By Tyler Waite
Last updated August, 2005
Activity Theory (AT) provides a descriptive framework for the
HCI professional through which they can better
understand and classify the cognitive, physical and social processes involved in
performing a specific task, and how those tasks can be related to a larger
motivating activity. AT illustrates how
actions and processes are divided and shaped by the larger community that is
involved in accomplishing a specific activity.
By applying the framework of AT to the accomplishment of a specific goal
within the framework of an overarching activity the HCI professional is forced
to consider how the individual user’s goal directed actions are influenced by,
relate to and feed into the larger community of which the individual is an
inextricable part. AT theory also
illuminates how human knowledge and physical artifacts are refined and automated
through a continual process of interaction and assessment. AT theory stresses the importance of
involving representatives of the target community in the process of designing
any type of computer mediated artifact so that the HCI professional has an
accurate understanding of the rules and tools that the community has developed
so that they can be accurately reflected and supported by the software
package. AT is a descriptive theory not
a predictive theory (Nardi, 1995), but by
understanding the cyclical components involved in any activity the HCI
professional can better predict what user groups the software should support as
well as some of the information needs of these user groups.
Activity theory was developed to illustrate the role of
society in shaping the mind of the individual, in contrast to psychoanalysis and
behaviorism. During the late 1980’s HCI
researchers noticed a similar problem with the research practices of the HCI
community which has been built upon the foundation of experimental and cognitive
psychology (Kuutti, 1995). Despite HCI’s
strong scientific foundation, the numerous journals, and hundreds of books that
had been written on HCI, the research was not driving practice (Kuutti, 1995).
Instead some of the biggest breakthroughs in interface design were being
made without any “explicit use of psychology” (Carroll, 1991, pg. 1). The controlled setting of classical
psychology experiments was not able to explain the larger issues that were
influencing how people used and interacted with computers or why one piece of
software was preferred over another.
The Human Factors approach to computer interface design viewed the human
as a component of a system that had behaviors that could be documented,
predicted and accurately modeled (Kuutti, 1995). It failed to take into account the role of
the community, personal motivation, or a specific setting on the individual’s
performance. Because of this researchers
like
Activity theory originates from the work of German philosophers Kant, Hegel, Marx and Engels, as well as the writings of Russian cultural-historical psychologists Vygotsky, Leont'ev, and Luria.(Engestrom, 1999). According to Leontjev (1978 as cited in Aboulafia, Gould, and Spyrou 1995) AT emerged as a way to explain the developmental processes where by a person is shaped by and shapes their environment through activities. It emerged in reaction to the natural science or materialistic view and the science of the spirit or idealistic view (Leontjev, 1978 as cited in Aboulafia, Gould, and Spyrou 1995). Through activities people work to achieve their idealistic or utopian visions by transforming the social and material world that surrounds them. Vygotsky developed the foundation of activity theory between 1920-1930. Vygotsky’s writings were heavily influenced by Soviet Marxism and Marx’s historical materialism (Bedny et al., 2000). The basic concept of this theory is that all human activities are mediated by culturally created signs or tools. Through external interactions with these signs the internal mental state of the individual is transformed (Aboulafia, Gould, and Spyrou 1995).
This framework has been used by Engineering Psychologists in
the
In AT the activity is the high level motivating concept. It is the general term that describes what the individual or group is trying to accomplish and typically indicates what outcome they are working towards. Some examples of an activity are fishing, learning to play piano, learning to write in kindergarten. Inside these activities are actions. Actions are performed in order to accomplished specific goals that will lead to the individual or group achieving the desired outcome of the activity. After each action is performed the situation is assessed to determine if the goal has been achieved. Actions related to fishing would be figuring out where to fish, loading the car, driving to the fishing hole, baiting your hook, catching a fish, cleaning the fish, and driving home with the catch. Some of the actions related to learning to play the piano would be finding a teacher, finding a piano to practice on, practicing every day, buying song books, and paying for lessons. Actions related learning to write would be going to class, writing the alphabet, having the work corrected, learning how to spell different words, learning how to put the words together. An activity can become an action. If you were on a week long camping expedition fishing may be just one of the actions involved in preparing dinner.
