Activity Theory and Web-based Training

David Peal and Brent Wilson
 

Peal, D., & Wilson, B. (2001). Activity theory and web-based training. In B. Khan (Ed.), Web-based training (pp. 147- 153). Englewood Cliffs NJ: Educational Technology Publications. Also available: http://ceo.cudenver.edu/~brent_wilson/acttheory.html

 

How is it that children in some cultures learn to do complex mental calculations, identify myriad plant types, and use star patterns to pilot canoes over long distances, all without the supposed benefits of formal schooling? Activity theory seeks to account, among other things, for the effectiveness of everyday learning environments. Researchers are beginning to identify how activity theory can inform the design of learning environments as well (Jonassen & Murphy, 1998). In this article we provide an overview of activity theory and show how this theory can serve as a framework for creating one particular type of designed environment: Web-based training (WBT). We also suggest ways in which designers can address the difficulties inherent in implementing realistic learning environments on the Web.

Activity Theory Essentials

Activity theory starts not with the individual learner, but with the activity system, a larger and more realistic unit of analysis. An activity system consists of a group, of any size, pursuing a specific goal in a purposeful way. For example, doctors practicing preventive medicine in a health-maintenance organization make up an activity system (Cole & Engestrom, 1993). Students in different countries collaborating on a project to build a Web site about the Himalayas make up another activity system. A telephone sales unit of a precision-parts distributor provides another example (Laufer & Glick, 1996). Even seemingly isolated activities are usually embedded in a larger system, as in the case of collaborating researchers who must negotiate differing approaches, align schedules, agree on a division of labor, and coordinate their actions with colleagues, editors, and others.

 Figure 1. Components of an activity system. See: http://www.edu.helsinki.fi/activity/6.htm

These examples can be analyzed into the elements shown in Figure 1, adapted from a diagram created by Finnish researcher Yrjo Engestrom (Cole and Engestrom, 1993, p. 8). The central relationship is between the individual participant (or subject) and the activity system's purpose (or object); this relationship is not direct, but is mediated by artifacts (tools). Participants are usually part of communities, a relationship mediated by rules for acceptable interactions. Communities in turn help accomplish system purposes and outcomes through a division of labor. The model depicted in the figure has the advantage over other situated-learning approaches of allowing for the identification of an activity system's elements and for the comparison of dynamic systems (Nardi, 1996). Activity systems are, in fact, in constant flux, always subject to change through the intrusion of new participants, purposes, and tools. Training itself can introduce change, in part by helping an activity system achieve its purposes in new ways, in part by revising participants' understanding of the activities in which they are engaged.

The Centrality of Tools

By bridging the worlds of participant and purpose, tools (depicted at the apex of the figure above) make activity possible in the first place. Tools can be both physical (notebooks, books, software) and cognitive (concepts, language, notational systems). Physical tools shape what people can and cannot do, as Norman (1988) shows in his critique of poorly designed doors, faucets, and other everyday things. Cognitive tools likewise enable and constrain activity. Instructional design procedures in the training world, for example, prescribe a way of doing (and not doing) design, thereby shaping associated ways of thinking (and not thinking) about design. Cognitive tools are of special importance because they free us from immediate stimuli, providing access to the past, to situations not physically present, and to imaginary worlds (Cole, 1996, pp. 121-122). Consider, for example, the centrality of tools for engineers, who could not function without tables, regulatory guides, design software, and a vast conceptual apparatus shared with other engineers (Perkins, 1993). For the engineer, tools can serve as "functional organs" experienced almost as personal properties (Kaptelinin, 1996, p. 51).

