Enacting Complex Relations
- Side: 357-368
- Publisert på Idunn: 2011-11-16
- Publisert: 2010-12-02
- Creative Commons (CC BY-NC 4.0)
How can actor-network theory (ANT) inform studies of the complexities of recent attempts to change education with the aid of new digital technologies? What are the methodological implications of ANT and how can interaction analysis be used as a resource for analyzing actor-networks? These questions are explored in this article where ANT is also contrasted with sociocultural theory, a theory with a certain impact in studies of ICT use for educational purposes in Norway. I argue that ANT provide insights and conceptual tools that enables analyses of how classroom practices are and become connected to other networks forming assemblages of people and things that produce knowing and learning.Keywords: Actor-network theory, sociocultural theory, methodology, complexity, ICT and education
In his book Utdanning for informasjonssamfunnet: Den tredje vei [Educating for the Information Society: The Third Way] Østerud (2004) employs insights from actor-network theory (ANT) to analyse complex negotiations between actors in a project where networked laptop computers were introduced into three classrooms in a city in the northern part of Norway. In particular, by drawing on the work of Star & Griesemer (1989), he scrutinises how the computer is metaphorically shaped into a boundary object, which is constituted differently by the various stakeholders in the project; the teachers, the researchers, a computer company, a publishing agency and an agency established for generating local innovation and development. Østerud eloquently demonstrates how the computer is drawn into and given meaning in a continuously circulating network with different actors that pursue contrasting goals, ideas and interests. But for the teachers, the computer at first fails to become an actor that stimulates development, since it was constituted as an agent that in and of itself would generate change and lead to innovative pedagogy and learning. I want to take some of Østerud’s insights as a starting point and push these further by elaborating on how concepts from ANT might be used as analytical resources in studies of ICT use in educational practices more broadly. I will contrast this with a sociocultural approach since in Norway this logic of inquiry has been quite influential in research on ICT in education over the last decade (see Furberg & Arnseth, 2009; Krange & Ludvigsen, 2008; Rasmussen & Ludvigsen, 2010). Whether ANT constitutes a theory or a theoretical framework is contested. It does not contain a conceptual core or a clearly defined conceptual framework. Instead it can be characterised as a “program of methodological provocations that constantly challenge the traditional categories in social sciences, introducing new sets of terms for their reconceptualising” (Miettinen, 2000, p. 171).
Perhaps it makes more sense to treat ANT as a methodology, as a way of doing and writing research that cannot easily be assembled into specific procedures or strategies (Latour, 1999). Nevertheless, it is possible to draw out some principles or rules of thumb for doing ethnographic research in ANT, but ethnographies that do not correspond to established ways of distinguishing between realism and interpretivism. As will become clear below, a distinction between fact and interpretation is not taken for granted in ANT, but is rather seen as an outcome or a network effect. This is also reflected in its methodology. When examining ANT it is therefore critical to discuss issues of methodology, since they are closely intertwined. In relation to this, I will discuss potential problems with relying on ethnographic strategies and argue that interaction analytic methods might be a resource in studies of actor-networks. Finally, I draw together the main threads of my argument and clarify how concepts and strategies from ANT can provide new insights into the manifold and complex processes involved in changing education with ICT.
The heterogeneous practices of ICT use in education
How to analyse and approach the heterogeneous relations between educational practices and ICT has been an important issue in educational research during the last couple of decades. This research agenda is made relevant by recent efforts to bring about change on many levels: from implementing new governmental policies to using new designs of learning environments and digital learning tools (see, for instance, Selwyn, 2011). Many researchers and policymakers have proposed that ICT should be employed in teaching and learning on a broad scale to equip students with the skills and competencies that correspond to the demands of the information society (Arnseth & Hatlevik, 2010). ICT has also been conceived as a means for changing pedagogy away from an information delivery model to a more learner-centred social constructivist model focusing on computer-supported collaborative and critical inquiry (Bereiter, 2002; Law, Pelgrum & Plomp, 2008). Examining the practical effects of these various efforts to change educational practices is a crucial goal for research.
