Thomas H. Rockwell

Strategies of Visual Search by Novice and Experienced Drivers

Six novice drivers drove a 2.1-mi. neighborhood route and a 4.3-mi. freeway route. Eye movements (including blinks and glances to the vehicles mirrors and speedometer) were videotaped. The visual behavior of a control group, consisting of four experienced drivers, was also videotaped on the same routes. The results showed that the novice drivers: (1) concentrated their eye fixations in a smaller area as they gained driving experience, (2) looked closer in front of the vehicle and more to the right of the vehicles direction of travel than the experienced drivers, (3) sampled their mirrors less frequently than the experienced drivers, and (4) made pursuit eye movements on the freeway route while the experienced drivers made only eye fixations. These results suggest that the visual acquisition process of the novice drivers was unskilled and overloaded. Thus, the search and scan patterns of the novice drivers may be considered unsafe in terms of impairing the drivers’ ability to detect circumstances that have high accident potential. On this basis it is recommended that novice drivers be prohibited from driving on public roads until they achieve an acceptable level of vehicle handling control and develop skill in acquiring visual information.

Mapping eye-movement patterns to the visual scene in driving: an exploratory study

Eye movements of eight drivers were filmed as the drivers traveled on a local expressway at 50 mi. per hour. Search and scan patterns of the drivers became more compact and the center of location shifted down and to the left as the drivers became more familiar with the route. The center of the final pattern was located above the right road edge marker and slightly higher than the horizon. The task of car following appeared to induce greater visual workload as indicated by increased sampling rates of lane markers and greater visual travel distances to examine road signs and other traffic. The search and scan patterns verified that the peripheral area of the eye is used for monitoring lane position, other vehicles, and road signs so that the fovea may be directed for a closer examination when the situation demands it.

Eye Movements in Curve Negotiation

Eye movements and fixations of five drivers were recorded and superimposed on a videotaped recording of the dynamic visual scene as they drove on a two-lane rural road. The results showed that (1) on curved roads, the fixation pattern follows the road geometry, whereas on straight roads, the search behavior is less active, and most of the fixations are close to the focus of expansion. The results indicate that in driving through curves drivers direct foveal fixations to lateral placement cues rather than rely on peripheral vision; (2) the process of curve scanning begins in the approach zone prior to the curve itself, suggesting that perceptually the curve negotiating process precedes the curve by several seconds; (3) the search patterns on right and left curves are not symmetrical; visual excursions to the right on right curves are greater than eye movements to the left on left curves; and (4) fixation duration statistics may be related to accident rates on curves.

Field dependence and driver visual search behavior.

This paper reports on two studies that examined the relationship between field dependence and on-the-road visual search behavior. In the first study, concerned with eye movements in curve negotiation, it was found that field-dependent subjects have a less effective visual search pattern. In the second study, young and aged drivers were compared on several information processing tasks and on their ability to maintain their eyes closed part of the time while driving. Of the various information processing tasks, only field dependence and visual search time correlated significantly with the mean time the drivers needed to maintain their eyes open while driving, Together the two studies indicate that field dependent subjects require more time to process the available visual information and are less effective in their visual search pattern.

Modeling fault diagnosis as the activation and use of a frame system

Twenty pilots with instrument flight ratings were asked to perform a fault-diagnosis task for which they had relevant domain knowledge. The pilots were asked to think out loud as they requested and interpreted information. Performances were then modeled as the activation and use of a frame system. Cognitive biases, memory distortions and losses, and failures to correctly diagnose the problem were studied in the context of this frame system model.

VISUAL INFORMATION SEEKING OF NOVICE DRIVERS

ALTHOUGH NOVICE DRIVERS LEARN TO CONTROL AUTOMOBILES WITH A FEW HOURS OF PRACTICE, THE DEVELOPMENT OF THEIR VISUAL INFORMATION SEEKING HABITS TAKES MANY MONTHS. FILMS OF NOVICE DRIVER EYE MOVEMENTS SUGGESTED THAT THERE ARE STAGES OF VISUAL LEARNING. TECHNIQUES NEED TO BE DEVELOPED TO TEACH DRIVERS WHERE TO LOOK BEFORE THEY ACTUALLY START DRIVING. /AUTHOR/

Evaluation of Alternative Alphanumeric Keying Logics

Four keying logics for the entry of alphanumeric characters were evaluated. Selection of the logics was based on their compatability with current keyboard designs used in the pilotaircraft interface, which involves communications, navigation, and other aircraft subsystems. Subjects in each of four groups learned one of the logics, and their performance was recorded on an alphanumeric keying task. Performance measures were keying speed and keying accuracy for alphabetic characters, numeric characters, character strings, and total list. Keying time for a keying logic that used 36 individual keys for each alphabetic and numeric character was significantly superior to other logics that used push-button telephone-type keyboards. Subjects were equally accurate on all logics, with an error rate of approximately 0.5%. Significant performance differences among the three matrix keyboard logics were inconsistent. Of these, performance on a keyboard with the telephone arrangement was most accurate.

