Kuutti, K. Activity theory as a potential framework for Human-Computer Interaction research. (in Nardi, Bonnie A. ed.), Context and Consciousness: Activity Theory and Human-Computer Interaction. MIT Press, Cambridge, MA, 1996, 17-44. Even though this paper is a bit old, I found interesting the part that discussed the "debate against the use of information processing psychology as the foundation of HCI":

Both the object and methods of the "mainstream" research has been questioned. Critics would like to add to the research object the users and their work tasks in real life. Methodologically the "Cartesian" ideal of cognitive science - continuing the use of experimental apparatus of laboratory-oriented classical psychology borrowed from natural sciences - has been seen unable to penetrate the human side of the interface. So now, in the midst of 1990s we have both the established, cognitive science - based "orthodoxy" and the emerging, although diverse "opposition".

And this gave rise to different traditions:

Thus we have three broad "traditions" in HCI research: the "technical" one, having roots already in the old "knob-and-dial" ergonomics, concentrating human perceptive abilities and motor skills and corresponding features of technical devices, the "conceptual" one that has formed the information processing psychology-based mainstream of HCI research, and the emerging new one searching new frameworks and theories in order to deal with the complexity.

Then the author describes how Activity Theory could enable to widen the spectrum and would be suitable as a pertinent underlying theory in CSCW. Why do I blog this? currently working on the theoretical chapter of my dissertation, I am trying to describe my framework which sits in between cogsci and broader theories (tough!).