Trial-and-error versus errorless learning: training, transfer, and stress.

Ninety-six student pilots were trained on a range-estimation task, either by a trial-and-error method or by an errorless method. After eight training trials, performance under the two methods was statistically equal. In subsequent performance, the trial-and-error subjects were superior on transfer (p < .05) and under stress (p < .01). Under a combination of transfer and stress, there was no significant difference between the two groups.