Introducing new technical artefacts can create tensions, ruptures and new complexities in teachers’ and students’ practices. The push to use ICT for learning is the result of the work of policymakers, school management and other stakeholders. It is not necessarily something teachers themselves want to use or find relevant for getting their work done. An additional consequence of the implementation of ICT, is that classroom practices become more intimately dependent on new sociotechnical systems. This contrasts with a classroom only containing textbooks, blackboards, pencils, notebooks and chalk, which over the centuries has become a stable practice with well-established and well-functioning ties to the producers of the artefacts used for teaching and learning. If the chalk breaks in two or a book is in bad condition there are well-known practices for fixing these breakdowns in the normal flow of activity, practices where teachers and learners exert a great deal of agency.
A textbook also provides the teacher with an ordered means of organizing and managing activities in time and space; that is, it provides a way for the teacher to arrange when he teaches what to students during the school year. Introducing ICT changes this, in that the Internet does not provide an ordered means of delivering curriculum content. These changes can be compared to what happens when, for instance, a hospital is equipped with a new ultrasound machine. Not only do physicians have to develop new practices for using the machine, including expertise in interpreting visual images, their practices also become dependent on the functioning of the software and the technical apparatus. Similarly, teachers become dependent on the technical functioning of computer networks and software; that is to say, on the functioning of a technical infrastructure. This new level of complexity is not specific to education, but constitutes a more general phenomenon (see Law & Mol, 2002).
What interests me here is how these new complexities can be approached theoretically and pursued methodologically and, by the same token, how we can empirically investigate the effects of these new complexities for teaching and learning. To be more specific, this new level of complexity is particularly about how pedagogies become dependent on or influenced by other sociotechnical systems, for instance, the installation and management of hardware and software, but also how computers as boundary objects provide access to applications and semiotic materials with varying degrees of relevance to the main goals of educational practices.
The connection between technical systems, digital tool and content providers, national and institutional policies, and the activities of teachers and students is not easily recognised. Their relationships are played out in multiple ways and directions, and what teachers actually do in classrooms is potentially connected to multiple networks and cultural resources that in varying degrees relate to and impact on their pedagogy. No study can cover the full complexity of these phenomena, and for practical purposes it is necessary to reduce complexity through establishing a clear research focus. Nevertheless, I will attempt to cover this issue more broadly and treat complexity as a research topic that can be studied through the use of specific analytical and methodological means. What concerns me is how we can conceptualise and study the formation and effects of these relationships. To do this I will combine, compare and contrast ideas from sociocultural theory and actor-network theory. Even though the approaches differ in how they conceptualise these relations; for instance in terms of the relative agency that is attributed to ICT tools, common concerns are how technologies mediate learning as well as how technological arrangements contribute in shaping pedagogies.
Analytical tools for studying heterogeneous relations
Even though sociocultural theory and ANT have different disciplinary and philosophical backgrounds, they share certain methodological principles. Both reject monocausal explanations, both attempt to overcome and transcend dualist notions of mind and matter and nature and society, both recognize the importance of material artefacts and both emphasize that resources for acting in the world are distributed among people, artefacts and environments (Miettinen, 2000). In addition, the notion of mediation is crucial in both approaches, even though they are used and interpreted differently. While ANT pursues a symmetrical interpretation of mediation, sociocultural theory pursues a dialectical interpretation of the relationship between subjects and objects, where artefacts and signs act as mediational means connecting them.
From a sociocultural perspective, human agency and learning is seen as mediated by cultural and material tools (Cole, 1996). To grasp and analyse human action we need to pay close attention to how people use tools in specific circumstances to realise their projects. Through acting on and transforming objects with mediational means the actor changes and develops in activity. Furthermore, human learning is conceived as the gradual mastery and appropriation of tools and becoming able to use them in rational and legitimate ways, ways that are recognized as such within a community of practice (Säljö, 2000; Lave & Wenger, 1991). According to Wertsch (1998), there is an inherent tension between humans and tools in the sense that there is always a possibility that ruptures and breakdowns in mediated actions might occur. Furthermore, there is never a perfect fit between human goals and the possibilities for action that tools offer the human actor. For Wertsch, however, the tools are not granted any independent agency. This is a theoretical assumption built into the sociocultural framework where human action is a function of the relationship between the subject, object and mediational means. Thus, a mobile phone, or a pole in pole vaulting, is not an actor in itself. They only become mediational means for human action when somebody uses them for something, but the tool does not determine what the actor uses them for. According to Wertsch (1998, p. 30):
Indeed, in and of themselves, cultural tools such as poles in pole vaulting and the form of syntax used in solving multiplication problems are powerless to do anything. They can have their impact only when an agent uses them.
