THE TEACHING OF SERIAL TASKS USING CHAINING STRATEGIES

S ummary . The experiment was concerned with teaching two types of chain or serial task. The chain types were paper‐folding, regarded as a motor assembly task, and numerical procedures. The aim of the experiment was to investigate the effect of chain length (short, medium and long) and teaching strategy (backward chaining, forward chaining and whole method) on the acquisition and retention of the two types of chain. 176 women students were randomly assigned to nine groups (3 × 3) defined by chain length and teaching strategy. Within each group a motor and a number chain were taught by means of self‐instructional programmes and followed by practice to a criterion of one promptless trial. One week later the tasks were relearned to the same criterion. No advantage was found for backward chaining in the case of the motor chains and the short and medium number chains. However, for the long number chain there was some indication that backward chaining was superior to forward chaining. Both methods were in general inferior to the whole method.