Is gesture-speech mismatch a general index of transitional knowledge?

When asked to explain their beliefs about a concept, some children produce gestures that convey different information from the information conveyed in their speech (i.e., gesture-speech mismatches). Moreover, it is precisely the children who produce a large proportion of gesture-speech mismatches in their explanations of a concept who are particularly "ready" to benefit from instruction in that concept, and thus may be considered to be in a transitional state with respect to the concept. Church and Goldin-Meadow (1986) and Perry, Church and Goldin-Meadow (1988) studied this phenomenon with respect to two different concepts at two different ages and found that gesture-speech mismatch reliability predicts readiness to learn in both domains. In an attempt to test further the generality of gesture-speech mismatch as an index of transitional knowledge, Stone, Webb, and Mahootian (1991) explored this phenomenon in a group of 15-year-olds working on a problem-solving task. On this task, however, gesture-speech mismatch was not found to predict transitional knowledge. We present here a theoretical framework, which makes it clear why we expect gesture-speech mismatch to be a general index of transitional knowledge, and then use this framework to motivate our methodological practices for establishing gesture-speech mismatch as a predictor of transitional knowledge. Finally, we present evidence suggesting that, if these practices had been used by Stone et al., they too would have found that gesture-speech mismatch predicts transitional knowledge.