In a sociocultural perspective, then, there is an asymmetric relationship between agents and tools. It is humans who act and tools mediate this action in various ways without exerting any independent agency.
In contrast, ANT proposes a more symmetric relationship between humans and tools or humans and non-humans. This methodological principle of generalised symmetry was developed by Callon and Latour during the 1980s (Callon, 1986; Latour, 1987). This move implies a radical form of anti-essentialism and anti-dualism. In any sense, the locus of agency in an activity is not given a priori to specific properties of either humans or non-humans. From an ANT perspective, devices or tools are treated as having programs for action, programs that in a sense anticipate what other actants may do. And for Latour “[a]ction is simply not a property of humans but of an association of actants” (Latour, 1999, p. 182, italics in original). Latour’s eloquent examples of the agency performed by door closers, speed bumps, hotel keys and fasten-your-seatbelt lights make visible how tools also contain values and norms (Latour, 1992). The action of putting on a seatbelt is the result of negotiation work between a human and a non-human. Through negotiations and associations actants are folded together and assembled and, from this perspective, actions such as learning or using specific tools for educational purposes are an effect of such local assemblages of human and non-human actors. In contrast, in sociocultural theorizing humans are constituted as the active agent. Humans are ontologically distinguished from non-humans, and the locus of agency in mediated action is in the human actor.
We might argue that, since Wertsch uses rather simple examples of tool use, such as pole vaulting, it follows that it is easier to conclude that humans need to pick up the pole and use it for some purpose. In contrast, Latour often use examples of material things that we as humans do not simply use or pick up, even though we of course pick up hotel keys. Put simply, while Wertsch probably would have studied how a key mediates the opening of a door, Latour is more concerned with the practical problem of making hotel guests leave their keys at the desk so the keys will not be lost. So the practical problem for hotel managers is to exert some kind of force with regard to guests, and the heavy key is recruited to exert this normative force on the guest. With regard to such things as speed bumps and door closers, it is perhaps easier to grant them some form of independent agency. For Wertsch, such things are perceived as constituting parts of the environment that the agent engages with using the various tools. To follow up on the example of pole vaulting, we might say that the bar constitutes part of the environment or goal that the agent will reach; it is the object in the triangular relationship between subject, mediational means and objects. For Latour, on the other hand, the bar would probably be conceived as an actor with a particular program for action; that is to say, the responsibility for providing a threshold has been delegated to a non-human actor. Through the fact that a bar is hoisted into the air at various heights it is recruited into a network and, through this recruitment, the relationship between the actors in pole vaulting changes. ANT, then, does not propose any distinctions between the material and cultural or human and non-human prior to their enrolment into networks. Instead these distinctions are conceived as network effects (Fenwick & Edwards, 2010).
After this more general introduction to the differences in these theories, and before I outline the possibilities for analysis offered by ANT, there is a need to be a bit more specific about how these concepts can be used to understand the heterogeneity of ICT use in education. Based on the previous account, it is obvious that sociocultural theory, as described by Wertsch, has it strengths in detailed studies of relatively discreet episodes of tool use. Sociocultural theory does not really offer analytical tools for studying more complex systems, including how such systems are constituted by human and non-human actors. Without going into detail, within the broader framework of sociocultural theorising, there are closely related theoretical models, such as activity theory, which offer conceptual tools for analysing action on a more systemic level, including boundaries and boundary crossing between intersecting activity systems (Engeström, 1987). Engeström conceives of tool use as mediated by a larger structural framework, including divisions of labour, rules and community. However, parallel to Wertsch’s model of mediated action, activity theory does not grant tools or non-human artefacts any independent agency (Miettinen, 2000). Thus, even though actions are directed at objects and mediated by the organisation of an activity system, the framework grants human intentionality a privileged position, even though this intentionality is dialectically related to the actors’ transformation of objects with tools (see Arnseth, 2008). This might of course provide for interesting analyses of human practices, including how technologies mediate such practices. What it does not offer are analyses of dynamic networks and flows, of how ideas, tools, infrastructures and people become enrolled into networks with sometimes unexpected effects. ANT allows for more flexibility in terms of analysing the more systemic nature of practices. In activity theory, activity systems change due to tensions between elements in the system, but the theoretical model remains relatively fixed. In contrast, through pursuing a symmetrical approach, ANT problematizes how distinctions between subjects and objects are brought into being and appear as fixed and stable entities in the first place.
