Book Review: Information Need - A Theory Connecting Information Search to Knowledge Formation

The front cover tells you succinctly what this book is about; 'A theory Connecting  - Information Search – to – Knowledge Formation.'  Equally bluntly, I shall set out my credentials for this review. I am not a library/informational professional but I have an interest in delivering digital and information skills to students. I have read and reviewed this book to further my own knowledge of the subject, as well as to see what (new?) ways there are for students to use search tools and methods as well as enhance both their digital and information literacies. Accordingly, I shall treat the review in this manner as the book itself falls roughly into these two parts.

Generalities: Information Proliferation

How people search for information depends of course upon the task in hand. Even before WW2, J. D. Bernal [1] was pointing out the information flood held in journals and books and that:

‘The kind of organisation we wish to aim at is one in which all relevant information should be available to each research worker and in amplitude proportional to its degree of relevance.  Further, that not only should the information be available but also that it should be to a large extent put at the disposal of the research worker without his having to take any steps to get hold of it.’

In the USA at the same time, Vannevar Bush was thinking along similar lines with his ‘Memex’ as a possible solution [2]. Linked to this was Bush’s selection concept of associative indexing. For everyday information searching people used encyclopedias or even visited libraries – as undergraduates still are exhorted to do. To search for the required information and indeed sift through the pile was not an easy task, even in a pre-computer world. Bush suggested an operational approach and procedure:

‘This process is simple selection: it proceeds by examining in turn every one of a large set of items, and by picking out those which have certain specified characteristics.’ [2, p. 106].

Charles Cole added the emphases to the above quotation in his ‘Information Need’. In this book he explores and seeks to interpret the rather dissimilar ways ‘computer science' and 'information science' deal with 'information need'. Thus (pp. 4-5):

‘The computer science perspective conceptualises the user's information need as the input into the information system …. which produces the answer output’; ………..  ‘Information science conceptualises the user's information need as a gap in understanding that opens the door to let in information from the textural environment – an information system, for example.’

Part one (Chapters 1-8) is mainly concerned with a definition of information need and Part 2 (Chapters 9-19) deals with the ideas developed in Part 1 and 'how it works'. The book is very much like an extended academic paper and builds progressively on its initial ideas. This review thus only sketches this development.

‘Definition of Information Need’

Chapter 1 provides a statement describing the importance of 'information need' and we progress rapidly to a history of information need from Bernal and Bush in the 1940s to the 1960s. Allen's six concentric circle model [3], derived from work in large organisations, represents formal and informal channels of information flow in user searches. This is then linked to ideas of ‘critical incident decision making’ developed at MIT.  Bush’s student, Claude Shannon, added his ideas of communication information flow and the concept of noise. Diagrams in Chapter 2 are used to explain and link these ideas and concepts. There is no mathematical or formal logical treatment, other than a basic mention of 'probability', really a general statement of ‘likelihood’.

In Chapter 2, Allen’s model is developed and linked to distinctions (or combinations) of ‘need’ and ‘demand’ and Taylor’s focus [4][5] on the psychological aspects of information need. These are developed in Chapters 3, and 4, as matter of levels from Taylor’s model. Modeling user-needs brings us back to Shannon and the introduction of ‘belief’ and Jerome Bruner’s ideas together with ‘feedback’ in Chapter 6. So far, so general and theoretical. Chapter 7 is concerned with ‘Adaptation, internal information flows and knowledge generation’ and uses a circle metaphor. Starting with a bottom-up environmental stimulus of circle 1 to that of circle 5, that encompasses knowledge revision and generation. Circle 5 is labelled ‘neurological determination’ and is linked to belief systems via Stevan Harnad’s categorical perception ideas [6] to ‘neurologically derived adaptation’ and the Neolithic era. The notion of Neanderthals as being cognitively different from modern humankind (with a well developed 'enhanced working memory') is used as part of an argument that we need better focus in searching. I am no expert on primate evolution, but it appears that the jury is still out on the 'differences' between humans and Neanderthals (‘Neanderthals are alive and well and living in Europe’ [7]). These ideas follow from rapidly developing work on cognitive psychology. However, we still have so much to learn about ‘memory’ and cognitive functionality that the ideas presented here are perhaps best kept away from definitive statements about memory. For example, long-term memory and working memory are mentioned and (p. 132) associative memory (usually thought of in terms of neural networks although not developed along these lines here), There is nothing about declarative memory and its components, cognitive origins and brain functions - aspects that may be important in linking to belief systems and rapid responses to given situations.

