Search

This chapter surveys the major research methods and techniques used in the study of color and its effects on human perception and performance. Although a great many research methods have been devised to obtain quantitative data on human vision, only a small subset of those methods are directly pertinent and useful in the study of color sensitivity and the effects of color. The more pertinent research methods can be generally classified into psychophysical, physiological, and behavioral methods. Psychophysical methods are those which measure perceptual capabilities of observers and relate the perceptual (psychological) processes to physical dimensions of the stimulus. Psychophysical methods include those that determine the magnitude of sensation attributed to the stimulus. Physiological methods include both central nervous system as well as sensory electrophysiological recordings. Behavioral methods are those which assess the performance capabilities of the observer in performing a task related to the visual stimulus, rather than attempting to measure the perceptual process alone., Pub. in Color in Electronic Displays, p95-135, 1992 (ISBN 0-306-44191-8). Prepared in cooperation with Virginia Polytechnic Institute and State University, Blacksburg, VA.

This study was conducted to investigate the effects of display failures and rotation of dot-matrix symbols on visual search and recognition performance. The type of display failure (cell, horizontal line, vertical line), failure mode (On, failures matched the symbols; Off, failures matched the background), percentage of cells failed (0,1,2,3,4%), and rotation angle (0 deg, 70 deg, 105 deg) were the variables examined. Results showed that displays that exhibit On cell failures greater than 1% significantly affect search time performance. Cell failures degrade performance more than line failures. Search time and accuracy are best when symbols are oriented upright. The effects of display failures and rotation angle on search time were found to be independent. Implications for display design and suggestions for quantifying the distortion caused by rotation are discussed.

This experiment is one of a series of studies about visual task performance with dot-matrix displays. In this study, the variables of matrix size and symbol rotation are evaluated for their effects on visual search time for familiar alphanumerics as well as for dot-matrix-formed U.S. Army symbols. The results show that increasing the size of the matrix from 7 x 9 to 9 x 11 or even to 11 x 15 dots significantly improves performance for alphanumerics as well as Army symbols, particularly for the more difficult and confusable symbols. Search accuracy improves in much the same manner as search time does.

When dot-matrix characters are rotated, as may be in a moving map display, their dot-matrix patterns are distorted and their legibility is thus affected. In this experiment, 16 subjects performed a random search task in which they were asked to look for a target in a random character pattern. The independent variables were (a) the direction (clockwise or counterclockwise), (b) the angle of stimulus image rotation, and (c) the target character's distance from the center of the screen, which was also the center of rotation. The dependent variables were response time and response correctness. Significant effects were found for the angle of rotation, the target character's distance from the center, and the target character. The results indicate that (a) no angle-dependent mechanism is involved in performing this task and the angle of rotation influences recognition mainly through the distortion of dot-matrix patterns; (b) the target character's radial distance from the center of the screen is the determining factor for search time, while the x and y coordinates of the target contributed to dot-matrix pattern distortion; and (c) the target characters interacted differently with the angle and distance factors to determine the extent of distortion and their legibility. Means to quantify the extent of distortion are discussed and the direction for future research is suggested.

This research is the second in a series of experiments undertaken to develop optimum design criteria for matrix-addressable displays. This study examined the effects of character size, luminance modulation, font, and polarity on both contextual task (reading) performance and performance in a random search task. The random search task consisted of locating a target letter or numerical within one of nine sectors on the display. Only upper case alphabetic characters and the numerals 0 through 9 were used. Using a font (3) by modulation (3) by polarity (2) by character size (3) within-subjects factorial design, only the main effects of character size, font, and polarity were found to be significant. There were no significant interactions. Post hoc analyses indicated that response speed increased as character size increased from a 7 x 9 matrix, to a 9 x 11 matrix, to an 11 x 15 matrix. Post hoc analyses revealed that response times were longest with the maximum dot font, but did not significantly differ between the Huddleston and Lincoln/MITRE fonts. Additional post hoc comparisons indicated that negative contrast (dark characters on a light background) produced significantly faster response times than did positive contrast (light characters on a dark background).

