Two Modes of Generalization: A Comparison

64 students were required to perform 2 independent generalization tasks to determine whether those who produced error responses on one task also produced error responses on the other task. The results show a significant inter-task correlation but the obtained relationship is too small to support individual case prediction.