Task analysis is the process of understanding the user's task thoroughly enough to help design a computer system that will effectively support users in doing the task. By task is meant the user's job or work activities, what the user is attempting to accomplish. By analysis is meant a relatively systematic approach to understanding the user's task that goes beyond unaided intuitions or speculations, and attempts to document and describe exactly what the task involves. The design of functionality is a stage of the design of computer systems in which the useraccessible functions of the computer system are chosen and specified. The basic thesis of this article is that the successful design of functionality requires a task analysis early enough in the system design to enable the developers to create a system that effectively supports the user's task. Thus the proper goal of the design of functionality is to choose functions that are both useful in the user's task, and which together with a good user interface, results in a system that is usable, being easy to learn and easy to use.

A key phenomenon in visual search experiments is the linear relation of reaction time (RT) to the number of objects to be searched (set size). The dominant theory of visual search claims that this is a result of covert selective attention operating sequentially to "bind" visual features into objects, and this mechanism operates differently depending on the nature of the search task and the visual features involved, causing the slope of the RT as a function of set size to range from zero to large values. However, a cognitive architectural model presented here shows these effects on RT in three different search task conditions can be easily obtained from basic visual mechanisms, eye movements, and simple task strategies. No selective attention mechanism is needed. In addition, there are little-explored effects of visual crowding, which is typically confounded with set size in visual search experiments. Including a simple mechanism for crowding in the model also allows it to account for significant effects on error rate (ER). The resulting model shows the interaction between visual mechanisms and task strategy, and thus it represents a more comprehensive and fruitful approach to visual search than the dominant theory.

An increasingly popular measure in the study of reading comprehension is that of reading time, or inspection time, the time a reader takes to process a piece of verbal input. In constrast to the more commonly used recall measure, which assesses the results of comprehension, the reading time measure taps an aspect of the comprehension process as it occurs. For this reason, the study of reading times can contribute uniquely to growing understanding of comprehension processes. In contrast to conventional theories of comprehension, the computer simulation models of comprehension and the allied efforts in artificial intelligence are committed to describing at a general level and in great detail exactly what must be done in comprehension. Simulation models show a strong split between being stochastic or deterministic. Stochastic simulations explicitly include a random process that ensures that the behavior of the model is variable.

With a simple demonstration model, Hulleman & Olivers (H&O) effectively argue that theories of visual search need an overhaul. We point to related literature in which visual search is modeled in even more detail through the use of computational cognitive architectures that incorporate fundamental perceptual, cognitive, and motor mechanisms; the result of such work thus far bolsters their arguments considerably.

An important application of cognitive architectures is to provide human performance models that capture psychological mechanisms in a form that can be "programmed" to predict task performance of human-machine system designs. Although many aspects of human performance have been successfully modeled in this approach, accounting for multitalker speech task performance is a novel problem. This article presents a model for performance in a two-talker task that incorporates concepts from psychoacoustics, in particular, masking effects and stream formation.

An important application of cognitive architectures is to provide human performance models that capture psychological mechanisms in a form that can be “programmed” to predict task performance of human-machine system designs. Earlier models accounted for some key aspects of performance in a two-talker task, but spatial separation of the speech sources produces complex effects not yet represented. Adding some first-principle mechanisms to the earlier models suggests that this fundamental aspect of multi-talker speech perception can be accounted for as well.

Speech recognition was measured as a function of the target-to-masker ratio (TMR) with syntactically similar speech maskers. In the first experiment, listeners were instructed to report keywords from the target sentence. Data averaged across listeners showed a plateau in performance below 0 dB TMR when masker and target sentences were from the same talker. In this experiment, some listeners tended to report the target words at all TMRs in accordance with the instructions, while others reported keywords from the louder of the sentences, contrary to the instructions. In the second experiment, stimuli were the same as in the first experiment, but listeners were also instructed to avoid reporting the masker keywords, and a payoff matrix penalizing masker keywords and rewarding target keywords was used. In this experiment, listeners reduced the number of reported masker keywords, and increased the number of reported target keywords overall, and the average data showed a local minimum at 0 dB TMR with same-talker maskers. The best overall performance with a same-talker masker was obtained with a level difference of 9 dB, where listeners achieved near perfect performance when the target was louder, and at least 80% correct performance when the target was the quieter of the two sentences.