From an ANT perspective, the relationship between ICT and human practices or, in sociocultural parlance, between subjects, objects and mediational means, cannot be conceived as a relationship between relatively discreet entities. On the contrary, they can become interconnected in various ways and they can expand and contract depending on the flow of movements and choices in space and time. The ordering of education, or the ordering of objects and actions, is a network effect brought forth and made visible through circulations among people and things (Fenwick & Edwards, 2010). When we start to unpack pedagogy or education we might start with what it produces. What are the effects that generally emerge from these networks and how can we trace and unpack this emergence?
One of the things that education produces is knowing; what actions students are able to perform which are institutionally acknowledged as knowing. Knowing is situated, embodied and distributed. From this starting point, we might try to trace the various negotiations involved in producing this knowing, including how the network of relations changes or is stabilized over time. In regard to education and what it produces, questions about what knowledge is, who produces it and how it is generated and circulated in a network are often hotly contested (Fenwick & Edwards, 2010). Let’s take the ideas about how ICT might support more innovative pedagogies and more personalised and flexible education for students. These ideas circulate in many networks – in policy documents produced by the OECD and the European Commission – and are again taken up by national governments and inscribed into plans and requirements for educators. They also circulate in academic circles that capitalise on this rhetoric to argue for the relevance for particular ideas about learning and pedagogy, emphasising the active learner that needs to be able to manage large quantities of information or collaborate in cross-disciplinary groups. I am not claiming that these ideas are necessarily flawed. I am interested in examining how these ideas become translated by practitioners and reassembled into their local practices. When we look at questions of implementation, the discourse on pedagogy and ICT becomes recruited into or excluded from the network that produces teaching and learning.
How these circulating ideas are taken up and responded to in educational practices can be unpredictable. For some educators new ideas become opportunities for changing their practice, something which in turn can enable them to become local or national experts. In this sense, they open up new opportunities for them to be enrolled into new networks (Nespor, 2006).
Inscriptions and infrastructures
When analysing the heterogeneity of ICT-use in educational practices, there are two concepts in particular that are crucial elements in understanding assemblages of actors and tokens, which together constitute pedagogical practice. The first is the notion of inscriptions and the second is the notion of infrastructures (Guribye, 2005). In contrast to sociocultural theory, which does not necessarily have the analytical means for studying relationships between contexts for action, the notion of inscriptions enables us as analysts to trace relationships between contexts and study in detail how inscriptions are oriented to and translated when they become assembled and recruited into new or already existing networks (see Silseth & Arnseth, 2011). According to Latour (1999, p. 311) the concept translation:
refers to all the displacements through other actors whose mediation is indispensable for any action to occur. In place of a rigid opposition between context and content, chains of translation refer to the work through which actors modify, displace and translate their various and contradictory interests.
In this sense, pedagogical models become inscriptions with a certain agency that educators need to pay attention to in one way or another, inscriptions that can be translated and modified by actors in and beyond the classroom. They can of course also ignore them, but this might prove difficult depending on their accountability to, for instance, parents or school leaders, who constitute important actors in their local networks. Thus, in some way or another they need to manage them.
The other concept with particular relevance to understanding the heterogeneity of ICT-use is infrastructures, which, in contrast to sociocultural concepts, constitutes environments as something more than tasks or obstacles that humans manage through the use of tools. Instead, infrastructures are treated as assemblies of actors that are interconnected in complex ways and that have their own programs for action that impact on what educators are trying to accomplish. Ruptures and breakdowns in infrastructural systems can in this sense have a serious impact on what educators do: for instance when they are conducting a presentation using digital technologies and the network breaks down or, from my own experience, when we tried to do design-based research in a classroom where the school’s firewall hindered communication with servers at the university, something which was crucial for using the new system for learning.
