Digest of Activity Theory Study

Nardi, Bonnie A. (19??)

Studying Context: A Comparison of Activity Theory, Situated Action Model, and Distributed Cognition.

Proceedings East West HCI Conference, St. Petersburg, Russia
System design will benefit from explict study of the context in which users work. The unaided individual divorced from a social group and from supporting artificts is no longer the model user. But with this realization about the importance of context come many difficult question. What eaxctly is context ? if the individual is no longer central, what is the correct unit of analysis? what are the relations between artifacts, individuals, and the social groups to which they belong. This paper compares three approaches to the study of context, activity theory, situated action models, and distributed cognition.

 Introduction

 A broad range of work in psychology (Leont'ev 1978, Vygotsky 1978, Luria 1979) anthropology (Hutchins 1991, Miller 1990) and computer science (Clement 1990, Mackay 1990) has shown that it is not possible to fully understand how people learn or work if the unit of study is the unaided individual with no access to other people or to artifacts for accomplishing the task at hand. Thus we are motivated to study "context" to understand relations between individuals, artifacts, and social groups.

 We mut develop an appropriate analytical abstraction that "discards irrelevant details while isolation and emphasizing those properties of artifacts and situations that are most significant for design (Brooks, 1991). Taking context seriously means finding oneself in the thick of the complexities of particular situations at particular times with particular individuals. .... How can we confront the blooming buzzing confusion that is "context" and still produce genfralizable research reults ?

 This paper looks at three approaches to the study of context from three dimensions (1) the unit of analysis (2) the categories offered to support a description of context and (3) the extent to which each treats actions as structured prior to or during activity. Situated Action Models.

 Situated action models emphasize the emergent, contingent nature of human activity. They do not deny that artifacts or social relations or knowledge or values are important, but they argue that the true focus of inquiry should be the "everyday activity or person action in a setting (Lave, 1988). As Suchman's (1987) statement that "the organization of situated action is an emergent property of moment-by-moment interactions between actors, and between actors and the environments of their action." Lave (1988) identifies the basic unit of analysis for situated action as "the activity of persons-acting in setting". a Setting is defined as "a relation between acting persons and the arenas in relation with which they act. In emphasizing improvisation and response to contingency, situated action de-emphasizes study of more durable, stable phenomena that persist across situations . A central tenet of the situated action approach is that the structuring of actiity is not something that precedes it but can only grow directly out of the immediacy of the situation (Suchman, 1987). To Suchman, the work in AI and cognitive science in which problem solving was seen as a "series of objective, rational pre-specified means to ends" (Lave, 1988), and work which over- emphasized the importance of plans, failed to recognize the opportunistic, flexible way that people engage in real activity. It failed to treat the environment as an important shaper of activity.

 Situated Action Models provide a useful corrective to these restrictive notions that put research into something of a cognitive straitjacket. Once one looks at real behavior in real situations it becomes clear that rigid mental representations uch as formulatic plans or simplistically conceived "rational problem solving" cannot account for real human activity.

Activity Theory

 In activity theory, the unit of analysis is activity. (subject, object, goal, actions, operations (Leont'ev, 1974).

 Operation becomes routinized and unconscious with practice (e.g, learning to drive a car and shifing gears). Operations depend on the conditions under which the action is being carried out. If a goal remains the same while the conditions under which the operation is to be carried out change, then "only the operational structure of the action will be changed" (Leont'ev,1974). (according to Churchman, change conditions can lead a purposeful subject to change his/her operation, but the goal could be the same.) Activity theroy holds that the constituents of activity are not fixed, but all levels can move both up and down (Leont'ev,1974).

 A key idea in activity theory is th enotion of mediation. Artifacts such as instrument, signs, and machines mediate activity and are created by people to control their own behavior. Artifacts carry with them a particular culture and history ( Kuutti, 1991) and are persistent structure that stretch across activities through time and space. Distributed Cognition

 Distributed Cognition asserts as a unit of analysis a cognitive sytem composed of individuals and the artifacts they use (Flor and Hutchins, 1991). It is concerned with structure - representations inside and outside the head-transformations these structure undergo.

 other major emphasis of Distributed Cognition is on understanding the coordintion among individuals and artifacts.