An extension of the auditory module in EPIC is introduced to model the two-talker coordinate response measure (CRM) listening task. The construct of an auditory stream is employed as an object in the working memory of EPIC’s cognitive processor. Production rules are developed that execute the two-talker CRM task. Analysis of these rules reveal two sources of possible error in the output of the auditory processor to working memory. Each is explored in turn and the production rules modified to provide a corpus-driven model that accounts for human performance in the listening task.

Being able to predict the performance of interface designs using models of human cognition and performance is a long-standing goal of HCI research. This paper presents recent advances in cognitive modeling which permit increasingly realistic and accurate predictions for visual human-computer interaction tasks such as icon search by incorporating an "active vision" approach which emphasizes eye movements to visual features based on the availability of features in relationship to the point of gaze. A high fidelity model of a classic visual search task demonstrates the value of incorporating visual acuity functions into models of visual performance. The features captured by the high-fidelity model are then used to formulate a model simple enough for practical use, which is then implemented in an easy-to-use GLEAN modeling tool. Easy-to-use predictive models for complex visual search are thus feasible and should be further developed.

An important application of cognitive architectures is to provide human performance models that capture psychological mechanisms in a form that can be “programmed” to predict task performance of human-machine system designs. While many aspects of human performance have been successfully modeled in this approach, accounting for multi-talker speech task performance is a novel problem. This paper presents a model for performance in a two-talker task that incorporates concepts from the psychoacoustic study of speech perception, in particular, masking effects and stream formation.

This chapter will first argue that model-based evaluation is a valuable supplement to conventional usability evaluation, and then survey the current approaches for performing model-based evaluation. Because of the considerable technical detail involved in applying model-based evaluation techniques, this chapter cannot include “how to” guides on the specific modeling methods, but they are all well documented elsewhere. Instead, this chapter will present several high-level issues in constructing and using models for interface evaluation, and comment on the current approaches in the context of those issues. This will assist the reader in deciding whether to apply a model-based technique, which one to use, what problems to avoid, and what benefits to expect. Somewhat more detail will be presented about one form of model-based evaluation, GOMS models, which is a well-developed, relatively simple and “ready to use” methodology applicable to many interface design problems. A set of concluding recommendations will summarize the practical advice.

Experiments on visual search have demonstrated the existence of a relatively large and reliable memory for which objects have been fixated; an indication of this memory is that revisits (fixations on previously fixated objects) typically comprise only about 5% of fixations. Any cognitive architecture that supports visual search must account for where such memory resides in the system and how it can be used to guide eye movements in visual search. This paper presents a simple solution for the EPIC architecture that is consistent with the overall requirements for modeling visually-intensive tasks and other visual memory phenomena.

Abstract : Investigations of possibly parallel dual cognitive decision behavior usually are designed so that the stimulus sensory channels for the dual tasks are independent. Experiments with dual audio-visual stimuli had interference delays of only 5-10ms and less than 1% errors by the best performers. This study used a dual decision paradigm but with auditory-only stimuli organized in a 'centersurround' presentation. 'Surround' stimuli were shaped white noise pulses presented over a headset in dichotic mode so as to be localizable in the space outside the listener's head. 'Center' stimuli were spoken words presented in diotic mode localized inside the head. Responses to the externalized sounds were made by button presses while memorized verbal responses were required for the internalized words. The best performers with this auditory-only organization had mean interference times of about 20ms with nearly 10% error rate. For these conditions, the center-surround arrangement did not support "virtually perfect" dual decision-making as well as the auditory-visual presentation. Further testing is planned with a simplified verbal task.