Reading and visual search speed and accuracy were used to evaluate the effects of simulated line and cell failures on a visual display. Five experimental variables (failure type CELL, VERTICAL LINE, AND HORIZONTAL LINE), PERCENT FAILURE 0%, 4%, 8&, OR 12% CELLS FAILED, display polarity LIGHT CHARACTERS ON DARK BACKGROUND OR DARK CHARACTERS ON LIGHT BACKGROUND, mode of failure FAILURES MATCH THE SYMBOLS OR FAILURES MATCH THE BACKGROUND, and matrix size 7 X 9, 9 X 11, OR 11 X 15 DOTS) significantly affected performance of both the reading and search tasks. Additionally, many of the interactions among these variables were significant. The experimental results indicate that as the percentage of failures increased, performance systematically decreased. For failure levels of less than 4%, little effect on performance was found. Performance increased as the size of characters increased. A 30% improvement in search time and a 7% improvement in reading time was obtained by using dot- matrix sizes larger than 7 x 9 pixels. In general, dark symbols presented on a light background (negative contrast) were read and searched for more quickly than light symbols presented on a dark background (positive contrast).

Research has demonstrated that the majority of antihistamines (H1 antagonists) have sedative effects and can impair psychomotor performance; however, it is claimed that astemizole (hismanal) does not possess central nervous system side effects. A two-factor, repeated measures, double-blind design was used to compare the effects of three treatments (two antihistamines and one placebo) on cognitive information processing, mood, selected physiological measures, subjective feelings of drowsiness, and subjective performance ratings in 28 healthy men. Evaluations were given at 1,3,5,7,9,11, 13, and 15 hours post ingestion. Time-of-day effects were evident in following directions, unstable tracking, code substitution, serial addition/subtraction, logical reasoning, manikin, and pattern comparison tasks. A general trend of improved scores through the day was observed and a temporal pattern of a low performance was suggested in the afternoon (2:00 pm and 4:00 pm). Temporal effects were noted for physiological measures. Keywords: Cognitive performance, Visual motor performance, Mood scales, Antihistamine.

Research has demonstrated that the majority of antihistamines (H1 antagonists) have sedative effects and can impair psychomotor performance; however, it is claimed that astemizole (hismanal) does not possess central nervous system side effects. A two-factor, repeated measures, double-blind design was used to compare the effects of three treatments (two antihistamines and one placebo) on cognitive information processing, mood, selected physiological measures, subjective feelings of drowsiness, and subjective performance rating in 28 healthy men. Evaluations were given at 1,3,5,7,9,11,13 and 15 hours post ingestion. Time-of-day effects were evident in following directions, unstable tracking, code substitution, serial addition/subtraction, logical reasoning, manikin, and pattern comparison tasks. (js)

This annual report describes the year's progress of a research program that will determine the validity of quality metrics of digital imagery and relate these metrics to image interpreter performance with both hard-copy and soft-copy displays. Separate technical reports are issued to cover major elements of work under this contract. This annual report simply summarizes work performed to date.

Several experiments were conducted to evaluate alternate unitary measures of video line-scan system image quality. A metric based upon the Modulation Transfer Function of the imaging system was derived, with emphasis placed upon the photometric properties of the system. This metric was shown to predict well the average effects of several imaging system parameters upon the ability of observers to extract information from both dynamic and static images. In attempting to predict the ability of observers to acquire specific targets in an air-to-ground search task; however, other target and background parameters become very important, and such image quality measures must therefore be refined. Relationships among alternate measures of line-scan image quality were discussed, and a conceptual model was presented for combining system noise, raster interference, scene content, and the visual requirements of the observer.

Ten medium scale (1:2700 to 1:4400) aerial photographs, typical of the imagery viewed by Air Force photointerpreters, were digitized to 4096 x 4096 files (20- micron aperture by 11 bits intensity) on a scanning microdensitometer. The image files were then multiplied by two Gaussian filter functions (Fourier domain) to yield two blurred and one ground truth level of each image and transformed to eight bits of intensity for output. One of four weightings of a 4096 x 4096 Gaussian noise file was added to each image file, yielding 3 Blur x 5 Noise x 10 Image combinations (150 images, total). Positive transparencies then served as the database for an information extraction task. Fifteen photointerpreters (PIs) from the 548th Reconnaissance Technical Group, Hickam AFB, Hawaii, served as subjects. Blur was a between-subjects variable with five PIs at each of three levels. Noise was a within-subjects variable (five levels). The Noise main effect was significant (p .01). The Blur main effect and the Blur x Noise interaction were not found to be statistically significant, although the Blur main effect was of the expected form. The data were correlated with image quality scaling values from a separate study using the same images and PIs. A significant correlation was found (r = .898)., See also ADA096950.

