Citation

Nataupsky M, McCloy TM, Bermudez JM, Tirnan VW, Buchta VG, Seeber CM, Colabufo JJ. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 1980; 24(1): 572.

Copyright

(Copyright © 1980, Human Factors and Ergonomics Society, Publisher SAGE Publishing)

DOI

10.1177/1071181380024001147

PMID

unavailable

Abstract

Recent studies have shown that criterion levels established in training directly affect later performance of subjects on experimental tasks. Approximately 20% of variance can be explained by these criteria. The purpose of this study was to determine if a similar relationship can be found in transfer of training situations. Twenty male Air Force Academy cadets were trained to one of two multiple criteria levels on a difficult flight manuever in a GAT-1 simulator. There was a easy criterion set and a more difficult criterion set. These two sets consisted of holding prescribed performance parameters in heading, vertical velocity, and altitude. After achieving their assigned criterion, all cadets in each of the two groups were then tested on the same task in a GAT-1 simulator, but this time the maneuver had to be performed under turbulent wind conditions. This wind condition served as the transfer task. Half of the cadets in each group had the same criterion in both the training and the transfer task. The other cadets had different criteria in the training and transfer tasks. Thus there were four experimental groups: easy-easy, easy-difficult, difficult-easy, difficult-difficult. One control group had the easy criterion while the other control group had the difficult criterion. There were five cadets in each control group. The dependent measure was the Transfer Effectiveness Ratio (TER), derived from trials of this criterion data. This index is an estimate of the amount of time saved in learning a transfer task when performance is adjusted to that of a control group. Several analyses of various tasks of derived scores yielded significant results, confirming that criterion levels established in training carry over to transfer of training situations. Moreover, the data showed consistency in accounting for 20% or more of the variance.

Language: en