Actions are made up of operations. Operations are behaviors that have become so well learned they do not require conscious effort to execute. Operations are automatic responses to perceived conditions of the current state of the object with respect to the actions and goals that are trying to be fulfilled. The perceived difference between the current state of the object and the desired outcome provides the motivation for the individual or group to develop goals and actions to transform the object into the desired outcome. A central principle of AT is that any activity can, in theory, become an action, or even become an automated operation. This is easily illustrated when we consider the previously mentioned example of a child learning the alphabet. Initially all of the child’s effort is focused on learning how to control the pencil so that it will make the desired mark on the paper. As they become more proficient at this they begin putting letters together to spell simple words. Soon the child is writing sentences with the words they have learned. Once these skills are mastered the activity of making letters or writing a word is no longer an activity. It is an automatic operation that the student performs to accomplish more complex activities such as writing a book report and one day perhaps to write their own book.
AT theory is an iterative model. As suggested by the examples above, the process of any activity involves a cyclical movement between the operations that are performed on the object of interest and assessing the outcome of those efforts. There is a constant evaluation and re-evaluation of the object against the conceptual model of the desired outcome. This feedback loop however is not limited to one individual. All activities have a social component. Since everything the human mind learns is shaped by a social matrix, and since nearly everything human beings do is dependent on the presence of a social structure, the role of the society or on a smaller scale the community in the accomplishment of an activity cannot be ignored. Through the community the activity is broken up into manageable parts that can be focused on by people specializing in that aspect of the problem. The child learning the alphabet needs a teacher to show them how to make the letters and to evaluate their work. There are also school boards and state sponsored organizations that help to determine the curriculum that will be taught to the child. The child will likely have class mates that they can consult with and compare their work to. The child will have a bus driver or parent to get them to school, and parents or guardians to feed and clothe them. The very language the child is learning is also a product of hundreds of years of refinement and social processes. It reflects the socio-political history of the area where the child lives. From the seemingly simple task of a child learning their letters, we can quickly see through the framework of AT that it is actually quite complex.
One of the primary roles of the community in AT, is to create rules. Rules inform the individual about the object that is the focus of the activity. Rules inform the individual about what the desired outcome is. Rules inform the individual about their place in the community and how they are to act as a member of that community. Rules also instruct the individual in how they should work on the object. Even though AT demonstrates how the individual is shaped and molded by the social matrix in which they are situated, the individual does have free will to decide which rules to apply and how to apply them. The individual also evaluates the rules that they receive and may suggest revisions to the rules, or new rules, to the community. Rules can be both illuminative and restrictive. They are illuminating when they provide the individual with a way of viewing the problem they are working on. They are restrictive when they prevent the individual from searching for alternative rules for solving the problem. When a rule is applied to an object it becomes a tool. In AT tools can be either a physical artifact such as a hammer, saw, or computer, or a tool could be an idea, a language, or a theoretical framework. Similarly the object that is the focus of the activity can be either a physical object like wheat that needs harvesting, or a mental problem such as finding a mutually agreeable political solution between opposing parties.
Figure 1. Engestom’s classic model of Activity Theory shows the relationship between the subject or individual, the object and the community, as well as how rules, tools, and the division of labor are used in the transformation of the object into the desired outcome. (diagram from Kuutti, 1995).
The model in figure 1 is very useful for visualizing the ways the different components of AT relate to and interact with each other. However, individuals rarely have only one object that they are working on. In the words of Engestom (1999, p.20) “Human activity is endlessly multifaceted, mobile, and rich in variations of content and form.” The individual’s role in a community may change depending on the goals that need to be accomplished at a particular point in the transformation of the object, and the individual always belongs to more than one community. The individual may be a programmer, a manager, a father, a soccer coach, and a husband. The demands of these other processes will have an impact on how well the individual can use the tools the HCI professional has designed for them. It should also be noted that a tool can be an individual or a group of individuals as would be the case with the CEO of a company. The object the CEO is focused on is running a company and transforming the company into a profitable enterprise. In order to do that the CEO assembles a hierarchy of trained employees to perform the day to day operations of the company under his direction and guidance. Within the company there are also managers and vice presidents that help assess the day to day operations of the employees that they manage. Similar to the CEO they also rely on the people they manage to work on their specific part of the company objective.