This report summarizes a five-year research program on the subject of quality evaluation of digitally derived imagery. High-resolution aerial photography was used to create digitized images with unclassified content comparable to that employed in military photointerpretation operations. The digitized imagery was used in several hard-copy and soft-copy interpretation experiments to assess the affects of image blur and image noise on both perceived image quality and the ability to extract information from the images. The soft-copy experiments included both nonprocessed and processed imagery. Finally, quality metrics of iamge quality were obtained for both hard-copy and soft-copy images and related directly to both information extraction performance and subjective quality scaling., See also 1, ADA135631.

This research examined the relationships between color contrast and legibility for digital raster video imagery. CIE colorimetric components were combined into three-dimensional color coordinate systems whose coordinates map one-to-one with the physical energy parameters of all colors. The distance between any two colors' coordinates in these 3-spaces is termed Color-Difference (Delta E). Delta A was hypothesized as a metric of the speed (RS) with which observers possessing normal vision could accurately read random numeral strings on one color displayed against backgrounds of another color. Two studies totalling 32064 practice and experimental trials were conducted. The first study determined that the CIE Uniform Color Spaces are inappropriate for the modelling of RS. Subsequently, a different 3-space geometry and colorimetric component scaling were empirically derived from the Study 1 data to produce a one-dimensional Delta E scale which approximates an interval scale of RS. This Delta E scale and others were then applied to the different stimulus conditions in Study 2 to determine the generalizability of such Delta E metrics.

This is the Final Technical Report of Task I: Human Engineering Survey and Analysis of the subject contract. The task is a survey of the pertinent visual performance, display system capability, and human engineering design requirements for flat panel visual displays, as applied to U.S. Navy Airborne, Shipborne, and Land-Based Systems. The report contains two application examples of the selection of flat panel displays for both airborne and ship- based information systems. It also surveys the current state of the art of flat panel display technologies, and the manner by which existing visual performance and theory can be applied to the selection and evaluation of flat panel technologies for various applications. The flat panel technologies surveyed are light emitting diodes, electroluminescence, liquid crystal, electrochromic, electrophoretic, and gas (plasma) discharge. A uniform set of performance variables is used to compare the various technologies, both among themselves and with the traditional cathode ray tube. Similarly, these technologies are evaluated against design criteria and current models of image quality which relate human performance to display characteristics. Pertinent human visual performance data are presented as the basis for selecting and applying image quality models. Data gaps and needs are summarized at the end of the report.

The role of flight crew number in the advanced conceptual bomber is likely to call rapid management of complex information. The elevated workload associated with these new informational tasks may be alleviated in part by the reduction of short term memory demands. Literature reviews were conducted to described the concepts of short term memory and workload. An experiment was conducted to describe the contribution of short term memory demands to operator workload. The Sternberg scanning model was used in a secondary task paradigm with various performance and subjective measures recorded. Sternberg choice reaction time and subjective ratings of workload increased as a function of visual memory set size. It was concluded that short term memory may significantly contribute to operator workload. Therefore, strategies for the reduction of short term memory demands in crew station information display design should be investigated. Keywords: Mental ability, Performance(Human), Operators(Personnel), Recognition, Memory(Psychology)., Prepared in cooperation with Virginia Polytechnic Inst. and State Univ., Blacksburg, VA.

The purpose of this research was to predict target-by-target acquisition performance of air-to-ground imagery from microdensitometrically determined photometric and geometric characteristics of the scene. Results showed that it is feasible to predict the ground range at which a given target will be detected by an airborne observer. This prediction could be made totally automatically, given reconnaissance imagery, a microdensitometer, and a small computer. Seventeen characteristics of targets, backgrounds, and target/background relationships which reliably correlate to target acquisition performance were identified. Sixty regression equations combining these variables into linear predictive models of target acquisition performance were developed from one set of targets and mission conditions and cross-validated against targets contained in three different reconnaissance missions.

Computers and their corresponding interactive display and control devices were, at one time, available to a limited user population. With technological advancement, human computer interaction has become accessible to a larger user population through the use of electronic displays and their associated input devices. In the early 1960s, the use of the display plane as an interactive surface was realized by E. A. Johnson at the Royal Radar Establishment in Hurn, U.K. (Orr and Hopkins, 1968). The input device subsequently designed and developed is referred to here as a touch entry device (TED). TEDs take advantage of the natural mode of pointing, as do light or sonic pens, without the need to use a stylus or additional cumbersome wirings. Many technological approaches to touch entry have been developed and implemented since its first inception. However, the operational characteristics of each TED must be considered in the application of these devices to disparate types of tasks.

