“A discriminability evaluation was performed on a proposed Space and Missile Warning symbol set. Our analysis focused on the discriminability of the symbols and the application of the information coding techniques. Inconsistent or inappropriate use of coding techniques can affect a user's interpretation of the symbol's intended meaning. Potential problems included the similarity individual symbols, use of alphanumeric markers and partially shaded symbols, and the lack of guidance on the minimum size of the symbols. After a lengthy review of previous research, we felt the literature could not provide adequate solutions. A two-part discriminability study was conducted to test the overall effects of the information coding techniques on discriminability, to identify individual symbols with low discriminability, and to determine an appropriate minimum size for these symbols. Search time was used as a measure of symbol discriminability. Size, shape, markers, and shading had significant effects on search time and errors. The experiments confirmed the suspected discriminability problems and modifications were made to the existing symbol set to create three new alternative symbol sets. Testing performed on these new symbol sets revealed that many of the problem areas from the original symbol set had been improved. Design guidelines and a new modified symbol set were proposed for review by the operational community.”
“Previous research has indicated that many persons have difficulty in describing the seriousness of hazards that are associated with some terms frequently used in warnings. Alternative explanations for this failure could be lack of understanding or simply inability to express their knowledge. Two studies were conducted in an effort to get more definitive information about what the general public knows about these terms. The obtained evidence suggested that many terms commonly used alone in warnings are not adequate to inform users of the extent of the hazards associated with those warnings. These results are discussed in terms of the need for completeness in warnings.”
“Existing research indicates warning labels are generally ineffective due to users ignoring them. One goal of the present experiment was to illustrate the importance that warning labels be as salient as possible. Features of salience examined in past research are size of the label, location, bold print, etc. The present study tests the effectiveness of warning label color and shape in terms of subject compliance, retention of label details, and perception of danger level. Results indicated that a written label surrounded by a shape resulted in higher compliance than a label with no surrounding shape. Color had significant effects only when used in conjunction with shape. A red label elicited a higher rating of potential danger, with green next, and black and white the lowest. A red octagon was significantly more effective than other combinations in terms of invoking a greater retention of label detail and also drawing higher ratings of perceived danger. A neutral shape elicited both lower subject compliance and fewer compliance points.”
“The effects of cost on warning compliance have been demonstrated in several previous studies. These studies have shown that cost reduction can dramatically increase compliance with a warning label's intent. The current paper describes two studies which support these previous findings under situations of household consumer product and recreational protective equipment use. In addition, these studies demonstrate that cost reduction can positively influence behavior under circumstances known to be detrimental to warning effectiveness. Such circumstances include low risk perception, familiar products, and inadequate warning labels/signs. These studies also show that increasing the cost associated with warning compliance, even a seemingly minor amount, can have devastating effects on compliance rates. These results indicate that the greatest effort possible should be taken to reduce compliance cost in hazardous situations when warnings are relied upon for hazard control.”
“Sneak Analysis was adapted for use in identifying human errors of commission. Flow diagrams were developed to guide the analyst through a series of questions aimed at locating sneak paths, sneak indications, sneak labels, and sneak timing. An illustration of the application of this methodology in a nuclear environment is given and a computerized tool to support Sneak Analysis is described.”
“A new method for modeling the influence of errors of commission is presented. This method extends the modeling done in HRA event trees to more accurately represent the population of potential human errors associated with errors of commission. The modified HRA event trees are called commission event trees (COMETs). The fundamental difference between COMETs and HRA event trees is that COMETs model errors of commission and deal with the problem of cascading errors often encountered when errors are either intentional or latent in nature. An illustration of the application of COMET is given for an error of intention in a nuclear control scenario.”
“Contemporary human reliability assessment techniques applied in the nuclear industry need to address more that simple errors of omission, errors of selection, and errors in task execution. The inadvertent selection of the wrong control from a bank of controls, skipping a step in a procedure by mistake, or the misreading of a display or indicator are representative of these classes of error. The distinction between simple errors and errors involving higher level cognitive processes has been acknowledged previously. As a next step, it is desirable to determine (1) the situations which promote the occurrence of low frequency high consequence decision based errors, (2) a classification scheme or framework for these errors, and (3) the expected failure rates for these errors once entry level conditions for their occurrence have been met. This presentation addresses 1 and 3 above, and presents a method for estimating errors associated with misunderstanding procedures, confusion regarding system response and inadequate problem solving. A function defining the relationship between composite performance shaping factors and human error probabilities is presented as well.”
“Advanced Driver Information Systems (ADIS) have the potential to assist drivers in choosing and reaching destinations by supplying information that is more complete and timely than is currently available. These rapidly developing information and communication technologies are also expected to bring about improvements in traffic network efficiency and safety. However, presenting new information to drivers raises a number of human factors concerns such as the potential for information overload and distraction from primary driving tasks. The challenge to driver/system interface designers is to maximize usability and learnability, minimize attentional demand and information density, and provide choices for individuals. The TravTek project is presented as an example of how human factors concerns can be addressed within the constraints of a particular ADIS system and of how human factors engineering and research can be integrated into the system design process.”
“This paper describes a 2-1/2 year project concerning human factors and future driver-information systems. The project goals are to (1) develop human factors guidelines, (2) devise test methods for safety and ease of use, and (3) develop a model that predicts driver performance when using these systems. A literature review has been completed and focus groups that identified driver needs have been conducted. In addition, a method was developed to select the most beneficial systems using accident reduction, congestion relief, and driver needs/wants as the criteria. Consequently, car phones, navigation, traffic information, vehicle monitoring, and hazard warning systems were selected for further study. Preliminary versions of each have been designed and laboratory tests are in progress. Usability tests in a driving simulator and on the road are scheduled.”
“This was a demonstration of a set of tools used to: 1) compare and evaluate software applications and prototypes; 2) evaluate documentation and instructional material; and 3) process video tape recordings of human-computer interaction (HCI). These tools include an event capture tool, which records events related to objects in graphical user interfaces, data filtering tools, which translate and aggregate user-generated events into meaningful characterizations of the interaction, and a multimedia data analyzer, which couples event logs and video recordings from HCI testing sessions.”
“An interactive learning environment was developed with the goal of empirically testing the effectiveness of various teaching strategies in improving problem solving performance. The domain chosen was transfusion medicine since it involves solving complex, multiple solution problems which are typically found to be difficult (Elstein, Shulman, and Sprafka, 1978) and because normal performance of this task calls for marking data sheets with intermediate conclusions, thereby improving the chances of the computer correctly inferring the student's reasoning. The testbed, called TMT (for Transfusion Medicine Tutor), monitors for errors, builds a model of what a student knows and can select teaching strategies based on human tutoring models that were developed from earlier studies. The testbed will be used to collect data of a student's performance in conditions where the degree of teaching and type of feedback are manipulated. A number of broadly applicable issues can be explored in this framework such as the difference between expert and student problem solving strategies, the effectiveness of different teaching strategies, and the importance of modeling student knowledge and providing visual feedback when developing an interactive learning environment. Preliminary results of our experiments, a demonstration of the testbed, and a discussion of how it was implemented will be presented in the demonstration session.”
“Although many automated tools support the productivity of professionals (engineers, managers, architects, secretaries, etc.), none specifically address the needs of the scientific researcher. The scientist's needs are complex and the primary activities are cognitive rather than physical. The individual scientist collects and manipulates large data sets, integrates, synthesizes, generates, and records information. The means to access and manipulate information are a critical determinant of the performance of the system as a whole. One hindrance in this process is the scientist's computer environment, which has changed little in the last two decades. Extensive time and effort is demanded from the scientist to learn to use the computer system. A common paradigm that meets the critical requirement of facilitating information access and retrieval by the chemist is demonstrated. This paradigm was embodied in EASI, a working prototype that increased the productivity of the individual scientific researcher.”
“This demonstration presents HSYS, a computerized methodology for analyzing human performance in complex operational settings. HSYS was developed in an attempt to better understand the interactional relationship between humans and operational systems (Hill, Harbour, Sullivan, and Hallbert, 1990) and to examine the many factors which influence Human-SYStem interactions. HSYS focuses on system interactions from the human's perspective and is built around a linear model of human performance, termed the Input-Action Model. The HSYS program automates much of the process of investigating human performance using Flowcharts based on the Input-Action Model. So far, HSYS has been designed and used primarily as an incident investigation tool. It is conceived, however, that HSYS might be useful as an analytical development tool where the Flowcharts would be used to examine whether all aspects that may influence human performance have been considered.”
“This paper documents the historical perspective of 'knowledge as design' as a natural human philosophy which necessarily resulted in user centered design praxis. A revival of this philosophy is called for in the presentation of the advanced knowledge and design acquisition methodology. The application of this methodology for overcoming bottleneck problems and alleviating brittleness in the Pilot's Associate is described and evaluated. The paper discusses three specific techniques designed to capture these perspectives: IDEF modeling, concept mapping, and design storyboarding. An integrative structure combining these techniques is proposed as an interactive way to let users, as well as other design team members, assimilate, progressively deepen, and combine knowledge for the purpose of developing intelligent systems and human-machine interface designs. Results indicate that pilots were able to successfully reveal their own comprehension of an air-to-ground mission and transform conceptual knowledge into actual designs for an intelligent pilot-vehicle interface.”
“What we know about the rationale for the requirements for the new design is limited by the imperfect understanding of the current system and our inability to divorce ourselves from that past. This provides the basis for a conundrum. How can we become sufficiently knowledgeable and avoid the biases inherent in the process of acquiring that knowledge simultaneously? Consider carefully the lessons of history and evaluate whether the legacy of the previous system and its predecessors accelerates or retards progress toward the future. To capitalize on our experience we need to recognize facts and mythology in our understanding of the way things are and need to be. The most important question to ask is "Why?". Without a true understanding of why, we can't develop effective, user-centered alternatives to the current system and our efforts will reduce to mere technology replacement.”
