HCI

Work Activity in Human-Computer Interaction

Neill A. Kipp

Martha L. Haigler

Linda S. van Rens

December 3, 1996
(go to table of contents)

(ABSTRACT)
GROUP FIVE bundles all the repertory of HCI together at the highest ontological level, tying together the science of activity theory, the sociological practice of ethnography, the software engineering design dogma called participatory design, and the relatively unbounded application field of computer-supported cooperative work.

Chapter 1.   Activity Theory

``To do is the only way we can know...'' the activity theorists say. These philosophers believe that people's actions cannot be seen separately from the environment in which they work (play, learn, etc.) and the circumstances under which that activity takes place. To the activity theorists, both work and play are versions of the learning activity.

1.1   Origins

This philosophy finds its beginnings in pre-revolutionary Russia in the work of Lev Semyonovich Vygotsky. Vygotsky was born in 1896 (100 years ago this year) in Byelorussia. He attended Moscow University and graduated in 1917. He taught young children as a schoolmaster, and then became active in studying developmental psychology. In 1934, he died of tuberculosis.

Vygotsky's greatest contribution is Thought and Language (sometimes translated Thinking and Speech ). This auspicious work was allowed to be translated only in 1962, approximately 30 years after it was written. In this book, Vygotsky sets forth the tenets of activity-based developmental psychology--- that children must speak the words to gain an understanding of them and that learning is a participatory process (not a conditioned one).

Two important concepts in Vygotsky view of activity theory are zone of proximal development (external path of social learning), and internalization (internal path of social learning) and topics related thereto, semiotic mediation, inspiration through conflict, and thought and language.

1.2   Zone of Proximal Development

Probably the most famous concept in active learning, the zone of proximal development is defined to be the distance between (1) what we know and (2) the potential we have for knowing. Actualizing that potential is fundamental to Vygotskian psychology.

For each zone, there are two characters, a novice, and a more knowledgeable other (MKO). Traditionally, the MKO is the teacher, but this casting is not required; peers of the novice often (and unknowingly) guide the novice through the learning activity.

For a single lesson or for the entire term, the MKO determines what each student knows and what each student is capable of knowing, and then invents some virtual `scaffolding'--- hints to get the student from one end of the zone to the other. With gentle nudging from the teacher and dialog with other students as well, the student explores the zone using the scaffolding provided, and achieves the goal.

1.2.1   Problems with the Zone

Setting the wrong zone could be disastrous. Should the zone be too narrow, the novice will achieve it too soon and be bored. Should the zone be too wide, the novice will become frustrated and disillusioned. Should the scaffolding be too obvious, the lesson becomes so pedantic to result in a loss of retention. If the class is large, the MKO risks misplacing the starting point. If members of the class have a wide range of ability levels, the MKO risks misplacing the achieving point. If the scaffolding is too thin or points the novice in the wrong direction, the novices might not survive the zone, or worse, they might learn incorrect or contradictory information.

1.3   Learning by internalization

The activity theorists believe that learning occurs through an active and socially-inspired dialog. This active dialog inspires learning by internalization, and that language is inextricable in this process: we can learn nothing without attaching it to a language, and that this language can be a language of signs [Blackler, 1995].

Internalization happens in four stages. This is best described using a scenario. Pablo knows that fractions can be reduced and certain numbers can be expressed as ratios and reduced like fractions. His teacher shows his team several circles and has them measure the circumference and diameter of each, then has them calculate the ratio of the circumference and diameter. Pablo finds that within a certain error tolerance, the ratio is always the same, no matter what size circle he is presented.

Through his activity, Pablo notices the pattern and begins to learn. He detaches (abstracts) the concept (the ratio is always the same) from his practical experience. He synthesizes it with his past knowledge (fractions and circles), he symbolizes it with a word (pi) and then he understands the concept--- it is placed in his conceptual system.

We call this process--- abstract, synthesize, symbolize, understand--- learning by internalization [cf. grok].

1.3.1   Semiotic Mediation

How we manipulate the pieces of language, or signs (semiotic mediation) in a social context to develop higher-order thinking is a tenet of activity theory [Dixon-Krauss, 1996]. Our learner in the example above can now use the abstract concept of pi in dialogs with his classmates to further his understanding of the mechanics of circles and spheres.

Additionally, the activity theorists believe that no learning but active learning can happen.

1.3.2   Inspiration through Conflict

What encourages a novice to learn? What motivates novices through the zone of proximal development?

We know that learners to not internalize concepts immediately. We see that above anything else, conflict promotes enunciation of ideas in a group situation. Processing the statements and reacting to them progresses learning--- we naturally conclude that thinking develops through social interaction.