 Difference between them

 Treatment of motive and goal:

In activity theory, activity is shaped first and foremost by an object held by the subject. In fact we are able to distinguish one activity from another by virtue of their different object. An object/goal is the beginning point of analysis. In a sense, situated action models are confined to what activity theory would call the action and operation levels.

 But in situated action models goals are retrospective and reflexive" and constructed (Lave,1988). Activity and its values are generated simultaneously. It explicitly rejects the very idea of an object generating activity, asserting that object (goals) and plans are simply "retrospective reconstructions," " artifacts of reasoning about action " after action has taken place. This can be explained by the old behaviorism stimulus-response model.

Persistent structures

 What role do persistent structures such as artifacts, institutions, and cultural values play in shaping activity ?

 Discussion:

 All three approaches to the study of context have merit. The situated action model has provided a much-needed corrective to the rationalistic accounts of human behavior from traditional cognitive science. But take an example, three individual going on a nature walk, a birdwatcher, an entomologist, and a meteorologist. The same situation, even they can take same action (watch a bird). The observable action would be identical from e.g, a videotape as in situated action models, but what differs is the subject's intent, interest, goals.... If we do not consider the subject's object, we cannot account for simple things such as the birdwatcher. One find oneself in a claustrophobic thicket of descriptive detail, lacking concepts with which to compare and generalize. The lack of conceptual vocabulary, the constant appeal to the situation itself in its moment-by-moment details, do not lend themselves to higher order scientific tasks where some abstraction is necessary. (this is equally true for PD). Two problems in this model(1) they do not account very well for observed regularities and durable, stable phenomena that span individual situations; (2) they ignore subjective.

 Conclusion:

 Activity theory eem the richest framework for studies of context in its comprehensivess and engagement with difficult issues of consciousness, intentionality, and history. It is the most difficult frameworks to master. Human-computer interaction studies as a corpus of knowledge that identifies the properties of artifacts and situations that are most significant for design, and which permits comparison over domains, generates high level analyses, and suggests actual design. A creative synthesis of activity theory as backbone for analysis seems promising.

 Holt, R. G. and Morris, A.W. (1993) Activity Theory and the Analysis of Organizations Human Organization, Vol. 52, No.1. The Society for Applied Anthropology Engeström points out that the unit of analysis in accounting for emerging institutions is neither that which occurs in the individual mind (the cognitivist position), nor the structure of the organization (the functionalist position), but the activiy through which both are continuously generated. Central to vygotsky´s effort was an approach that denied the strict separation of the individual and the social were conceived of as mutually constitutive element of a single, interacting system; cognitive development was treated as a process of acquiring culture. Descartes´dualism is frequently characterized as an incommensurability between mind and body. Engeström explicitly conceptualizes "activity" as a minimal unit of analysis. Activity, defined as systems of collaborative human practice(Eng. 1988), becomes the generator of a continuously emerging context. This formulation rejects any view that holds context as a "given," that is, preset conditions to which participants respond. Were not the paradox that consumption necessitates production, and vice versa, activity would not exist. All activity systems are the result of the dynamic interplay of contradiction, need states and inevitable. The impetus to system change arises from four types (levels, or layers) of contradictions: primary (within each constituent component of the activity); secondary (between the constituent components of the activity); tertiary (between the activity itself and a culturally more advanced form of the activity); and quaternary (between the central activity and its neighboring activities) (Egn. 1987) An activity is not a homogeneous system. It always contains sediments of earlier historical types, as well as buds or foots of a future type. These sediments and buds are found in and between the different components of the activity system, including the mental models of the subjects." in the left of the paper, nothing useful except some description of contradictions involved in the explosion of Challenger which has shifted purpose from research to economoc benfit and introduce economic pressure (contradiction) into the system.

Engelstad, P. H. and Gustavsen, B. (1993)

Swedish Network development for Implmenting National Work reform strategy

Human Relation. Vol 2(46): pp 219-247
After 1970, the research on system design in Scandinavia has been focused on those affected by the changes as acting subjects rather than as passive entities to be seen as "objects" in experiments. One of the problems in this approach is how to achieve broad scope in the changes.