The first section ends with a lengthier chapter (Chapter 8) on bringing these ideas together. These relate back to Taylor’s levels and that, ‘information need manifests itself to the user in a more specific way as the user nears completion of a search (p. 92) and the division into three stages: pre-focus, focusing and post-focus. These are linked to Carol Kuhlthau’s stages of her Information Search Process model [8].

How It Works

The second part of the book deals with the implementation of Cole’s theory. However, it is still, in the main, theoretical. Chapter 9 starts with the ideas that there are three types of information search based on six propositions (based upon Taylor’s ideas seen in the pre-focus, focusing and post-focus phases). Chapters 12-15 elaborate on these phases with Chapter 16 providing ‘corroborative research’ from previously published studies. However, if a theory is developed (in whatever area) then scientific methodology suggests it should be tested (rather than corroborated) in practice. Unfortunately, in my view, the theory is not tested here and the chapter tells us little about how we should achieve the focus or when we know we have achieved it (and with what resolution).

Chapter 19 poses the questions of the book, ‘what is information need and how does it work’. A subtext is the recognition of information as ‘thing’ and ‘process’ as well as (some sort of) ‘knowledge’. These recognitions also operate in time, although Cole argues that it is the achievement of ‘focus’ that is the important aspect.

To obtain a simple fact, a basic search may be all that is required. Yet on a page of web-derived information it is necessary to place value judgments on the quality of information searched for (as distinct from the quality of the source of the information). This returns us to the uncertainty for the searcher (student). The searcher cannot always know or recognise the quality of the information being searched. The search terms themselves might be wrong. I referred to Shannon's concept of 'noise' mentioned in Chapter 4 and this is explored a little further in Chapter 11. It is here that I feel there is need of more exploration when it comes to implementation. For within the concept of noise (vagueness, inexactness and imprecision [9]) also come the human failings via mistakes and errors as well as misdirection. This is where even quasi-intelligent computer behaviour takes the place, naturally and adaptably for much knowledge formation, especially where working on subject boundaries and when terms are used in unfamiliar ways. This leads of course to ontologies - although I could find no reference to ontology in the book (certainly not in the index). As an example of a mistake; on page 11 is a list of five energy-generating solutions to a problem. One term given is ‘Ranking’. I know this to be a mistake for ‘Rankine’ as it is next to ‘Brayton’ which is a correct and appropriate term in the list. This is the sort of incorrect known (as opposed to a ‘known known’) that might send a searcher in the wrong direction unless the noise has been recognized from prior knowledge.

An example illustrating the ideas presented in Chapter 18 is that of researching and writing an undergraduate social science essay. This essay is left to the last minute and we again meet Kuhllthau’s Information Search Process ideas involving uncertainty and mental state [8]. Chapter 18 envisions ‘The Astrolabe’; ‘an information system for stage 3 information exploration’ which is suggested as the tool for ‘breaking the back of producing an essay’. The ‘astrolabe’ is used here as a metaphor for a navigation device and for using Google searches. Aside from the noise and errors induced by ‘Astrolabe (an actual device used widely in medieval times by navigators and astronomers to determine latitude, longitude, and time of day’, p. 179) it is not a system I would recommend to students for doing research. Aside from difficulties of knowing the veracity of statements and providing the focus to give a coherent whole (essay), the procedure gives no indications of time to completion. The necessity to use planning procedures (such as Concept Maps) to plan searches from an existing knowledge base are not covered, despite being a commonly-used tool for this purpose.