1.3.3   Thought and Language

If learning can only occur through dialog in a social context, how do we learn on our own? Dialog can be intrapersonal as well as the Vygotskian interpersonal. Through conflict that inspires dialog in ourselves, we work concepts through the process of internalization, through reading, inspiration, and experiment.

Mental functions begin as language and then internalize to be thoughts. Early in life we concentrate on interpersonal dialog, then, as we internalize more and more, we have more intrapersonal dialog, and our knowledge grows.

1.4   Summary

We see that learning is an active, social process. Through dialog we can achieve the zone of proximal development. Through internal dialog we proficient ones can set up zones for ourselves. Through signs, we can internalize concepts that present themselves in our activities.

How did Vygotsky study his theories? How do we best study activity to determine the most efficient ways to teach and learn? Clearly, we must travel to the workplace, or the learnplace, or the playplace to find out... because that is where the working and learning occurs. Studying activity requires an in-place the strategem of doing an ethnography.

By doing ethnographic research, we can see how humans and computers interact in the classroom and office. We can use that information to make those situations more efficient and more productive without placing undo strain on the human participants [Bødker, 1991].

Finally, we have found several open topics in the field of activity theory. First, one may study how novices become experts in the workplace. We may study how modern workers in a non-routine workplace can share priorities. We may study the emotional impact of the communication overload of an interactive workplace. And lastly, we may study the emotional impact of the increase of personal accountability in the modern workplace.


footnotes for chapter 1
F.GROK
The term grok, coined by Robert Heinlein in Stranger in a Strange Land (1961) describes the situation where one conceptualizes something so well that it becomes part of one's being--- that one lives a concept so well that one becomes it. This author places ``grok'' as the fifth level of internalization. For example, we can understand that life under the rule of a whimsical monarchy could be difficult but until we live in that situation, we cannot ``grok'' it.

Chapter 2.   Ethnography and HCI

Ethnography is the work of describing a culture. The main objective of this activity is to understand another way of life from the native point of view. It is used to refer to both the work of studying a culture and the end product ``an ethnography'' or written text. ``Field work, then involves the disciplined study of what the world is like to people who have learned to see, hear, speak, think, and act in ways that are different. Rather than studying people ethnography means learning from people'' [Spradley, 1979].

2.1   Background

The origins of ethnography can be traced back to anthropology. Its primary use is as a way of gaining insights into the life experiences of people whose daily lifestyle was different from those living in Western developed societies. The ethnographic approach has also been used in sociology where the focus is primarily on ``sub-groups'' within industrial societies (e.g., youth, addicts, ethnic groups, etc.). The practice of ethnography is based on four presuppositions:

Ethnographic representations include textual accounts and summaries, videotape, storyboard and images as well as artifacts generated by the work process.

2.2   Ethnography's Impact on HCI

The socialization of Ethnography and HCI began in the 1980s and was motivated by three emerging truths. First, there was a gradual awareness that the narrow focus on isolated individuals using computational artifacts was inadequate. Second, there was gradual agreement that human intelligence was socially constituted and achieved. And lastly, there was a growing interest in developing computer technologies that acknowledged and supported the cooperative nature of human activity. Ethnography was viewed by some as a way of gaining insights into the nature of human activity that could provide the grounds for the design of new technologies.

Prior to this most studies of human-machine interaction amounted to experimental or laboratory evaluations of individual performance on isolated tasks, analysis of the adequacy of selected user interface features, and comparisons between different user interface strategies. Such experiments provided valuable information such as the limits of individual performance capabilities, differences between novice and skilled users, and trade-offs among specific user interface choices. However, these technologies were no longer sufficient to handle the everyday experience of using (interacting with) new technologies.

Also, very few of the social resources routinely available to users are present in the laboratory setting. For example, Jeanette Blomberg (1988) discovered that co-workers experiencing trouble with an unfamiliar photocopier often consulted co-workers known to be more knowledgeable about the operation of the machine before checking with the available manual. In laboratory studies the tasks given to users are highly constrained and they must perform under unrealistically short timeframes in order to get the quantitative measurements. Thus laboratory studies alone were unable to extend our understanding of how the larger social environment impacts the experiences individuals have with new technologies. ``The HCI professionals often need to develop new, non-traditional techniques that yield an understanding of real customers solving real problems in the real world'' [Blomberg, 1995]. The idea behind this is that the better we understand existing work practices the better the design of systems tailored to the needs of their users. Ethnography was viewed as a way of getting insights into the nature of human activity that could provide the grounds for the design of new technologies.

