Activity Theory

Sam Rajkumar
School of Library and Information Science
Indiana University
Bloomington, IN 47405-3907

srajkuma@indiana.edu

Summary
Activity Theory is a cultural-historical theory of activity initiated by Russian psychologists S.L. Rubinstein and A.N.Leontiev in the early 20^th century. It seeks to explain social and cultural work practices by relating them to the cultural and historic context in which the work activity is taking place. The basic unit of analysis as proposed by Activity theory is an activity. The rapid expansion in information technology and subsequently in HCI has lead to the focus on the end user. Designers often have difficulties in identifying the user’s problems or the complex organizational conditions of the implementation of new technology, where the user’s needs are the main focus. Activity Theory can provide a framework to first, reinterpret the concept of user needs, and second, to study the development of activity systems and related to processes that are innovative in nature. User needs should be understood both historically and as something to be constructed collaboratively and Activity Theory can play an important role in achieving this. This paper will present a case study that uses Activity Theory for studying user needs and the success of the theory at achieving this.

Motivation
In the 1970s computers were entering into work life in a major way and there was soon a need to create a better working relationship between users, their computers and their environment. In the early 1980s concerns evolved for the maintenance and development of skills of the involved workers and for technological alternatives. (Bertelsen and Bodkaer, 2003) Many of these concerns were addressed using insights from social psychology and industrial sociology. With the advent of the personal computer, researchers turned to the new field of HCI to answer questions about human behavior, interaction with the environment, influence of culture, history and interaction with other humans. Earlier cognitive science- based theories could not adequately address issues that were cropping up related to better understanding of work context, levels of expertise, role of the artifact and focus on individual user in an increasingly workspace characterized by co-operative work among many people. These shortcomings made the HCI community look outside of cognitive science based HCI for a framework that could address these issues and also offer concrete conceptual tools that could be used in designing better work practices using computers. Activity Theory shares with earlier and also contemporary approaches an interest in actual material conditions of human activity. However these approaches sometimes lack a concern for motivations of actions and also an in-depth level of analysis. Activity theory on the other hand shares the idea that a hierarchical analysis of human action is valuable with means and ends analysis, task analysis alike. (Bertelsen and Bodkaer, 2003). However the focus is on flexible hierarchies rather than on static breakups of wholes into parts. Also unlike say Norman’s (1988) seven stage model, Activity Theory emphasizes that activity takes place at different levels at the same time and not necessarily in sequence. (Bertelsen and Bodkaer, 2003).

Thus the conceptual framework of Activity theory has been seen as a way of providing a means of analyzing the actions and interactions with artifacts within a historical and cultural context – an approach which was lacking in the existing cognitive theories (Rogers, 2004). Activity Theory is thus seen as being useful to design interfaces and systems that take into account the context of use and thus can play an important role in better user experiences.

History and Background
Activity Theory was a result of efforts by Russian psychologists to develop a new psychology based on Marxist philosophy. The Marxist philosophy which came into being after the Russian revolution of 1917 redefined the view of psychology as it were and sought to imbibe psychology with Marxist thinking. The early efforts are mostly associated with L.S. Vygotsky whose work is seen as providing the foundations to this new way of approaching human behavior and psychology. An early result of Soviet psychologists was the "principle of unity and inseparability of consciousness (i.e., human mind) and activity". The meaning of this principle was “that human mind comes to exist, develops, and can only be understood within the context of meaningful, goal-oriented, and socially determined interaction between human beings and their material environment.” (Bannon, 1997)

The psychologists associated with the development of Activity Theory are S.L. Rubinstein and A.N. Leontiev. Between the two it was Leontiev’s work that received more prominence and consequently became better developed (Rogers and Scaife, 1997). Sergey Rubinstein, was the first who formulated the notion of human action as a unit of psychological analysis. Later Alexey Leontiev, who was a student of Lev Vygotsky, developed the conceptual framework which is known as Activity Theory. Since then Activity Theory became a leading theoretical approach in Russian psychology leading to many studies being carried out that used this approach.

Activity theory over the next few decades became part of the larger research community. The focus has moved from the psychological to also take into account the social context as well. In the recent past the HCI community has raised concerns about the lack of a unifying theory and perspective of human computer Interaction. In this regard Activity Theory has gained popularity as an approach that take into account the cultural and organizational context and also directly focuses on day to day practical work thus providing an alternative unifying approach. The researchers who were instrumental in bringing Activity Theory into Human Computer Interaction (HCI) and Computer Supported Cooperative Work (CSCW) include Bodker (1991), Engestrom (1993) and Nardi (1996).

