Search

The aim of this project is to provide basic knowledge of the methods which may be used by a man-computer system to detect the presence of a target, using data from a passive sonar receiver. This research consists of analytical studies to evaluate important system parameters and experimental investigations measuring operator performance under various operating conditions. The first two reports in this volume describe the effects of pattern variations on human pattern recognition. The results measured the operator's ability to visually detect patterns differing in shape and to detect patterns generated by statistically dependent sequences. The second two reports deal with basic human information processing and describe the testing of a predictive model for reaction time to visual stimuli and a test of the effects of number of stimuli on memory span., See also Volume 1, AD653278.

The use of a unique computer-controlled (PDP-5) CRT display system with light pen facility in the study of a wide range of signal detection and pattern recognition problems is described. The displays used to study the capabilities of the human operator to detect signals embedded in noise are described and illustrated. In addition to illustrating displays to study the effects of various signal and display parameters, pre-processing and real-time, operator-directed (via light pen) processing are also shown. The use of this system for tachistoscopic stimulus presentations to study basic human information processing capabilities is also described. (Author), Prepared in cooperation with Connecticut Univ., Storrs. Dept. of Electrical Engineering and Dept. of Psychology.

The aim of this project is to provide basic knowledge concerning the methods which may be used by a man-computer system to detect the presence of a target using data from a passive sonar receiver. This research consists of analytical studies to evaluate important system parameters and experimental investigations to measure operator performance under various operating conditions. The first two reports evaluate the loss in detection capability if clipped data rather than continuous data is used for detection. Under the alerted operator assumption it is shown that, on the average, fewer signal samples are required to achieve a given detection capability using sequential detection techniques instead of fixed sample size detection techniques. The next two reports describe methods of improving detection probability through pre-processing of the clipped signal before it is presented to the operator in the form of a binary bearing-time display. The last two reports investigate methods to analyze the statistical properties of correlated random digital sequences. It is shown that the complete distributions of a large class of random digital sequences can be described as either multinomial processes, Markov processes, or linearly dependent processes. (Author)

The study is the third in a series which is concerned with man as a monitor. These studies are related to the problems encountered in automatic and semiautomatic systems, and in the modern submarine in particular, where a man must detect and control incoming signals. The study was designed to obtain performance measures when Ss had a two-fold task, that of compensatory tracking on one display (keeping a pointer on target) while monitoring two other displays for discrete, steady-state signals.

The aim of this project is to provide basic knowledge of the methods which may be used by a man-computer system to detect the presence of a target, using data from a passive sonar receiver. This research consists of analytical studies to evaluate important system parameters and experimental investigations measuring operator performance under various operating conditions. The first report is a description of the overall research capability of the display facility at the University of Connecticut facility. The next two reports describe experimental investigations using a simulated sonar search task. The results of the operator's ability to detect curved targets and to use computer aids under light-pen control are described in one while the effects of transforming the dots to vertical line segments are discussed in the other. The next three reports describe a series of experiments on basic human information processing characteristics. (Author)

An experimental program to determine the capabilities of a man-computer system in processing target-in-noise displays is described. A general detection model, based on digital processing and a time varying operator display of the type found in sonar systems, is discussed. The theoretical effects of a number of detection model parameters are analyzed and their implications for the human operator explored. An experimental study of human operator processing of simulated Time-Bearing data under varying signal to noise ratios (S/N) and clipping thresholds (T) are described. Results indicate that human operators can maintain a consistently high detection performance level over the range of S/N and T studied. Performance, as measured by the number of trials to reach a decision, follows the theoretical optimum predicted by analytical studies. (Author)