Newell and McDonald (1992) suggested that the effects of learning strategies on parameter learning (a coordinative relationship among the limbs is already formed and the goal of learning is to acquire suitable movement parameters) may differ from the effects on coordination learning (the coordination among the limbs is not yet formed yet and needs to be learned). The purpose of this research is to examine the effects of learning strategies for acquiring a new bimanual relative-phasing pattern.

In the experiments, the effects of four strategies, (1) advanced organization (giving to learners some conceptual clues about whole task structure or feedback structure), (2) observational learning, (3) guidance method, and (4) physical practice only, were examined for acquiring a new bimanual relative-phasing pattern. The experimental task involved continuous back-and-forth movements of both upper limbs whereby one limb led the other limb by 1/4 of a cycle (90 degrees relative phase). Subjects of the advanced organization group were provided with, before practice, the following explicit instructions: "The movement of the right hand reflects the horizontal traces in the monitor and the left reflects the vertical". A description of the in-phase pattern (0 degrees relative phase) and anti-phase (180 degree relative phase) was also given to this group. Subjects of the guidance group were guided from behind by an expert in the 90 degrees relative phase pattern.

An ANOVA was conducted on the retention test for strategy groups. The main effect for group was significant (F(3,28) = 16.761, p<.001). The advanced organization group performed with significantly lower RMSE scores than all other groups. It was demonstrated that the most suitable strategy was a combination of the advanced organization and actual physical practice. Such a cognitive plus active-movement strategy could drive learners to learn the movement actively and connect the actual movement performance with summarized terminal visual feedback about their movements. Schmidt and Lee (1999) suggested that although the effect of the guidance method was denied to occur in parameter learning, it was speculated that the guidance method was effective in coordination learning. However, Experiment 1 showed that the guidance method was not effective on coordination learning. This may be because there is a difference between active movements performed by the learners themselves in normal physical practices and the passive movement imposed by the expert in the guidance method. It may also be because the guidance group may not have regarded the terminal visual feedback available during the learning trials as important feedback information.

In regard to the observational learning, both an expert model and a learning model were compared (the novice model is a learner who learns the given task and improves his/her skills—the novice model is also named a "learning" model). These two models were thought to be similar to each other in the effect on parameter learning. However, Experiment 2 showed that the learning model was more effective than physical practice only on coordination learning, although the expert model was not more effective than physical practice only (F(4,40) = 4.468, p<.01). This showed that observing the problem-solving processes experienced by the learning model is greatly more effective than observing excellent performance repeatedly presented by the expert model on coordination learning. Observing the learning model also seems to have enhanced the learners’ motivation because the learning model progressed from a novice to an expert. Therefore the observational learning with a learning model is a useful strategy because of both observing a problem-solving process and enhancing learners’ motivation.

The effects of the learning strategies on coordination learning were suggested to be different from those on parameter learning in some instances such that the coordination learning should need much stronger engagement of cognitive factors than does parameter learning. In conclusion, the employment of strategies such as advanced organization and observational learning with a learning model is effective on coordination learning.