Explanation
Activity Theory is not strictly speaking a ‘theory’ as theory is commonly understood. It seeks to provide a set of basic principles which provide a broader conceptual framework with which to understand the goal oriented, socially and culturally influenced work practices of humans using computers. Activity theory uses an ‘activity’ as the basic unit of analysis. Human activities are driven by human needs in order to achieve certain purposes. The activity in question is usually mediated by tools and this concept of mediation of activity using artifacts is a central theme of this theory. The role of the context in the understanding of the individual actions must is also crucial as human activity is shaped by the context and also the history (i.e. the present and the past contexts). Thus historical understanding of work context is also needed to fully understand its contemporary context fully.

In order to understand Activity Theory this explanation will address the following:
a) The levels of activity and the Activity Theory Framework
b) The role of artifacts in Activity Theory
c) The basic principles of Activity Theory
d) Methodology – proposed methods

a) The structure and levels of activity
Activity Theory seeks to answer the basic question of ‘what is activity?’ in the following way. The basic modeling of activity is done in a hierarchical structure with three distinct levels. Leontiev (1978) proposed three levels in an activity: the activity level, the action level and the operation level. Starting from the top here is a brief description of each level of activity.

(i) Activities consist of distinct actions or series of actions, which in turn consist of operations. In order for there to be an activity there has to be first a human motive that gives rise to the need for this activity. Activities are realized as individuals and cooperative actions, and chains and networks of such actions which are related to each other by the same overall object and motive (Decortis, Noirfalise and Saudelli). Activities can thus be differentiated based on their motives. An example of an activity is building a house.

(ii) Actions are basic components of activities. They are subordinate to the larger activity. The goal of an action is a conscious goal that guides the action. Different actions may be undertaken to meet the same goal. If the activity is building the house then transporting building materials and laying the foundation are at the action level.

(iii) Operations are ways of executing actions. They correspond with the way of goal achievement in that they are the concrete conditions required to achieve the goals. Operations which may start out as conscious acts may over time become routinized and unconscious. For the activity of building the house at the operational level using the jackhammer or nailing the various pieces together are operations.

These three levels of activity and their relationship can be seen in figure 1.

Figure 1. Three levels of activity (Kutti, 1996)

The model however is not a rigid one but allows a flow between levels. Thus operations over time can become unconscious actions. Also the line between action and activity is difficult to define as goals and motives can often overlap or be used interchangeably. Activity theory thus allows the constituents of activity to dynamically change as conditions or context change. All levels can possibly move both up and down as the need may arise.

The basic Activity Theory Framework- mediated structure of an action
In the original framework an activity is oriented by an object (also called motive). This concept refers to the "raw material" or "problem space" at which the activity is directed and transformed into the outcome. The activity is acted out by a subject, the individual or group that is the focus of the activity. And the relationship between the subject and the object of activity is mediated by a tool. A tool can be anything used in the transformation process, including both material tools and tools for thinking.

Figure 2. The basic activity Theory Framework (Kuutti, 1996)

This theory development was first used to explain individual activities. Nevertheless, most of human activities are collective ones taking place in rich communal and social environments. The earlier structure, presented above is too simple to show the systemic relations between an individual and his or her environment in an activity. Thus a systemic model (Figure 3), proposed by Engeström (1987), is used to explain collective activities and cooperative work. An activity is undertaken by a human agent (subject) who is motivated toward the solution of a problem or purpose (object), and mediated by tools (artifacts) in collaboration with others (community). The structure of the activity is shaped and constrained by cultural factors including conventions (rules) and social divisions (division of labor) within the context. Engeström emphasizes the mediational role of the community and that of social structures including the division of labor and established procedures.

