Wednesday, March 11, 2009

What theory in HCI is about

I am developing a theory of interaction (!). In the process of doing that, I am trying to categorize my theory. To do that I need to know what are the (common) categories of theories in HCI. So, this is where I need some help. I have decided that the categories should be based on what the theory is about, that is, defined by its object of study. As a first draft, it is possible to distinguish three kinds of theories with three distinct objects of study.

The first category are theories that has human computer interaction as the object of study, that is, theories that say something about the interaction between humans and interactive artifacts.

The second kind of theories are those that have the design process as a core object of study. Based on intuitive statistics (!) I think this is where we find the major part of theories in the field.

The third kind are theories that address how interaction and interactive technology changes society and environments, that is, theories that have the relation technology--society as the object of study.

So, what else? Any ideas?


Christian Briggs said...

Hi Erik,

How about theories that have to do with the relationship between the designer and the user - like Participatory Design?

Webfoot said...

How about visual and form theories?

xythian said...
This comment has been removed by the author.
xythian said...

Currently you have theoretical bins for ..

User <-> Technology
Designer <-> Technology
Technology <-> Society

And a suggested category of ...

Designer <-> User

However, what about Technology <-> Technology?

It seems odd to strip out any human element in a theory of HCI, but given the recent import and early development of ecological and material culture theories into HCI it might be interesting to consider theories of Technology <-> Technology.

Also, many ecological theories imply a more diverse notion of agency than HCI is used to which makes a Technology <-> Technology theory even more compelling.

Erik Stolterman said...

Thanks Christian, Webfoot and Xythian for your comments. They do help me. I will try to defend my three categories and argue that there no others :-)So, my answer back would be that all your proposals are covered in my three.

Christian: I see PD as one of many theories that has the design process as the object of study. The purpose is to improve the design process.

Webfoot: I would place visual and form theories as part of the human computer interaction category. These theories are focused on aspects of the designed artifact and are aimed at improving the interaction by knowing more about the artifact.

Xynthian:First I think the designer-user category is a sub-category to the user-technology as I replied to Christian. The technology-technology is not clear. I think ecological theories are still in the first user-technology category, just a refined sub-category, but it still has the relationship user-technology as the object of study. If there are other aspects of the techology-technology category I am not sure.

Thanks again!

xythian said...


I completely agree that ecological theories as they stand fall under the user <-> technology or technology <-> society categories.

However, as an intellectual exercise I can see something fruitful coming from considering technology<->technology theories. It feels unnatural to remove the human in such a way, but that unnatural feeling could be a sign of something interesting on the horizon. :)

Erik Stolterman said...


Yes, I agree. I think there are many other possible theories with other objects of studies out there, but my argument is that in the field, these three are the ones we can find today (I think). And the idea of theories where the object of study is devoid of humans is quite interesting!

xythian said...

So, in the good tradition of critical reflexive practice, I thought about inverting turning technology <-> technology and seeing where that leads.

What about theories of user <-> user? User <-> user theories though suffer from the same drawbacks as technology <-> technology. However, what if you extend user to include designers and researchers?

When you think about designer <-> designer or designer <-> researcher are you now looking at theories which discuss how the field of HCI should, could or does work?

I think one potential "user <-> user" theoretical area might be design research via Forlizzi and Zimmerman. Their theory isn't directly aimed at improving or understanding user <-> designer or designer <-> technology, but proposes a different way of doing HCI itself. A sort of meta-theory not aimed at interactions embedded into the lives of actual users, but a theory for how to operate at a higher, disciplinary level.

Erik Stolterman said...


Good comments...

I think that user--user sounds like sociology and not as hci or informatics.

As soon as you bring in designer and theories about designers, I would label it as the design process category, it is all about the process.

However, maybe there is a category that can be labeled research, which means that theories have HCI reseach as the object of study, and I think the Zimmerman paper on design as research is such a paper.



Balakrishna Chennupati said...

Hello Erik,

Maybe not so much a theory, but how about a design history of interaction artifacts. It will probably be a study of the combination of the qualities of the interaction and what processes were used to get there. Theories of how maybe a particular artist/designer or design movement influenced new design? Or maybe that's process?


shveta said...

Hi Erik,

This is a great research on theories that you are doing - will help in adding clarity to what design is as a discipline.

I'm thinking out loud here - I feel a need to have something that scopes out design in a way that is clearly understandable and accepted by all disciplines. All the time, I see conflicting ideas and hugely different perceptions that different people have about what design is and how it can shape a product. How about a theory that talks about, lets say, the relationship of design with business research, or product management?

tanykim said...

To me your categorization seems to be from micro(cognitive perspective?) to macro level following from HCI, to Design Process to technology--society?
Or it may be related to the spectrum from objectivism to subjectivism.

My humble opinion is positing the various theories along the axis above?

Jodi said...

Erik, great post, and I am late to the party due to my time zone. :> What about characterizing the theory as it shapes the field? Early theories create new constructs and ask questions. Intermediate theories have some constructs proven, and some agreement in the field. Mature theories define a field and are largely agreed upon. Can we ever have these in design?

Erik Stolterman said...

Thank you all for your comments. I am writing a new post where I will comment on these proposals!

Anonymous said...

Hi Erik,

What about theories of HCI as a discipline itself? Ie stuff in the vein of Paul's Implications for Design - theorizing about what knowledge means in HCI and how to approach it.

This may not be helpful for the scheme you are trying to come up with but in my own work I find it useful to make classification systems that are not only about the object of study but also the orientation to the object of study. We have a tendency in HCI to think of areas in HCI to be about their topic of study - e.g. ubiquitous computing studies ubiquitous technologies, affective computing studies emotion - but this tends to collapse the object of study with the orientation to the study, as though the only possible way to do "workplace studies" is through the lens of CSCW/ethnomethodology and the only way to do build hardware is through the lens of Computer Science. I would guess that cognitivist theories would look very different from design theories and those in turn would differ from a variety of ethnographic/anthropological theories.


Erik Stolterman said...

Hi Phoebe

Thanks for your comments. And you raise a really important point. In my other post I allude to the fact that I will later add some aspects and not only focus on the "object of study", and one of the other aspects are, what I simply call "methodology" which I think relates to your point. So, for instance, phenomenology is an example, and so are the ones you mention, ethnographic and cognitivistic, etc. Each of these can be seen as theories of course but are in my mind more methodologies. To some extent they are theories, but they are not theories about interaction per se, instead they are ways to approach whatever research object in question. They are epistemological theories that makes claims on how to approach reality, but they don't really say anything in particular about the object of study.

Ok, this needs much more space to be developed fully. I will come back to this.

Again, thanks Phoebe for raising this aspect!!

Featured Post

Why Design Thinking is Not Enough

If you go to Youtube and look for "design thinking" you will find a large number of videos with TED talks and other talks all expl...