An activity, then, is simply what the activity system does: people using tools for more or less well-defined purposes. Any activity consists of deliberate actions, which can be further decomposed into automatic operations (Leont'ev, 1974). In the case of telephone sales, for example, order-processing constitutes an activity carried out through deliberate calculations (actions) and automatic order-writing (operations). These three levels (activity, action, operation) are situationally defined and subject to change. For management, order-processing is but a lower-level action in a larger corporate activity system. And we've all experienced how otherwise-ignored operations can become disrupted. When software crashes (ceasing to be a tool and forcing itself into consciousness as a troublesome object), automatic operations are halted as one tries to figure out which actions are required to resume work (Nardi, 1995, p. 75).

Everyday Learning in Activity Systems

To account for the relation between learning and human development, Lev Vygotsky, the Soviet psychologist who laid the groundwork for activity theory, developed the concept of zone of proximal development (ZPD) in 1934; the term has been popularized in the U.S. since the translation of his Mind and Society in 1978. Vygotsky defined the ZPD as the distance between a child's current development (as measured by independent problem solving) and that child's potential development (as measured by what can be accomplished "under adult guidance or in collaboration with more capable peers"; Vygotsky, 1978, p. 86). Implicit in this definition is a prescription for adults, teachers, and others to engage learners one at a time, at the limit of their potential. Under guidance, with practice, novices "gradually increase their relative responsibility until they can manage" on their own (Cole, 1985, p. 155). Skills, rules, knowledge, and tools are thereby internalized, forming the basis of the cognitive tools used in problem-solving and self-directed learning.1

Vygotsky's concept has helped anthropologists explain how, in non-Western cultures, complex skills such as weaving and midwifery pass between generations (Lave & Wenger, 1991; Rogoff, 1990). In the American classroom, educators have even designed zones of proximal development (e.g., Newman, Griffin & Cole, 1989; Campione et al., 1984). Such designs have at least four elements:

  1. Learning activities that are part of real or simulated activity systems, with close attention to the tools and interactions characteristic of actual situations;
  2. Structured interaction among participants;
  3. Guidance by an expert; and
  4. Eventual surrender of learning control to increasingly competent learners.

"Guidance by an expert" requires a more active role than suggested by "guide on the side," today's jargon for the teacher who no longer merely purveys information in a didactic fashion. To be effective, instructors must be sufficiently expert in their domain to judge individual learning needs, sufficiently experienced as instructors to adjust dynamically as needs change, and sufficiently sensitive as people to continuously juggle the novice's and expert's perspectives, with the steady goal of nudging the novice toward mastery (Wertsch, 1984; Newman, 1997; Wells, 1996). In the ZPD learning takes place not "in the head," through a one-way process of knowledge transmission. Some theorists discount individual cognition altogether, arguing that thought and intelligence are embedded in and "stretched across" the larger structures of activity (Pea, 1993; Lave & Wenger, 1991; Salomon & Perkins, 1998).

A final general point: activity systems are not normative, and the "natural" way of learning is not necessarily the best. Instead, the concept of activity system helps the designer identify the elements and dynamics of often undesigned systems, which can be rife with contradictions--differences, for example, among participants about purpose, division of labor, and tool selection and use. By analyzing the troubles of current activity systems, designers can help emerging ones develop successfully.

Implications for WBT

WBT can focus at any level of an activity system--activity, action, or operation. The lower the level, the looser the ties to a specific activity system and more transferable the skill, since activities are unique to systems while operations can be applied in many systems. WBT itself can be seen as both a tool and a simulated activity system within which participants are introduced to and refine their use of actions and operations. Purposive, coordinated learning can be organized and led by an instructor, automated by a computer-based tutorial, or created by the learners themselves, depending on the WBT design. In this respect, the Web, like any other medium, affords a variety of interactions in support of activity, depending on the use to which it is put.

Using WBT merely to teach routine behaviors (operations and actions) fails to do much more than automate current practice or lower costs, thereby shoehorning what is often procedural knowledge into a hypertext format. The novel contribution of activity theory is to focus attention on the ways in which WBT can support higher-level activity by enabling reflection on current business processes and facilitation of new teams, rules, and divisions of labor. WBT in support of activity can be delivered as a course, but perhaps more effectively as part of everyday activity systems. Further, what Kuutti (1996) says of information technologies applies to WBT as well. WBT can make new, distributed activities possible and allow organizations to take on new purposes (pp. 35ff).