“Most studies of decision making in user interface design have been based on post-hoc interviews. To convey the realm of user interface design practice we conducted a longitudinal study of one designer in an organisation while designing an in-house database system. We applied the participant observation method. The observations revealed that the design took place in a highly turbulent organisational context, the working situation was extremely fragmented, and the information available on the users' tasks was incomplete and contradictory. Under these circumstances the designer adopted an ad-hoc design strategy. No specifications and plans were made. Instead, prototypes were developed aiming at getting feedback from users; however the users were much more concerned with organisational consequences of the system. In the prototypes, the user interface was literally designed from the upper left corner of each screen. Little explicit evaluation was made and drafts became promoted to the real system. Thus, the decisions became disjointed incrementals to the existing systems and work practise, i.e. the decision strategy was muddling through.”
“Incorporating the decision processes used by people in complex decision tasks is one of the most significant challenges facing designers of information systems. The need to capture and represent those decision processes is a fundamental part of the development of any information system and requires input from users. The purpose of this paper is to present an approach to the design of information systems that identifies the information needs of the user, reveals the reasoning process and decision strategies employed by users to make decisions, and represents those processes and information requirements in such a way as to enhance system development. The research combines user needs analysis with cognitive modeling to provide the basis for the design of a prototype information system. The paper shows how user needs analysis provides the knowledge and information necessary to cognitively model the decision processes of users in a particular problem domain.”
“This paper addresses some of the difficult and elusive problems associated with validating human performance simulation models. Simulation validity can be subclassified into input validity, structure validity and output validity. Of these, output validity is the most objective and also the most important, because it determines whether or not the purpose of the modelling effort can be met. In testing for output validity analysis of variance alone is not sufficient for validating human performance simulation models, as is often taken for granted by many researchers. A more systematic approach is proposed and implications discussed. The approach is based on considering analysis of variance in terms of the power of the test and a predetermined level of acceptable differences between model and reality.”
“A procedure is described for integrating data sets from several studies based on a sequential research strategy. A data bridging technique is used so that the results of three previous studies can be combined to build integrated second-order empirical models. The previous studies investigateed 10 independent variables, but 16 uninvestigated interactions from the complete set of items in a second-order model were not investigated in these studies. To select the best data points for examining these uninvestigated interactions, the maximization of |X'X| criterion was used. A computer program was developed to calculate the determinant value of X'X matrix for each candidate data point. Each selected data point was then analyzed to determine multicollinearity of the uninvestigated interactions. From these analyses, six additional data points were selected to examine the 16 uninvestigated interactions. Results from six additional data points along with the previous data sets were combined to build second-order empirical models using polynomial regression. The empirical models were then analyzed to choose the best models in terms of statistical properties such as variances of coefficients and prediction variances of the models. This procedure is discussed as a method for integrating data collected through sequential experimentation into an empirical model describing the functional relationships among independent variables. This approach increases the generalizability of data used in the design and evaluation of human factors interfaces which involve a large number of factors.”
“Different types of knowledge (experimental, empirical, and experiential) combine with assumptions, beliefs, models, and theories to determine what Human Factors (HF) researchers and practitioners study and do. In research, problems of epistemology and meaning arise because one must interpret the meaning of effects produced in a study. In design, epistemology underlies the relationship between behavioral design variables and their anticipated performance effects.”
“Though we are quite clever in developed ways to measure and express system performance, including human system performance, we frequently, stop short of understanding what our results mean. One reason for this is that we do not have cost effective and efficient ways to use the richness of data available in written or verbal information. Techniques of protocol and content analysis provide ways to use such data, but are usually too expensive and time consuming to use in applied settings. This paper explores the possibility of human factors specialists using commonly available personal computers and software in ways that allow practical analysis of verbal and written material as part of a T&E effort.”
“This paper: 1) describes the need for a human-factors taxonomy; 2) identifies existing taxonomies from the scientific, training, test and evaluation, and mission modeling disciplines; 3) lists the rules used in combining these taxonomies into a single, coherent taxonomy; 4) presents the taxonomy at its top levels; and 5) provides a source for obtaining a copy of the AIAA Human-Factors Taxonomy standard.”
“A framework is presented for classifying the cognitive skills exercised in deploying and managing resources in competitive situations. Behavioral task descriptions are partitioned into four cognitive levels: functions, procedures, skills, and abilities. A cognitive procedure is defined as a goal-directed, semi-structured, flexible sequence of actions in an emergent situation, adapted to deviations from nominal values for local conditions, terrain, environment, and adversarial actions. A procedure is represented as Cognitive Procedure Script (CPS), a sequence of actions required to achieve an operational goal. Actions in a CPS must be reduced to detectable response/event sequences for carrying out the actions: Implementing Steps (IS). To infer a candidate CPS being executed a computer maintains a temporal record of IS and searches for patterns that correspond to an action in a CPS. The CPS is represented also as a sequence of states and subgoals instrumental to a final goal. Each state is defined by situational conditions, subgoal, state maintenance behavior, action to transition to the next state, and the IS for executing the action. Implications for task analysis and measures of task complexity are discussed. A long range strategy is presented for developing and validating a Cognitive Skills Taxonomy to support decisions about manpower, personnel, and training during system design.”
“According to a within-subjects design, 83 naval pilots and flight officers were administered computer-based and paper-based tests to assess recognition of aircraft silhouettes in order to determine the relative reliabilities and validities of these two measurement modes. Estimates of internal consistencies, equivalences, and discriminative validities were computed for multiple performance measures. It was established that the relative reliabilities and validities derived for these two assessment schemes were contingent upon the employed multivariate measurement criteria, i.e., percentage correct responses, average response latency, and average degree of confidence in recognition judgments, as well as the statistical criteria used to ascertain the comparative quality of these two modes of testing.”
“The rationale for a discipline of test and evaluation (T&E) is that systems are developed to achieve certain purposes. This is true whether the systems in question are purely hardware, software, training, socio-political, or some combination of these or still other system types. The major objectives of test and evaluation are to determine whether the systems achieve their purposes and, if they do not, to determine how they can be improved. The human factors results of this effort must be both usable and useful if the T&E practitioner is to have credibility. The usability of T&E results is a function of the adequacy of the methods used for collecting and analyzing data in the context of real-world constraints, a world in which there may be frequent departures from an ideal test and evaluation environment. The measurement problems inevitably encountered include those associated with the disparity between the engineering and behavioral science disciplines, the complexities of experimental control in operational situations and operational fidelity in non-operational settings, and the absence of direct translations between human performance and the performance of the larger system of which the human is a part. The usefulness of T&E results is a function of how relevant they are to the decision makers who are managing the system development program, as well as those other individuals who are designing, developing, operating, maintaining, or using the outputs of the system. Unlike the human factors specialist, the T&E "customer" is generally not concerned with a search for the truth. Rather, the customer frequently needs only that information necessary to answer very specific questions. We will fail in our responsibility as human factors professionals if we do not provide the type of information, in the proper format, that will influence the decisions and behavior of the user of that information. Both the usability and the usefulness of human factors test and evaluation efforts must be properly addressed by T&E researchers and practitioners. This is true if for no other reason than for the sheer pragmatic need to get continuing support and funding for our discipline. The first two papers in this symposium will emphasize the usability of results in the real-world context of the T&E environment and the last two will emphasize the usefulness of the results to the sponsor or user. The participants have had multiple opportunities to reconcile and more clearly address differences in the positions they espouse through exchanges and critiques of their respective papers.”
“Using "robust" testing and evaluation (T&E) components will ensure system T&E success despite unanticipated resource and schedule shocks. Selected tools for building robust T&E plans, methods, and analysis components are described. These robust tools provide a beginning framework that the practitioner can build upon to achieve successful T&Es.”
“A system development process, called the User Oriented Evaluation (UOE) process, and an evaluation tool were created to place greater emphasis on user needs during computer system development. The UOE process is an iterative method for design that emphasizes the role of the user as the initiator of system requirements; evolutionary design requirement definition by enabling users and developers to experiment through the use of prototypes at all phases of system design; and appropriate utilization of developer and user areas of expertise. The evaluation tool is an integral part of the UOE process and provides the ability to solicit on-line meaningful feedback from users in real-time, and a means to capture a user's on-going experience with the computer system. The paper contains a description of the UOE process and the evaluation tool, the capabilities of each and the history of their development.”
“Approximately 9 to 10 billion calls are handled by telephone operators of the Bell Operating Companies throughout the U.S.A. on a yearly basis. Bell Communications Research has an Operator Services Ergonomics district that employs a multifaceted approach to improving operator efficiency and comfort, and providing guidance on both the operator and customer human interfaces for new services. The approaches taken represent a recognition of several key factors that dictate and/or influence the type of test and evaluation conducted and how the results from such tests are conveyed to decision makers.”
“There are many tasks hazardous to human life which can be accomplished remotely through telerobotic control. Robot technology has advanced to the stage where teleoperated manipulators are versatile and effective enough to be used successfully in a wide variety of circumstances. As telerobotic systems become more sophisticated, it is important to ensure that the human-machine interface is adequate for the task. One very important type of feedback information that is missing from standard telerobotic control stations is the immediate and compelling binocular coding of depth, which is thwarted through the use of standard monoscopic ("2D") video systems, making the operator dependent on other less salient visual depth cues. This is unfortunate, since most telemanipulation tasks require operators to have a good sense of the relative locations of objects in the remote world. To that end, a practical Stereoscopic Video (SV) system was developed that is compatible with standard video display and recording equipments. An experiment was conducted to examine the potential benefits of SV for teleoperation. The results showed that SV can aid teleoperation by reducing task execution times, reducing error rates, and reducing the time needed for training.”
