to interpret why they did it.”

PhD candidate Annie Li told the Journal in an e-mail that future studies would hope to confirm the findings of this study.

“In the future, we hope to replicate our findings with a larger sample of children, add in a measure to examine prosocial lie-telling ability, and include several other measures to clarify what our results mean on a theoretical level,” she told the Journal in an e-mail.

All final editorial decisions are made by the Editor(s)-in-Chief and/or the Managing Editor. Authors should not be contacted, targeted, or harassed under any circumstances. If you have any grievances with this article, please direct your comments to

When commenting, be considerate and respectful of writers and fellow commenters. Try to stay on topic. Spam and comments that are hateful or discriminatory will be deleted. Our full commenting policy can be read here.