As noted earlier one of the problems of previous research on human machine interaction was the attempt to view humans and human actions as discrete, predictable components of the system. However as noted by Engestom (1999, p.32) “Actions are not fully predictable, rational, and machine-like. The most well-planned and streamlined actions involve failures, disruptions, and unexpected innovations.” In any complex social system there will be competing goals, limited resources, differing values, and a variety of desired outcomes. Engestrom calls these conflicting forces within activities contradictions. Through AT the researcher may be able to identify some of these underlying contradictions that may produce failures, disruptions or necessitate innovations, and by identifying them be more effective at developing ways to mediate the effects of the contradictions or encourage the production of innovations to overcome the contradictions.
When researching various social groups it is inappropriate to impose personal assessments regarding the superiority or inferiority of one social group compared to another social group. What one person may see as primitive and backward another may see as sustainable, and eco-friendly. However just as it is a mistake for the researcher to impose their own value structure on their evaluation of another social group, Engestrom (1999) points out that it is also a mistake to ignore the social values that the group uses. Social values help decide what objects are worked on, how labor is divided, what rules should be imposed, and what outcome the society is working towards. Rather than seeing one type of society or social activity as primitive and another as advanced the researcher should try to understand the goals and values of the group and how their actions and activities support or contradict those values.
How then can AT be used to inform HCI and interface design? By applying the AT model to the target user population the HCI professional is forced to consider the larger complexities of any job that extend beyond just the single user interacting with the system. By placing the work that that computer program is supporting in this broader framework the HCI professional will be able to see what forces may be involved in deciding whether a particular software solution is adopted or not. Through AT the HCI professional will be more aware of the need to consult existing rules and practices in the application domain so that they can better support the knowledge and culture of the user’s community. They will explore what value systems the community has established to direct their activities and facilitate their decision making process. The cyclical nature of how people interact with and refine their tools as illustrated by AT points out the need for an iterative design process, and the development of an interface and computer architecture that can grow and change as the needs of the community change. AT also demonstrates the importance of the community in shaping and refining the tools that they use to change object problems into outcome solutions, and the importance of involving representatives of the community in the design process. If the software package that the HCI professional develops for a community does not support the existing rules, values, goals, and tools of the community or if it fails to support effective and efficient interaction between labor divisions it will not be adopted as a tool by the community. Thus when developing a computer application the HCI professional should have a firm understanding of how the target community is currently accomplishing their work and some understanding of the history that lead to the current practices. It is likely that the target community will have developed numerous rules, methods, and value structures for approaching the problem that are not apparent to the novice observer. Where one community may value efficient processing of information and a uniform no-nonsense interface, another may value a lively, colorful, fun and perhaps even somewhat ambiguous user experience.
Computers have already established themselves as one of the
most important tools mankind has for transforming the world. Computers are
connecting people together into communities that are not bound by geographic
boundaries but by common ideas and goals. Computers are automating processes
that used to take hours or days. Computers are allowing individuals and
organizations to have a clearer understanding of the object they are interested
in and to test out various possible outcomes and ways to achieve that outcome.
Computers are able to disseminate knowledge and rules rapidly to the community
members. Computers are facilitating more efficient transfer of knowledge
between the various divisions of labor. Computers are helping us heal our sick,
grow our crops, and build our homes. Computers play an important role at each
stage of the activity process. They are an object to be transformed by the
computer science fields. They are modified and changed to meet the needs of
various divisions of labor. They facilitate the creation of communities. They
accelerate the dissemination of rules. They are a powerful tool for the
individuals to transform their work object and to then communicate these
transformations back to the community. Computers have embedded themselves at
every level of human activity. Computers are also reshaping our moral and
social values as they facilitate access to and discussion with people who have
radically different views than our own. AT provides a powerful framework for
understanding those activities. It provides the HCI professional with a
framework for converting rich ethnographic data into illuminating descriptions
of transformational processes. The traditional usability lab can be viewed as
an exploration of how well the individual user is able to use the software as a
tool to transform the object that is defined by the tasks that the HCI
professional provides them. Instructional technology professionals can use AT
to think of ways to support the classroom activities being transformed into
automated operations. CSCW professionals can use AT to think of how the work
object is divided between the different workers and how to best support those
actions.