“The present investigation examined the effects of training context on the development of a skill in a semantic category visual search task. Thirty-two subjects were trained in a visual search task which allowed the separate examination of feature learning and attention strengthening. Subjects were trained in a Consistently Mapped (CM) condition which allowed both feature learning and attention strengthening, "Attenuated Strength" search conditions which allowed only feature learning, and a Variably Mapped (VM) condition which allowed neither feature learning or attention strengthening. The present experiment also examined the temporal characteristics of feature learning by manipulating the training context which was defined as the number of trials in a row that a particular search condition appeared. There were four different training contexts used (1, 5, 10, and 50) and eight subjects were assigned to each. Each subject performed 11,000 training trials. Analysis of the reaction time (RT) data indicated that within each training context, RT was fastest for the CM condition, intermediate for the Attenuated Strength conditions, and slowest for the VM condition. The results also suggest that 10 trials were sufficient for feature learning to occur. Furthermore, while there was evidence of attention strengthening in the CM condition there was no evidence of attention strengthening in the Attenuated Strength conditions. For inconsistent tasks that allow feature learning, the rate at which the context changes is a critical variable in determining final task performance.”
“This paper describes the Language Technology Project at the University of Central Florida. The Goal of this project is to develop, evaluate, and commercially produce language courseware and training techniques for personal computers (PCs), including IBM and Macintosh, equipped with voice interfaces. The primary emphasis of the Language Technology Projects has been in the development of demonstration courseware. The second area is the evaluation of courseware, including research regarding effective computerized instructional strategies. Informal field-testing of one package in an Orange Country (Florida) elementary school suggests that the style of instruction is readily compatible with students as young as the first garde. Several studies have also provided insight into instruction design.”
“If skill learning is evaluated in terms of performance on retention or transfer tests, recent evidence has shown that systematically reducing the frequency of augmented information feedback in training facilitates learning, as compared to practice with augmented feedback on every trial. However, in past experiments feedback frequency and scheduling have been confounded. In the present experiments, we separated these variables, showing that both reduced feedback frequency, and the particular scheduling of feedback presentations, are important variables for maximizing learning.”
“Investigations of students and practicing medical technologists indicate that both groups make significant numbers of errors tasks such as antibody identification. One potential solution to help with this problem is to provide access to a computerized learning environment in which users can get exposure to a larger and much broader set of problems than would otherwise be possible. This paper describes such a learning environment, the Transfusion Medicine Tutor, and discusses the ways in which it supports guided discovery learning.”
“This paper describes the process by which an AI knowledge engineering technique was used for performing knowledge acquisition within the context of Instructional System Design. The method is a type of cognitive task analysis in which conceptual graphs are developed using several complementary knowledge acquisition methods. The graphs can be used to explicitly represent a variety of knowledge types including semantic, rule, and implicit knowledge. We outline the process by which we applied the method to represent expert knowledge for instructional design in the domain of engineering mechanics, and describe modifications and lessons learned from the application.”
“The effects of physical image quality on reading and on perceived image quality from CRT and hard-copy (photograph) displays were studied. The results showed that as the image quality of a display increased, indicated by an increase in the value of the modulation transfer function area (MTFA), the reading speed and subjective image quality ratings increased. This change in reading speed and perceived image quality occurred similarly for both hard-copy and soft-copy conditions. If the image qualities of the displayed text are similar, hard-copy and soft-copy displays will yield equivalent reading speeds.”
“The forgoing experiment was designed to observe and measure users' perception of flicker, also referred to as Critical Fusion Frequency (CFF) at a variety of controlled ambient light levels in a setting that was designed to approximate a range of realistic user environments. The goal of the study was to determine the phosphor type and refresh rate required to eliminate flicker perception for the majority of our sample of forty-eight (48) participants.”
“Spatial direction errors during interpretation of perspective images, such as 3D map displays, may originate from misjudgment of the orientation of the viewing direction used to make the display. One source of these errors could be perceptual evidence of the display surface. Two experiments are reported in which the same judgement exocentric task was presented, but the cues to the picture surface were reduced or eliminated by presenting the task as a stereoscopic, virtual image or by a geometrically matched physical model. A theory developed to model exocentric direction errors on perspective displays has been fitted to the data from these two experiments. The parameters estimated from the fit in both experiments indicate that the subjects may be more correctly estimating the viewing direction than in ordinary perspective displays. Consequently, in some real world or stereo viewing conditions, errors in estimating the viewing direction are not likely to dominate exocentric direction errors.”
“A critical question in simulation is the degree of correspondence or alignment between what is seen in a generated graphics world and it's intended actual physical structure. An allied concern, is the role of perceptual experience as a potential influence on this perceived spatial layout. The present experiment examined the potential for these perceived distortions. Eight observers estimated the location of nine object locations embedded in a graphical scene of a traffic intersection. Participants were asked to judge the relative and absolute distances between objects in the display. Half the subjects viewed the intersection after driving for thirty minutes through the simulated traffic world while the remaining participants did not receive any prior experience. A multidimensional scaling analysis revealed differences between simulator and no simulator experience groups for absolute and relative 3-D solutions. Scaled representations of actual and perceived spaces are discussed in the framework of a model of perceived spatial layout and with respect to human factors issues in the use of both simulated and virtual worlds.”
“The present research investigated the effectiveness of rotational and stereoscopic display techniques as applied to practical applications in computer graphics. In the described experiments, depth perception was evaluated by examining accuracy in a 3D path-tracing task, with stimulus displays resembling the structure of cerebral angiograms. In the first experiment, task performance was determined without 3D enhancements, with either technique used alone and with both techniques combined. The results indicated that performance improved using either technique, however, performance with rotational displays was superior to stereoscopic displays, and performance was best when both techniques were combined. The results of a second experiment revealed that rotational displays were no better than stereoscopic displays enhanced with multiple static viewing angles. The experimental findings are discussed in terms of the visual information available using either 3D display technique as well as with respect to the hardware requirements of both display systems. The results are also related to the weighted additive model of depth perception.”
“This paper describes a number of aspects of a display technology under development, which involves the integration of stereoscopic computer graphics and stereoscopic video displays. The background and justification for this development are discussed, based on the need of operators of remotely controlled vehicles and/or manipulators to estimate absolute sizes and locations of objects at a remotely viewed site. The basic technology involves superimposing an interactively controllable computer-generated stereographic cursor onto a stereoscopically viewed video image. Absolute measurements can be made with this system, based on relative comparison of cursor position with target object location. Experimental results are presented in which the ability of subjects to perform such tasks was investigated. In general, results were promising; subjects were able to align virtual pointers with real targets essentially as well as they were able to manipulate real objects. A number of implications of this technology for the enhancement of three dimensional video displays are discussed.”
“The development of colour Head-Up Display (HUD) technology has encouraged enquiry into the benefit of colour coding conventional HUD formats. Colour could be used to warn crew of danger, highlight changes in onboard systems or assist in target acquisition and combat. Specifically for this experiment colour encoded the accuracy of a subject's performance, redundant colour codes indicated the success with which a subject was maintaining a requested flight profile. Contemporary research has often found a subjective preference for colour coded displays, but little evidence of an objective advantage. The results of this experiment followed such a trend. Colour coding failed to provide any advantage in performance terms, yet it significantly reduced subjective workload, as measured by the NASA TLX (p<0.05). The initial indications are that colour coding was not optimised in this task and other techniques, e.g. recovery from unusual attitudes, and applications, e.g. target information, might be more appropriate.”
“This study examined the relative effects of color, shape, and color+shape codes on time to recognize whether or not targets were updated in sequentially-presented diagrams. These relationships were examined with different retention-intervals (RIs) and tasks, with and without updates. Response-time was faster with color+shape or color than shape, across tasks and across RIs, whether or not targets were updated. However, the benefit of both color conditions was greater when targets were updated and subjects performed a distractor task during the RI; i.e., color showed a greater resistance to interference effects.”
“This study examined the effects of display luminance on the ability of human observers to recognize color symbols displayed against similar color backgrounds. The Signal Detection paradigm was utilized and subject sensitivity, as measured by d', was the primary measure of interest. The symbol colors were red, green, and blue. Background colors were .01 to .07 1976 CIE/UCS units distant from the symbol color. Luminance levels ranged from 11.85 cd/m{squared} to 127.25 cd/m{squared}. The symbols were presented on a cathode ray tube (CRT) under ambient lighting of two lux. Display luminance was found to affect subject sensitivity, d', as a function of symbol-background color combination. The results imply that display luminance for the presentation of blue symbology on bluish backgrounds is optimal at 19 cd/m{squared}. For the red and green symbol-background conditions, display luminance between 56 and 93 cd/m{squared} yields the best performance.”
“Although a large amount of research has been performed to develop principles and guidelines for the effective use of color in display coding, very little of it has dealt with the color coding of continuously varying data -- also known as pseudo-coloring. This research is investigating the application of pseudo-coloring to the display of data from sonar systems, in order to enhance target detection and classification performance. A scheme has been developed for adaptable pseudo-coloring of sonogram displays, called an Adaptable Acoustic Color Coding Interface System (AACCIS), in which the sensor operator adjusts one or two dimensions of control of the color coding scheme in order to optimize performance in operational detection and classification tasks. Two experiments were performed to: (1) select specific adjustment parameters for user control, and (2) compare task performance between a few alternative adjustable and non-adjustable color codes.”
“This paper describes a study which employed eye tracking to investigate visually searching lists. An experiment was designed to test if "cognitive style" influences search strategies. We examined the search patterns of subjects as they viewed lists of sorted and randomly ordered columns. We found that cognitive style was a good predictor of the scanning strategies of ordered lists. For the unordered lists, efficient, systematic searching was performed by college students but not by high school students.”
