The emergence of computer software in the Army's tactical equipment on a relatively large scale raises the question of how to best train the diverse populations of soldiers on this software. This paper presents the results of an experiment comparing computer-based programs for training the knowledge and "digital" skills required in using a map interface typical of many of these systems. A prototype Land Warrior map interface was modeled. It was used to display and find individuals and units, and to determine the range and azimuth between objects and individuals/units on the map.

First, soldiers learned a coding system for identifying individuals and units displayed on the map. This system combined the Army's standard weapon and unit symbols with the battle roster numbering system. Two training conditions were compared. In one condition, much information was presented before soldiers had an opportunity to apply the knowledge and skills they had acquired. This placed high demands on working memory. In the other condition, the same information was presented in smaller chunks, placing lower demands on working memory.

In the map phase of the experiment, three ways of training map skills were compared. Again, low and high demand conditions were implemented, and an exploratory condition was added. In the exploratory condition, soldiers were informed of the map functions they had to learn, but not how to execute the required steps.

A total of 168 soldiers from four Infantry courses participated: those in initial Infantry training, the Infantry Officer Basic Course, and the Basic and Advanced Noncommissioned Officer Courses. These soldiers represented the soldiers in an Infantry platoon who will use the future Land Warrior system. Within each course, soldiers were randomly assigned to the experimental conditions.

Consistent differences in the soldier populations occurred in both experimental phases, with officers typically achieving the highest scores in the shortest amount of time, and Infantry trainees scoring the lowest, taking the most time. In both training phases, differences in favor of the low demand condition occurred for the instructional segments where the amount of information presented was the most discrepant between the high and low conditions. On the map final exam, soldiers in the exploratory condition had the lowest scores, with those in the low demand condition achieving the highest scores. In sum, the low demand condition was the most effective for code and map skills. Combining this traditional mode of instruction with an exploratory mode might also prove very effective in acquiring the interactive skills and insights required to work with digital interfaces.