Regardless of the level of training, activity theory can inform the key aspects of WBT design: what is being learned, how training takes place, and what is expected of the learner.

What is being learned

For activity theorists, community knowledge lies in activities and tools, not syllabi and performance objectives. WBT designers should arguably spend more time designing activities that use the resources of activity systems and less time worrying about whether some sort of purified expertise has been "captured" from a subject-matter expert. One could safely assume that it hasn’t, since the needed expertise will likely emerge from activity itself. Ideally, assessment should include a way of tracking students' day-to-day activities and judging the adequacy of their participation.

WBT makes the most sense when activities are already on the Web, such as learning HTML, designing e-commerce sites, and programming in JavaScript. Also, it makes sense when activities are easily represented in virtual environments, such as researching a subject or collaboratively writing a paper. In addition, the discrete information required in professional continuing education (in medicine and accounting, for example) lends itself well to WBT, especially when an instructor is not required.

In other cases WBT should arguably be integrated with real-world training. A WBT course on data-entry skills could pair novices with experienced colleagues who know the tricks and shortcuts as well as the rule-based techniques. A WBT course on coaching for managers could include Web-based tutorials on rules and skills, but significant time should be spent in actual performance settings, with learners working alongside experienced practitioners. Here they can learn the ways in which experts recognize and grapple with the messier problems that emerge in everyday situations.

How training takes place

American educators prefer scaffolding to zone of proximal development, but both terms denote the guidance learners need as they acquire new skills. In traditional activity systems, the community and especially the adult guide provide much of the scaffolding for novices. The same can be true in WBT environments, which have the benefit of providing access to distributed expertise (Brown et al., 1993). However, because the Web medium is presently limited to text, sound, video, pictures, fewer cues and far less everyday information can be shared than in face-to-face encounters; "live" communications usually communicate no more than words. This inherent deficit can leave learners feeling isolated and adrift.

With some ingenuity in implementation, WBT can compensate for media limitations. Instructors will themselves need guidance in the art of scaffolding as they learn to use e-mail, chat, and message boards to engage students regularly, substantively, and supportively as required. An effective WBT environment will also include a variety of performance supports such as job aids and reference sheets to help learners pick up community practices. Above all, realistically presented practices will allow participants to master tools efficiently and thoroughly--not just as skills, but as "personal properties."

Virtual learning communities even hold some advantages. Because WBT interactions can be archived, students can consult previous work samples. A searchable and browsable library of such samples (as in a Lotus Notes discussion database) can make a workgroup's history more retrievable than it ever has been. By making a community's knowledge tangible, collaborative visualization technologies help participants see and resolve differences, building joint understandings (Edelson, Pea, & Gomez, 1996). In the text world of Usenet newsgroups, compilations of frequently asked questions (FAQs) have successfully met this need for years by representing shared knowledge in a compact format, efficiently bringing novices up to speed and reducing the burdens placed on the community's most experienced members. Such digital archives do not, of course, encapsulate past human interactions in quite the way Vygotsky envisioned. Even so, networked storage, retrieval, and visualization capabilities can help compensate for the serious limitations of WBT and effectively embody virtual workgroups.

What is expected of the learner

Traditional and virtual communities differ in the expected roles of students. In WBT environments, instructors cannot micro-monitor students’ progress, or even be aware of learning needs on a daily basis­a real obstacle to implementing ZPDs on the Web. Students must be encouraged to take the initiative and seek out needed resources. Because WBT constrains opportunities for the direct observation and modeling of expert participants, designers should expect failures--not so much in content transmission as in learners' inadequate adaptation to new roles.