In considering the relation between ethnography and systems design, there are three general directions:

Studies of work provide a deeper understanding of work in settings where new technology might one day be used. Their focus is on explicating how work is accomplished in varied and complex settings. Studies of technology already in use include the use of spreadsheet software and CAD systems as a means of designing more successful end user programming systems. For example, Blomberg [1995] studied the introduction of a computer-based design tool, Trillium, into the user interface design community of a technology development company. The implications of this study are that ``design'' continues after a system is introduced and that working systems include the hardware and software and the organization of social resources. The study also raises design issues regarding the ease to which users are enabled to modify a technology.

2.3   Future directions

The promise of integrating technology design and field studies of work is yet to be fully realized. Ethnographically-based design projects are just beginning to provide concrete examples of the value of bringing knowledge about specific work practices and work analysis closer, and of the requirements for creating an environment wherein the worlds of design and work analysis can come together.

One future challenge will be finding ways of moving from specific sites of ethnographically-based design to the design and development of generic products. The challenge will not just be that the requirements placed on a technology vary across settings of its use but that multiple voices will provide input that will shape the final product. Therefore it will be necessary to accommodate a wider range of perspectives (and participants) if we expect a final product to resemble design concerns that originate from sites of intended use.

Secondly, the issue of how ethnographically-informed design projects that take place over years and that involve many geographically distributed, people also need to be addressed. For example, are there times in the overall development effort when a close relation to end-user sites would be most productive?

Lastly, technology designed with the needs of the user versus the everyday political realities of the workplace will need to be addressed. Achieving the goals of ethnographically-informed design often requires confronting unequal distributions of power within the workplace.

2.4   Summary

Ethnography refers to a method used to develop understandings of everyday work practices and technologies in use. While not new to the fields of anthropology and sociology, ethnography has now found its way into technology and human-computer interaction. Ethnography and HCI promise to integrate technology design and field studies of work.


Chapter 3.   Participatory Design

Participatory design is a way of designing in which the systems engineer and those who will be affected by the new technology (usually the end-users) work together, thereby combining the knowledge of two different fields. Participatory Design is based on the philosophy that the user of a system is the one that knows best what requirements the system should meet, because he is the only one who really knows in what way the system will be used. Seen in this light, Participatory Design is directly based on activity theory, the theory which says that people's actions cannot be seen separately from the environment in which they work [see activity theory].

3.1   About Participatory Design

Participatory approaches first emerged in Scandinavia, in a time where there was an emphasis on workplace democracy, quality of working life, and skill preservation [Blomberg, 1995]. In Participatory Design the end-user is involved in the process of developing a system. This means that both the traditional systems designer and the user work together in all or some of the stages of the systems development cycle. The designer and the user share responsibility for the quality of the system they are designing.

This cooperation of two people with different points of view makes it possible for mutual learning to take place. In this process of mutual learning, the user learns from the designer about the technological aspects of designing, e.g. the possibilities and impossibilities of technology. The designer learns from the user about the application domain [CACM, 1993].

3.2   Different ways of Participatory Design

To operationalize the principles of Participatory Design, there is a large number of methods to choose from. These methods differ on two dimensions: (1) whether user participates in design activities or designer participates in user's world; (2) best phase in development cycle for applying the method.

Two examples of Participatory Design methods are:

3.3   Why Participatory Design

There are 2 main reasons for using Participatory Design. The first reason is pragmatic. In the life-cycle of a computer system, 60% to 80% of all problems are due to inadequate requirements specifications. Because Participatory Design combines the knowledge of two different areas and because it triggers discussion between these two areas, it can help better specify systems requirements.

The second reason for using Participatory Design is political. Many people believe that it is one's right to influence his own workplace. Participatory Design is one way to give people the opportunity to exert influence on the features of the system they will be working with in the future [CACM, 1993].

3.4   Goals of Participatory Design

Participatory Design has two main goals. Its first goal is the empowerment of workers, which corresponds with the political reason for using Participatory Design [see politics]. The second goal which Participatory Design aims to achieve is the improvement of the information system, both by better system requirements and increased user acceptance as a result of their early involvement in the design process.

3.5   Difficulties

There are several possible problems which might inhibit the process of Participatory Design from realizing one or more of its goals.

