Throughout history, individuals believed to have extraordinary capabilities were generally highly ranked in their communities; this suggests a universal “extraordinary-dominant expectation” in human minds, which may play a key role in religious thought, even in modern societies. This study shows that 5–6-year-old children, who begin to understand real-world causalities regarding how the body and mind of human beings work, predict that individuals who exhibit extraordinary capabilities have higher social status in interactions with individuals who exhibit ordinary capabilities. In Experiment 1, we showed children two individuals achieving goals using either humanly possible or impossible methods, the latter involving simple forms of violation of intuitive psychology (knowing without seeing), physics (flying), or biology (fire breathing). The children clearly judged the latter as surprising and unusual. More importantly, the children predicted that individuals showing extraordinary capabilities will gain contested resources and play a dominant role in interactions with ordinary individuals, indicating a higher social status. Further investigations suggested that the children specifically linked extraordinary capacities to social status, as they did not attribute dominance to individuals who apply surprising/unusual but possible methods (Experiment 2), and that they did not indiscriminately attribute positive characteristics to extraordinary capabilities despite a strong extraordinary-dominant expectation being replicated (Experiment 3). These findings demonstrate that extraordinary-dominant expectations can be observed in childhood across different intuitive knowledge domains, helping understand the cognitive mechanisms of religious thought and the cognitive foundations of hierarchical social systems.

Reciprocal interactions require memories of social exchanges; however, little is known about how we remember social partner actions, especially during childhood when we start forming peer-to-peer relationships. This study examined if the expectation-violation effect, which has been observed in adults’ source memory, exists among 5–6-year-old children. Forty participants played a coin collection game where they either received or lost coins after being shown an individual with a smiling or angry expression. This set-up generated congruent (smiling-giver and angry-taker) versus incongruent (smiling-taker and angry-giver) conditions. In the subsequent tasks, the children were asked to recall which actions accompanied each individual. The children considered the person with incongruent conditions as being stranger than the person with congruent conditions, suggesting that the former violated the children’s emotion-based expectations. However, no heightened source memory was found for the incongruent condition. Instead, children seem to better recognise the action of angry individuals than smiling individuals, suggesting that angry facial expressions are more salient for children’s source memory in a social exchange.

Remembering whether a person is cooperative is essential in social interactions. It has been shown that adults have better memory of a person who showed an incongruence between emotional expression and expected behavior (e.g., smiling while stealing). To examine whether children would show similar emotional incongruity effects, we examined 70 children aged 5 or 6 years. They obtained coins that could be exchanged later for rewards (stickers) by answering quiz questions. Then, they participated in the coin collection game where individual persons with smiling or angry expressions appeared one at a time on a computer monitor. These same individuals then either gave coins to or took coins away from the children, leading to congruent (smiling giver and angry taker) and incongruent (smiling taker and angry giver) conditions. After the game, children needed to choose between two faces to indicate which one previously appeared in the game. Participants recognized faces better under the incongruent conditions. In particular, the smiling taker was recognized significantly better than the angry taker, whereas no difference was observed for the smiling and angry givers. Evidently, 5- and 6-year-olds better remember individuals whose facial expression or appearance is incongruent with their expected behavior.