Learning in Man-Computer Interaction: A Review of the Literature (Research Reports Esprit (1)).

In the beginning of the 1980's it became evident that the European IT industry needed a substantial collaborative effort in the area of pre-competitive research and development. At that time several studies indicated that in the 90's and later a strategic lack in the economic power of European industry as a whole might result from the lack .of a competitive position in the key area of information technology. Therefore a strategic research program in this area was initiated by the Commission of the European Communities. This program was carefully planned and several hundred experts were involved in the process of definition and implementation. The program was named ESPRIT - European Strategic Programme for Research and Development in Information Technologies. At a very early phase of the program it was obvious that a substantial collaborative multinational effort in the direction of human factors in information technology should be undertaken. It became clear that only with the incorporation of human-factors-oriented knowledge into the product design technologies of the 90's could later success in the market be assured. Moreover, it was felt that Europe especially is well equipped to deal thoroughly with user issues in relation to information technology.

1. Introduction.- 1.1. Objectives.- 1.2 Perspectives for application.- 2. The Problem.- 2.1 The concept of ‘User-Friendliness’ ― What is the ideal information-processing system from the user’s point of view?.- 2.2 Does the ideal system require no learning?.- 2.3 Theoretical foundations.- 2.4 Representation of knowledge.- 2.5 Prior knowledge of the user.- 2.6 Design options: Requirements for learning-support systems.- 2.7 Three faces of the user-model.- 2.8 Cost and benefit of training and practice.- 2.9 Objectives for further research.- 3. Modelling the Human-Computer System.- 3.1 The components of man-computer systems: Task / Device / User.- 3.2 Are separate descriptions of Task / Device / User knowledge possible?.- 3.3 Human information processing and cognitive modelling.- 3.4 Learning processes and learning-theories.- 3.5 Definitions of knowledge, skill and learning.- 3.6 Motivation, needs and decision-making.- 3.7 Known user needs and motivation.- 3.8 Formal modelling of the user.- 3.8.1 Task and knowledge in man-computer interaction.- 3.8.2 Task-analysis and cognitive task-analysis.- 3.8.3 Cognitive task-analysis and task-action mapping.- 3.8.4 GOMS models.- 3.8.5 KIERAS & POLSON’s Cognitive Complexity Theory.- 3.8.6 Transfer.- 3.8.7 Models of learning and skill acquisition.- 3.8.8 Other man-machine models involving learning.- 3.8.9 Resumee.- 4. Methodological Considerations.- 4.1 Empirical methods for the description and analysis of knowledge and learning of computer users.- 4.2 Describing learning processes by learning curves.- 4.3 Estimation of expenditure for learning.- 4.4 Benchmark-tests.- 4.4.1 The structure of real tasks.- 4.4.2 Task change as a consequence of computer usage.- 4.5 Methods for data collection.- 4.5.1 Data-logging.- 4.5.2 Protocol analysis.- 4.5.2.1 Analysis of verbal reports with the aid of a model versus without a model.- 4.5.2.2 Free verbalization versus probed elicitation of verbal reports.- 4.5.2.3 Concurrent verbal reports versus retrospective comments on records of verbal reports.- 4.5.2.4 Studies employing protocol analysis.- 5. Empirical Studies.- 5.1 System variables in empirical studies.- 5.2 Command language (input language).- 5.2.1 The ‘Learnability / Usability Tradeoff’ for command languages.- 5.2.2 Command languages.- 5.2.2.1 (Restricted) natural language as command language.- 5.2.2.2 Specific versus general command names.- 5.2.2.3 User derived command languages and ‘Aliasing’.- 5.2.2.4 Syntax of command languages.- 5.2.2.5 Abbreviations of commands.- 5.2.2.6 Discussion of natural language I/O.- 5.2.3 Query languages.- 5.2.4 Menu interfaces.- 5.2.4.1 Arrangement of items in menus.- 5.2.4.2 Optimizing breath and depth of menus.- 5.2.4.3 Icons versus command words.- 5.3. Input devices.- 5.3.1 Selection and pointing devices.- 5.3.2 Keybords.- 5.3.3 Discussion of input devices.- 5.3.4 Speech input.- 5.4 Format of output to the user and system modes.- 5.5 User support functions.- 5.5.1 Documentation.- 5.5.1.1 Documentation needs of beginners and experts.- 5.5.1.2 Reading skill and documentation.- 5.5.1.3 User-centered design of documentation.- 5.5.1.4 Task-oriented versus device oriented documentation.- 5.5.1.5 Advance organizers and mental models.- 5.5.1.6 Conclusion.- 5.5.2 Error messages and help functions.- 5.5.3 Help systems.- 5.5.4 Advice giving systems.- 5.5.5 Adaptive systems.- 5.5.6 Training systems / Computer aided instruction.- 6. Users.- 6.1 User classification.- 6.1.1 Prior knowledge and user classification.- 6.1.2 User classification schemes.- 6.1.3 ‘User-groups’.- 6.1.4 Users task knowledge.- 6.1.5 Discussion of user classification.- 6.1.6 What is the ‘real’ user?.- 6.2 Education and training for computer use.- 6.2.1 Computer literacy and the development of qualification.- 6.2.2 Overview of training and education.- 6.2.3 Conclusion.- 7. Areas of Application.- 7.1 Text-editing.- 7.1.1 Initial training requirements.- 7.1.2 The learning curve.- 7.1.3 Asymptotic value.- 7.1.4 Prior knowledge & individual differences.- 7.1.5 Concept formation.- 7.1.6 Errors due to transfer of knowledge.- 7.1.7 Transfer of training in text-editing tasks.- 7.1.8 Conclusion and discussion.- 7.2 Training for CAD.- 7.3 Learning to program.- 7.3.1 Performance in programming.- 7.3.2 Concept acquisition.- 7.4 Other areas of applications.- 7.5 Resumee.- 8. Perspectives for Modelling.- 8.1 Basic assumptions and definitions.- 8.1.1 Device.- 8.1.2 Task.- 8.1.3 User.- 8.1.4 Learning.- 8.1.5 Is the computer a tool? Man-computer interface or man-task interface?.- 8.1.6 Main assumptions differentiating our approach from others: Salient properties of the proposed conceptualisation.- 8.2 Formal structure of our model.- 8.2.1 The task space.- 8.2.2 States.- 8.2.3 Formal representation of the task-space.- 8.2.4 State transitions.- 8.2.5 Work and tasks in the task-space.- 8.2.6 Modelling user knowledge.- 8.2.7 Practical modelling.- 8.2.8 The representation of learning processes.- 8.3 Validation of cognitive models.- 8.3.1 Estimating the parameters of learning curves.- 8.3.2 Methods for empirical validation of models of user knowledge.- 8.3.2.1 PUM defines which knowledge must be probed.- 8.3.2.2 Usefulness of data-logging for model validation.- 8.3.2.3 Using verbal protocols to identify the descriptive user model.- 8.4 Scenario for the use of a model based evaluation methodology.- 8.5 User models in MCI.- 8.5.1 Mental models.- 8.5.2 Embedded user models as a basis of adaptive systems.- 8.6 Motivation in user-models.- References.- A Classification.- B Literature.