This report summarizes the literature on target acquisition with emphasis on target detection and collision avoidance in the air-to-air environment. A comprehensive, annotated bibliography is included. Major factors in air-to-air acquisition are discussed, pertinent data are presented, and potential improvements through training are listed. The report is concluded with a description of a research program designed to improve air-to-air detection performance. (Author), Prepared in cooperation with Virginia Polytechnic Inst. and State Univ., Blacksburg. Human Factors Lab., Rept. no. VPI-HFL-79-10.

A television image is formed by a series of parallel luminous lines called a raster. The visual prominence of the raster structure has been shown to interfere with the extraction of information from the image. However, the raster may be suppressed experimentally by a deflection process called spot wobble. Experiments were conducted to determine the effects of raster structure suppression on visual sine-wave contrast sensitivity thresholds, dynamic target acquisition performance using noise-free and noise-degraded imagery, and alphanumeric recognition performance using noisy and noise-free static displays. Results of the dynamic experiment indicate raster structure suppression and improvements in sine-wave threshold sensitivity are correlated and that a suppressed raster significantly improves target acquistion performance for noise-free conditions. Results of the static experiment were inconclusive, as no spot wobble effect was obtained, although display noise had a significant effect upon search time. Significant correlations with the modulation transfer functions area (MTFA) image quality metric were obtained, although the correlations were neither large nor consistent enough to be strongly advocated for detailed design evaluations.

This report contains the results of two experiments which investigated the effects of the quality of a televised image on eye movements and search-related dependent measures. The first experiment search task involved having subjects perform an air-to-ground search during a simulated flight. The quality of the image presented was varied by either passing, low-pass filtering, or attenuating the video signal and by adding electrical white noise to the video signal. The results of this experiment indicate that (1) at the highest level of electrical noise added, the percent of correct target acquisitions was decreased moderately, (2) the larger the target, the higher the percent correct responses, (3) the low-pass filtering of the video signal led to shorter ground ranges at acquisition for the large-sized targets, and (4) that the larger the target, the longer the fixation duration. Low to moderate correlations between modulation transfer function area (MTFA) and performance measures generally indicated that as MTFA increases performance improves, and that as MTFA increases fixation duration decreases. The search task of the second experiment consisted of having the subjects search for a designated letter or numeral across a televised picture of randomly positioned letters and numerals. The quality of the picture was varied by either passing, low-pass filtering, high-pass filtering, or attenuating the video signal and by adding electrical white noise to the video signal.

This report summarized the work performed during the first year, dealing with two tasks. The first task, entitled 'Human Engineering Survey and Analysis,' reviews the current flap panel display technologies, a summary of the human operator-critical characteristics of each technology, a technical summary of relevant human operator visual characteristics, and an evaluation of each technology by current visual display theoretical evaluation approches. The second task, entitled 'Hue/Luminance Contrast Tradeoffs,' involved laboratory research designed to develop a metric by which luminance contrast and chrominance contrast can be traded off in display design to provide a metric of total effective display contrast. Work to date on both these tasks is summarized in this report. (Author)

This report describes the results of two studies designed to measure dynamic noise, photometrically, on a raster-scan display. One study evaluated spot microphotometry of a single raster line; the other evaluated microdensitometric power spectral analysis of close-up photographs of raster lines. The spot microphotometry technique proved superior and very reliable, yielding results that are highly correlated with input noise spectra. (Author)

A new model for temporal and spatial encoding in the visual system is developed and presented. Three series of experiments were conducted to validate the model. The experiments used (1) a TV display of random, dynamic noise, and (2) a specially designed stimulus generator which produces very large homogeneous visual fields which can be easily modulated to reproduce a large variety of temporal waveforms having a rise time longer than 1 ms. The obtained results support the proposed model. The principal findings are: (1) Time integration of the eye is locally controlled at the retina and has very fast dynamics. (2) The obtained CFF curves suggest a correlation between the frequency at which maximum sensitivity is obtained and the sensitivity itself. (3) As predicted by the model, temporal bands are developed in the visual system for stimuli showing temporal discontinuity points. The width of the temporal bands was measured and a strong correlation was found between the temporal band width and the integration time. The width of the temporal bands is a function of the luminance level of the bands; the width is not dependent on the stimulus slope. The apparent brightness of the temporal band is, however, dependent on the slope of the stimulus.

