Cognitivism vs Social Constructionism:

'Knowledge consists of a mix of scientific or expert knowledge and unconscious, selective and culture-specific background assumptions. In certain social contexts and within certain social interests these assumptions appear as factual or valid, whereas in other social contexts they are seen as questionable'. [25, p.4]

Knowledge has some aspects of particularity to an individual, but there is no such thing as the subjective knowledge structures as espoused by cognitivists. 'Knowledge and knowledge structures are neither objective nor subjective, but intersubjective'. [25, p.4].

Both Dervin and the social constructionists view information seeking behaviour as situational and variable according to time and space, but the latter focus much more on this situationality and variability within the context of society and power. The individual is perceived as primarily influenced by their membership of society and of groups within society. The individual is neither a 'site of power' within society as Dervin suggests [21, p.732], nor an individualistic builder of knowledge structures with, subjective meaning. Intersubjectivity means that meanings are shared with others.

The arena of knowledge construction 'consists of several competing discourses, based on incompatible and contradictory assumptions. When the lense of the production of knowledge changes, the facts also change' [25, p.54] .

From my reading there exists a stark contrast between the two viewpoints, reflected simply in the definitions of information and knowledge. It is almost an updating of the now historic marxist-capitalist divide. On one side is the social production of knowledge for which it is easy to construct a machine and factory metaphor, on the other the individual searching for sense and personal fulfilment.

Dervin's metaphors approach those of the 'hero's journey' [27] , that staple of the Hollywood screenwriter, where the protagonist, faced with a series of challenges, discovers reserves of resolve to overcomes the challenges, as well as a renewed confidence to take on the next inevitable heroic struggle on the journey. If Dervin's viewpoint reflects her American-ness, then I suggest that social constructivist viewpoints reflect the advocates' predominately European socio-cultural perspective, characterised by a modern history of massive outmigration, war, socially interventionist governments and significant marxist tradition. In a criticism of social constructionists Dervin considers that they have skewed the debate toward social rigidity rather than individual creativity:

'Because we have sought only across time-space understandings, we have missed so much of the whole range of human existence that involves struggling with, breaking with, coming to terms with, and changing whatever structure the human finds oneself in. In essence, we have done better at developing understandings of human rigidities than of human creativities' [19, p.4].

6. View of the Information Seeker

Cognitism is fundamentally a individual-centred approach. Social constructionism is not.

The are a variety of parameters for assessing assumptions and declarations about the place, orientation and role of the individual within information behaviour. Here are a few:

• whether the individual is to be the main focus of information behaviour research or not;
• whether the individual is a constructor of knowledge or a processor of information;
• whether the individual is unknowing or an expert;
• whether the individual is a needy seeker filling a 'gap' or 'anomalous state' or a habitual information gatherer;
• whether the individual is a user , a client, a customer , a partner, or a contributor, and;
• whether the individual is isolated and private or connected and social

Brookes and Belkin both view the information seeker as an active processor of information and a knowledge builder. For Brookes the person dealing with objective knowledge changes it somehow while integrating it into their knowledge structure. The person does more than just transmit. Belkin again seems to share Brookes perspective. Both can be seen to consider knowledge as less individualistic and more socially constructed than (at least early) Dervin, as individuals were assumed to share characteristics of common knowledge structures in a discipline [12, p.13].

Belkin seems to view the individual as the problem area in IR. If we could only understand the cognitive aspects of knowledge creation we could build the appropriate machine. The view of the information seeker is as purposeful but inarticulate, unable to specify exactly what is needed to resolve the anomaly [14, p.62], and thus requiring the intervention of the mechanised form of a personal reference librarian.

While Belkin's ASK and Dervin's 'gap' are concepts with similarities, their focal point is quite different. Dervin takes the machine and the system out of the picture, placing the camera firmly on the individual. The key is to make space for the individual's narratives. Sense-making characterises individuals as actors, experts and theorists. They are constructors of knowledge and developers of strategies. The sense-making interview methodology stresses that the interviewee should describe their own world and the actions taken within it rather than the interviewer expressing any assumptions about that world and strategies. [21; p.740].

For Belkin individuals are units within large-scale research projects in the practical application of ASK and as a such are sources for pattern recognition and cognitive models rather than individuals in their own right.

