The study by Smith and Yu (2008) investigates how infants learn word-referent mappings in early language acquisition. Traditional theories suggest that infants resolve referential ambiguity by unambiguously deciding the referent in a single word-scene pairing. However, the authors propose that infants can also learn through cross-situational statistical learning, where they accumulate statistical evidence across multiple ambiguous word-scene pairings to determine the correct mappings. The experiment involved 12- and 14-month-old infants learning six word-referent pairs through individually ambiguous trials. Each trial presented two words and two potential referents, with no information about which word matched which referent. Across trials, the correct word-object pairings were consistent, allowing infants to learn the mappings over time. The results showed that both age groups of infants looked longer at the target object when it was associated with the auditorily presented label compared to the distracter. This indicates that infants were able to learn multiple word-referent mappings by accumulating statistical evidence across trials. The study suggests that cross-situational statistical learning is a significant mechanism in early lexical learning, potentially contributing to the rapid and robust acquisition of vocabulary in young children.The study by Smith and Yu (2008) investigates how infants learn word-referent mappings in early language acquisition. Traditional theories suggest that infants resolve referential ambiguity by unambiguously deciding the referent in a single word-scene pairing. However, the authors propose that infants can also learn through cross-situational statistical learning, where they accumulate statistical evidence across multiple ambiguous word-scene pairings to determine the correct mappings. The experiment involved 12- and 14-month-old infants learning six word-referent pairs through individually ambiguous trials. Each trial presented two words and two potential referents, with no information about which word matched which referent. Across trials, the correct word-object pairings were consistent, allowing infants to learn the mappings over time. The results showed that both age groups of infants looked longer at the target object when it was associated with the auditorily presented label compared to the distracter. This indicates that infants were able to learn multiple word-referent mappings by accumulating statistical evidence across trials. The study suggests that cross-situational statistical learning is a significant mechanism in early lexical learning, potentially contributing to the rapid and robust acquisition of vocabulary in young children.