“Recently Navon (1984) suggests that much of the dual task data that lend support to resource theories are methodologically flawed and that there are non-resource related mechanisms such as outcome conflict that could explain dual task data equally well. If Navon's concerns can be validated, the current view of workload as a resource-related concept would have to be revamped. The present research examined performance tradeoff -- demonstration of resource allocation -- with the optimum-maximum method. The optimum-maximum method was a variation of the secondary task technique proposed by Navon to encourage maximum joint task performance without conveying to the subjects that task performances must tradeoff. A continuous tracking task and a Sternberg memory task were used; three priority levels were used to induce resource allocation. An absence of performance tradeoff with the optimum-maximum method would support Navon's claim that performance tradeoffs may not be true indications of resource competition or task demand and therefore not reliable reflection of mental workload. Performance tradeoff was observed when the tracking task was optimized. When the Sternberg task performance was optimized, Sternberg performance was sensitive to the priority manipulations but the concurrent tracking performance was not. Peripheral vision was proposed to have protected the tracking performance when subjects were visually focusing on the Sternberg stimuli. However, the peripheral vision hypothesis could not account for the graded performance changes across priority levels. The present data suggest that the resource view is still a useful concept and the secondary task technique still could provide informative disclosure about task demand and mental workload.”
“The present experiment examines the effects of on-line feedback in the optimum-maximum procedure proposed by Navon (1985). Twenty Clemson University students performed a dual-task consisting of two compensatory tracking tasks. Subjects performed three different dual-task combinations in which the tracking dynamics of the two tasks was varied. Each tracking task was optimized at three levels. One group received on-line feedback on both tasks, the other group did not. Results indicated modest performance trade-offs between tracking tasks. The performance trade-offs appeared to be stronger when on-line feedback was used. These data demonstrate the effects of on-line feedback on dual-task performance and suggest that conclusion concerning resource trade-offs may depend on whether performance feedback is provided.”
“Electro-optical imaging systems have been integrated into rotorcraft operations, allowing pilots to fly at very low altitudes and avoid obstacles in reduced visibility. The hardware characteristics of these systems result in visual displays which differ significantly from unaided, daylight vision. The impact of these differences on perceptual performance (and, ultimately, on pilotage) is poorly understood. In this paper, we identify critical human factors concerns suggested by field data and review empirical studies of performance on flight-relevant perceptual tasks, notably depth and distance perception. Hardware modifications to improve man-system performance are suggested.”
“Development of a computer-based system for generating simulated infrared (IR) imagery is described. The system provides realistic representations of IR targets and backgrounds for training soldiers in combat vehicle identification (CVI). Simulations of several fielded IR sensors enables system users to generate training imagery sets, both snapshots and animated sequences, showing realistic sensor effects. The system is workstation-based and has a user interface that permits a non-expert to generate desired imagery sets from menus of available models and scenarios.”
“The U.S. Army's CECOM Center for Night Vision and Electro-Optics (C2NVEO) has been conducting a series of human perception tests to complement results of field tests. The perception tests score military observers on their ability to identify and recognize simulated thermal images of combat vehicles. The results of these tests are used to develop and refine the Night Vision FLIR Performance Model (FLIR90/ACQUIRE) which predicts range performance for human target acquisition using different sensor designs. During the analysis of an early group of tests it became apparent that observers needed to be trained to equal levels of competence before taking the tests. For this purpose, we created a software package designed to teach observers about thermal signatures and to help the observer become more familiar with the target set used in our tests. In the following months we trained and tested dozens of soldiers and civilians with various backgrounds. These training and testing sessions provided a large database of observations for analysis. The paper includes discussions of the simulation techniques used to produce the test and training imagery and a description of the software package used for training. Conclusions are discussed in light of the patterns of learning achieved by the training and some implications for the human perception of tactical targets.”
“An important issue in implementing Rapid Communication (RAPCOM) display technology is the manner in which information is coded within a given display sequence. This study compared performance on a search task for four single coding strategies as well as four redundant code combinations at five presentation rates. Results indicated significant differences in accuracy and reaction time for task performance using single and redundant codes at various frame durations. These findings helped to identify several potential coding formats that can be utilized in real-world settings.”
“This study investigated the effects of statistical training and various perceptual characteristics of scatterplot displays on intuitive estimates of correlations. University professors and first year undergraduate students estimated correlations from scatterplots (1) with three different levels of correlation in the data, (2) with or without a regression line, and (3) with three different types of dispersion of the data point clouds. The faculty generally used higher and a wider range of values, but both groups perceived a higher correlation when a regression line was present and both groups were equally influenced by the different types of dispersion of the point cloud. These findings indicate that the estimation of correlations from scatterplots is a perceptually based process, which is largely independent from formal statistical training.”
“This experiment investigated effects of response symmetry in a two-handed aiming task. A first-order system with isometric controllers was used to present the task. The Motor Program hypothesis predicts better performance in a symmetric condition because only one set of parameters is required to control both hands. The Divided Attention hypothesis predicts better performance in an asymmetric condition because attention can be more readily switched between hands. Results supported the Divided Attention hypothesis because (1) accuracy was better for the asymmetric condition, and (2) the relative advantage of the asymmetric condition increased as targets became more difficult.”
“An experiment was conducted to assess the relative influence of several factors on performance with orthogonal stimulus-response arrays. Subjects responded to the onset of one of three aligned light circles with a press of one of three aligned response keys. The response array was aligned parallel, angled, or orthogonal to the stimulus array. The results indicated that performance with orthogonal arrays is worse than with parallel or angled S-R arrays. For the orthogonal arrangements, the results also indicate that each hand prefers a mapping directly opposite to the other hand, and that this mapping reverses when the orientation of the stimulus and response arrays are transposed. In addition, the results also revealed that the relative costs of orthogonal S-R arrangements are somewhat attenuated when the assigned mapping associates (i.e, colocates) a given display with its closest control.”
“The principles of stimulus-response compatibility were used to provide a performance evaluation of an experimental notation for the keyboard in which pitch varies horizontally in visual space. Subjects performed a choice reaction time task using either the horizontal notation or a more traditional vertical notation. Half the subjects in each notation group were provided white noise and half pitch-varying feedback with responses. Responses were reliably faster with the horizontal display. Further, the superiority of the horizontal display was not dependent on the nature of the feedback. In a second experiment, performance of musicians and nonmusicians was compared using the horizontal and vertical notations. Performance by musicians was equivalent with horizontal and vertical displays and uniformly better than nonmusicians. The findings fail to support the argument that horizontal visual representations will be less useful in musical tasks because they will be incompatible with the mental representation of pitch as varying vertically from 'low' to 'high'.”
“This study examined biases, sources of difficulty and display support in strategic planning. Eight subjects performed a strategic planning "rescue" video game, which required them to make a series of choices regarding which node to "fly to" in order to rescue simulated casualties. After making each choice, subjects needed to fly a challenging tracking dynamics along a path to reach the next "node" in the decision space where the casualties were "rescued." The dynamics along each path could be at one of four levels of difficulty. The difficulty determined the probability that the corridor would be flown successfully and therefore casualties rescued at the other end. To maximize their score, subjects had to consider the number of casualties at each node, the length of the path to the node (the shorter the better), and the probability they would fly the path successfully (an estimate of their own performance based on past experience). Periodically subjects were asked to give explicit estimates of those probabilities, such data provided in order for us to evaluate the calibration of estimated with true probability of success. Half the flights were flown with restricted preview of only the two nodes of the immediate choice. The other half offered full preview of the whole map. The results revealed that (a) subjects' decisions were less optimal when full preview was offered, (b) this deficiency appeared to result because full preview led subjects to rely too much on the simple strategy of choosing routes with most casualties, and neglecting the use of more abstract probability values in guiding their choices, (c) subjects appeared to be well calibrated in their confidence of traversing paths correctly.”
“The human factors benefits of the Head Up Display (HUD automobile applications are improved visibility, and prevention of view-obstruction when collecting driving information. We evaluated several HUD's considering such factors as location (in-plane location and distance), character size, brightness and color through investigations of view-point distribution, measurement of recognition time and subjective evaluation. The results made clear an appropriate range for each factor. It was also confirmed that the HUD recognition time is significantly shorter than that of the conventional instrument panel display by more than 0.1 second.”
“Research into the format of visual displays of multidimensional data has revealed relative superiority of certain formats in relation to the nature of task demands such as fault diagnosis versus fault detection. Previous research also suggests that order of training or experience in working with alternate display formats might also influence the ability of operators to gain the full benefits of these display formats. This last factor comes into play during the redesign of existing displays and the design of multiple display environments in domains such as nuclear process control, tactical and commercial flight, and anesthesia monitoring. The experiment described in this paper sought to further examine and clarify the types of performance decrements that can emerge after switching visual display formats used to support a complex decision-making task (Purcell and Coury, 1988). Accuracy and response time performance demonstrated that transfer of training was asymmetric after the switch. These results indicate that the order in which operators learn to use a given format of visually displayed data can enhance or hinder their subsequent use of alternate display formats. Redesign efforts and/or the use of multiple display formats in complex decision-making settings must address these effects.”
“When completing tasks in complex, dynamic domains individuals must consider both high-level issues (e.g., relationships among several variables) and low-level data (e.g., the values of individual variables). An important issue in display design is to determine those graphic forms that allow the efficient extraction of information at both levels. One display that has the potential to achieve these dual design goals is the configural (object) display. Research on configural displays has indicated that this type of display will facilitate the extraction of information about high-level issues if the emergent features directly correspond to the critical data relationships that exist in a domain. On the other hand, designing configural display to offset the cost that is usually associated with the extraction of low-level data has proven to be more difficult. One potential design strategy to accomplish this is to increase the perceptual salience of the lower-level display elements and color coding is one technique to achieve this. Performance for color-coded configural and separate displays was compared in two empirical investigations. For the extraction of information regarding high-level issues the configural display significantly increased accuracy with no cost in latency. For the extraction of low-level data there were no differences between the configural and separate display for accuracy, but there was a significant decrement in latency associated with the configural display. However, the results indicate that this decrement in performance dissipates with experience, and under certain conditions was not significant.”