My assessment of Belkin's perspective is in part determined by his degree of success. If he could inform systems design in such a way as to result in significant IR improvements then a cognitivist machine/human focus would obviously inform the theory of practice. The summary of the current 'MONGREL' research proposed on his website indicates that after twenty-something years success has been limited:

'… tools to support information retrieval and filtering…performance is mediocre. …make mistakes that are obvious and aggravating to users, and relevant documents are usually mixed with many others that are totally unrelated…..lower the productivity and effectiveness of people … fundamental issue … is the lack of adequate models of the user and the domain…..need … more information about the context….better models of the user…more knowledge about the domain….' [16].

I also suggest that much if not most information behaviour does not fit the ASK scenario. Information seekers are frequently experts in their subject domains and are clear on the semantics and syntax within that domain and even the usual patterns within automated information systems. Not knowing exactly what they don't know doesn't mean that they don't have the language skills within the discourse to frame the questions appropriately.

Further, I suspect that most knowledge creation is to do with continuous social networking, habitual reading from a position of strength, and 'plugging into' and exploring certain channels of information. Information behaviour is then also about gathering information, adding to existing repertories of knowledge and 'keeping up' and pursuing passions and hobbies rather than overcoming a sense of anomaly.

Talja considers that the cognitive viewpoint positions the individual as the point of chaos, 'the central uncertainty factor from the point of view of information transfer' [25, p.3] whereas they are more accurately characterised as knowing experts.

The social constructionist viewpoint is interested in the social-cultural context of information behaviour. Within their social milieu people have various areas of expertise. They are information givers as much as they are information seekers. By not focusing on preconceived narrow notions such as ASK, gap, needs and problems, the social constructivist position actually strengthens the focus on the individual by broadening the scope of the observations, to enable the observation of the person in their full social environment. The cognitive viewpoint leaves the scope narrow and relies on mental constructs such as models and narratives. It's epistemological basis is insufficient to underpin a holistic approach to information behaviour.

Hjorland also makes a valid point that the user-centred approach actually focuses on wants rather than needs. 'The information demanded by users is an expression of their subjective information needs, which may thus be different from their real or objective needs' [28, p.264]. As Frohmann says 'not everyone knows what they want to prevent AIDS, not everyone wants what they need' [ 18, p.98]. While Dervin declares a focus on information needs she is clearly meaning information wants.

7. View of the Information Professional, or the Information Mediator

A key contentious issue social constructionists have with the cognitivist viewpoint is the conceptualisation of the mediator, the information professional (often a librarian), as the neutral expert dispensing wisdom.

Garry Radford has criticised the traditional and persistent idea of the role of librarians as the guardians of objective knowledge implementing systems which emulate the ideals of 'scientific method' [5]. The metaphor is the authoritarian bespectacled librarian captured in our childhood imaginations. The users role is to learn their place and the rules of the system.

Bernd Frohmann contends that the librarian is conceptualised by cognitivists as all-knowing experts enjoying 'the clarity of complete knowledge' while 'users' are 'mere supplicants to the system' , 'enlightened about their image-gaps but only to the extent of being aware of their existence' [9, p.379].

How do people suffering from gap or ASK address their needs in a library setting? I suggest they will take the following approach. They will use an information seeking strategy developed over time, bearing in mind institutional and power contexts, such as how reference librarian's behave towards clients, and an agreed institutional way of preceding. Different information seekers can be expected to display similar patterns of behaviour, but with certain individual differences in style and effectiveness.

I contend that the reference librarians operate similarly. They will use a familiar strategy developed over time, based on general principles of how information seekers frame questions and tempered by an agreed institutional way of proceeding. Different reference librarians can be expected to display similar patterns of behaviour, but with certain individual differences in style and effectiveness.

Brenda Dervin in recent works is alert to the issue, '…just as sense-making assumes that ordinary human sense-makers struggle with multiple verbings in myriad conditions with myriad outcomes and uses, researchers as human sense-makers are likewise involved' (Here she means 'researchers' as people carrying out sense-making research) [21, p.735].

Kimmo Tuominen claims that the cognitivist positioning of the librarian in the ASK or gap situation as the expert mediator assumes information seeking is an otherwise private activity. [29, p.357]. To me it is simply commonsense to appreciate that much information seeking is actually undertaken outside of the mediator-user relationship, such as in social settings with colleagues and friends. Further, in my experience effective reference librarians view the reference interview as a collaborative exercise of mutual discovery.

