Children's causal inferences from conflicting testimony and observations

Area of Science:

• Cognitive Development
• Causal Inference
• Social Learning

Background:

• Preschoolers learn causal relationships through direct observation and testimony.
• Investigating how children integrate potentially conflicting information from these sources is crucial.

Purpose of the Study:

• To examine how 4-year-olds integrate conflicting statistical and social information about causal relationships.
• To understand children's evaluation of informant reliability based on observed evidence.

Main Methods:

• Children observed probabilistic or deterministic evidence contradicting an informant's testimony about a machine's activating blocks.
• Informant reliability was manipulated (knowledgeable vs. naïve).
• Children's judgments of block effectiveness and informant trust were assessed.

Main Results:

• Children favored the informant's endorsed block more when the informant was knowledgeable, but still considered the observed statistical data.
• Evidence of conflicting data reduced children's trust in the informant and their willingness to select her endorsed novel block.
• Children demonstrated rational integration of social and physical cues, adjusting beliefs based on informant confidence and data strength.

Conclusions:

• Preschoolers rationally integrate social testimony and direct observation, even when conflicting.
• Children dynamically assess informant reliability and data strength to make causal inferences.
• This study highlights the sophisticated reasoning preschoolers employ in learning causal relationships.