“We have carried out three experiments which explore the effectiveness of various visual depictions of tactile data for the task in which subjects view a robot gripping an object and attempt to predict if the robot will be able to pick the object up. The display types tested range from simple total closing force values through to graphic three-dimensional contour plots, overlaid with force-torque vectors, with the subject having dynamic viewpoint control over the display. Our results suggest that an operator's utilization of such tactile displays depends on the effectiveness of the model of the gripping configuration which can be obtained through visual images of the scene. In the case of simple planar contact between gripper and object, a total force value produced maximum performance enhancement. In more complicated, or awkward non-planar gripping, the total force value was of no use, but spatially distributed force representations could still yield enhanced performance. Further experimentation was carried out with complex displays which subjects reported as depicting the contact forces better. It was found that these displays actually reduced performance, drawing into question the telepresence assumption that more realistic information will lead to better telerobotic operator performance.”
“Automation of previously human-controlled systems is accelerating. Although various proposals have been made regarding the allocations of functions among human and machine components of systems, the human factors discipline is not keeping pace with automation. Using examples primarily from aviation, a number of questions are raised regarding the relative roles of people and computers.”
“Based on modifications of Fitts' Law suggested in the literature, 121 unique formulas were tested against reciprocal tapping data from 1,318 subjects (1,047 males and 271 females) who participated in the Baltimore Longitudinal Study of Aging from 1960 to 1981 in order to determine the best formula (based on the standard error of estimate) and to examine age and sex differences using this formula. The best formula for males differed from that found for females, resulting in a set of new formulas which take into consideration age and sex and which fit the experimental data better than past formulations. While females were faster than males and young were faster than the old, a substantial portion of age and sex differences might be explained by a speed-accuracy tradeoff.”
“We evaluated several human factors issues involved in applying AT&T Text-to-Speech (TTS) technology to Dual Party Relay (DPR) services for hearing- or speech-impaired end-users. Covered in this paper are: 1) lexical studies that were conducted to gather abbreviations and typographical errors generated by users of the Telecommunications Device for the Deaf (TDD); and 2) a laboratory trial that was performed to evaluate the feasibility of the TTS application and to examine associated user interface issues. Information gathered from the lexical studies was used for the laboratory trial and for constructing the field trial system. The laboratory trial consisted of several studies, including a comparison between the comprehension of TTS and human voices, identification of a desirable TTS presentation rate, examination of the effects of pauses, and observation of the difficulties related to TTS relay calls. Recommendations were made based on the lab trial results. We concluded that applying TTS to DPR service is feasible. That is, hearing and hearing-impaired subjects can carry on DPR conversations with the use of TTS technology. A field trial was subsequently conducted in California.”
“Mobile and wireless telecommunication systems are expanding rapidly, and many new systems are in development. To address the impact on speech communication, this paper describes a new conversation-based test of transmission quality designed to be sensitive to degradations common in mobile systems. An experiment using this conversation test added roundtrip delays of 0, 600 and 1200 ms to the transmission path. Three types of dependent variables were collected: subjective ratings of quality and listening effort, measures of conversational dynamics, and scores of task performance. Although neither the subjective ratings or task performance varied across delay conditions, there was a significant effect on one measure of conversational dynamics, the number of interruptions. Arguments are presented that changes in conversational events may be superior to traditional subjective measures for evaluating mobile systems.”
“What makes a customer manual look attractive and easy to use? How do typeface, margins, tabs, color, style of headings, and so on, contribute to this perception? We believe that a manual that looks hard to use may discourage users from even trying to use it. Why struggle to make a manual accurate, well-written, and complete if users never turn to it in the first place? In the study reported here, eighteen subjects ranked six different customer manuals on nineteen dimensions, for example, use of color and overall attractiveness. The manuals were actual production manuals that were written to be used by full-time administrators of complicated electronic equipment. Analyses of the data clearly show that those features that help the user to find information (i.e., tabs, headings, and an impression of good organization) are important in determining judgments of perceived ease of use.”
“The results of four sequential experiments were combined into integrated empirical models using data bridging. The resulting regression models can be viewed as quantitative design guidelines for telephone-based information systems. A total of ten independent variables involving environmental, hardware, dialogue, and user factors were considered across the four sequential studies that evaluated a telephone-based interface. Three dependent variables including total search time, user added keypresses, and message transcription accuracy were evaluated in each separate study. Polynomial regression was used to generate an integrated second-order empirical model for each of the three dependent variables. The major contributors to total search time were the time delay between the presentation of each menu item (input timeout) and the structure of the menus. Age of the user and menu structure were the primary contributors to user added keypresses required to recover from errors. Overall, the accuracy of message transcription was influenced primarily by background music, presentation rate of the synthesized speech and the age of the user. Total search time in this sequential type of information delivery system is primarily dialogue driven. Critical dialogue parameters for this system were input timeout and the number of alternatives in menus. Due to the small number of errors in searching for information, the need to minimize user added keypresses for error correction was not of primary design importance. Accuracy of speech message transcriptions was primarily dependent upon the acoustical environment of the listener. Overall, the use of integrated empirical models offers several advantages including a basis for generalization across several studies and the ability to conduct specific interface design tradeoffs.”
“The Intelligent Interfaces Group at NYNEX Science and Technology has evaluated numerous phone based interfaces (PBIs) during the course of iterative design. Many of the PBIs studied allow users to truncate spoken prompts by pressing keys on their touch-tone telephones. We have found that mistaken assumptions about how and when users will truncate spoken prompts may lead to large discrepancies between the expectations of system designers and the behavior of users. In order to study truncation behavior, we created the Task Oriented Taxonomy of Truncation (TOTT). This taxonomy can be used to describe the behavior of users in truncating spoken prompts in PBIs. TOTT was found to facilitate our understanding of users' truncation behavior and allowed us to change the PBI prompts to better fit this behavior. We found that many users did not interrupt the spoken prompts and we speculate that they may be using a model of turn-taking from human conversation. Future areas of research and applications of TOTT are discussed.”
“Phone-based interfaces allow users easy access to information stored in distant computers. Applications include gaining credit card information, information about financial aid and registration at universities, and banking services. One interface tool that is often used in these applications is the menu. Unfortunately, the use of menus in these types of interfaces is not well investigated. The current study was developed to investigate issues related to the use of menus in auditory interfaces using computer synthesized speech as the output medium. Variables of interest included: 1) menu-page configuration (organized versus random); 2) menu-hierarchy configuration (top-down versus bottom-up); 3) individual differences in short-term memory capabilities; and 4) user anxiety. Subjects performed 20 menu selection trials from one of four menu arrangements. Subjects were given a digit-span task to assess short-term memory capacity and a Sternberg task to assess speed of recall from short-term memory. Finally, subjects were given the State-Trait Anxiety Inventory to assess subject trait and state anxiety. Results suggest that menu performance is a more complex process than VDT menu performance. Short-term memory abilities contributed a major portion of the variance in menu performance. User interest in the menu task also accounted for a good deal of the variance in menu performance and interest was found to be correlated with user anxiety. These results and the results of the menu configuration variables are discussed in terms of possible implications for phone-based menu use and design.”
“The primary goal of this research was to examine the relative effectiveness of help messages presented aurally versus those presented visually. Additional goals were to evaluate help messages baselined for a next-generation product and to identify improvement opportunities for these messages. Three formats were evaluated: paper instructions (user card), visual help messages presented on a computer-simulated telephone's display screen, and auditory help messages. Thirty-six subjects performed two different tasks using a computer-based on-screen simulation of a telephone. Dependent measures consisted of time to complete the tasks, error rate, and subjective measures of how easy the tasks were and how well subjects liked the instructions they received. An analysis of variance indicated that tasks performed using auditory help messages had a significantly lower error rate than did those tasks performed using visual instructions. No significant differences were found among the formats for task completion time or task ease, but there was a significantly greater preference for instructions that included display help messages accompanied by a user card.”
“The Human Factors community has virtually ignored one of the most important components of the extended user interface, user documentation. Last year at the 34th Annual Meeting in Orlando there was only one paper on either the design or use of documentation. By comparison, there were approximately a hundred papers related to screen design for video display terminals. The same disparity shows up in the Human Factors Journal. Last year there were no papers on documentation and eleven on accessing information from video display terminals. This lack of research in the human factors of documentation is surprising for several reasons: First, documentation is ubiquitous and it is big business. Just about every software package and every piece of hardware comes with its own manual today. Second, the importance of documentation to the usability of a product is not a secret. Alphonse Chapanis, in his presidential address delivered at the Eighth Annual Meeting of the Human Factors Society, told the audience "A very large and important area of human factors engineering [has been] almost entirely neglected. This area consists of the language and the words that are attached to the tools, machines, systems, and operations with which we are concerned." Third, there is no shortage of interesting applied research questions in this area. To name a few: -- When and why do users turn to a manual? -- Does format affect perceived usability? How? -- How do people access information in a document? How can that access be made easier? -- How do technical writers create technical instructions? -- How can documentation be better integrated with online help and error support systems? Although all the participants on the panel have been actively studying the use of documentation, the focus of the panel will not be a recounting of past research. Rather the panelists will offer us direction, based on their extensive experience, for the next decade and beyond. Where are we going? What do we need to know so that we can build a science of documentation? Why have so many human factors practitioners shied away from documentation? How can we make significant improvements to the usability of documents? What form will the documentation of the year 2000 take, and how can human factors contribute to its usability?”