Tuominen is also correctly concerned at the use of the user-centred viewpoint in order to exert even further inequality in the power relations between mediators and users. In critiquing leading cognitivist Carol Kuhlthau's book Seeking Meaning he claims that she posits the repositioning of librarians as information counsellors for their anxiety-ridden clients in the age of rapid information expansion and cognitive overload. Kuhlthau's model for the modern librarian is a physician - the client is informationally unwell. Not only is this approach obviously not really user-centred, it also is likely to reinforce prevailing notions that the problem is the user's inarticulate and anxiety-ridden state rather than an poorly designed IR system or inadequate service provision [29].

Assumptions about information seekers and information professionals are important in the theoretical underpinnings. It is my view that if both parties are to benefit from advances in information science that the epistemological basis must position the information seeker as a knowing person and the information professional as a collaborator rather than mediator. Ironically the user-centred cognitive approach ultimately ends up being profession-centred.

8. View of the System

I use the concept of system here to include mechanised systems, such as IR systems, bureaucratic and other enabling systems within institutions, but not social systems.

For Belkin focus on the system is basic. The problem for Belkin is building more 'intelligent' and interactive systems [14, p.68;16]. IR systems need to resolve rather than match requests. The key is structured dialogue between the system and user. Bryan Allen has described much cognitive systems-based work undertaken beyond Belkin but utilising similar approaches: cognitive models, user stereotypes, semantic modelling and various algorithmic exercises [30].

Dervin is wary of any focus on the system, even when the purpose of the exercise is to inform better systems design. Research with a purpose to '…inform system design and operation must adopt a utopian methodology for it is only in this way that the research can break free of unstated assumptions embedded in the normally accepted defining discourses of the system' [21, p.734].

Sanna Talja wants to conceptualise information systems as 'participation systems in the organisation and systemisation of social knowledge' [25, p.7] . It is thus important to study the socio-cultural aspects of the systems as well as the users. She considers that the inadequacies that cognitivists ascribe to information seekers, such as uncertainty, knowledge gaps, lack of understanding and limited constructs, can also be ascribed to information retrieval systems [25].

In IR people will have variable levels of skills related to their domain knowledge and how familiar they are with engaging in the retrieval process within a given domain. Their confidence and strategy would vary. In such an environment building an IR to support the variability between information seekers would be particularly challenging. It seems to me that the system at a minimum would need to have been encoded with significant data about domains and the language of those domains. The human-computer interactive environment would require considerable (artificial) intelligence at the system end. This currently looks like a big ask (pun intended). I think Talja is correct is concluding that 'the central problem facing information systems is how to incorporate multiple viewpoints into the system'[25, p.7], but suspect it is very difficult to achieve.

Belkin's work not-withstanding we have seen a decisive shift in research orientation from systems-centred to user-centred approaches. Heidi Julien makes the point that most research and discussion is related to 'users' of IR systems or institutions such as libraries, rather than 'non-users'. Her analysis indicates that most users are from the top strata of information rich groups in society. In making the call for attention to non-users, she argued that the user-centred focus of cognitivists is in fact a system-centred approach, because the people are studied because they use the system. 'Ultimately this focus perpetuates the vilified 'systems-centred' approach to our field. [31, p2]. The solution to imperfect systems then are actions like better training in how to use them.

9. View of Knowledge Creation

If as Brookes and Belkin state, information science studies 'the interaction between people and objective knowledge, in order to know more about subjective knowledge structures' [12, p.12], what are knowledge structures and how might the conception and study of them inform information behaviour? Knowledge structures are about 'knowledge representation and classification practices' according to Sanna Talja [25, p.5] . For Dervin it is about creation of the individual's 'world view', 'about making and unmaking worlds' [21, p.9]. Both viewpoints deny the process is more than transmission and processing.

It seems to me that Belkin and other cognitivists have not been successful in their 'outside looking in' approach to understand knowledge structures via cognition. Dervin's approach is more 'inside looking out' and does seem useful to inform studies of what people need in INS, and can lead to improvements in systems, and how representatives of the systems relate to their clients. While sense-making is applied in studying information use as well as INS there is no interest in knowledge per se, and little on issues such as 'information literacy'. Ross Todd has decried the lack of information utilisation studies by fellow cognitivists [11].