The theory and language of AT may seem daunting and foreign to many designers. Also applying it to design may seem like it would be a time consuming and lengthy processes. However the structure of AT can also be used as part of a mental exercise to think through what is involved in any activity that the designer is attempting to support; or to classify and categorize ethnographic data that they may have already gathered through user interviews and observation. For this exercise we will use the framework of AT to look at issues related to developing a mobile computing tool to assist people with low vision in a museum setting. The author of this document has recently conducted research on such a device. The knowledge gained from that research will be used as the data source for this exercise.
First of all what is the activity we are trying to support? Exploring an art museum.
Who is the individual we are designing the tool for? A visually impaired patron.
What tools will they need?
Understanding of the languages
available on the device.
A way to know where they are.
A way to know how to get from where they are to where they want to go.
A way to know what is on exhibit.
A way to interact with the tool that contains the information they need to
access to learn about the artifacts on display.
What are the objects that they are working on?
A museum that they wish to explore.
Lack of knowledge about artifacts.
Spending their leisure time.
What is the transformational outcome they are working toward?
Having successfully explored the
museum.
Having learned about the artifacts on exhibit in the museum.
Having experienced an enjoyable outing with friends and family.
Who is the community?
The immediate community is the museum staff, and the
patrons exploring the museum.
The larger community may extend to curators, board of directors, state federal,
and corporate funding agencies, as well as organizations set up to support the
needs of people with vision impairments.
What is the division of labor in the community?
The museum has people at the information desk to assist
people with questions.
The museum has people at the ticket counter to let people into the museum.
The museum has docents to assist patrons with their questions.
The museum has curators to create and manage the exhibits.
The museum has people who build the exhibits.
The museum has people that create the information available at the museum.
The museum has people that create the signage that helps people learn about the
artifacts and navigate the museum.
The museum has security guards to protect the artifacts.
The museum has maintenance staff to make sure the facility is clean and in
working order.
Patrons interact with each other to decide what to look at.
Patrons interact with each other and docents to understand what they are looking
at.
Patrons interact with each other and museum signage to decide where to go in the
museum.
Patrons use their schedule, and the schedule of the museum to decide how much
time they will spend at the museum.
What are some typical rules in a museum?
For Patrons:
Do not touch the artifacts unless a sign says
it’s ok.
Speak softly.
Walk don’t run.
Hours of operation.
Price of
admission.
Off
limit areas, and public areas.
For Staff:
Assist Patrons with their questions.
Prevent
patrons from damaging artifacts.
Dress in a
professional manner, perhaps in period costumes.
Arrive at
museum 1-2 hours prior to museum opening to patrons to ensure everything is in
order.
What are the motives for this activity?
For Patrons:
To have fun with friends
and family.
To
learn about other cultures and people.
To enjoy beauty and artistic creations.
For exercise.
For
relaxation.
For Staff:
To educate the patron about
art and culture.
To inspire
the patron.
To protect
and preserve rare artifacts.
To create a
safe and friendly environment.
To
support the information needs of a more diverse patron population.
What are some of the contradictions inherent in this activity?
Low vision patrons desire to touch artifacts may be in
opposition to museums goal of protecting and preserving artifacts.
Staffs desire to work normal business hours may conflict with patrons’ available
free time or the free time of person who can drive them to the museum.
Museums desire to present an unusual and visually complex environment may
interfere with the patron’s ability to orient to and navigate within the museum.
Transitory nature of museum exhibits makes it difficult for low vision patrons
to learn layout of galleries.
Artistic design of labels may cause them to have low contrast levels or unusual
font styles making them difficult to read.
Limited space for label information necessitates the use of small font sizes, which are difficult for low vision patrons to read.
Patron’s ability to see artifacts is hindered by glare of lighting off display case that the museum uses to protect and preserve artifacts.