Figure 3. Engestrom’s extended Activity System Model. (1997)

All the elements of this system are continuously changing. The human beings not only use instruments, they also continuously adapt them, consciously or otherwise. They obey rules, but also transform them. They follow rules of division of labor but in doing so constantly help evolve them as well. Transformation is then crucial to this model – and the interplay between the various elements constantly leads to the various new outcomes being created.

b) The role of artifacts in Activity Theory
Human activities always contain artifacts. These artifacts are not only physical as traditionally understood. They can be sign systems, instruments, procedures, machines, methods, law’s, processes and anything else that shapes work. The key role of artifacts in Activity Theory is that they have a mediating role in that there is rarely a direct relationship between elements of an activity but that these are mostly mediated by artifacts. Artifacts in turn get created and transformed during the development of the activity. They also are influenced and shaped by the culture and history of use. Thus tools are not only used by humans when they enter a work environment but also are created and transformed by such used throughout any activity. The concept of mediation is crucial to understanding activity as a whole.

c) The basic principles of Activity Theory (Kaptelinin and Nardi, 1997)
The adoption of Activity Theory in HCI is a set of basic principles that constitute a general conceptual system. The basic principles of Activity Theory include the hierarchical structure of activity, object-orientedness, internalization/externalization, tool mediation, and development. The following are some of the more important principles:

Hierarchical structure of activity
The unit of analysis is an activity directed at an object which motivates activity, giving it a specific direction. Activities, actions and operations make up the three levels of activity.

Object-orientedness
The principle of "object-orientedness" states that human beings live in a reality that is objective in a broad sense: the things that constitute this reality have not only the properties that are considered objective according to natural sciences but socially/culturally defined properties as well.

Internalization/externalization
Activity Theory differentiates between internal and external activities. It emphasizes that internal activities cannot be understood if they are studied separately from external activities, because they transform into each other.

Mediation
Activity Theory emphasizes that human activity is mediated by tools. Tools are created and transformed during the development of the activity itself and carry with them a particular culture - historical remains from their development.

d) Activity Theory Methodology
Review of the existing literature of Activity Theory shows the existence of a conceptual model for explaining levels of activity and the relationships between them. Surprisingly there is no clear step by step method to ‘carrying out activity theory in practice. There exist many structured approaches but no one unified approach. Some of the more prominent methods as suggested by researchers are as follows:

i) Engeström (1993) has noted that Activity Theory does not offer ready-made techniques and procedures for research; rather, it is a conceptual tool. These conceptual tools must be adapted to the specific nature of the object being studied. Engeström himself describes three principles of activity theory:
• Making the collective activity system as the unit of analysis
• Searching for internal contradictions as the driving force behind disturbances, innovations, and change in the activity system
• Analyzing the activity and its constituent components and actions historically.

ii) Nardi (1996) describes four methodological considerations for activity theory:
• Allow for research time frames long enough to understand users' objects. The changes over time in objects and their relationships must be studied.
• Pay attention to broad patterns of activity rather than narrow episodic ones that fail to reveal the overall direction and import of an activity.
• Use varied sets of data collection techniques including interviews, observations, video, and historical materials, without undue reliance on any one method.
• A commitment to understanding things from the users' viewpoint – user centering the inquiry process.

iii) Bodker (1996) examined Activity Theory with a focus on artifacts. The observation focuses on breakdowns and changes in focus caused by shifts in focus or activity.
Regarding new technology, Bodker (1991) argues that the interface can only be revealed through its use in real situations. The interface which is otherwise ‘invisible’ only becomes evident when breakdowns occur, and these are useful for subsequent improvement by the designer.

A case study - Articulating user needs in collaborative design
The case study (Miettinen and Hasu, 2002) looks at the collaborative design of a high-technology product, a neuromagnetometer used in the analysis of the activity of the human cortex. The producer, Neuromag Company tried to transform the device from a basic research instrument (working prototype) into a means of clinical practice. The device was implemented in basic research and the challenge of the producers was to transform it into a tool of clinical practice and create a clinical market. This kind of transition from design to user activity is a critical phase in an innovation process during which it will be resolved whether the artifact is going to be established as a tool in user activity. Several approaches in the past have tried to address this issue - Economics of innovation, actor network theory, but each has had it’s limitations.

