Flexibility of theory of mind in a matrix game when the opponent's level changes.

This study used a matrix game to explore adolescents' and young adults' flexibility of theory of mind (ToM) and to examine whether prediction could promote a person's ToM reasoning when the opponent's level of ToM changed during the course of the game. A total of 202 participants of different ages (13, 16, 19, and 24 years) were presented with the task. Results showed that the ToM flexibility in prediction and decision making was not significantly different across the age groups. A person's flexibility of ToM was better when the opponent's level was stable than when the level changed. It was even more difficult for a person to adapt when the opponent changed from a higher-order to lower-order level. The results showed prediction to be a kind of information clue that would prompt people to think further. It is both feasible and discriminative to explore even higher levels of ToM with matrix games.

Age-related schema reliance of judgments of learning in predicting source memory.

Source memory refers to mental processes of encoding and making attributions to the origin of information. We investigated schematic effects on source attributions of younger and older adults for different schema-based types of items, and their schema-utilization of judgments of learning (JOLs) in estimating source memory. Participants studied statements presented by two speakers either as a doctor or a lawyer: those in the schema-after-encoding condition were informed their occupation only before retrieving, while those of schema-before-encoding were presented the schematic information prior to study. Immediately after learning every item, they made judgments of the likelihood for it to be correctly attributed to the original source later. In the test, they fulfilled a task of source attributing. The results showed a two-edged effect of schemas: schema reliance improved source memory for schema-consistent items while impaired that for schema-inconsistent items, even with schematic information presented prior to encoding. Compared with younger adults, older adults benefited more from schema-based compensatory mechanisms. Both younger and older adults could make JOLs based on before-encoding schematic information, and the schema-based JOLs were more accurate in predicting source memory than JOLs made without schema support. However, even in the schema-after-encoding condition, older adults were able to make metacognitive judgments as accurately as younger adults did, though they did have great impairments in source memory itself.