A sensitivity analysis of empirically derived car-following models☆

CONSIDERABLE interest has been evidenced in recent years in traffic flow and car-following theory. Two classes of models are used to describe these phenomena: macroscopic models usually involve structures which relate volume, speed and density for a large number of vehicles; microscopic models concern relation of the driver-vehicle system to intervehicular dynamics. Analogies to kinetic theory of gases, diffusion processes, compressible fluids, etc., are just a few of the approaches to macroscopic modeling that have been studied. This paper concerns itself with microscopic models of the car-following situation and how the driver of the following car responds to the action of the lead car. Microscopic models have importance in (a) better understanding traffic flow concepts, (b) designing automatic systems which closely match human performance and (c) designing systems to aid or assist the driver in the car-following task, i.e. partially automated systems and sensory supplementation. At least three approaches can be taken to the study of the driver-vehicle car-following function. One is to use physical models, e.g. Newtonian-forcemass-acceleration models (Herman, 1961). A second approach, and the one taken in this paper, is curve-fitting from empirical data in order to describe the driver-vehicle system in terms of intervehicular dynamics. Lastly, this paper will conclude with an introduction to a third approach to car-following models, which is presently underway at the Systems Research Group at The Ohio State University. This combines traffic dynamic information with driver psychophysical data to predict control movements from intervehicular dynamics. The purpose of this study was to examine the sensitivity of empirically derived carfollowing models to (a) time delays on predictor variables, (b) lead-car velocity programs, that is, the forcing velocity function (open road versus traffic), (c) subjects and (d) replications within subjects. Emphasis in this work was with regression models of the driver-vehicle system as shown in Fig. 1 using acceleration of the following vehicle as the dependent variable and headway (spacing) lead-car velocity, following-car velocity and lead-car acceleration as possible predictor variables. Time was also used as a predictor variable in the original experiment to measure learning effects. Models predicting the change in following-car acceleration, A& were also investigated. In general, time did not improve data fit which suggests little

THE EFFECT OF DISCRETE HEADWAY AND RELATIVE VELOCITY INFORMATION ON CAR-FOLLOWING PERFORMANCE

STUDIES WERE MADE TO DETERMINE WHETHER A DISPLAY PRESENTING CHANGING INFORMATION BEFORE THE DRIVER CAN SENSE CHANGE MIGHT IMPROVE CAR-FOLLOWING PERFORMANCE. THIS DISPLAY WOULD PROVIDE INFORMATION WHETHER THE DRIVERS HEADWAY WERE WITHIN A BANDWIDTH OR ERROR TOLERANCE ESTABLISHED BY THE EXPERIMENTER OR ELECTED BY THE DRIVER HIMSELF. ANALYSIS OF VARIANCE WAS USED TO TEST THE DATA. SUBJECTS, BANDWIDTH, AND TYPE OF DISPLAY WERE CONSIDERED AS MAJOR INDEPENDENT VARIABLES IN THE EXPERIMENT AS WELL AS THE INITIAL TARGET HEADWAY. THE DEPENDENT VARIABLES WERE THE HEADWAY VARIANCE AND RELATIVE VELOCITY VARIANCE OVER THE TRIAL. RESEARCH INDICATES THAT A DISCRETE LIGHT DISPLAY PRESENTING HEADWAY AND RELATIVE VELOCITY INFORMATION CAN IMPROVE CAR-FOLLOWING SYSTEM PERFORMANCE. BEST DISPLAY COMBINATION WOULD APPEAR TO BE THE SMALL BANDWIDTH FOR BOTH DISPLAYS. ABSOLUTE PERFORMANCE IMPROVEMENT AT CLOSE TARGET HEADWAYS SEEMS RATHER SMALL. AT CLOSE TARGET HEADWAY CONDITIONS, THE DRIVER MAY PREFER TO USE OTHER CUES THAN THOSE PROVIDED BY THE DISPLAY.

An Experimental Evaluation of Method and Tool Effects in Spike Maul Use

An experiment was performed to study the effects of hand-tool design factors and methods of tool use upon tool force generation and loading of the back. Forty novice and experienced subjects were tested on their ability to drive a spike with a railroad spike maul. Method of tool use affected the spike-driving performance of novice subjects, whereas tool striking surface area influenced the ability of experienced trackmen to drive spikes. Method of tool use also affected the components of spine loading. Tool force generation was considered as a function of spine loading indices to create efficiency measures. Efficiency was used to evaluate the cumulative trauma effects of the hand tool.