One of the reasons for this approach is to achieve "critical mass" of changes between organizations. The LOM (Leadership, Organization, Co-determination, in swedish) projct shifts from earlier Scandinavian work reform which focus on design work, less communication. The LOM project did not ignor design but emphasized very strongly the way in which the participants in a development process communicate., i.s good communication or good processes rather than good systems since communication exerts a major influence on the design process and consequently on what system the actors are able and willing to creat.

A main reason for this approach is the rcognition that today all successful productivity strategies rely on active and involved workers. there is no controversy; the issue is how to creat those structures and processes which can bring forth this activity and involvement.

It comes to focus more strongly on methodology and strategy and less on specific topics.

The dialogue-oriented conferences had provided visions, encouragement, a feeling "that we are all in the same boat and should work together. In this way a substantial expansion of the learning potential inherent in the clerical work roles is achived, and proceed quite far in the application of new technology.

 The starting point would for obvious reasons have to be with the communicative links between the researchers and practioners. It quite clear that the probability of generating successful project outcomes would depend on the existence of relevant competences and entreprenership as well as mutual understanding and trust among the participants.

 In action reasearch "learning by doing" is the main point, although the learning is to take place in the form of reflection on experience departing from theoretical premises of the type generally applied in reasearch. It was created to "diffuse" or "disseminate" changes from one workplace to the next.

 This tendency towards "every body being able to learn from every body" suggests that the development of a new forms of organization now can become more and more disassociated from such factors as technological basis or market conditions.

 

Greenwood, D. J. , Whyte, W. F. and Harkavy, I. (1993)

Participatory Actioon research as a process and as a goal

Human Relation. Vol 2(46): pp 175-192.
We argue that the degree of participation achived in any project is the joint result of the character of the problems and environmental conditions under study, the aims and the capacities of the research team and the skills of the professional researcher.

 We believe that participatory action research strongly encourages continuous learning on the part of professional researchers and the members of the organizations

involved. Participatory action research encourages integrative, interdiciplinary social science. Democracy in knowledge production gives the participants a stake in the quality of the results, increasing the reliability of information and likelihood that results will be put into practice. Participation is also important because self-management is a moral and political value that we seek to promote.

 participatory action research is an ongoing organizational learning process, a research approach that emphasizes co-learning, participation, and organization transformation. To be scientific, a model, following the parsimony principle, must endeavor to explain the largest amount of variance in the data with simplest possible models. The key is that we must seek the simplest possible, not merely simplest model. Since we believe that reality itself is a complex, multi-causal web of force that are historical dynamic . The social sciences have developed a kind of dialogical routine that permits academic and applied researchers to utilize each other as foils while doing as they please. Each sterotypes the other. Neither reads much of the other´s work. Each feels superior to the other. In this division of labor, each needs the other as the "straw man." (the dichotomization)

 

Holt, R. G. and Morris, A.W. (1993)

Activity Theory and the Analysis of Organizations

Human Organization, Vol. 52, No.1. The Society for Applied Anthropology
Engeström points out that the unit of analysis in accounting for emerging institutions is neither that which occurs in the individual mind (the cognitivist position), nor the structure of the organization (the functionalist position), but the activiy through which both are continuously generated.

 Central to vygotsky´s effort was an approach that denied the strict separation of the individual and the social were conceived of as mutually constitutive element of a single, interacting system; cognitive development was treated as a process of acquiring culture. Descartes´dualism is frequently characterized as an incommensurability between mind and body.

 Engeström explicitly conceptualizes "activity" as a minimal unit of analysis. Activity, defined as systems of collaborative human practice(Eng. 1988), becomes the generator of a continuously emerging context. This formulation rejects any view that holds context as a "given," that is, preset conditions to which participants respond.

Were not the paradox that consumption necessitates production, and vice versa, activity would not exist. All activity systems are the result of the dynamic interplay of contradiction, need states and inevitable.