To summarise, while sociocultural theory provides useful tools for analysing meaning making and instances of tool use, including how tools mediate specific tasks and problems, ANT offers analytical tools for tracing connections and flows in and between systems. ANT enables analyses of how local enactments are connected to other systems and it enables us to scrutinize the flow of inscriptions and tokens, but also how various human and non-human actors become enrolled into educational practices, where it has important implications for how these practices produce teaching and learning. While this opens up possible directions for research, there are also problems involved, particularly in terms of methodology. It is this question that I now turn my attention to.
Methodological implications and challenges
Latour and ANT’s ethnographic practices contrast with many established ways of doing ethnography. He subscribes to neither a critical realist stance towards participants’ accounts and actions nor to an interpretive or reflexive stance towards doing and writing ethnography. For this reason, his method has received a lot of criticism from established ethnographers (see Austrin & Farnsworth, 2005 for an elaborate discussion). However, Latour instead insists on following the various actors and traces their movements and translations to see how they are connected and become assembled and stabilized or blackboxed. Alternatively, it is possible to retrospectively trace and deconstruct stable entities and reconstruct how they became stabilized and solidified. Even though Latour has received substantial criticism for not being sufficiently aware of the importance of interpretation, few have compared or contrasted this methodological approach with methods that focus more on registering action and interaction among people, tools and environments using video or audio-recording techniques. My aim is to discuss the potential relevance of interaction analysis for a methodological practice using concepts from ANT.
Even though Latour relies heavily on ethnographic research practices, ANT does not necessarily come with a particular predefined method or research strategy, even though there obviously are some methods that are better suited for tracing the flow and assembly of human and non-human actors and the effects of these assemblies. As mentioned above, it makes more sense to treat ANT as a methodology, as a way of using various concepts for investigating actor-networks. Still, this research program would imply that certain methods are more useful than others. In particular, quantitative methods would be problematic because they tend to blackbox the actual production or enactment of practices. In terms of qualitative methods, interviews are problematic, since we need to rely on peoples’ accounts of their actions and not what they actually do in practice. This problem is not always acknowledged by, for instance, Latour, and he has been criticised for relying too much on the accounts of certain human actors whose power and agency in a network can easily be exaggerated in interviews. This might for instance be the case in the Aramis study (Latour, 1996). In a sense, Latour is an eclectic methodologist who shifts between data collection strategies and analytic and textual genres. Still, there is a risk that relying too much on important actors’ accounts might seriously affect alternative analytic stories that would show how negotiations and translations are enacted.
Furthermore, methods that primarily rely on analyses of discourse and language use are also problematic for ANT, since we do not get access to the performativity of non-human actors. ANT urges us to follow the actors. “Follow the actors, both as they attempt to transform society and as they seek to build scientific knowledge or technological systems.” (Callon, Law & Rip, 1986, p.4).
This leaves us with what we might describe in broad terms as an ethnographic approach, where participatory observation and close scrutinization of action and relationships between elements in networks is at the core of the research strategy. I will not go into any detail, but instead focus on some of the practical and methodological challenges involved when doing these kinds of studies. The first challenge involves what we might term a stopping rule for doing data collection. For Latour and ANT it is impossible to tell when a relationship or process starts and where it ends. There are only flows and circulations that sometimes become temporarily stabilized, but where even this stability also requires active work by various agents.
Still, even though there are no easy methodological or theoretical ways of limiting the scope of a study, this is very much felt as a practical research problem. When do we as researchers know when we have enough data, and how do we establish boundaries that in a sense single out the research object that we want to study? In ANT this becomes both a practical and a theoretical problem since this cannot be decided in advance. It is nevertheless a pertinent problem, since in principle everything can be connected to everything.
