RESUMO
This study investigated young children's ability to draw psychological and nonpsychological inferences during reading comprehension. Whereas nonpsychological inferences require the retrieval of general background knowledge, psychological inferences rely on more contextualised knowledge relating to mental states. Based on several pretests, children, who were able to read fluently, aged 7-8 years (second graders; n = 42) and 8-9 years (third graders; n = 46) were assigned to either a skilled comprehenders group or a less skilled comprehenders group, based on their listening comprehension. They were then given short stories to read, followed by comprehension questions. Some questions required the drawing of psychological or nonpsychological inferences. Generalized linear mixed models revealed that (a) psychological inferences were more difficult to generate than nonpsychological inferences for all the children, skilled and less skilled comprehenders alike, (b) both types of inference skills (psychological and nonpsychological) were associated with similar improvements as grade level increased, and (c) less skilled comprehenders had greater difficulty than skilled comprehenders generating psychological and nonpsychological inferences. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
