Bayes nets and babies: infants’ developing statistical reasoning abilities and their representation of causal knowledge

Abstract

A fundamental assumption of the causal graphical model framework is the Markov assumption, which posits that learners can discriminate between two events that are dependent because of a direct causal relation between them and two events that are independent conditional on the value of another event(s). Sobel and Kirkham (2006) demonstrated that 8-month-old infants registered conditional independence information among a sequence of events; infants responded according to the Markov assumption in such a way that was inconsistent with models that rely on simple calculations of associative strength. The present experiment extends these findings to younger infants, and demonstrates that such responses potentially develop during the second half of the first year of life. These data are discussed in terms of a developmental trajectory between associative mechanisms and causal graphical models as representations of infants' causal and statistical learning.

The Development of Diagnostic Reasoning about Uncertain Events between Ages 4–7

“…Several studies suggest that children can engage in diagnostic reasoning about causal structures at early ages [2]–[5]. This previous research suggests that such capabilities are present during the first year of life [6]–[8]. By the preschool years, children's actions suggest they engage in hypothesis testing of the potential causal structures they diagnose, particularly when faced with data that suggest there are multiple possible causal structures [4]–[5], [9]–[11].…”

Person as Scientist, Person as Moralist

“…Some research examined early causal reasoning -intuitions about spatio-temporal relations (Leslie & Keeble 1987;Oakes & Cohen 1990), causal mechanisms (Bullock et al 1982;Schulz 1982), and the use of statistical cues in causal judgments (Gopnik et al 2001;Sobel & Kirkham 2006). Other research focused on children's "mind-reading" abilities -what they knew about the intentions, desires, beliefs, and knowledge states underlying human actions (e.g., Lutz & Keil 2002;Repacholi & Gopnik 1997;Wellman 1990;Woodward 1998).…”

Bayesian generic priors for causal learning.

“…From a very young age, humans display a remarkable ability to acquire knowledge of the causal structure of the world (e.g., Bullock, Gelman, & Baillargeon, 1982), often learning causeeffect relations from just a handful of observations (e.g., Gopnik, Sobel, Schulz, & Glymour, 2001;Sobel & Kirkham, 2007). Causal knowledge is particularly valuable in guiding intelligent behavior, making it possible to make predictions, diagnose faults, plan interventions, and form explanations (see Buehner & Cheng, 2005).…”