I suggest that sense-making's efficacy and epistemological claims are weakened as we moved beyond INS into areas such as the communication of knowledge (use, sharing) and studies of how people construct meaning. For Dervin the critical questions have been whether people are processors of information or active constructors of information? She seems to have convinced the information scientists of the latter, and extended her focus to assert that this construction can occur in a socially influenced context. But she doesn't seem to address the difference between the knowledge structures people have or what they consist of.

For social constructionists the individual creation of knowledge is not of direct interest. The interest is on 'collective reality' rather than individual subjectivity. Talja claims that most cognitivists end up theorising about subjectivity rather than 'information processes' , but that Dervin has replaced the focus on 'mental representations and subject knowledge structures with the concept of dialog' [25, p.3]. For Talja the critical question is how do you measure the difference between people's knowledge structures and thus the way they create knowledge?

The social constructivists viewpoint considers it is 'mentalism' to try to understand knowledge structures from a cognitive perspective because there is nothing that can be measured. As well cognitivism misses the point, because knowledge structures can be shown to be similar between people according to their membership of social groups with a common domain of knowledge. The differences are not caused by people interpreting the world in individually different ways, or related to levels of comprehension or skill, but rather they interpret it in socio-culturally defined way which vary between cultures and groups and knowledge domains .

Two other concepts fundamental to social construction are language and discourse. Talja asserts the basic relationship of language (spoken, written, subtext) with knowledge - 'No concepts, thoughts or meanings can exist outside language' [25, p.3]. Language is really the only way we can research knowledge creation and knowledge structures, or what she terms knowledge formations. As language and knowledge as basically synonyms we can observe knowledge through language. We know that there are different ways of looking at the world as observed from use of language, and different ways of limiting what is conceived as agreed truth or acceptable with a domain by the use of language by its members.

With 'intersubjectivity' we make individual choices when creating knowledge but these choices are strongly bound by what is perceived as truthful or appropriate within the discourse of the domain. This means that there is a self-reinforcing shared subjectivity within groups. 'Meanings, values and ethical principles are not constructed by individuals, they are constructions that have been created in social interaction' [25, quoting from Raymond Williams].

10. View of the Relationship between the Individual, the Social Environment and Information Behaviour

One of the most consistent criticisms of cognitivism has been that it fails to recognise the social aspects of information behaviour. If it can't conceptualise the socio-cultural context then it simply remains as Talja puts it 'a theory of how individuals process information' [25] It is my view that cognitivism thus has thin epistemological claims and is an insufficiently robust theory for the understanding of information seeking.

Dervin has refocused sense-making away from cognitivism:

'Sense-Making assumes that issues of force and power pervade all human conditions; that humans are impacted by the constraining forces of structural power (both natural and societal) and that as individuals in specific situations they are themselves sites of power, to resist, reinvent, challenge, deny and ignore'. [21, p.2].

She also has moved towards a social constructionist view of knowledge creation - termed 'communitarianism'. The basic assumption of communitarianism is that knowing is processual negotiation of meaning between people. 'It assumes that knowing is made and remade, reified and maintained, challenged and destroyed in communication: in dialogue, contest and negotiation [quoted in 32, p.7]'.

Whilst sense-making has moved decisively from a cognitive perspective toward social construction, there does seem a significant area of difference - and that is in perceptions of context.

When Dervin says that sense-making has evolved into a methodology useful in any context [21], she is using context as situated in background, the situation of the information seeker at the information seeker sees it. For Sanna Talja context is external to the person, fundamental and foreground [24]. The context is social. Hjorland writes 'it is much safer to study the users' subjective experiences, mental states and behaviour as autonomous phenomena than to study (social) reality and how it effects the users' [2, p.17]. In introducing his domain activity model Hjorland takes this further: 'the best way to understand information in IS is to study the knowledge domains as discourse communities, which are part of society's division of labour' [2, p.6].

Social constructionism then seems better positioned to understand the individual's information seeking behaviour within society. This promises more pattern recognition across studies than micro-studies of individuals within time-space. However for me the major challenge of the social constructionist viewpoint is to develop a methodology that will enable practical outcomes for research. To date my readings indicate that social constructionism is an excellent viewpoint and methodology for envisaging the individual within power structures and socio-cultural and historically based environments. As well discourse analysis has application in exposing the weaknesses of cognitivism. However, I can't see it informing the construction of better IR systems.