  1. The designer overrules the users. Because of the way computer systems are traditionally designed in which the designer is used to working mostly on his own and making his own decisions, it is not unlikely that the designer might take over most of the design process. Some reasons why this might happen are: (1) the designer not wanting to realize the possible valuable impact of users, and (2) a passive attitude of users, either due to their viewing the designer as superior to them or due to their feeling of not being able to really influence any major decisions.
  2. The designer uncritically does what the users want. It is a possible danger if the designer assumes that the users always know what is best for them, because this is not always the case. Sometimes users do not know enough about the possibilities of technology or are too conservative in their way of thinking about the work processes to reach the best solution on their own, without discussion with the system designer.
  3. Do users know what they need? The approach of Participatory Design is based on the assumption that users know best what their needs are. However, it still has to be shown that users are really the best ones to determine their own requirements. It is very well possible that this assumption does not hold, or does not hold in every situation.
  4. Users and designer have different objectives. Whereas the goal of the designer is to meet a deadline for implementing the system and the development process is his priority, the priority of users often lies within their own work and meeting their own deadlines.
  5. Participation does not necessarily further acceptance. Although one of the goals of Participatory Design is increased user acceptance [see goals], user participation does not always guarantee that the gap between user and designer will be bridged and hence that users will be more acceptant of the new system.
  6. Participation often stops after the design project is stopped. One of the reasons for using Participatory Design is the belief that people should be able to influence their own workplace [see politics]. Although Participatory Design enables people to influence the design of the systems they work with, very often the active involvement of users stops shortly after the design process is stopped, thereby ending the possibility for users to exert influence. Seen in this light, Participatory Design does not really meet its political goals.

Chapter 4.   Computer-Supported Cooperative Work (CSCW)

Greif and Cashman coined this umbrella-like term in around 1986 to unite a growing interest in the application of computer science and psychology to use of a computer in any setting with more than one user [Bannon, Schmidt, 1991].

The core issues of CSCW are as follows:

Cooperative work in this context defines when individuals work together using computer systems [Bannon, Schmidt, 1991]. Indeed, cooperative work has a history that predates computers. It is the entire focus of the Marxian philosophy in the 1800s!

Furthermore, groupware is a limiting term collecting all software that supports groups. CSCW claims more turf, however, by not restricting itself to the technology that supports multi-user applications, rather CSCW opens itself to the philosophies that allow integration of technology into the workplace [see activity theory] and the study of the workplace itself [see ethnography].

Coverage of this topic is minimal due to a complete course offering on CSCW in the Spring of 1997 by Professor Mary Beth Rosson.


Bibliography

BANNON91
Bannon, Liam J., and Kjeld Schmidt. ``CSCW: Four Characters in Search of an Author'' Studies in Computer Supported Cooperative Work: Theory, Practice and Design J. M. Bowers and S. D. Benford, editors. The Netherlands: Elsevier Science Publishers, B.V., 1991.

BLACKLER95
Blackler, Frank. ``Activity Theory, CSCW and organizations,'' Perspectives on HCI - Diverse Approaches, Chapter 10, pages 223-248. Andrew F. Monk and G. Nigel Gilbert, editors. San Diego: Academic Press, 1995.

BLOMBERG95
Blomberg, Jeanette L. ``Ethnography: aligning field studies of work and system design,'' Perspectives on HCI - Diverse Approaches, Chapter 8, pages 175-197. Andrew F. Monk and G. Nigel Gilbert, editors. San Diego: Academic Press, 1995.

BODKER91
Bødker, Susanne. Through the Interface: A Human Activity Approach to User Interface Design, ``Human Activity and Human-Computer Interaction.'' Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1991.

CARROLL96
Carroll, John M. ``Encountering Others: Reciprocal Openings in Participatory Design and User-Centered Design,'' Human-Computer Interaction, Lawrence Erlbaum Associates, Inc. Volume 11, pp 285-290, 1996.

CACM93
Communications of the ACM, Volume 36, Number 4, June 1993, Special issue on Participatory Design.

DIXONKRAUSS96
Dixon-Krauss, Lisbeth. ``Vygotsky in the Classroom,'' White Plains, New York: Longman Publishers, 1996.

JS79
Spradley, James P. The Ethnographic Interview, New York, Holt, Rinehart and Winston, 1979.

Appendix A.   Glossary


Goto: [1] [1.1] [1.2] [1.2.1] [1.3] [1.3.1] [1.3.2] [1.3.3] [1.4] [2] [2.1] [2.2] [2.3] [2.4] [3] [3.1] [3.2] [3.3] [3.4] [3.5] [4] [A]

ETD-to-HTML formatting system
etd2html (prototype) Version 3.0 Beta, November 10, 1996
This document formatted on Tue Dec 3 19:29:27 EST 1996.

Comments and questions? etd@etd.vt.edu.