For WBT to be effective, students need to see learning goals in terms of problems to be solved. The metacognitive skills of self-monitoring, question-asking, decision-making, and self-evaluation must be constantly in play. A key task for instructors in this environment is both to cultivate such skills and to encourage students to model experienced peers. Since many students do not arrive well-prepared in these areas, WBT environments must include support for high-level skills. Scaffolding thus extends beyond support of technical skills and into reflection, troubleshooting, and self-guided training.
 

Activity theory and WBT: Some design guidelines

Cultivate the social. Communities are central to normal training practices and should also be central to WBT. Even individualized modules need to be conceived as part of a broader network of interacting people.

Get people talking. Learning happens when people get practice using the community's tools, with all the associated jargon and tricks of the trade.

Provide continued scaffolding. Learners benefit most when they stay engaged at that "stretching" level, with support from multiple experts.

Tap inherent intentionality. Every activity system has its purposes, which inform all underlying actions and operations. WBT environments will be successful to the extent that they tie learners’ goals to these larger purposes.

Build around meaningful activity. The power of WBT lies in the cases, projects, and problems--not in objectives and content outlines. Activity-based WBT draws on the power of authentic collaborative activity to convey what needs to be taught.

Work with the available tools. Some WBT tools will be richly simulated; others will be lower fidelity, text-only case studies. Learners and trainers need to improvise and adapt as they accustom themselves to the online environment. Even so, impressive learning can result through creative use of online tools and resources.

Use collaboration to allow different levels of participation, support, and learning. Not every learner has to engage in the same activity in the same way; such standardization can kill a creative activity system. Learners need to be encouraged to find a place where they can learn--at the center, periphery, or somewhere in between. In collaborative activities directed at challenging, complex problems, room can be made for everybody.

1The apprenticeship process presented schematically here, and only characterized at a high level by Vygotsky himself, has been closely analyzed by Barbara Rogoff (1990) in her cross-cultural study of child development--a more readable account of apprenticeship than Lave & Wenger's popular theoretical study of situated learning (1991). Though dealing with children, Rogoff's work bears close reading by designers of training for adults. Gordon Wells (1996) provides many insights into teacher-student dynamics in the classroom ZPD. Kieran Egan (1997) has developed Vygotsky's hints about the genesis of cognitive tools into a highly imaginative theory of human development and education.

References

Brown, A. et al. (1993). Distributed expertise in the classroom. In Salomon, G. (Ed.), Distributed cognitions: Psychological and educational considerations (pp. 188-228). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Campione, J. et al. (1984). The zone of proximal development: implications for individual differences and learning. In Rogoff, B. & Wertsch, J. (Eds.), Children's learning in the Zone of Proximal Development (pp. 77-91). San Francisco, Washington, London: Jossey-Bass Inc. Publishers.

Cole, M. (1985). The zone of proximal development: where culture and cognition create each other. In Wertsch, J. (Ed.), Culture, communication, and cognition: Vygotskian perspectives (pp. 146-161). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge, Mass.: Belknap Press of Harvard University Press.

Cole, M., & Engestrom, Y. (1993). A cultural-historical approach to distributed cognition. In Salomon, G. (Ed.), Distributed cognitions: Psychological and educational considerations (pp. 1-46). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Cole, M., & Wertsch, J.V. (1996). Beyond the individual-social antinomy in discussions of Piaget and Vygotsky. Human Development, 39, 250-256

Edelson, D., Pea, R., & Gomez, L. (1996). Constructivism in the collaboratory. In Wilson, B.G. (Ed.), Constructivist learning environments: Case studies in instructional design, (pp. 151-164). Englewood Cliffs, NJ: Educational Technology Publications.

Egan, K. (1997). The educated mind: How cognitive tools shape our understanding. Chicago and London: University of Chicago Press.

Engestrom, Y. (1996). The activity system. On the Web at http://www.helsinki.fi/~jengestr/activity/6b0.htm.

