Causality

Theoretical Landscape

One general approach for causal learning and reasoning is that children possess innate predispositions for domain-specific causal knowledge that allow for sophisticated reasoning abilities such as prediction and explanation. One such approach, advocated by those who take a core knowledge approach, suggests a level of continuity in development. Children have access to knowledge and reasoning systems driven by evolutionary mechanisms that do not change much in development and serve as the foundation for adult knowledge. Perhaps the most stripped-down version of this approach suggests that human beings are predisposed toward possessing a causal learning system, but no specific causal knowledge is “hardwired” into the system. The acquisition of causal knowledge comes from environmental interaction.

A second approach is advocated by proponents of theory theory, who suggest that children are born with initial domain-specific knowledge, but these initial structures change via observation and interaction with the environment. Most proponents of this account suggest that the child is born with the capacity to represent and reason about causal knowledge. However, some interpretations of this account suggest that the ability to represent and reason about causal knowledge itself undergoes development; in this view, these abilities develop during the first year or few years of life.

A third approach is that children are born with the capacity only for statistical learning, which allows them to recognize complex associations among events and thus leads to prediction and explanation. These approaches suggest that the learning mechanisms are present early on, but young children do not represent specific knowledge. In some ways, this view is most similar to the stripped-down core knowledge view, except that in this view, the representation of such knowledge is never causal but merely regularities among events.

A final approach—one based on sociocultural processes—is that causal knowledge is not specifically represented in the mind of the child but, rather, develops as a function of social interaction and cultural processes. Integrating this idea with approaches like theory theory, one can think of social information as another form of input to a causal learning system. Indeed, many pieces of causal knowledge—such as the nature of germs, the shape of the Earth, and the definition of gravity—are gained through social means, not observation. However, other researchers reject this view and suggest that such causal knowledge is co-constructed between the child and others via social or cultural processes.

Each theoretical account of how causal knowledge is acquired specifies the ways in which people make predictions, generate explanations, and reason counterfactually. Some argue that each of these abilities is independent from one another. For example, when making an inference about another’s mental states, young children (3-year-olds) can often explain another’s actions in terms of false beliefs that predict how another person will act given that children possess a belief that is [Page 322]not true. Others suggest that there are coherent representations of causal knowledge that allow for each of these reasoning abilities. Explanation emerges before prediction in this case because the prediction tasks require additional cognitive mechanisms—such as inhibitory control—for children to demonstrate success.

...