Gavan Lintern

Part-Task Training for Tracking and Manual Control

Part-task training was defined as practice on some set of components of the whole task as a prelude to performance of the whole task. Part-task procedures are intended to improve learning efficiency and to reduce costs. Our review focused on the instruction of tracking skills for manual control. Transfer of training was emphasized and crucial features of the methodology and of means of assessing transfer were discussed. The part-task procedures of segmentation, fractionation, and simplification were explained, and procedures for reintegrating parts into the whole task were summarized.

Display principle, control dynamics, and environmental factors in pilot training and transfer

Sixty-four flight-naive men were tested in a fractional factorial, quasi-transfer experiment to examine the effects of four display factors, one control response factor, and one environmental factor on acquisition and transfer of aircraft landing skills. An additional 12 trainees served as experimental controls. Transfer was measured from each of 64 experimental training conditions to a criterion condition with a conventional inside-out pictorial contact display, normal simulator control dynamics, and a 5-knot crosswind. Transfer was better following training with pictorial displays than with symbolic displays, and with normal rather than reduced bank control order. Interactions of crosswind with predictive augmentation and with bank control order showed that for some conditions, transfer benefited from training with predictive augmentation and from training without crosswind.

An informational perspective on skill transfer in human-machine systems

Differentiation of perceptual invariants is proposed as a theoretical approach to explain skill transfer for control at the human-machine interface. I propose that sensitivity to perceptual invariants is enhanced during learning and that this sensitivity forms the basis for transfer of skill from one task to another. The hypothesis implies that detection and discrimination of critical features, patterns, and dimensions of difference are important for learning and for transfer. This account goes beyond other similarity conceptions of transfer. To the extent that those conceptions are specific, they cannot account for effects in which performance is better following training on tasks that are less rather than more similar to the criterion task. In essence, this is a theory about the central role of low-dimensional informational patterns for control of behavior within a high-dimensional environment, and about the adjustment of an actors sensitivity to changes in those low-dimensional patterns.

Transfer of landing skill after training with supplementary visual cues.

An aircraft simulator, with a closed-loop computer-generated visual display, was used to teach flight-naive subjects to land. A control training condition in which subjects learned to land with reference to a skeletal airport scene consisting of a horizon, runway, centerline, and aiming bar was tested against training with constantly augmented feedback, adaptively augmented feedback, and a flightpath tracking display. A simulator-to-simulator adaptively of training design showed that adaptively trained subjects performed best in a transfer task that was identical to the control groups training condition. Several subjects attempted six landings in a light airplane after they had completed their experimental work in the simulator. They performed better than another group of subjects that had not had any landing practice in the simulator.

Transfer of landing skills in beginning flight training

Beginning flight students from the University of Illinois flight training program were given two sessions of landing practice in a simulator with a computer-animated contact landing display before they commenced intensive landing practice in the aircraft. For each experimental student there was a control student, paired with the same instructor, who received no landing practice in the simulator. Experimental students required significantly fewer presolo landings in the airplane than did the paired controls, representing a potential saving of about 1.5 presolo flight hours per student. These data show that pretraining with a moderately detailed, yet relatively inexpensive, computer-animated landing display can offer worthwhile savings in flight time. Some students were provided adaptive visual augmentation during their simulator training, and there was evidence of incremental transfer attributable to this instructional feature.

Adaptive Training of Perceptual Motor Skills: Issues, Results and Future Directions.

This review has two aims; the first is to assess AT as a method for teaching control skills, and the second is to establish a conceptual framework that will allow a detailed analysis of adaptive manipulations and their influence on skill acquisition. The major studies in adaptive training research are described and evaluated. A critical examination of the various experiments reveals that there is less support for the application of adaptive manipulations to applied motor skill training than is generally believed. Some apparently favorable experiments have methodological and interpretive flaws that seriously weaken their conclusions. Other experiments that provide tenable support have characteristics that are unique in adaptive training research so that the generality of their data is in doubt. The limitations of the data prevent any firm conclusions being drawn about the efficiency of adaptive training. However a detailed analysis of motor skill theory and research indicates that some adaptive manipulations could be effective. Methodological and conceptual issues that are critical to successfully testing those manipulations are clarified in a discussion of the adaptive training concept. In addition, that discussion outlines several empirical tests that are needed to enable a more effective analysis of adaptive training.

Simulator design and instructional features for air-to-ground attack: a transfer study

A transfer experiment was conducted to define simulator design requirements and instructional procedures for an air-to-ground attack trainer. Two levels of scene detail, three levels of field of view, and three levels of simulator training trials were manipulated. Student naval aviators were pretrained in the Visual Technology Research Simulator (VTRS) on a 30-deg bombing maneuver before their performances were assessed in their standard weapon training course. Other students not pretrained in the VTRS were used for control comparisons. Prior experience in the VTRS helped student pilots use their weapon flight time more effectively. The most substantial transfer gains were produced by the first 24 simulation training trials; there was little additional benefit from further training. There was no evidence of differential transfer from the different simulator conditions. These trends indicate that physical fidelity is not a requirement for positive transfer and lend further support to the need for a concept such as psychological fidelity or functional equivalence.

Content, variety, and augmentation of simulated visual scenes for teaching air-to-ground attack

The Visual Technology Research Simulator was used for a quasi-transfer-of-training study in which 32 military pilots were taught to deliver bombs from a 30-deg dive. Scene content had a strong and consistent effect on performance and on differential transfer. A landscape scene that contained buildings, roads, and rectangular fields was better than a schematic grid pattern for both training and transfer. Scene variety in training did not benefit transfer, and there is a distinct possibility that it can interfere with early learning. Augmented feedback proved to be a potent instructional variable, but one that showed complex effects. It helped inexperienced pilots with their dive pitch control, and it helped the more experienced pilots with their longitudinal bombing error. The data presented here have strong implications for design and use of flight training simulators in that they indicate the importance of scene content and augmented feedback as training variables.

Explicit and implicit horizons for simulated landing approaches

In a flight simulator experienced pilots flew landing approaches to a representation of an airport scene in which various sources of information had been distorted or removed. Reasonably accurate approaches could be made to a scene that contained only an aimpoint and a horizon. The addition of a runway outline did not enhance accuracy or stability, which lent credence to the hypothesis that the invariant angle between horizon and aimpoint can support glide slope control. Explicit distortion of this angle by simulation of up-sloping or down-sloping terrain beyond the runway had predictable effects on glide slope control. Implicit specification of a veridical horizon with texture lines parallel to the runway centerline weakened the effect of distortions in the explicit horizon. Thus both explicit and implicit specifications of the horizon contribute to perception of the glide slope angle. Implications of these results for the design of visual scenes for flight simulation are discussed.

Quasi-Transfer as a Predictor of Transfer from Simulator to Airplane

Simulators have emerged as important components of flight-training programs. Nevertheless, the development of design principles that can maximize training transfer and cost-benefit trade-offs are not well established. The most significant challenge to research that would bear on simulator design principles is the difficulty and expense of flight transfer experiments. This difficulty and expense can be reduced by the use of an insimulator transfer design, designated here as a quasi-transfer study, in which transfer is to a high-fidelity configuration of a simulator. Of primary concern for such studies is whether the implied assumption of correspondence between quasi-transfer and transfer effects is well founded. In this article, we review evidence that bears on this issue. The evidence is not entirely supportive but does indicate some correspondence between quasi-transfer and transfer.