“As technology improves, portable computers become smaller and more compact. A clear design challenge is to provide a system that is as compact as possible without degrading system usability. The keyboard is still the primary input device for compact computers. Previous research has indicated that reduced key spacing adversely affects skilled typing. Therefore, a portable computer system should provide a keyboard with full-sized keys in the primary typing area. The purpose of this study was to determine if reducing key size and spacing adversely affects the usability of a numeric keypad. Skilled keypad operators compared a standard-size numeric keypad to two keypads that had reduced center-to-center key spacing. One of these keypads achieved its reduction primarily by reducing the key spacing. The other reduced both key size and spacing. (Note that the small changes in key size and spacing have little effect on the overall device dimensions of a numeric keypad.) Operators typed numbers faster with and preferred the standard keypad over the keypad with both reduced key size and key spacing. If a numeric keypad is offered as part of a portable computer, every effort should be made to provide full-sized keys. If reduced key spacing is unavoidable, wide keys are preferable to narrow keys.”
“The objective of this research was to develop and test an integrated cursor control and clicking device called a KeyMouse. The bottom of a single key on a standard keyboard was instrumented with pressure transducers. When the key was fully depressed pressure variations across the surface of the key, caused by a rolling motion of the finger, could be used to control the cursor much like a traditional mouse. A usability study was conducted to determine the optimum layout and configuration of the KeyMouse and its associated click keys. Both subjective preference and performance data revealed a strong preference for a two handed configuration with cursor control via the dominant hand and operation of the click keys with the other.”
“A reasonable assumption might be that persons with limited physical ability (persons with cerebral palsy) will experience greater mental effort in performing a physical task. Three computer input devices were used to determine if differences in physical ability result in differences in mental workload estimated by working memory capacity, keystroke rate and efficiency, and subjective workload estimated by the Task Load Index (TLX). The three devices were a voice input system, trackball, and two-degree-of-freedom keyboard with headstick. Subjects with cerebral palsy used the three devices to perform a dual-task and their results were compared with those obtained from a sample of subjects without physical disabilities. Results showed no significant difference in normalized memory capacities between the two groups. However, there were significant differences in Normalized Working Memory Capacity between devices and trials for both groups. As expected, differences in physical ability produced significant differences in keystroke rate, keystroke efficiency, and subjective workload.”
“Four years ago a group of usability test specialists held a panel discussion about this relatively new technique ("Usability testing in the real world," Mills, 1987, SIGCHI Bulletin, 43-46). One of the objectives of that panel was to expose professionals in the usability community to testing methods. That panel assumed that the topic of testing would be new to its audience. Since that time, the number of laboratories that conduct tests has exploded. While no one seems to know for sure, there appear to be in excess of 100 labs in the U.S. alone. Yet, there is no formal organization for people who do testing. This panel session focuses on the current status of testing and its future. The panel will assume that the audience is at least somewhat familiar with the principles of testing. The primary objective of the panel is to make the case that usability testing is a method that is evolving and changing. It is changing because product development processes are changing and because its strengths and weaknesses are more obvious now. The panel's second objective is to provoke a discussion about usability testing and testing methods. The panel will focus on several themes: * Usability tests are becoming less formal. The typical evolution of a human factors method is that it becomes more structured and formal over time. Usability testers, however, have had to adapt to changes in the product development process. More functions than ever are being implemented in software. More flexible software tools have made it possible to withhold freezing the components of user interface until very late in the design process. Consequently, conducting quick, informal tests is becoming more typical because the formal test to verify usability often comes too late to influence the product design. * Usability testing is moving beyond testing the user interface itself. With the introduction of object-oriented programming and contextual design there is more interest in understanding users' cognitive processes. There is also renewed interest in understanding how testing fits with other methods for improving usability. * Usability testing can improve managerial and organizational process. Usability tests reveal more than just flaws in product designs. They also can identify problems in the design process. When test results point to poor management practices and groups that are not cooperating, they can lead to changes in the way organizations develop products. Viewed from this perspective, a usability test has benefits beyond the improvements to the product being tested. It can be used to evaluate management practices in addition to competence in applying effective technical practices.”
“The degree to which a knowledge-acquisition technique (KAT) extracts useful information from a human expert varies, in large part, as a function of the type of expert system being developed. The present study employed two KATs, ARK (a structured-interview technique) and Repertory Grid (a similarity-judgment technique), to elicit knowledge from U.S. Army helicopter pilots with expertise in mission planning. The resultant knowledge bases were compared in terms of their applicability to two tasks: attack route evaluation and attack route planning. The study revealed important qualitative and quantitative differences in the knowledge bases elicited by the two KATs. The Repertory Grid method elicited dimensions with which the efficiency of attack routes could be classified and, in general, appears to be well-suited for use in development of "convergent" expert systems intended to perform tasks involving categorization or evaluation. In contrast, ARK elicited procedural information (e.g., goals, strategies, and rules) with which route planning could be performed, and appears better suited for obtaining information to be used in the development of "divergent" expert systems designed to conduct tasks involving planning or scheduling. The results have implications for the selection of KATS and the design of knowledge-elicitation sessions, and suggest that further analyses of the knowledge-acquisition process be conducted.”
“It is popular to talk about using a problem-driven approach to guide the design of computer-based tools. It is not always so easy, however, to determine just what the significant problems are within some existing work environment. A thorough cognitive task analysis can provide valuable guidance in understanding the nature of existing problems and in designing solutions. In this paper, the use of multiple sources of data are explored in completing such a task analysis in order to design tools to support individuals preparing input for a full-text database.”
“New system architectures, such as model-based reasoning and neural networks, have increased the difficulty of expert system design specification. In this paper, I suggest that to identify the appropriate subtask allocation and expert system architecture, certain preliminary questions should be asked and the data evaluated. The front-end analysis described here is a framework loosely based on three levels of human cognition; analytical, rule-based, and implicit processing. Keeping these different types of cognitive task performance in mind, the framework specifies a set of factors to evaluate in the front-end analysis. Once data is collected for these factors, it is possible to evaluate whether an expert system should be designed for each specific subtask, and if so, what type of system architecture should be implemented. A suggested guideline for architecture choice is presented.”
“Two knowledge elicitation tools for cognitive task analysis are described and compared: Concept Mapping and the Critical Decision method (CDM). Concept Mapping is a procedure that can be used to represent the interviewee's conception of a task by developing a graphical schematic of these perception of the task's components. It is appropriate when one needs to capture the interviewee's cognitive organization of the task's routine elements and how these elements fit together. CDM is highly effective at eliciting tacit knowledge about perceptions, expertise, and aspects of a domain that are often difficult for experts to articulate. It has proven to be an effective tool for capturing the deeper, difficult-to-articulate knowledge that separates experts from novices. Used together, these techniques can be very complementary and effective. Concept Mapping provides an overview of the user's image of the task including information about the clustering of and flow between concepts. CDM is an effective tool for identifying decision strategies, critical cues, situation assessment, goals and intent, expectancies, mental simulation strategies, and improvisation. Used in combination, the techniques can effectively generate recommendations for training and display design.”
“This paper reports a behavioral evaluation of touch interface techniques intended to be used with highly interactive, graphical, windowing software environments. Previous research (Mack and Lang, 1989) indicated that touch interface techniques can produce levels of task performance in such environments comparable to that obtained using conventional mouse pointing devices. The touch technology in that study enabled users to emulate the basic interactions associated with using a mouse: that is, to emulate click, double click and dragging techniques using taps, double taps and tap, hold and drag. While encouraging, problems remained, especially when the finger was used as the input device. The purpose of this study was to compare mouse and touch techniques using an alternative to mouse emulation for controlling touch interactions. Instead, users selected one of three possible touch "modes". In each mode, a simple tap (contact and lift-off) was interpreted by the software in ways corresponding to the three basic mouse interaction techniques. Performance on realistic office task scenarios using a finger and stylus touch techniques and the new touch control method, resulted in comparable performance between mouse and touch stylus when the stylus (but not mouse) was used with a tilted display. Experience with mouse pointing devices, or graphical interfaces enabling direct manipulation, did not affect performance.”
“This paper reports a behavioral evaluation of touch interface techniques intended for use with highly interactive, graphical, windowing software environments. Previous research (Mack and Lang, 1989) indicated that touch interface techniques can produce levels of task performance in such environments comparable to that obtained using conventional mouse pointing devices. In the Mack and Lang study, touch interactions emulated mouse techniques such that single and double taps (contact and lift-off), and tap, hold and drag corresponded to mouse clicks and double-clicks, and dragging with the mouse. While the results were encouraging, problems with the touch techniques remain. The purpose of this study was to compare two alternatives to direct mouse emulation for controlling the interpretation of touch interactions using a stylus or finger as input devices. This study also evaluated two touch-tailored interface techniques: gesture-based commands and pop-up (display) keyboard. Finger performance improved to levels comparable to those for stylus when an alternative protocol for controlling interaction modes was employed. Gestures led to performance comparable to conventional menu bar and pull-down menu techniques. The pop-up keyboard significantly slowed performance.”
“This study was conducted to determine the relative perceived difficulty of performing different mouse tasks (pairings of mouse actions with button combinations). Right-handed individuals with various types of computer experience used a 3-button mouse to perform 49 simple target acquisition tasks. Perceived difficulty judgments varied with tasks. Significant groupings of tasks along the difficulty dimension were also apparent. For most mouse tasks, the left and center buttons were judged to be relatively easier to use than the right button. Additionally, chorded use of the left and center buttons was perceived to be easier than use of the right button alone. The results suggest that systems should both rely upon the right mouse button as a default setting for system-related functions. Other considerations for mouse usage are discussed.”