Some formal logic might help in Cole’s analysis. The addition of Bayesian methods would be of great assistance in approximating the requirements of a search and starting with ‘prior’ knowledge’. It would show how a focus could be achieved and indeed modified, according to the level of search. Are we any the wiser with the 'astrolabe system' presented in Chapter 18? My view is that we are not. Way back in Chapter 8, Cole provides a brief discussion on ‘Evolutionary adaptation and information foraging’. Without embarking on another review and comparing the two ideas, my preference is for Pirolli’s [10] foraging theory as it does investigate human-technology interactions in an adaptive manner and has web searching as its metier.

Conclusions

Should you read this book? Of course, it depends what you are after. If it is looking at an updating of Taylor’s and Kuhlthau’s ideas and what was then known as the 'man-machine interface', as well as developing what we would now might call 'workflows', then possibly. If you are interested in the general problem of tackling a semantic problem, fine. If you are involved with developing better search models for library users, then I fear it has little to offer over Kuhlthau on a practical level, and Pirolli on a theoretical. Can we combine the two? I sympathise with what Cole is trying to do:  (p. 194-5). ‘Only the levels perspective on information need will achieve the linkage of knowledge formation and information access in information system design that we believe necessary to overcome the information overload issues we all face in a pre-focus stage of performing an information-based task, and then information literacy issues the informationally disadvantaged in our society face because they see information, information use, critical thinking, and the information age as strange abominations that only others can use.’

In a Higher and Further Education sector, where I believe that students are poorly served by academics in delivering digital and information literacies [11][12], instructional roles tend to fall on library and information professionals. Cole’s book is a start to help the practitioner but for me the book needs to provide more practical assistance beyond the theory. In particular, there is a need to look towards the Semantic Web and perhaps (Ward) Cunningham’s Law, ‘The best way to get the right answer on the Internet is not to ask a question, it’s to post the wrong answer’ [13].

References

1. Bernal, J.D. The Social Function of Science, Faber and Faber, p118
2. Bush, V. 1945. As we may think. Atlantic Monthly 176, 101-108.
   http://engl114-pressman.commons.yale.edu/files/2008/09/vbush.pdf
3. Allen, T.J., 1969. Information needs and uses. In C.A. Cuadra (Ed) Annual Review of Science and Technology.
4. Taylor, R.S., 1962. The process of asking questions. American Documentation, 13, 391-396.
5. Taylor, R.S., 1968. Question-negotiation and information seeking in libraries. College and Research Libraries 29 (3) 178-194. http://www.dtic.mil/cgi-bin/GetTRDoc?AD=AD0659468
6. Harnad, S. 1987. Category induction and representation. In S. Harnad (Ed). Categorical Perception: The Groundwork of Cognition. Cambridge University Press.
7. The Neandertal Genome, Science/AAAS, 7 May 2010
   http://www.sciencemag.org/site/special/neandertal/
8. Kuhlthau, C.C.,1993. Seeking Meaning: A process approach to Library and Information Services. Norwood, NJ. Ablex. and London: Libraries Unlimited. ISBN 1-59158-094-3 and see: http://library.humboldt.edu/ic/general_competency/kuhlthau.html
9. Swinburme, R.G., 1969. Vagueness, Inexactness, and Imprecision. British Journal for the Philosophy of Science, 19, 281-299.
10. Pirolli, P., 2007. Information Foraging Theory. Oxford University Press, pp. 204.
11. JISC, 2009/2012. Learning Literacies in a Digital Age. Briefing Paper.
    http://www.jisc.ac.uk/publications/briefingpapers/2012/developing-digital-literacies.aspx
12. JISC 2012, Learning in a Digital Age
    http://www.jisc.ac.uk/publications/programmerelated/2012/learning-in-a-digital-age.aspx
13. Ward Cunningham http://en.wikipedia.org/wiki/Ward_Cunningham

Author Details

W. Brian Whalley
Visiting Professor
University of Sheffield
Sheffield, UK

Email: b.whalley@sheffield.ac.uk
Web site: http://brianbox.net/

Brian Whalley has retired from formal education and research in HE but now spends time catching up on previous sins of omission in writing up his glaciology and geomorphology research as well as contributions to student use of computers in fieldwork. He is a co-investigator on an Higher Education Academy-funded project on ‘Enhancing Fieldwork Learning’. He is also involved with developing metadata for fieldwork images and data and personal learning environments focused on tablet computers.