TY - CHAP

T1 - Self-Explanation, Feedback and the Development of Analogical Reasoning Skills: Microgenetic Evidence for a Metacognitive Processing Account.

AU - Cheshire, Andrea

AU - Ball, Linden J.

AU - Lewis, Charlie

PY - 2005

Y1 - 2005

N2 - It is generally accepted that providing explanations during a task can facilitate problem solving performance in both adults and children. This paper aims to answer two important questions. First, can current theories of explanation be generalised to children’s explanations of self-generated answers? Second, what is the impact of such self-explanation on the development of children’s analogical reasoning skills? One-hundred-and-ten six- and seven-year-old children took part in seven sessions of matrix completion trials in one of five conditions: (1) explanation plus feedback; (2) explanation only; (3) feedback only; (4) practice; and (5) control. Analysis revealed that, contrary to existing theory, explanation of selfgenerated answers is not the most effective way to encourage the development of analogical reasoning. Rather, feedback on response accuracy is necessary for the attainment of heightened levels of performance. Results also indicate that children shift from using surface-level perceptual cues as a basis for their responses to a more sophisticated strategy involving an understanding of deeper-level relational structures. It is argued that these results support a metacognitive processing account of the development of analogical reasoning skills rather than an account emphasising changes in mental representations.

AB - It is generally accepted that providing explanations during a task can facilitate problem solving performance in both adults and children. This paper aims to answer two important questions. First, can current theories of explanation be generalised to children’s explanations of self-generated answers? Second, what is the impact of such self-explanation on the development of children’s analogical reasoning skills? One-hundred-and-ten six- and seven-year-old children took part in seven sessions of matrix completion trials in one of five conditions: (1) explanation plus feedback; (2) explanation only; (3) feedback only; (4) practice; and (5) control. Analysis revealed that, contrary to existing theory, explanation of selfgenerated answers is not the most effective way to encourage the development of analogical reasoning. Rather, feedback on response accuracy is necessary for the attainment of heightened levels of performance. Results also indicate that children shift from using surface-level perceptual cues as a basis for their responses to a more sophisticated strategy involving an understanding of deeper-level relational structures. It is argued that these results support a metacognitive processing account of the development of analogical reasoning skills rather than an account emphasising changes in mental representations.

M3 - Chapter

BT - Proceedings of the Twenty-Seventh Annual Conference of the Cognitive Science Society

A2 - Bara, B. G.

A2 - Barsalou, L.

A2 - Bu, M.

PB - Sheridan Printing

CY - Alpha, New Jersey

ER -