The impetus to system change arises from four types (levels, or layers) of contradictions: primary (within each constituent component of the activity); secondary (between the constituent components of the activity); tertiary (between the activity itself and a culturally more advanced form of the activity); and quaternary (between the central activity and its neighboring activities) (Egn. 1987)

 An activity is not a homogeneous system. It always contains sediments of earlier historical types, as well as buds or foots of a future type. These sediments and buds are found in and between the different components of the activity system, including the mental models of the subjects."

 in the left of the paper, nothing useful except some description of contradictions involved in the explosion of Challenger which has shifted purpose from research to economoc benfit and introduce economic pressure (contradiction) into the system.

 

Osmo Saarelma
 The wotking Health center project, Helsinki, Finland
 & Yrjö Engeström
 Laboratory of comparative Human Cognition, University of California, San Diego.

New Work-New Tools: Information system development in the context of developmental wotk research

Research and development of information system in Scandinavian countries has for some years focused on interralations between works and system design. This approach has been called "Scandinavian tradition" or even "Scandinavian challenge" (Bjerknes, Enn & Kyng, !988).

 As we begin to see information system as tools of work, the object of design is no more a computer application but the work itself.

 The design of new tools was seen as an emancipatory process where workers learn new skills through participation in the design.

 When the system was implemented in the real workplaces, contradictions with the interests of employers and other professional groups emerged.

 A quit simple system was used as a tool which structured the work in a new way and provided the work community with a new communication tool. this created new work practices and forms of information system use which were not even anticipated when the system was developed.

 Concrete analyses of work as well as analyses of the works´conceptions of their work were scarce or absent in the Utopia and Florence project. In contrast, the Medea project tried from the beginning to analyze the actual work and concept concerning it. The project aims at developing a computerized decision support system for primary health care.

 Developmental Work Research (DWR) differs from the Scandinavian approaches described above in that it aims both at changing the work practice and at designing information system on the basis of systematic analysis of work. In DWR the object of analysis and development is the whoe collective activity system of the workplace. The concept of activity system was initially formulated by A. N. Leontév (1978;1981) as an important step in the evolution of the cultural -historical and mediational approach to mind initiated by L.S. Vygotsky (1978). The concept of activity system has been further elaborated and modeled in the work Engeström.

 Collective activity system: The whole triangle. Collective activity is manifested in individual goal-directed actions.

An individual goal-directed actions may be depicted as the uppermost sub-triangle (subject-tools-object).

 When an action is repeated and routinized it becomes an automatic operation. In an operation, the tool becomes an organic and self-evident extension of the acting subject, this implies that the uppermost corner of the the uppermost triangle is gradually lowered and the triangle collapses into a direct, seemingly unmediated line from subject to object.

tools: refer to external material artifacts as well as to signs, concepts and internalized mental models.

 Activity systems are in constant internal disequilibrium. They change and develop by resolving their historically evolving internal contradictions. For example, in medical care the problems of patient (the object of work) have been changing from strightforward diseases and traumas to more complex socio-psychological problems, and patient typically expect more information and discussion than before. The available tools for handling biomedical problems are not enough. A contradiction emerges between the novel object and the traditional tools. Such a systemic contradiction manifests itself at the action level in the form of disturbances. ... Thus, whatever new tools are designed, they are not aimed only at facilitating work as it is but at resolveing systemic contradictions tha give rise to the dilemmas. This means that all the components of the activity system (whole triangle) are transformed, not only the tool component. Such a transformation is seen as a process of expansive learning, and described as a cycle.

DWR approach is not likely to prioritize technically ideal information systems and tools. Changes in organization, in division of labor and in rules may be more rapid and dramatic than changes in tools. The main achievment from the view point of information systems is the foration of a collectively shared comprehensive perspective which guides the long term development of tools as integral components of activity system moving from its zone of proximal development.

Guohua, Bai. Dept. IES, Luleå University

Study of Activity Theory

1995.10.31-1995.11.01, Umeå University

 By: Prof. Yröj Engeström

1995.10.31, Kl 13:00

 Introduction (Yröj Engeström ):

 It is the third times for Yröj to visit Umeå, development work research, Training for change (tomorrow).

 the fundmental idea is mediation of human behaviour by cultural artifiats. Human being is always a cultural being. And human always focus on the object (subject or object world).

