Acquisition and retention effects of reduced relative frequency of knowledge of results

Three experiments examined the acquisition and retention of a motor skill by manipulating the amount and timing of information feedback or knowledge of results (KR). In the linear positioning task used here, KR is the distance moved on the just-completed response. In Experiment 1 the effect of consecutive KR trials was examined by presenting separate groups in acquisition with sequences of either two or three consecutive KRs in a reduced relative frequency protocol. There were no statistically reliable differences in error scores (absolute or variable) in either short-term retention (2 min, R1) or long-term retention (24 h, R2), between five groups (n = 13); 100% KR, 20 KRs at a fixed interval (every fifth trial), 2 consecutive KRs, 3 consecutive KRs, and no consecutive KRs (20 KRs presented quasi-randomly). Experiment 2 examined subject-controlled KR by employing a group given 20 KRs on request, a KR presentation-trial matched control, a 100% KR group, and a 20 KRs at fixed interval group. There were no statistically reliable between-group differences in error scores for either R1 or R2. Experiment 3 replicated Winstein and Schmidt's (1990) KR fading procedure with the same results. These experiments confirm earlier findings (Sparrow & Summers, 1992) that reduced relative frequency in acquisition does not enhance performance in retention. It is suggested that reduced relative frequency effects are task specific and are not found with linear positioning tasks.