Do children understand what information is most helpful for identifying good sources? (4- and 5-year-olds)
Quite a bit of research has demonstrated that children rely on various cues to determine who might be a good source of information. For example, someone's previous behavior, and even that person's appearance, may affect whether a child chooses to trust them. However, it may also be the case that some kinds of cues (like whether someone knows the names of a lot of different objects) may be better indicators of knowledge and reliability than other cues (like whether they simply look knowledgeable). In this series of studies we are exploring whether children can make this distinction between different types of cues. In other words, do they understand that some pieces of information may be more helpful in identifying a good source of information than others? This work helps us understand whether children have a metacognitive understanding of how they learn about other people.
What do children think about different source types? (10-year-olds)
This study explores what children think about ChatGPT as an informant. Prior research has revealed that as children age, they perceive the internet and VAs (voice assistants) as more trustworthy with factul information (e.g., scientific and historical facts) than human sources. Prior to this technology, children also demonstrated preferences for nonhuman, print-based sources (via written text) as compared to human informants. In the current study, we are exploring the intersections of these areas of research to analyze how children think about text-based AI and whether they deem it a more or less reliable source than humans or other text-based sources (e.g., textbooks). This work helps us to understand: 1) the capabilities of AI as an educational tool and 2) how children's interactions with them may develop over time.
Does anonymity affect children's assessment of others? (5- and 10-year-olds)
Previous research has shown that young children often assign positive traits to others, regardless of what information they have about those people. In this study, we are seeking to understand more about the reasons behind this so-called 'positivity bias'. Specifically, does answering questions about others in an anonymous way influeence how positive children will be towards others? If children are less positive when they are reporting answers anonymously, then it is possible that social pressure to maintain positive relationships with others may be one factor in why children are typically positive about others' attributes. This works helps us understand what factors children consider when assigning traits to others.