The second challenge concerns issues of credibility in research and how the quality of the research can be discussed and assessed in the research community. A problem with ethnographic studies is that to a large extent we need to take the researchers’ interpretations on trust, since we do not have access to how the data were actually produced in situ (Silverman, 2001). The solution to this problem might be to combine methods. In particular, it might be fruitful to combine the use of observations and field notes with video recordings of talk and action. Video analysis enables us to closely scrutinise the interconnections between actions, but also how material entities are oriented to or impact on what human actors are trying to accomplish (Jordan & Henderson, 1995). Furthermore, the transcribed data can be made available to other researchers so that in principle they are able to assess the interpretations and come up with alternative ones. From an ANT perspective, the data are potentially more easily recruited into the fact-producing research network. It constitutes more convincing evidence because other researchers can more easily assess the researchers’ accounts and, by the same token, the analytic story can become less easily refuted (see also Billig, 1987).
The problem with detailed studies of interaction, however, is that they are time-consuming and limiting, in the sense that they only enable detailed analysis of relatively discreet episodes of interaction. When the objective is to trace the emergence, flow and assembly of actants, this is a serious limitation. Such an analytical strategy requires more flexible means of collection and analysis of data. This problem can be addressed by combining detailed analysis of interaction with analyses of texts, models, or inscriptions of any kind and observations of action in and between situations (Baker, Green & Skakauskaite, 2008). I am not claiming that this allows for triangulation of findings. This validation technique does not fit well with ANT’s anti-essentialism and constructionism, because the data do not simply represent any underlying or stable phenomenon. What it might offer is an alternation between different levels of analysis, where interaction analysis can demonstrate detailed instances of emergence and where observations can provide insights into broader changes and flows. Detailed studies can enable us to make some of the broader claims problematic or provide the analysis with richer and more detailed descriptions of more generic patterns. On the other hand, observations over time can provide us with knowledge that might force us to reconsider the analytical claims arrived at over a much shorter time span.
In this article I have tried to discuss some of the analytical strategies afforded by ANT and tried to compare this to important concepts in sociocultural theory. I have also tried to address some of the methodological challenges involved in doing studies inspired by concepts from ANT. When studying technologies and how they are used in education, I have argued that ANT might provide new insights into the complexities of using ICT in education, in that we as researchers are not limited to studying rather discreet episodes of action, but can trace how local practices become connected to other systems that at times have a crucial impact on the local enactment of education. Knowing and learning are in this regard not discreet products of local meaning making practices, but are network effects – networks that are never fixed, but where inscriptions from other practices can become enrolled and have an impact on teaching and learning. From this perspective the relationship between pedagogy and ICT emerges as more unstable and unpredictable. To be able to gain knowledge about the impact of technology in education we need to pay close attention to how this interconnection is played out in practice. This might offer more nuanced views of what ICT actually has to offer, as well as the complexity involved in developing pedagogies supported by ICT. Simply introducing ICT does not necessarily have an impact, since technologies can be translated and incorporated into or simply excluded from already existing practices. The interactive whiteboard can simply be switched off. Having said that, technologies can exert considerable resistance in regard to such attempts and in that sense create ruptures and breakdowns that educators in one way or another need to manage.
Arnseth, H. C. (2008). Activity theory and situated learning theory: contrasting views of educational practice. Pedagogy, Culture & Society. 16(3), pp. 289- 302.
Arnseth, H. C. & Hatlevik, O. E. (2010). Challenges in Aligning Pedagogical Practices and Students Competencies with Demands of the Information Society: The Case of Norway. In Siran, Mukerji & Purnendu, Tripathi (Eds.). Cases on Interactive Technology Environments and Transnational Collaboration. Concerns and Perspectives. IGI Global.
Austrin, T. & Farnsworth, J. (2005). Hybrid genres: fieldwork, detection and the method of Bruno Latour. Qualitative research. 5, pp. 147-165.
Baker, W. D., Green, J. L. and Skukauskaite, A. (2008). Video-enabled ethographic research: A microethographic perspective. In G. Walford (Ed.). How to do Educational Ethnography. London: The Tufnell Press.
Bereiter, C. (2002). Education and mind in the knowledge age. New York: Routledge.
Billig, M. (1987). Arguing and thinking: a rhetorical approach to social psychology. Cambridge: Cambridge University Press.
