TRAINING 



509 



seek to determine what the limits of such 

 aids are. On the other hand, most aids are 

 used as part of a curriculum which also in- 

 cludes lectures, drill sessions, motivating ses- 

 sions, and other methods of instmction. 

 Research on how to make a good film will 

 not itself tell how to make a film which func- 

 tions effectively in a total program. When 

 considered as only one technique of teaching, 

 one must ask, what can films (or other aids) 

 do better than other instiiictional methods? 

 Shall we use five minutes of film footage to 

 motivate men to learn, or shall we leave that 

 to the instructor and spend the costly footage 

 on getting across complex ideas more thor- 

 oughly? How can films and trainers be 

 most effectively used along with drills on 

 actual equipment? Should all explanations 

 and trainer drills precede drills at sea, or can 

 training aids be used more effectively for 

 review of difficult topics? These and other 

 questions are open to research, but it is 

 essential that studies go beyond demonstrat- 

 ing that training aids are good. They must, 

 instead, find ways of making sure that the 

 time and effort expended on training mate- 

 rials is fully productive. Too often, in the 

 past, techniques have been adopted which 

 did not produce enough gain to justify their 

 cost in effort and money. 



Research must also be directed to inci- 

 dental aids. Do booklets, posters, and the 

 like change men's behavior? Veiy little is 

 known on this question, yet such aids oc- 

 cupied a great deal of manpower in wartime. 

 Evaluation of incidental devices is not diffi- 

 cult. If it should be shown that posters and 

 similar devices have any effect on behavior, 

 research could then profitably turn to the 

 question of how to design them for greatest 

 potency. 



Transfer of Training 



Research on transfer has shown that what 

 a man learns in one situation will transfer to 

 a new situation if the two situations are simi- 

 lar, if the man understands a general rule 

 applying to both situations, or if the original 



training employed a variety of stimuli. 

 These principles were useful in designing 

 military training, but they were effective in 

 some problems and not in others. The diffi- 

 culty is that present theory of transfer does 

 not make clear just when situations are 

 similar enough for transfer. So far, only 

 experimental study of each new problem 

 can determine for certain whether training 

 will help. Under some circumstances, train- 

 ing may be positively harmful because the 

 man is trained in one situation and fights in 

 another. If a telephone talker learns to 

 enunciate in a quiet room, he may be un- 

 intelligible when he relays messages on a 

 noisy ship. Tracking exercises with a gun 

 mounted on a classroom floor may lead to 

 improvement; when the graduates try to 

 track a target from a rolling deck, their 

 training may prove helpful, unhelpful, or a 

 source of confusion and interference. If the 

 pips presented by a synthetic radar trainer 

 do not exactly resemble signals from actual 

 targets, men may learn to respond to false 

 cues. Such problems can be reduced by 

 training carefully designed so that the man 

 responds to stimuli like those he will face on 

 duty with responses like he will make on 

 duty. When tests of the transfer value of 

 the training are made, errors will be dis- 

 covered and eliminated. But research 

 which shows exactly how much similarity be- 

 tween training and duty is required, and just 

 what aspects of similarity must be con- 

 sidered, would be especially helpful. Some 

 studies have drawn attention to factors of 

 this type (26, 35, 36, 42, 48, 59, 61). 



If a training experience could transfer to 

 a large number of situations, it would be 

 most helpful. For example, it would be 

 an advantage if training a man on one radar 

 or one fighter plane would equip him to 

 handle other models with only a brief check- 

 out. This is in part a problem of equip- 

 ment design. When the landing flap con- 

 trol is sometimes to the right of the pilot, 

 sometimes to the left, serious accidents re- 

 sult because of negative transfer. All gear 