This report summarizes a research program which has investigated the effects of numerous design parameters of alphanumeric dot-matrix displays upon operator performance. Among the parameters investigated experimentally are dot size, dot shape, dot contrast (or modulation), dot spacing, matrix size, character size, word context, ambient illuminance, character (dot) luminance, and character font. Operator performance in reading and search tasks was predicted by a linear regression model and subsequently cross-validated by additional experiments. Evaluations of several specific flat panel displays were made, with operator performance predicted acceptably from the previously derived regression model of display quality. Optimal capital alphanumeric fonts were determined for 5 x 7, 7 x 9, and 9 x 11 matrices. (Author)

This report presents the results of an air-to-ground television target acquisition experiment which investigated the effects of mission profile, video system line rate and bandwidth, and video noise level. The target acquisition performance data are related to these variables and to a measure of display image quality, the Modulation Transfer Function Area (MTFA), which is measured microphotometrically at the display surface. The target acquisition performance results are largely as expected. For a camera field of view of 18.8 x 14.2 deg, the mean ground ranges of correctly acquired targets are 28,661, 24,376, and 12,171 ft, respectively, for mission profiles of 23 deg depression angle, 500 ft/s velocity; 23 deg, 3000 ft/s; and 45 deg, 500 ft/s. As depression angle decreased, there was a large decrease in acquisition range; as velocity increased, there was a smaller decrease in acquisition range. Altitude was 10,000 ft.

A computerized analysis technique was developed for the evaluation of eye movements. This techique, when applied to eye movement data of 1 millisecond temporal resolution, is considered useful to evaluate such eye movement parameters as fixation duration, interfixation distance, and number of fixations per trial. This analysis technique was evaluated for displays having a single target and up to 192 nontargets in a static display. The results indicate that fixation duration is unaffected by the density of nontargets, but that the mean interfixation distance decreases linearly with increases in nontarget density. Due to this decrease in interfixation distance, the search time and the number of fixations per trial increase linearly with the density of nontargets. An overall evaluation of the eye movement measuring device is offered. (Author)

This report summarizes the research on user performance and display quality metrics pertinent to flat-panel displays. The report is divided into five sections. Section 1 is a brief introduction. Section 2 provides a description of each of the current flat-panel technologies and compares the technologies based on nine display parameters. Section 3 summaries research investigating the effects of various display variables on user performance. The research described in this section is divided into three types; alphanumeric research, cartographic/symbolic research, and literal research. Section 4 defines and discusses different quality metrics which have been used by researchers to predict user performance. One purpose of this literature review was to determine variables requiring further investigation. Based on the findings, an experimental plan is presented in Section 5. An annotated bibliography of references pertinent to these topics is included at the end of the report. (Author)

Television displays generate an image composed of a number of parallel raster lines. These lines, when visible, act as an interfering pattern and detract from operator performance in obtaining information from the video system. One way to reduce line visibility is to deflect the scanning spot vertically as it scans; this technique is commonly termed spot wobble. An experiment was conducted which evaluated changes in operator performance as indicated by the ranges at which targets were acquired and the number of correct responses to target presentation in a simulated air-to-ground search task. These performance parameters were evaluated at four spot wobble amplitudes and three viewing distances. The main findings were that spot wobble had no significant effect on the number of correct responses, but that large-amplitude spot wobble significantly increased the ranges at which targets were acquired. Additionally, several subjective indicators of preferred image quality were evaluated, and show that there is wide variance among subjects as to what image characteristics they prefer.