Callon, M. (1986). Some elements of a sociology of translations: Domestication of the scallops and the fishermen of St. Brieuc Bay. In J. Law (Ed.). Power, action and belief: A new sociology of knowledge? London: Routledge & Kegan Paul.
Callon, M., Law, J. & Rip, A. (1986). Don’t throw the baby out with the bath school: A reply to Collins and Yearley. In A. Pickering (Ed.). Science as practice and culture. Chicago: The University of Chicago Press.
Cole, M. (1996). Cultural psychology. A once and future discipline. Cambridge, MA: The Belknap Press of Harvard University Press.
Engeström, Y. (1987). Learning by expanding: An activity-theoretical approach to developmental research. Helsinki: Orienta-Konsultit Oy.
Fenwick, T & Edwards, R. (2010). Actor-Network Theory in Education. London: Routledge
Furberg, A., & Arnseth, H. C. (2009). Reconsidering conceptual change from a socio-cultural perspective: Analyzing students' meaning making in genetics in collaborative learning activities. Cultural Studies of Science Education, 4, pp. 157–191.
Guribye, F. (2005). Infrastructures for learning. Ethnographic inquiries into the social and technical conditions for education and training. PhD thesis. University of Bergen. Norway.
Jordan, B. & Henderson, K. (1995). Interaction analysis. Foundations and practice. The Journal of the Learning Sciences, 4(1), pp. 39-103.
Krange, I., & Ludvigsen, S. (2008). What does it mean? Students’ procedural and conceptual problem solving in a CSCL environment designed within the field of science education. International Journal of Computer Supported Collaborative Learning, 3, pp. 25-51.
Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Cambridge, MA: Harvard University Press.
Latour, B. (1992). Where are the missing masses? The sociology of a few mundane artifacts. In W. Bijker & J. Law (Eds). Shaping technology/building society: Studies in sociotechnical change. Cambridge, MA: MIT Press.
Latour, B. (1996). Aramis, or the love of technology. Cambridge, MA: Harvard University Press.
Latour, B. (1999). Pandora’s hope: Essays on the reality of science studies. Cambridge, MA: Harvard University Press.
Lave, J. & Wenger, E. (1991). Situated learning. Legitimate peripheral participation. Cambridge, NY: Cambridge University Press.
Law, J. & Mol, A. Editors (2002). Complexities. Social studies of knowledge practices. Durham: Duke University Press.
Law, N., Pelgrum, W. J. & Plomp, T. (2008). Pedagogy and ICT use in schools around the world. Findings from the IEA SITES 2006 study. Hong Kong: Springer.
Miettinen, R. (2000). The riddle of things: Activity theory and Actor-network theory as approaches to studying innovations. Mind, culture and activity, 6(3), pp. 170-195.
Nespor, J. (2006). Technology and the Politics of Instruction. Mahwah, NJ: Lawrence Erlbaum Ass.
Rasmussen, I. & Ludvigsen, S. R. (2010). Learning with Computer Tools and Environments: A Sociocultural Perspective. In K. Littleton, C. Wood & J. K. Staarman (Eds.). International Handbook of Psychology in Education. Emerald Group Publishing Limited.
Selwyn, N. (2011). Education and technology: Key issues and debates. London: Continuum.
Silseth, K. & Arnseth, H. C. (2011). Learning and identity construction across sites: A dialogical approach to analysing the construction of learning selves. Culture & Psychology, 17(1), pp. 65- 80.
Silverman, D. (2001). Interpreting Qualitative Data. Methods for Analysing Talk, Text and Interaction. London: Sage.
Star, S. L., & Griesemer, J. R. (1989). Institutional ecology, translations' and boundary objects: Amateurs and professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39. Social studies of science, 19(3), pp. 387-420.
Säljö, R. (2000). Lärande i praktiken. Stockholm: Prisma
Säljö, R. (2010). Digital tools and challenges to institutional traditions of learning: technologies, social memory and the performative nature of learning. Journal of computer assisted learning, 26(1), pp. 53-64.
Wertsch, J. (1998). Mind as action. New York: Oxford University Press.
Østerud, S. (2004). Utdanning for informasjonssamfunnet. Den tredje vei. Oslo: Universitetsforlaget.