“Three velocity control interfaces to three dimensional virtual environments are compared. The interface devices are: a six degree of freedom position sensor, a six degree of freedom isometric joystick, and a conventional mouse in conjunction with a soft control panel displayed on the monitor. In each interface the devices are used to control velocity, and all make use of a quadratic function to map the input to the viewpoint velocity. We use two structured exploration tasks to assess the usability of the different interfaces. In the first task an interviewing technique is used in conjunction with an exploration task which involved examining widely spaced details of the 3D scene. The second task is designed to reveal how well users can interact at different scales using the different devices. Subjects are required to navigate through a tube which varies over four orders of magnitude in size. The results show that subject's behavior is highly constrained by the local size of the tube: they maintained a constant velocity relative to the local size of the tube. They also showed differences in the effectiveness of the different devices in determining traversal rate that the positioning device and the control panel were about equally effective for fast navigation, and both are better than the isometric joystick.”
“One important element of hypertext design is the presentation of the hypertext cue. This study compared learning relationships about animals in an oriental zodiac using one of two online, hypertext presentation styles and printed text. In one hypertext condition, the cues were embedded in the text; in the other condition, the cues were located in the margins. A control condition contained unrelated study materials. The database on the zodiac consisted of a multiple hierarchy. Two measures of learning were used. Reaction time before and after a brief study period was used to assess knowledge about the relationships among animals in the zodiac. Participants' ability to correctly reproduce the underlying model of the zodiac in the database using a card sorting task was used to assess the degree to which they learned the underlying model. Analysis of variance revealed a significant interaction between learning (pre- and post-study period) and presentation condition (hypertext embedded in text, hypertext in the margin, printed text, unrelated text material). An analysis of the results of the card sort data revealed that participants more accurately reproduced the underlying model after study with hypertext, regardless of the location of the cue, than with text. The results are discussed with respect to the design of hypertext information and the need for future research.”
“Most discussions of hypertext usability are not founded in empirical measurement but more on conjectures based on personal experience. In this paper we report on two empirical analyses of hypertext usability, focusing on the quality of links produced by different means. We conducted two experiments to test the predicted relevance and the evaluated relevance of links, that is, where links are evaluated either before or after they are traversed. In order to evaluate these two kinds of relevance, we conducted two experiments where a hypertext document was created from a printed text. In each experiment we compared the relevances of three different sets of links. One set was created by a human author, whereas the second set was created automatically using the HEFTI (Hypertext Extraction From Text Incrementally) model for converting text into hypertext. We also generated a third set of links by assigning links randomly between nodes. The main goal of this research was to develop empirical tests that evaluate the usability of hypertext links. A second goal was to test the validity of automatically generated links using the HEFTI model. In this paper we detail the two experiments, and discuss their implication for methods of hypertext usability assessment and design.”
“The advent of graphic-based display environments such as those found on the Macintosh, Windows, OS/2 Presentation Manager, and X Windows, has the potential for providing barriers for individuals with visual impairments, particularly those who are blind. The problems stem from three factors. First, the information being displayed on the screen has shifted from a character-based format that was stored in an ASCII text buffer to a pixel-based format. This makes it much more difficult for screen reading software to determine what characters are on the screen. Second, text-based systems used a single font and relatively few attributes (bold, underline). On the graphic displays, text can assume a very large variety of sizes, fonts, and attributes (bold, underline, italic, crossed out, etc.). When these font changes or attributes contain information, they complicate the process of presenting information via speech or braille. Third, the advent of the graphics-based system has led to much more widespread incorporation of graphic elements, such as charts, diagrams, and pictures, within the text. The new systems also introduce a number of advantages or opportunities for individuals with severe visual impairments or who are blind. Both the consistency of the human interface among applications and the use of system tools in the display process hold the potential for providing access to a broader range of applications for persons who are blind, if an effective human interface to these operating systems can be developed. The Systems X project is using a multi-sensory approach to explore techniques for providing an effective and efficient interface to graphic-based by people who are blind. A prototype system, dubbed "Systems 3," has been developed which uses speech input/output, the keyboard, and a virtual tactile tablet to allow individuals to access both text and graphic elements in the system. Using the prototype, individuals who are completely blind have been able to use a Macintosh computer, access and read text documents, and handle simple to moderate graphic elements such as bar charts. A series of research studies is now ongoing to test the limits of the access and to quantify the relative efficacy of various approaches to providing a nonvisual interface to these graphic-based operating environments.”
“Although video has been used for many years to record data, few tools have been developed to manipulate enhance video data. Current multimedia interfaces have severe cognitive and attentional limitations, reflecting technology-centred designs which have not profited from human factors theory and user-centred design. This results in cumbersome and expensive systems with serious usability problems. This paper describes a prototype Video ANNotation and Analysis system (VANNA), which integrates video, non-speech audio, voice, textual and graphical data, and which incorporates emerging technology, user-centred design and human factors theory.”
“The design of useful, easy to use application software requires early and sustained involvement of end users in the design and evaluation of the user interface. The present paper describes the experiences of the authors during the design and evaluation of a prototype user interface for an accounts receivable application. Some of the issues inherent in prototyping, and the anticipated, as well as unanticipated, benefits of prototyping are discussed.”
“With the recent development of a number of powerful end-user interface (EUI) prototyping tools, the typical software usability department has never before had such a golden opportunity to take a lead role in product development. However, there is more to establishing a successful EUI prototyping effort than simply possessing a background in EUI design and having access to a prototyping tool. Indeed it has been our experience that the success of a prototyping effort is dependent on many factors -- some obvious, some not so obvious, and others learned only through postgraduate training at the School of Hard Knocks. The purpose of this paper is to describe 22 tips we believe will help ensure a successful prototyping effort.”
“Over the past decade, a multi-phased project supported by agencies of the Department of Defense, FAA, and NATO has been underway to understand and remediate problems in the transitioning of ergonomic research to system design applications. Efforts to enhance the usability of ergonomic data in system design have resulted in the present R&D project, which is concerned with developing a multi-media ergonomics database on CD-ROM. The Computer Aided Systems Human Engineering (CASHE) system, Version 1.0, will contain the Boff and Lincoln (1988) Engineering Data Compendium, MIL-STD-1472D and the Perception & Performance Prototyper. The Perception & Performance Prototyper will allow the user to experience and manipulate the technical data found in the Compendium and MIL-STD-1472D. The CASHE tool will also include specialized data retrieval, scaling, and analysis capabilities as well as state-of-the-art in information retrieval, browsing, and navigation.”
“Groups of individuals working together effectively and efficiently are key to successful organizations, and ways in which new technologics can be exploited to support groups at work are being explored. The Decision Support Center (DSC) provides one approach to enhancing group productivity. The DSC consists of a series of networked PCs running software programs (TeamFocus) that provide automated support for group communication and decision making. The present evaluation was designed to provide some initial data on the potential benefits that may be realized by utilizing the DSC in the development of new products. Output from traditional methods employed during the requirements phase of the AD/M process was compared to output from a parallel Decision Support Center (DSC) requirements gathering session. The results indicated that the DSC session, in addition to validating already known system requirements, generated a substantial number of new requirements. The results provide sufficient evidence to warrant further exploration of areas in which utilization of the DSC may prove valuable.”
“It is proposed that current user interface research is not being incorporated in a timely way in new systems. Some reasons for this are proposed and discussed. Possible solutions include (a) better selection of research topics, (b) conducting rapid studies, and (c) writing research reports specifically for practitioners.”
“Two studies are reported of the usability of icons. Firstly, an experiment was carried out to investigate the transfer of performance between sets of icons for different computer functions. The aim was to determine whether users would benefit from common underlying elements within concrete icon sets, compared with unrelated abstract icons. A cross-over study was employed whereby subjects performed an icon identification task with one set of icons (a concrete or abstract learning set) and then transferred to another set of icons (a concrete or abstract transfer set). It was hypothesized that transferring from concrete to concrete icons would yield the best results due to the use of repeated elements in the concrete sets. Identification time data did not produce significant results to support the hypothesis. Significant results were yielded from the error data, this demonstrated the superiority of concrete icons, but no advantage came from previous experience with a concrete set. A second study examined the rating of the icons used in the first study in terms of concreteness and appropriateness. Appropriateness was found to be a reasonable predictor of icon identification time.”
“In an effort to reach markets in which a keyboard/mouse interface is difficult or inconvenient to use, manufacturers are now beginning to introduce light-weight portable computers which recognize hand-printed characters. The recognition accuracy that these new computers are able to achieve will be a critical factor in determining their acceptance by users. There are, however, few published studies of handwriting recognition accuracy and the variables which affect accuracy. The purpose of this study was to assess recognition accuracy as a function of a number of factors which might vary in the real-world use of handwriting recognition systems. These factors included style of writing, amount of training, interval of disuse, and alphabet. The findings suggested that recognition accuracy reached a steady-state level with a relatively small amount of training and remained at that level for as long as a month (the longest interval tested in this study). For an 82-character alphabet, character recognition accuracy was 92.7% for discrete writing and 87.6% for run-on writing. Accuracy with alphabets restricted to digits or uppercase only was quite high, with mean recognition rates ranging from 95% to”
“The large number of typists and secretaries in the work force who type documents for others are evidence that a substantial segment of the population can not or will not produce their own documents. There are, however, significant pressures in the work place for shifting some or all of the responsibility for document production to the producer of the document content. Automatic speech recognition systems would make this possible if they were to provide a means for handling formatting operations as well as transcribing content accurately. This study investigated user reactions to six formatting methods (e.g., for creation of a list structure, for beginning a paragraph) in a simulated "document creation by voice" application. The results of this study indicate that subjects preferred the method in which a spoken command was used to produce formatting and immediate recognition feedback was received. The second highest ranked condition was one which required that, in addition, the spoken command be preceded by a mouse click. An argument is made in favor of implementing the second-most favored method because it would allow for the inclusion of a number of additional features requested by users.”