What are some of the actions the device could support?
Locating their position in the museum.
Locating bathrooms, elevators, galleries, exits, café, gift shop, stairs, etc.
Locating artwork.
Learning about artwork.
Enlarging the artifact.
Locating people such as docents, or partners.
What are some of the goals that vision impaired patrons have?
To have a comparable experience to that of sighted
patrons.
To be able to explore the museum independent of a friend, tour guide, or
partner’s assistance.
To be able to decide what they want to learn about and how much they want to
know.
To be able to envision the objects more clearly either with
their limited sight or in their minds eye.
To learn about the history behind the artifact.
To learn about the artist that created the artifact.
To share an enjoyable experience with a friend or partner.
What are some of the conditions that the patron may encounter?
Not knowing where to find a particular piece of artwork, or
place in the museum.
Not knowing the relationship between their current position and where they want
to go.
Not being able to read the signs.
Not being able to see the artifact clearly.
Not being able to see obstacles in their way (e.g. a bench or rope and
stanchion).
Having trouble navigating museum due to crowds.
Small children, unaware of their impairment, running in front of them.
Not knowing if it is ok to touch the artifact.
Becoming lost or disoriented.
What are some of the operations that the patron may need to perform?
Searching for the location of a
specific artifact, type of artifact, or gallery.
Searching for the location of facilities such as restrooms, information desk,
gift shops, or café.
Identifying their current position in the museum.
Identifying the best route to take to get to the desired destination.
Identifying potential obstacles in their path.
Searching for the location of a friend.
Searching for the location of an exit.
Zooming an image of an artifact.
Playing an audio file about the artifact.
Increasing the font size of the information on the artifact.
Adjusting the color of the font and background to maximize visibility.
Controlling the volume, pitch, and speed of the audio file.
Browsing the artifacts around them.
Indicating which artifact they want to know more about.
Stopping the information being presented to them.
What activity level support is the device going to provide?
Tool- Mobile computing device supports low vision patrons
ability to freely explore museum.
Object- Mobile device provides a means for assessing how much of museum has been
explored.
Individual- Mobile computing device helps them overcome visual impairments that
prevent independent exploration.
Rules- Mobile computing device can be used to deliver rules based on context.
Community- Mobile computing device can be used to share information between
patrons and between the patrons and the museum staff
to improve the museum experience.
Division of Labor- The device acts as a personal docent or tour guide, and information desk. The device could be used by museum curators and staff to identify areas that are popular or problematic.
What action level support is the device going to provide?
Tool- Device will provide an
interface (audio, tactile, or visual depending on patron need) for accessing
information about surroundings.
Object- Device will provide an way to enlarge the
artifact to make it more visible
Individual- Device will be personalized to the individuals needs.
Rules- Device presents rule reminder when it detects specific context (e.g.
It is ok to touch the artifacts in this gallery).
Community- Device provides a way for patron to request assistance.
Division of Labor- Device will provide a way for patron to locate assistance or
assistance to locate patron.
What operation level support is the device going to
provide?
Tool- Device will automatically update information it presents based on
patrons location in the museum.
Object- Device will provide information about the artifacts and surroundings.
Individual- The information the device presents will be tailored to match the
patrons interest, background, and past experiences
at the museum or museums using the device.
Rules- Device will alert patron if they are attempting to enter a restricted
area.
Community- Device will support sharing of information between the patron and
their friends as well as with the larger museum community
through a digital message board.
Division of Labor- Device will require the creation of new jobs and operations
to maintain the database and information delivery network.
What are the contradictions that may create barriers for implementing this
solution?
Large amount of data that needs to be created contradicts with the museums limited staff time and financial resources.
Cost of purchasing technology, and developing and maintaining system contradicts with patron’s unwillingness and inability to pay extra for service.
Desire of patrons to have a small lightweight device contradicts with the need for the technology to have sufficient memory or processing power.
Desire of patrons to have large screen contradicts with their desire for the device to be lightweight.
Desire of museum to ensure that the technology is rugged
and durable contradicts with patron’s desire for device to be lightweight.
