When do children start to use syllogistic reasoning?

Author(s): Sydney Foss

Mentor(s): Sabine Doebel, Psychology

Abstract
Deductive reasoning is typically associated with adolescence although some literature is pointing to children being able to use it in certain contexts. This effect was first demonstrated by Harris and Dias (1988) who found evidence that children are able to use syllogistic reasoning in a playful context compared to when premises are read to them in a matter-of-fact manner. For example: All pigs fly; Percy is a Pig; Does Percy fly? The current study seeks to test an alternative explanation for what is occurring: in a play context, children may be more likely to base their answers on the content provided to them in the premises, whether or not an answer follows logically from those premises. Children ages 4 and 5 years will be randomly assigned to hear premises and conclusions that are presented in either in a play context or a serious context. The conclusions will either follow logically from the premises or not. Children will also be asked to justify their response. It is expected that children in the play context will be more likely to respond in ways consistent with drawing an inference that follows logically from the premises than children in the serious context, as previously found; however, they will also be more likely to respond inaccurately when the conclusion does not follow (e.g., All pigs fly; Percy flies; Is Percy a pig?). Additionally, children in the play condition will use ‘theoretical’justifications of the kind reported in previous research even when affirming the consequent, suggesting that these justifications are not indicative of deductive reasoning.
Audio Transcript
Hello! My name is Sydney Foss and I am currently a senior majoring in Psychology with a concentration in Development. This semester I participated in OSCAR’s USRP program where I have been studying at what age children start being able to use deductive reasoning, specifically through the use of syllogisms. So, if I were to say something like, “All cows quack. Susie is a cow. Does Susie quack?” You would most likely say “yes” and if I asked you why, you might say something like “because it follows the premise” or “because you said so.” This is an example of a syllogism. Syllogisms are a form of reasoning where a conclusion is drawn from a set of premises. If you were to ask a young child the same thing, they will respond with “no” instead of “yes.” This is because they are basing their conclusion off of past experience instead of what is in the syllogism itself. They haven’t seen a cow quack so how could that be true? But could there be a way that children would reach the correct response? In 1988, researchers Harris and Dias sought out to find just that. What happens if you change the context around the syllogism? They found that by putting the syllogism in the play context by saying “Pretend I am on another planet” beforehand, children were more likely to give the correct response. But is this because the play context helps with deductive reasoning or because of something else? My project investigates whether in a play context children are more likely to base their answers on the information given, even when the information is not logically correct. Additionally, it acts as a replication study to balance out the play and serious conditions which in the original work were unbalanced in representation. We are planning on having children aged 4-5 come into the lab space and hear 8 different syllogisms in a play or serious context. Additionally, we are using two different types of syllogisms that afford different inferences. The first is Modus Ponens which should look a little familiar. This is “All cows quack. Susie is a cow. Does Susie quack?” Saying ‘yes’is a correct deductive inference. The second is affirming-the-consequent. For example: “All cows quack. Susie quacks. Is Susie a cow?” Saying ‘yes’is the incorrect inference. A key question is whether children will be more likely to answer yes to this last type of syllogism in the play context versus the serious context or whether they will instead be more likely to realize it doesn’t logically follow from the premises. If they are more likely to say “yes” in the play context it could suggest that the play context encourages simply responding in terms of the information provided as opposed to deductive reasoning since ‘yes’is the incorrect answer to this reasoning type. Another way we are testing children’s reasoning is by taking the modus ponens statement and having the answer to it be “no.” For example: “All cows quack. Susie is a cow. Does Susie moo?” Saying ‘no’is the correct deductive inference here. After each “yes” or “no” response, children will be asked for a justification. “Why did you say yes?” or “Why did you say no.” We want to see the reasoning behind their answers. Due to the COVID-19 pandemic, we have made it so that the inferences will be said in a video format and a researcher will ask the questions. This is so that children can see the mouth movements and not question that they are just hearing things incorrectly. We predict that children will be more likely to draw valid conclusions from modus ponens syllogisms in the play context versus the serious context, but they will also draw more invalid conclusions from affirming the consequent syllogisms. We also expect they will use similar justifications across modus ponens and affirming the consequent syllogisms. I would like to thank my mentor Sabine Doebel, my lab manager Nicole Stucke, OSCAR for giving me this opportunity, and all of you for watching.

3 replies on “When do children start to use syllogistic reasoning?”

Excellent. Nice description of two types of syllogisms. Do you think the play context makes is harder for children to follow the logic or during play, logic is just not an issue because they are pretending? I hope that makes sense. Interesting stuff.

Thank you! They are imagining the information they are being told and responding on the basis of the information in an intuitive way rather than using deductive logic. They are more likely to do that in the play context because real life isn’t relevant in that instance. For example, with “All cows quack. Susie is a cow. Does susie quack?” They are just thinking of cows quacking and then they are inclined to say yes, susie quacks.

Hi Sydney, very interesting work. How will you generalize your findings as to the age when using syllogistic reasoning starts? How did you decide your sample size to account for inter-personal differences among children?

Leave a Reply