“374 Research examining display format and highlight validity (Tullis, 1984; Fisher & Tan, 1989) have shown that these factors affect visual display search; however, these studies have been conducted on small, artificial alphanumeric displays. The present study manipulated these variables, applying them to realistic, complex Space Shuttle displays. A 2 (display type: Orbit Maneuver Execute, Relative Navigation) x 2 (display format: current, reformatted [following human-computer interface design principles]) x 3 (highlighting validity: valid, invalid, no-highlight) within-subjects analysis of variance found significant main effects of these variables on search time and a significant format by highlight interaction. Search through the current, poorly-formatted displays benefited from valid application of highlight, and showed no cost of invalid highlight. Reformatted displays demonstrated neither reliable cost nor benefit of highlight application. Significant correlations were found between observed search times and search times predicted by Tullis' Display Analysis Program (1986): the relationship was highest with non-highlighted displays and was less predictive with valid and invalid highlight applications. Issues discussed include the enhancement of search through format and highlighting, and the necessity to consider several factors when predicting search performance.”
“These studies suggest that for certain menus there may be performance advantages in having choices placed on the right side of options. This holds for measures of both speed and accuracy. However, for some menus there is a slight advantage for having choices on the left side. Accuracy levels tended to be better with choices positioned to the right of options. The results suggest that practitioners should consider the relative positioning of options and choices in the design of user interfaces.”
“The design of control centers is advancing toward totally computer-based man-machine interfaces. Computer based interfaces offer many potential advantages over traditional hardwired control panel interfaces including greater flexibility regarding the type of data displayed and its presentation. However, achieving this potential will require development of new interface concepts that will change the way operators interact with the plant. Extensive evaluation throughout the design process will be required to verify and validate the interface concepts. This paper describes a process for uncovering evaluation issues related to the computer-based control room concept and its relationship to cognitive activities of plant control. Important evaluation issues are presented.”
“The current study explores the relationship between the display representation of a multidimensional data base (separated vs. integrated) and the information processing demands of different types of tasks. Subject answered a range of questions about a complex economic data base while viewing either a 2-D or 3-D (perspective) display of subsets of the data. Animated vs. static representations of dynamic data were also examined. The 3-D display supported better performance than did the 2-D display on the more integrative questions, while no difference between formats was found for questions demanding focused attention. The 3-D display also supported a better visual model of the data space, but did not improve subjects' ability to verbalize the rules underlying this space. Animation did not provide a better means for understanding change information than did the static presentation mode. Implications for the design of data visualization display interfaces are discussed.”
“Although many automated tools support the productivity of professionals (engineers, managers, architects, secretaries, etc.), none specifically address the needs of the scientific researcher. The scientist's needs are complex and the primary activities are cognitive rather than physical. The individual scientist collects and manipulates large data sets, integrates, synthesizes, generates, and records information. The means to access and manipulate information are a critical determinant of the performance of the system as a whole. One hindrance in this process is the scientist's computer environment, which has changed little in the last two decades. Extensive time and effort is demanded from the scientist to learn to use the computer system. This paper describes how chemists' activities and interactions with information were abstracted into a common paradigm that meets the critical requirement of facilitating information access and retrieval. This paradigm was embodied in EASI, a working prototype that increased the productivity of the individual scientific researcher.”
“An experiment examined how people use an online retrieval system. Subjects solved general topical search problems using a database containing the full text of news articles (e.g., find articles about the "Background of the new prime minister of Great Britain"). Time, accuracy and content of the searches were recorded. Of particular interest was the use of two iterative search methods available in the interface -- a Lookup function that allowed users to explicitly specify an alternative query; and a LikeThese function that could be used to automatically generate a new query using articles the user marked as relevant. Results showed that subjects could easily use both query reformulation methods. Subjects generated much more effective LikeThese searches than Lookup searches. An analysis of individual subject differences suggests that the LikeThese method is more accessible to a wide range of users.”
“This paper describes the use of the operator function model (OFM) to represent troubleshooting of printed circuit board assemblies. The model is potentially useful for understanding the troubleshooting process and providing the knowledge required by either a computer-based decision support system or an intelligent tutoring system. The model was derived from empirical data based on observations and concurrent protocols of troubleshooters. The 'raw' data were converted into cases. Based on analysis of the troubleshooting cases, an operator function model of troubleshooting was proposed. Model validation entailed comparison of model predicted troubleshooting with actual operator troubleshooting.”
“In the supervisory control of highly automated dynamic systems, the human operator is responsible for monitoring system state and compensating for system failures. Increasingly, the human operator interacts with one or more knowledge-based (e.g., expert) systems that assume responsibility for a portion of the supervisory control task. Thus supervisory control is shared by both human and machine agent(s). To date, however, there is little research addressing the interaction of human and knowledge-based machine agents in supervisory control. This paper describes research that explores issues in the design of the cooperative human-machine interaction in the control of a complex dynamic system. The paper presents a set of prescriptive principles that define the human-computer interaction in supervisory control systems. In addition, the paper summarizes a case study of a NASA satellite ground control system in which human operators work with an expert system. The prescriptive principles and case study results form the basis of an architecture for cooperative problem solving for real-time control of dynamic systems.”
“The research presented in this paper is concerned with the planning component of project management and describes the use of interviewing and observational techniques to develop cognitive models of planning for project management systems. Interviews were conducted with six project managers from six different problem domains. Protocol analysis was used to develop two types of cognitive models: process models that provided detail descriptions of planning actions; and a higher level model of the planning process based on the Hayes-Roth theoretical model of planning. A visual representation of the planning process, called DMAP, was created to identify the type planning that occurs at each stage of the planning process in project management. The discussion focuses on the use of cognitive models in the design of knowledge-based systems for project management.”
“Characteristics of transaction-processing tasks are described in the context of retail point-of-sale systems (cash registers). Proliferation of functionality coupled with a lack of emphasis on interface design have led to systems that require extensive training and result in high error rates. A state-action representation of a transaction-processing task is described. This approach to task modeling is appropriate given the deterministic nature of transaction-processing tasks. To illustrate this approach, a task model for a particular retail point-of-sale system and an interface design based on this model are described. A preliminary analysis of subjects using the new interface suggests that the model-based design will result in significant reductions in training time and error rates.”
“Current research suggests that embedded user models would allow more intelligent and helpful human-computer interaction. However, until recently, the cognitive modeling technology capable of generating models sufficiently complete and detailed as would be needed to drive development of embedded user models has not existed. COGNET (COGnition as a Network of Tasks) is a new cognitive modeling framework focusing on real-time, multi-tasking domains, and has been successfully applied and validated in several domains. This paper describes research to explore COGNET for embedded user models. The research took an existing COGNET model and implemented it in software as an embedded user model for an intelligent human-computer interface. The HCI used the resulting embedded model to produce significantly enhanced human-computer interface functionality. Specifically, the model was used to provide attention aiding, dynamic task prioritization, context-sensitive decision structuring, and context sensitive task automation in a complex real-time vehicle monitoring task. The software used to implement the example model also proved generalizable to other domains and COGNET models. This software has been named BATON (Blackboard Architecture for Task-Oriented Networks). Similarly, the interface architecture is broadly applicable to providing these same functions in other real-time, multi-tasking domains.”
“This paper reports an investigation into issues concerning the user population for a complex interactive systems domain, namely image processing. A variety of image processing systems are available, attempting to provide broad applicability, but are often complicated to understand and hard to use. One difficulty in producing an image processing system is defining its user population. In the image processing domain, this is a challenge because image processing has a diverse population of users, with great variances in user expertise and expectations. To better understand the process of classifying users of complex interactive systems, we conducted a study to define the user population of image processing systems, and to determine their common goals and tasks. First, we produced a task-based model of the user population of image processing systems, with levels in the model representing classes of tasks users perform with an image processing system. We then conduced a semi-structured interview and an ordering task to gather data to validate this model. Subjects were a broad variety of image processing users at Virginia Tech. We also had each subject complete a questionnaire to help identify common goals and tasks among image processing users. Results indicate validity of the model for classifying image processing user populations. Results provide the basis for continued research in a domain that requires highly flexible and adaptable interfaces to functionally complex systems. Further development of the model can provide image processing system developers with requirements for baseline functionality of an image processing systems.”
“Automated machines facing the general public's daily life (i.e., ATMs) have become common elements of environments. Even with the continued growth, some institutions still find it difficult to get more than half of their customers to use the machine regularly. This paper presents an optimized interface for an automated teller machine designed to attract even the extremely resistant user. To accomplish this task, information was gathered from non ATM users in addition to the regular customers of such machines. Several existing machines were critiqued based on the needs of non ATM users in addition to the generally accepted human factors criteria. Design solutions were generated for all desirable elements, such as input and output devices characteristics, aesthetics, etc. Designed were tested using the method of pair-comparison. The chosen ATM design is believed to enhance ease of use, accessibility, and overall productivity of such units for virtually the entire spectrum of ATM users.”
“There are many situations in assessing product or system usability where it is desirable to know where a person is looking. Currently, eyetracking systems that provide this capability are rarely found in organizations that are involved in usability engineering. In contrast, one of the unique features of the Contel Intelligent System's Usability Laboratory (recently merged with GTE Laboratories) is the Eyegaze System, an eyetracking system, developed by LC Technologies. Our desire is to implement an approach for characterizing a computer user's ocular "behavior" in a way that supplements the measures of performance that cab be derived from observations of overt behavior. This presentation will describe an eye movement study conducted using PRODIGY Display Screens, highlighting the use of the Eyegaze System as a usability tool. A short video will be presented (1) showing the Eyegaze System as an integral facet of the usability laboratory, and (2) showing results of the type of study that can be performed using an eyetracking system for usability assessments.”