From this set of questions and answers it is clear that AT can be a useful framework for outlining some of the major issues that need to be considered when developing a mobile device to support the exploration of a museum by people with low vision. It can just as easily be applied to any other design process. AT will not tell you how the interface should look, or what specific technology to use, but it will provide you with a list of issues the technology should address and a clearer sense of the social framework in which the technology will be used.
There are a large number of links on the web about activity theory. Here are a few of them.
Activity Theory by Yvonne Rogers and Mike
Scaife
http://www-sv.cict.fr/cotcos/pjs/TheoreticalApproaches/Actvity/ActivitypaperRogers.htm
What is Activity Theory
http://carbon.cudenver.edu/~mryder/itc_data/activity.html
The Center for Activity Theory and Developmental Work
Research
http://www.edu.helsinki.fi/activity/
International Society for Cultural and Activity Research (ISCAR)
http://www.iscar.org/
Enlarging the
Research Object: An Introduction to the use of
Activity Theory as a Framework for Human-Computer Interaction Research
http://www.comp.glam.ac.uk/pages/staff/srharris/presentations/Introduction%20to%20Activity%20Theory_files/frame.htm#slide0024.htm
Activity Theory as a Basis for Design for All
http://www.stakes.fi/tidecong/213hyppo.htm#4.%20Tools%20offered%20by%20activity%20theory%20for%20defining
Activity Theory – an introduction
http://osiris.sunderland.ac.uk/~cs0car/hci/3_con_at.htm
While AT theory provides us with a useful framework for thinking about and identifying the larger societal influences on the individual’s conception of their tasks and how to use specific tools to accomplish their larger activity driven goals; as Nardi (1995) states it is a descriptive theory not predictive and so does not dictate how the descriptive findings should be applied to the design of the interface. The effectiveness of the theory is dependent on the researcher and whatever community representatives are included in the design process being fully aware and able to explicate what tools they use, how they are applied to the object, as well as role of various social groups on the activity, and what rules these groups impose on the process. Thus the application of AT to HCI does not ensure that the system it is used to help design will be free of all interaction problems. Still the cyclical nature of human interaction with, and refinement of, its tools, as illustrated by AT, indicates that design is never finished. Design is an ongoing dialogue between the individual and the larger society as they assess the results produced through interacting with the tools in moving them towards the accomplishment of their motivating activities.
Aboulafia, A., Gould, E., & Spyrou, T. (1995). Activity theory vs. cognitive science in the study of human-computer interaction . Proceedings of IRIS .
Bedny, G. Z., Seglin, M. H., & Meister, D. (2000). Activity theory: history, research and application. Theoretical Issues in Ergonomic Science, 1(2), 168-206.
Carroll, J. M. (1991). Designing interaction: Psychology at the Human-Computer Interface. Cambridge: Cabridge University Press.
Engestrom, Y. (1999). Activity theory and individual and social transformation. Y. Engestrom, R. Miettinen, & R.-L. Punamaki (Eds.), Perspectives on Activity Theory (pp. 19-38). Cambridge, UK: Cambridge University Press.
Henderson, A. (1991). A development perspective on interface design and theory. in J. M. Carroll (Ed.), Designing Interaction; Psychology at the Human-Computer Interface (pp. pp. 254-268). Cambridge: Cambridge Univ. Press.
Kuutti, K. (1995). Activity Theory as a potential framework for human-computer interaction research. in B. Nardi (Ed.), Context and Consciousness: Activity Theory and Human Computer Interaction (pp. 17-44). MIT Press.
Leontjev, A. N. (1978). Activity, Consciousness, and Personality. Prentice-Hall.
Nardi, B. (1995). Context and Consciousness: Activity Theory and Human-Computer Interactions. Interactions of the ACM , 2(4).
Nardi, B. A. (1996). Activity theory and human-computer interaction. in B. A. Nardi (Ed.), Context and Consciousness Activity Theory and Human-Computer Interaction (pp. 7-16). London, England: MIT Press.
Rogers, Y., & Scaife, M. (1997) Activity Theory [Web Page]. URL http://www-sv.cict.fr/cotcos/pjs/TheoreticalApproaches/Actvity/ActivitypaperRogers.htm.