A study was undertaken to test and evaluate the human factors design aspects of an automobile moving-map navigation system. The primary objective of the study was to assess the driver attentional demand required by the navigation system during vehicle operation. A secondary objective of the study was to assess design specifics and determine whether or not the design was optimal in terms of efficiency of use in an automotive environment. Thirty-two driver-subjects drove a specially instrumented 1985 Cadillac Sedan de Ville on public roadways for this research. A cross-section of driver-subjects (both genders, ages 18 to 73, and driving experience from 2,000 to 40,000 miles per year) participated, and a cross·section of roadway types (residential, two-lane state route, and limited·access four-lane) and traffic conditions (light and moderate) were used as part of this research. The driver-subjects were asked to perform a variety of tasks while operating the research vehicle. These tasks included navigation tasks normally performed while using the navigation system, as well as a wide variety of conventional automotive tasks (e.g., tuning the radio or reading the speedometer) normally performed during vehicle operation. The purpose of asking the driver-subjects to perform a variety of conventional automotive tasks was so that direct comparisons in attentional demand could be made between tasks performed daily in an automotive environment and the navigation tasks. Twenty-one performance and behavioral measures were collected and analyzed for this research. These measures included eye—scanning and dwell-time measures, task-completion-time measures, and a variety of measures indicating driver performance and behavior. The data analyses for these measures focused on two major goals. First, the analyses determined which tasks (both navigator and conventional) required the highest attentional demand. Second, the analyses were used to determine groups of tasks which, for all pr

The purpose of this report is to present in orderly fashion the key factors in the design, analysis, and characterization of low-light-level devices. The parameters of interest are those that have the greatest effect on the transfer of information from the scene before the lens of the television camera to the human looking at the picture tube. Part I examines specifications for low-light-level devices. Part 2 discusses criteria for image quality. Part 3 examines the human visual process and optical aids to that process under conditions of low illumination. Part 4 discusses the image intensifier as a device to aid vision at low light levels and lists the parameters of a variety of available image intensifiers. Part 5 introduces television camera tubes and develops the concepts leading to display signal-to-noise ratio, Rosell's signal- to-noise ratio at the display. It is this, the authors believe, that is the most powerful means of evaluating 'resolution,' a term usually used loosely and incorrectly. Part 6 compares camera tubes on the basis of lag and display signal-to-noise ratio and discusses adverse factors such as 'burning' and 'blooming.' Part 7 sums up with some brief comments on important tube parameters and their specification., Prepared in cooperation with Westinghouse Defense and Space Center, Baltimore, Md. and Virginia Polytechnic Inst., Blacksburg.

Two modes of gun turret operation were compared in a study of facilitating tracking performance. Twenty subjects were used, arranged in two separate groups of ten subjects. One group was assigned to each mode of gun turret operation. One mode of turret operation was termed fully articulated; the manned turret moved both in azimuth and elevation in response to operator hand control movements. The second mode of turret operation was termed partially articulated; the manned turret moved in azimuth, but the sight alone moved in elevation. The results of the data analysis indicated that, over the 56 target runs each subject tracked, there were no significant differences in tracking performance between the two modes of turret operation. However, the interaction between mode and trials was et operation on tracking performance. Turrets which move in Azimuth and elevation vs. those which move in Azimuth only, using the sight for elevation. No significant differences in performance noted. (Study 2.)

The paper deals with aliasing, an important effect of one-and two- dimensional sampling on image quality. Aliasing changes the normal criteria of utility of the modulation transfer function (MTF) as a measure of system quality. Aliasing can be eliminated by letting the MTF fall to zero at one-half the sampling frequency. This, of course, markedly reduces the signal-to-noise ratio at the display (SNR(D)) near cutoff and thus reduces operator performance. Several factors control the amount of information an observer extracts from an image and the rate at which he extracts it. The paper presents the results of a series of experimental programs relating observer performance to the modulation transfer function area (MTFA) and to the SNR(D) and derives the relationship between MTFA and SNR(D). The paper discusses the tradeoffs as they are presently understood.

A laboratory simulation experiment was performed to determine the effect of the TV camera lens field of view upon air-to-ground target recognition by closed-circuit television. Measures of performance were probability of correct target recognition, range of correct recognition, and number of errors committed. It was found that, as the field of view decreased (1) probability of correct recognition decreased, (2) range of correct recognition increased, (3) incorrect target recognitions did not vary, and (4) number of no-response targets increased. The results were discussed in terms of their applicability to tactical airborne situations. (Author)

This item is from the William Crawford Gorgas Papers. It includes correspondence, articles, and other items that document the numerous awards and tributes Gorgas received during his life, collected by his widow, Marie Gorgas for her book, William Crawford Gorgas: his life and work., The digitization of this collection was funded by a gift from EBSCO Industries.

This item is from the William Crawford Gorgas Papers. It includes correspondence, articles, and other items that document the numerous awards and tributes Gorgas received during his life, collected by his widow, Marie Gorgas for her book, William Crawford Gorgas: his life and work., The digitization of this collection was funded by a gift from EBSCO Industries.