Jonassen, D. H., & Murphy, M. (1998, February). Activity theory as a framework for designing constructivist learning environments. Paper presented at the meeting of the Association for Educational Communications and Technology, Saint Louis MO.

Kaptelinin, V. (1996). Computer-mediated activity: Functional organs in social and developmental contexts. In Nardi, B. (Ed.), Context and consciousness: Activity theory and human-computer interaction (pp. 45-68). Cambridge, MA and London, England: MIT Press.

Kuutti, Kari (1996). Activity theory as a potential framework for human-computer interaction research. In Nardi, B. (Ed.), Context and consciousness: Activity theory and human-computer interaction (pp. 17-44). Cambridge, MA and London, England: MIT Press.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, UK; New York, Port Chester, Melbourne, Sydney: Cambridge University Press.

Laufer, E., & Glick, J. (1996). In Y. Engestrom and D. Middleton (Eds.), Cognition and communication at work (177-96). Cambridge UK; New York; Melbourne: Cambridge University Press.

Leont'ev, A. (1974). The problem of activity in psychology. Soviet Psychology, 13(2), 4-33.

Nardi, B. (1996). Studying context: A comparison of activity theory, situated action, and distributed cognition. In B. Nardi, (Ed.), Context and consciousness: Activity theory and human-computer interaction (pp. 69-102). Cambridge MA and London: MIT Press.

Newman, D. (1997). Functional environments for microcomputers in education. In Cole, M., Engestrom, Y., & Vasquez, O. (Eds.), Mind, culture, and activity: Seminal papers from the Laboratory of Comparative Human Cognition (pp. 279-291). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Newman, D., Griffin, P., & Cole, M. (1989). The construction zone: Working for cognitive change in school. Cambridge: Cambridge University Press.

Norman, D. (1988). The design of everyday things. New York, London, Toronto, Sydney, Auckland: Doubleday/Currency.

Pea, R. D. (1993). Practices of distributed intelligence and designs for education. In Salomon, G. (Ed.), Distributed cognitions: Psychological and educational considerations (pp. 47-87). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Perkins, D. N. (1993). Person plus: A distributed view of thinking and learning. In Salomon, G. (Ed.), , Distributed cognitions: Psychological and educational considerations (pp. 88-110). NY: Cambridge University Press.

Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. New York and Oxford: Oxford University Press.

Salomon, G., & Perkins, D. N. (1998). Individual and social aspects of learning. In P. D. Pearson & A. Iran-Nejad (Eds.), Review of Research in Education, 23, 1-24.

Wells, G. (1996) The zone of proximal development and its implications for learning and teaching. Presented at the 2nd Conference for Sociocultural Research, Geneva, September 1996. To appear in G. Wells (in press), Dialogic Inquiry: Towards A Sociocultural Practice and Theory of Education.

Wertsch, J. (1984). The zone of proximal development: Some conceptual issues. In Rogoff, B. and Wertsch, J. (Eds.), Children's learning in the "Zone of Proximal Development" (pp. 7-18). San Francisco, Washington, London: Jossey-Bass Inc. Publishers.

Author Notes

Brent Wilson is associate professor, Information and Learning Technologies, School of Education, University of Colorado at Denver. He can be reached by e-mail (brent.wilson@cudenver.edu), regular mail (Campus Box 106, P. O. Box 173364, Denver CO 80217-3364), telephone (303.556.4363), and fax (303.556.4479).

David Peal has written computer books for Sybex, Osborne/McGraw-Hill, Lycos Press/Macmillan, and AOL Press. He has a Ph.D. in history and is working toward an MA in Educational Technology Leadership at George Washington University. He can be reached by e-mail (davidpeal@aol.com), telephone (301.718.7355), and fax (301.718.7329).

For more information: Martin Ryder has collected a large set of Web links relating to Activity Theory (http://www.cudenver.edu/~mryder/itc_data/activity.html).