11. Conclusions

I agree with Sanna Talja that context is central. While Brookes considered (objective) information to be everywhere, I submit context is everywhere. Information behaviour can be best understood by an understanding of context. So too can information behaviour by information scientists. Below are some of my impressions of the context for the development of the epistemological frameworks covered in this critique.

For Bertram Brookes the context included the failure of information science at that time to exhibit a robust and independent epistemological basis. Information science was (and is) a fledging science. Brookes work can be perceived as an attempt to harden its epistemological base in order to increase power and influence for his profession. His allegiance to cognitivism, drawn from psychology, and positivism, drawn from the (non-social) sciences can be seen as his recognition of the dominant theoretical constructs operating at that time in the dominant areas of discourse within universities.

Nicholas Belkin's interest in IR places his focus narrowly on information searching behaviour - the interrogation of machines. His context includes his role as a leading member of the Association for Computing Machinery. A further context has been leading the large scale computer based research projects that dominate his curriculum vitae [33] is context does not require him to develop a holistic epistemological basis for information behaviour. The IT industry which he serves lacks interest. Whether information behaviour is socially constructed on not has little relevance to issues such a feedback, user enquiry modelling and the like.

In 1986 Dervin and Nilan wrote, 'It is hoped that the analysis of gap defining would finally reveal some universal elements of the ideas humans must construct to guide their movements' [8]. Dervin has sought to develop a theory that can be applied from the top level as a metatheory (a broad and loose set of ideas and constructs) as well as the level of a research methodology with inter-disciplinary application. She decries that sense-making has been used primarily as method and not for it's theoretical robustness: 'It is fair to say that sense-making's metatheory has never been fully realized either in research or in explication of method' [a href="#21">21]. There seems to be an emerging practice of INS researchers utililising a social constructionist epistemological framework yet implementing sense-making methodology. This leads me to the following assessment based on context.

The increasing influence within IS of social constructionist epistemology has exposed weaknesses in Dervin's efforts to maintain the theoretical high-ground within the domain. In her quest for epistemological and professional status Dervin has moved her theoretical basis away from the individualist-cognitive framework and towards the social constructivist position.

The theoretical basis of sense-making is rooted in information behaviour, and more specifically in information seeking behaviour. The conceptual basis is still focused on the individual creating knowledge. Where the focus of the information science activity may need to be on the relationship between man and machine (such as in Belkin's IR and HCI research) or the field of 'knowledge transfer' (or information utilisation), or the wider social context, Dervin's theoretical perspective assumes a marginal importance.

The context for social constructionists includes the tradition of marxism. The steady weakening of the theoretical basis of marxism within post-industrial society has resulted in marxists seeking new theoretical underpinnings. In a similar way that Dervin and Nilan's took up polarised position vis-à-vis positivism in the 1986 paper, social constructionists have taken up polarised positions towards cognitivism. This may be seen in context as a reaction against the dominance of US-capitalist socio-cultural values embedded in adherence to the individual at the centre, both as hero and blameful victim. In my view they have aimed well. However while the epistemological basis is robust, there is yet little evidence of social constructionism providing the methodological basis to inform research that will make a difference to the delivery of information services.

Information scientists are engaged in external and internal domain competition to assert socio-cultural power and influence. I believe a key contextual issue within IS is the relatively higher profile of computer science, relabeled as information technology. One danger is an inward pre-occupation within IS, between cognitivists and social constructionists leading to loss of focus on the bigger battle, with the result that dominating domains such as information technology, communications and engineering are able to trivialise IS.

It may be easy to trivialise IS if it does not construct practical improvements which the information economy desperately needs. Uppermost in importance is the improvement in organisation and retrieval systems. This call ramps up in urgency with the rapid move of much of the world's current document creation to the mega-collection that is the web.

The challenge for IS is for relevance. Relevance requires IS to develop a strong theoretical environment - robust sciences are expected to have theoretically based research and discourse frameworks. As Kimmo Tuominen states 'the survival of a certain scientific life-form depends to a great extent on how effectively its supporters can convince a sufficient number of influential individuals and groups of its usefulness' [23, p.3]. It's thus somewhat irrelevant which viewpoint is most epistemologically robust. It's about marketing.