 In the earily 1980, Finnish reesarchers formulated the development work research as the applicaton of activity theory.

 AT has been focus on psychology problems, Vygovsky work become well known in the west in 1980'.

 Compuetr carries history, to understand the role of computer, you have to understand the hitory where they are used for history of the object and tools).

The principles of development work research:

 1) The colective activity as unit of analysis.

 2) Diversity and multi-voicedness as key characteristics of activity system

 3) historycity as fundation for understanding development dynamics

 4) Contradictions as driving force of development

 5) Expansive learning as basis of developmental methodology.

 An activity can never be ended, but actions do; activity is historically continued.

 Activity -------> Motive ----------> collective --------> Why ?

 Actions -------> Goal ----------> Individual or group ---------->What ?

 Operation ----------> Condition ----------> Organism or automatic device ----------> How ?

 In the perspective of the above levels, it is dangeous for designers to become rutin (only in operation level), because you forget to ask you why (motive) and people can do wrong things more efficiently.

 I have asked myself, how to make a line of an activity if we have to use it as unit of analysis.

 A collective acticity should be the unit for analysis, but how to draw a line of the unit; put this in another words, what make a distinguish between any two kinds of acivities. If it is motive, then how do we study motive? through actions ?

Take the hunting exemple, if we compare two hunting activities, ancient hunting (motiv for food) and nowdays hunting (motiv for fun), there are almost no differences in action and operation levels (they are all in cooperative in shooting, fritening, etc.), but qualitatively different from the activity level (motiv).

A dilemma for me is that we can find the different motives (two different activities) only outside the current hunting activity. We have to investigate what the hunters do after hunting. Probably, people will say I should expand the line of hunting activity to include those studies after the hunting (if I could still call it an activity). But I am sure this expanding will be never ended, or I could never find the line for the unit to my analyis. This line is especially difficult to draw when we have a real case. So, practically, how do we define an activity ? or should we define it ?

 Three generations of activity theory

 1) The first generation of activity theory is represented by triangle Subject, object mediated by tools.

 2) The second generation of activity theory is represented by triangle in which community, rules, and labor of division are included.

 3) The third generation of activity theory is to study the relationship between the central activity and its neighbor activities.

 Traditionally, psychological study focus on feedback between a subject (in most case, it is an individual)and outcome (which Singer called second order feedback, or I called it mental model reformation. Acording to activity theory, this is not enough, you have to consider to reconstruct the whole activity in which a another feedback process is involved, namely the first order feedback to Singer (feedback from outside of the pair subject and object). In this case, you have to consider to activly reform the world, not only reform your mental model. The first order feedback is not a black box strategy (as in the second order feedback to only change mental model), instead, we must open the black box and to reconstruct the inside of black box (rule, div. of labor, community, etc.).

 Neighbour activities could be, tool producing activity, rule producing activity, object activity, competing activity, collaborative activity.

 1995.11.01 Kl 10:15:

 Training for change: (focus on learning):

 Conflict and contadiction (problem for an approach which is based on aggrenment to develop). Activity theory is not such an approach which is based on consensus (to answer Kristo's question).

 Two kins of learning problem in practice.

 1) to exclude the important role of mediator and merely focus on the pair of learner and practice, which is so called "to drop into the water to learn swimming"

 ............ mediator (artifacts)---------

 Learner < ----------> Practice

 In this so called practical learning mediator is not attended, typically in the "situated learning legitimate peripheral participation" (Lavc & Wenger)

 2) "to cut the tree into the classroom", school learning in which practice is not attended. This kind of learning forget that the tree in the classroom is only a mediator for learning.

 ............ mediator (artifacts)............

Learner < ----------> Practice

 A good learning should be experimentation learning which combines theoretical hypothesis and practical testing on going circle among three points. To start from practical problem, and form hypothesis, and then test them with help of artifact (mediator). This can be shown as in the figure.

 The end


Return to Bai's Web