UC San Diego

The purpose of this dissertation was to investigate preschool children's selective trust and understanding of reliability. These experiments present new key findings in the area of selective trust. Chapter 1 investigated children's reasoning about the reliability of sources based on honesty. Children received advice in a search task from speakers with either a history of helpfulness or deception. Three- and four-year-olds did not discriminate between helpful and deceptive sources, even though 4-year- olds identified and predicted the future behavior of deceptive sources. Five-year-olds systematically preferred advice from helpers. These results suggest children distrust deceptive sources by 5 years of age, and that children may have knowledge of mental states they do not apply to their trust judgments. Children's ability to selectively trust honest sources was also related to their theory of mind understanding. Chapter 2 investigated whether children make relative or absolute judgments about the reliability of inaccurate sources. Children heard testimony about novel object names from speakers with a history of correctly or incorrectly labeling familiar objects. Children tended to accept labels provided by inaccurate informants, unless another source (with a history of accurate labeling, or no known history) also provided conflicting testimony. These results suggest young children use prior accuracy to determine the relative reliability of informants, but still trust a single informant alone, regardless of that informant's past reliability. Chapter 3 investigated whether children's selective trust is influenced by the reason a source is unreliable (bad knowledge or bad intentions), and the strength of cues about a source's unreliability. Children heard a puppet give inaccurate advice in a search task as a consequence of bad knowledge or bad intentions. With only minimal cues to speakers' mental states, children distrusted incompetent sources more than deceptive ones, but with stronger cues to unreliability, children distrusted both sources equally. This suggests children have more difficulty reasoning about unreliability resulting from bad intentions. Collectively, this research highlights the importance of understanding children's trust judgments as well as the reasoning behind them, and suggests children's decisions to trust sources may not always reflect knowledge they have about the mental states of others