My sense is that social constructionism places the study of information behaviour in the only context that has the potnetial to provide the full picture of the individual, the system and the social, political and historical environment. I also sense that with many hundreds of researchers utilising sense-making that it can provide an effective methodology for at least INS research, rather than a robust epistemological framework

A dominant intersubjectivity may emerge within the discourse of information science. It may already be emerging with a blending of the viewpoints of Dervin and the social constructionists. Pertti Vakkari indicated this when summing up the feeling of consensus and convergence at the Information Seeking in Context Conference in 1996:

'Information is [now] seen as a social construct created in the interaction of people and messages. The research object is conceptualized more and more as a social phenomenon leading to the use of variables from the organizational and societal level' [32, p.463].

Prominent information scientist Tom Wilson, who is neither a cognitivist nor a social constructionism, made a similar observation when discussing his own viewpoint in 1999:

'Just as positivism arose out of rejecting metaphysical speculation, so an alternative school has arisen out of rejecting the view that scientific empiricism can be applied to the social world. There is no one philosophical basis, but hermeneutics and phenomenology are seen as providing the basis for what is generally called Interpretative (or Interpretive) Research where the assumption is that social reality can only be understood through social constructions such as language, consciousness and shared meanings. Interpretive research does not predefine variables, but explores human sense-making in naturalistic settings' [34, slide 2].

Sixteen years on from Dervin and Nilan perhaps a new paradigm shift is called for - the social construction/social sense-making viewpoint replacing the cognitive/user-centred viewpoint. While this may do much to advance the epistemological rigour of IS, I suspect that the IR-focused cognitivists like Belkin will still attract substantial grants and maintain robust power structures within research institutes.

IS is on the margins of an enormously energised and dominant environment of the information economy. To reduce this marginalisation, a clear and effective epistemological basis is essential, but the development of research and practice which is perceived by those beyond IS to inform the development of better solutions for document organisation and retrieval and information services is critical.

The continued development of the information economy demands that IS does better. There is enormous tension around documents, corporations, profits and access. Y2K and the dot.com bubble burst has weakened the power basis of IT. Information science has a space to fill in determining the relationship between people, systems and society within the information economy, and promises the potential for provision of better services, improved access, improved systems design, possibly even solutions for fast-tracking nations away from 'information poverty', and the development of librarians as the analysts and brokers of the web delivered information environment.

References

1. Frohmann, B (1994). Discourse analysis as a research method in library and information science. Library and Information Science Research 16: 119-138.

2. Hjorland, B. (1997). Information seeking and subject representation : an activity-theoretical approach to information science. Westport, CT : Greenwood.

3. Hjorland, B. (1998). Theory and metatheory of information science: a new interpretation. Journal of Documentation 54: 606-621

4. Julien H. (1996). A content analysis of the recent information needs and uses literature. Library and Information Science Research 18: 53-65.

5. Radford, G. (1992). Positivism, Foucault, and the fantasia of the library: conceptions of knowledge and the modern library experience. Library Quarterly. 62: 408-424.

6. Lesk, M. (1997). How much information is there in the world? http://www.lesk.com/mlesk/ksg97/ksg.html (visited 04.04.2002)

7. Lyman, P. & Varian, H.R. (2001). How much information? http://www.sims.berkeley.edu/research/projects/how-much-info/ (visited 04.04.2002)

8. Dervin, B. & Nilan, M. (1986). Information Needs and Uses. Annual Review of Information Science and Technology. 21: 3-33.

9. Frohmann, B. (1992). The Power of Images: A Discourse Analysis of the Cognitive Viewpoint. Journal of Documentation 48 : 365-386.

10. Brookes, B. (1980). The foundations of information science. Part 1: Philosophical aspects. Journal of Information Science 2: 125-133.

11. Todd, R. (2000). A theory of information literacy: in-formation and outward looking. In. C. Bruce and P. Candy (eds). Information Literacy around the World. Wagga Wagga: Centre for Information Studies: 163-175.

12. Belkin, N. J. (1990). The cognitive viewpoint in information science. Journal of Information Science. 16: 11-15.

13. Belkin,N.J. (1978). Information concepts for information science. Journal of Documentation, 34: 55-85.

14. Belkin, N.J., Oddy, R.N., & Brooks, H.M. (1982). ASK for information retrieval: Part I Background and theory. Journal of Documentation, 38: 61-71.