In this case study, the researchers use activity theory to study the simultaneous transformation of an artifact, the local network related to it and user activities. Some other approaches have tried to capture user needs but each approach has had its limitations. For example the user cannot have an articulated need of something that differs radically from their present practice and knowledge (Lindell, 1991; Vicario & Troilo, 1998). Second, user needs are mainly analyzed as individual preferences and opinions on product characteristics (Heiskanen & Niva, 1996). This is a limited view, especially when the users are work communities or other collective groups of users. The idea of a contextual analysis of the usability and human-computer interaction takes a step in that direction (Beyer & Holzblatt, 1998). The analysis of usability either experimentally, or only as a situated human-machine interaction, is also too narrow an approach. User needs should be understood both historically and as something to be constructed collaboratively in the course of the design and implementation process. In this case study Activity Theory was used to analyze the users, artifacts and the context in which they are situated. The researchers used Activity Theory to propose a three-level analysis to capture the multidimensionality of the constructing of the user needs: (1) the analysis of use value in a historical perspective, (2) the complementary means (software, standard data, scientific verification) needed for the implementation and use of the product, and (3) the analysis of the situated use of the product. (Miettinen and Hasu, 2002)

The research team organized a seminar where all the end users (or actors) were part of the collaborative design process. It was found that different actors of the innovation network have different points of view, interests and goals in relation to the artifact and these can be seen as contradictions. The case study shows that the concept of a local activity system is fruitful in studying the dynamics of this type. The interests and points of view of the participants became understandable on the basis of the history and current problems of their activities. Outside of this, understanding user needs has severe limitations. Activity Theory was an integral part of this case study in understanding the different points of views, interests and goals in relation to the artifact.

Pros and Cons
As an alternative theory, the conceptual framework of Activity theory is a good way of providing the means of analyzing the actions and interactions with artifacts within a historical and cultural context – an approach which was lacking in the existing cognitive theories. In this context it is a descriptive tool and not a prescriptive one. Activity Theory can be useful to design interfaces and systems that take into account the context of use and thus could play an important role in better user experiences. However it is not easy to find any such examples where Activity Theory has been used. It offers concepts and a language to discuss and describe human activity in a manner that has a real world context though sometimes the distinctions of levels of activity are not very clear. If designers could understand the Activity Theory view of human activity i.e. a holistic view of activity that takes into account context and culture, it would be beneficial to a better understanding of user needs.

Some of the principles of Activity Theory such as internalization/externalization are difficult to find in recent research practice and case studies using Activity Theory. It thus seems that researchers who use Activity Theory only use a part of what the theory offers. While a common complaint in the HCI community is that the language is difficult to understand and use of certain descriptors is confusing, it might be useful for the community at large to arrive at a simpler, more readily usable set of terms within the same framework. This might ensure that Activity Theory can be more useful and accessible to a wider audience.

The usefulness of using Activity Theory in the case study has been mentioned in the earlier section - mainly the interests and points of view of the users become understandable on the basis of the history and current problems of their activities. The transition from design to user activity in development of new technology is a critical phase in the innovation process. Using Activity Theory, this case study showed that the future or fate of the artifact as a tool in user activity can be evaluated fairly well. Activity Theory was able to show where the tool could be an enabler and where it could prove to be restrictive when understood from the viewpoint of the user’s history and problems with current work activities. The users current activities and problems provided a context of use against which to evaluate the future tool. Some of the limitations faced by the research team was in using the concepts in the articulation of the future potential needs, which are not yet already there or evident. Further, the data of this study consists mainly of the discourses in the user seminar, issues related to the situated use of the device was not very clear and the role of Activity Theory in such is not expressed.

In this case study, the dialogue with the participants of the innovation process emerged in a natural way out of the research process. Having interviewed the key actors at length, the researchers had a fairly good overall conception of the motives of the participants. The user seminar produced valuable data for the researchers. All this points to the need for extensive time and resources being spent- and the question is whether this kind of process can be followed in design practice where time and resources are allocated judiciously.

In terms of how exactly Activity Theory was operationalized in this study was not clear. This is a common problem in the practice of Activity Theory – how exactly it is used in case studies is often blurred. Never the less this case shows that Activity Theory can provide a framework to first, reinterpret the concept of user needs, and second, to study the development of activity systems that are created and related to processes that are innovative in nature.

Useful Links
Cooperative Technologies for Complex Work Settings Web site
http://www-sv.cict.fr/cotcos/pjs/cotcos.htm

Short overviews on Activity Theory:
http://carbon.cudenver.edu/~mryder/itc_data/act_dff.html

http://www.acm.org/interactions/vol2no4/depts/book.htm

An extensive web resource on Activity Theory includes readings, conferences and other resources:
http://carbon.cudenver.edu/~mryder/itc_data/activity.html

Special issue on CSCW - Activity Theory and the Practice of Design
http://www.ics.uci.edu/~redmiles/activity/final-issue.html

Center for Activity Theory and Developmental Work Research
http://www.edu.helsinki.fi/activity/

http://www.marxists.org/archive/leontev/works/1978/intro.htm
This is an on-line version of the introduction to the whole book Activity, Consciousness, and Personality (Leont'ev, 1978) with links to other relevant resources.

