End-point focus manipulations to determine what information is used during observational learning

We required two groups of participants to observe an end-point model (ENDPT) while another two groups viewed a full-body, point-light model (FULL) to determine the role of relative motion information in acquisition of a multi-limb, whole-body action. One ENDPT and one FULL group also bowled a ball. Following retention, all groups observed the FULL model. The participants’ movements were compared to the model and outcome attainment was quantified. There was no difference in shoulder–elbow coordination between groups in acquisition or retention. The FULL groups replicated hip–knee coordination more accurately than did ENDPT groups in early acquisition only, with no significant differences in late acquisition or retention. Both bowling groups became more accurate at the task across acquisition, but the ENDPT group was more accurate and consistent in retention. Providing intra-limb relative motion in re-acquisition did not improve coordination for the ENDPT groups, but it did facilitate movement control (peak wrist velocity) and outcome attainment (target accuracy). The acquisition of coordination during observational learning is not only a result of copying relative motion information, but also involves copying of end-point trajectory information from the primary effector.