15. Savage, P.A, Belkin, N.J., Cool,C., and Xie,H (1997). An investigation of mental models and information seeking behavior in a novel task. http://mariner.rutgers.edu/tipster3/sigir97.html (visited 13.04.2002)

16. MONGREL: Supporting effective access through user- and topic-based language models. Collaborative research with W.B. Croft and J. Allan, University of Massachusetts, Amherst. October 2000 - September 2003. http://scils.rutgers.edu/mongrel/ (visited 13.04.2002)

17. Savolainen, R. (2000). Incorporating small parts and gap-bridging: two metaphorical approaches to information use. The new Review of Information Behaviour Research 1: 35-50.

18. Frohmann, B (1990). Rules of Indexing: a critique of mentalism in information retrieval theory. Journal of Documentation 46: 81-101

19. Dervin,B. (1992). From the mind's eye of the user: the sense-making qualitative-quantitative methodology. In: J. Glazier & R. Powell. Qualitative Research in Information Management. Englewood, Co.: Libraries Unlimited: 61-84. Read as draft at http://www.mcc.ufc.br/etagi/projetobb/zendervinpowell92.html
Note that page numbers in paper refer to A4 printout of web version.

20. Murphy, T.P. et al. (1997). Sense-making methodology: trying to make sense of it all. EETAP Resource Library, no.24. http://www-comdev.ag.ohio-state.edu/eetap/pdf/info24.pdf

21. Dervin, B. (1999). On studying information seeking and use methodologically: The implications of connecting metatheory to method. Information Processing and Management, 35: 727-750. Also available at: http://communication.sbs.ohio-state.edu/sense-making/art/artdervin01.pdf

22. Savolainen, R. (1993). The sense-making theory: reviewing the interests of a user-centred approach to information seeking and use. Information Processing & Management. 29:13-28.

23. Tuominen, K. & Savolainen, R. (1997). A social constructionist approach to the study of information use as a discursive action. In P. Vakkari, R. Savolainen & B. Dervin (eds). Information Seeking in Context. London: Taylor Graham: 81-98.

24. Talja, Sanna, Keso, Heidi, Pietilainen, Tarja (1999). The production of 'context' in information seeking research: a metatheoretical view. Information Processing and Management 35: 751-763. Also available at http://www.uta.fi/~lisaka/taljakeso.pdf

25. Talja, S. (1997). Constituting 'information' and 'user' as research objects: a theory of knowledge formations as an alternative to the information-man theory. In P. Vakkari, R. Savolainen & B. Dervin (eds). Information Seeking in Context. London: Taylor Graham: 67-80.
Read as: http://www.uta.fi/~lisaka/ISIC.HTM
Note that page numbers in paper refer to A4 printout of the web version.

26. Savolainen, R. (1995). Everyday life information seeking: approaching information seeking in the context of 'Way of Life'. Library and Information Science Research. 17: 259-294.

27. Vogler, C. (1993). The writer's journey: mythic structure for writers. Studio City, Calif.: Wiese.

28. Hjorland, B. (2002). Epistemology and the socio-cognitive perspective in information science. Journal of the American Society for Information Science and Technology 53: p.257-270.

29. Tuominen, K (1997). User-centered discourse: an analysis of the subject positions of the user and librarian. Library Quarterly. 67: 350-371.

30. Allen, B. (1991). Cognitive research in information science. Annual Review of Information Science and Technology. 26: 3-37.

31. Julien H. (1999). Constructing 'users' in library and information science. Aslib Proceedings 51 6: 206-209.

32. Vakkari, P. (1997). Information seeking in context : a challenging metatheory. In P. Vakkari, R. Savolainen & B. Dervin (eds). Information Seeking in Context. London: Taylor Graham: 451-466.

33. Belkin, N.J. (2002). Nicholas J. Belkin : curriculum vitae http://scils.rutgers.edu/~belkin/newcv.htm (visited 12.04.2002)

34. Wilson, T. (1999). Structure in research methods: a new approach to the categorisation of research methods. [Doctoral workshop, Oslo College]. http://informationr.net/tdw/publ/ppt/ResMethods/ (visited 13.04.02)