Comments
Personally from the perspective of someone who practices usability I would find it is useful for identifying problem areas outside of the one-on-one usability lab context. Lab observation is no replacement for an understanding of work in the real world context and Activity Theory provides a good framework for such analysis. The case study shows how this can be done, though how exactly this should be operationalized is not very clear.

Some of the other problems one faced are those cited in existing literature – that of the models, principle and terminology being difficult to use in everyday design work. In spite of reviewing several viewpoints on the levels of activity one still finds it difficult to distinguish between ‘action’ and ‘activity’. It does seem in certain complex organizational contexts what is an activity to one person or role may be an action to another. Further since these levels evolve and move up and down it would take a designers view over time to be able to track and understand these movements.

In conclusion, if Activity Theory is to be used more widely by designers of information systems we would need more Activity Theory based application ‘tools’ or more clearly operationalized methods such as Nardi’s Activity Checklist. Just as contextual design has been organized into a ‘how to’ method, it would be useful for researchers and practitioners to come up with a similar ‘how to’ method to applying Activity Theory. Finally greater discussion, debate and actual use of Activity Theory within the HCI community will further clarify and shape the role of Activity Theory as an alternative theory.

References

ActivitypaperBannon.htm on November 1, 2003

Beyer H. and Holtzblatt K., Contextual Design: defining customer-centered systems. San Francisco, Calif.: Morgan Kaufmann Publishers, 1998.

Bødker, S. (1991) Through the interface: A human activity approach to user interface design. Hillsdale, NJ: Lawrence Erlbaum

Bodker, S. (1996). Applying activity theory to video analysis: how to make sense of video data in HCI. In B. A. Nardi (Ed.), Context and consciousness: activity theory and human-computer interaction, Cambridge and London, MIT Press, 69-103.

Bertelsen and Bodker (2003) Activty Theory, HCI Models, Theories, & Frameworks: Toward a Multidisciplinary Science. Carroll, J (ed)

Decortis F., Noirfalise S., Saudelli B. Activity Theory as framework for cooperative work. Retrieved from http://www sv.cict.fr/cotcos/pjs/TheoreticalApproaches/Actvity/ActivitypaperDecortis.htm on October 29, 2003

Engeström, Y. (1987). Learning by expanding, Orienta - Konsultit, Helsinky.

Engeström, Y. (1993). Developmental studies of work as a testbench of activity theory. In S. Chaiklin and J. Lave (Ed.), Understanding practice: perspectives on activity and context, Cambridge, Cambridge University Press, 64-103.

Heiskanen, E. and Niva, M. (1996): A bird’s-eye view on users and usefulness. Take a look at the users. Seminar papers September 3, 1996. Helsinki: Finnish National Consumer Research Centre, pp. 5-21.

Kaptelinin and Nardi B.A. (1997) Activity Theory: Basic Concepts and Applications, CHI tutorial, 1997

Leont'ev, A. L. (1978) Activity, Consciousness, and Personality. Englewood Cliffs, NJ, Prentice Hall, 1978.

Miettinen, R & Hasu, M. (2002). Articulating user needs in collaborative design: Towards an activity theoretical approach. Computer Supported Collaborative Work, 11, 129-151.

Nardi, B.A. (1996). Activity theory and human-computer interaction. In B. A. Nardi (Ed.), Context and consciousness: activity theory and human-computer interaction, Cambridge and London, MIT Press, 69-103.

Norman, D.A. (1988) The Design of Everyday Things. MIT Press

Kuutti, K. (1996). Activity Theory as a potential framework for human-computer interaction. research. In Nardi, B. (ed), Context and Consciousness. Chapter 2, 17-44.

Rogers, Y. (2004) New theoretical approaches for HCI. ARIST: Annual Review of Information Science and Technology, 38.

Rogers Y. and Scaife M (1997). Activity Theory. Retrieved from http://www-sv.cict.fr/cotcos/pjs/cotcos.htm on October 25, 2003