Accommodating the User's Information Search Process: Challenges for Information Retrieval System Designers

The model of the information search process (ISP) described here was developed in a series of studies of the experience and behavior of library users involved in extensive research projects and more recently people in the workplace using information for complex work-related tasks. This research, discussed in numerous publications, presentations and seminars, has had considerable impact on library services but relatively little impact on information retrieval (IR) system design. With the current interest in users and user-centered systems, the time seems right for considering the implications of this research on the design of user-centered IR systems. This article is an introduction to the ISP for those involved in the design of information systems and services. It briefly presents the problem addressed, summarizes the research findings, discusses theoretical perspectives emerging from these findings, describes the impact on library and information services and suggests some implications for IR system design. This research has developed through a series of studies on the user's perspective of information seeking in libraries. In the first study in the series, common patterns in users' experience were identified and a model of the ISP was developed. Four subsequent studies further verified and refined the model. Readers interested in studying these research findings in more depth can begin with the references in For Further Reading. A full description of the model is published in Seeking Meaning: A Process Approach to Library and Information Services, while strategies for guiding students in each stage in the process are presented in Teaching the Library Research Process. Information seeking in the workplace is discussed in my latest article in the Journal of the American Society for Information Science (in press.) People using a variety of sources of information to learn about a particular subject, complex problem or extensive issue often have difficulty in the early phases of information seeking. Even when they begin with enthusiasm and initial success, many become confused and uncertain about how to proceed after a short period of time. This is particularly noticeable with students who have been assigned a research paper, but is not characteristic of students alone. Hesitation, confusion and uncertainty in the early stages of information seeking are reported by people in all types of libraries and are noted by people engaged in complicated information seeking tasks in the work place as well. In fact, although there is no way of knowing just how many people give up after initiating a search because they become uncertain and feel confused about how to continue, I suspect the number is substantial. This research into the user's process of information seeking grew from my own experience with students when I was a high school librarian. I noticed a recurring problem: no matter how well students were oriented to the library and its resources, there was a common pattern of behavior when students came to the library for the first few days to begin an extensive research project. They were confused and disoriented, often expressing annoyance at the assignment, the library and themselves. I have come to understand that this point, when people encounter an information system early in their research, whether it be a library or other type of database, is the most difficult stage of the search process. Rather than experiencing a steady increase in confidence from the beginning of a search to the conclusion, as might be expected, a dip in confidence is commonly experienced once an individual has initiated a search and begins to encounter conflicting and inconsistent information. A person "in the dip" is increasingly uncertain until a focus is formed to provide a path for seeking meaning and criteria for judging relevance. Recent study of the ISP in the workplace reveals that the user's experience of the stages in the search process is related to how much the person knows about the problem and the degree of construction that needs to be undertaken during information seeking. In more routine tasks, where the goal is to answer a simple question or to monitor periodic change, people do not usually experience stages in their information seeking. However, in more complex tasks, where the goal requires considerable learning, people are likely to experience a process as described in the ISP model. A dip in confidence seems to be a natural stage in the ISP. In Seeking Meaning, I discuss how, when the ISP is viewed as a constructive process, the work of George Kelly, John Dewey and Jerome Bruner becomes pertinent. They explain similar situations in which people are actively engaged in learning. Their work then provides a useful framework for understanding the process of information seeking. Each of these theorists describes the constructive process as occurring in a sequence of stages or phases to be actively worked through by the individual in the context of the situation. Uncertainty common in earlier stages is caused by the introduction of new information that conflicts with previously held constructs. Advances in information systems that open access to a vast assortment of resources has not eased the user's dilemma and in many cases has intensified the sense of confusion and uncertainty. New information systems may deepen the problem by overwhelming the user with everything all at once, rather than offering a few well-chosen introductory pieces for initial exploration. Can mechanisms be developed that accommodate a range of different types of information seeking tasks from routine to complex? Can mechanisms be developed that accommodate different stages in the ISP in complex tasks? 1 Model of the Information Search Process Wiley Periodicals The ISP may be thought of as occurring in six stages: Initiation, Selection, Exploration Formulation, Collection and Presentation. Common patterns of thinking, feeling and acting were found in each stage. The names of the stages represent the primary task at each point in the process. Although the sequence of tasks may appear somewhat recursive rather than strictly linear, the process proceeds from the initiation to the completion of the project. The first stage in the ISP is initiation, when a person first becomes aware of lack of knowledge, information and understanding to solve a complex problem or accomplish an involved project. Feelings of uncertainty and apprehension are commonly associated with the vague or ambiguous thoughts in this early stage of information seeking. Selection is the second stage, when the task is to identify and select the general area or topic to be investigated. Feelings of uncertainty often give way to a brief sense of optimism after a selection has been made and there is a readiness to begin the search. The next stage is exploration, which is often the most difficult stage for users and the one most misunderstood by providers of information services and designers of information systems. Feelings of confusion, uncertainty and doubt frequently increase during this time. The task is to investigate information on the general problem in order to extend personal understanding. At this stage an inability to express precisely what information is needed makes communication awkward between the user and the system. Thoughts center on becoming oriented and sufficiently informed about the topic to form a focused perspective that will guide the search and lead to accomplishing the user's goal. The fourth stage is formulation, which is the turning point of the process when feelings of uncertainty diminish as understanding increases. The task is to form a focus from the information encountered in exploration. Thoughts become more clear and defined as a focused point of view is formed. The fifth stage is collection, when interaction between the user and the system functions most effectively and efficiently. At this point the task is to gather information pertinent to the focused problem. The sixth stage is presentation, when the task is to complete the search and resolve the problem. This research has revealed that people experience the process of information seeking holistically as an interplay of thoughts, feelings and actions. These studies were among the first to investigate the affective aspects, or feelings of a person, in the process of information seeking along with the more commonly studied cognitive and physical aspects. One of the more surprising findings was "the dip" described above after a search had been initiated during the exploration stage. However, this experience is also one of the most recognizable to both librarians and library users when the model is presented to them. Users tend to think that they are the only ones to experience "the dip" before they become aware that it is a common occurrence in the ISP. In summary, this research found that from the user's perspective forming a focus is a central task of the early stages of the search process. Searching is a process over time rather than a single event. Searching is a holistic experience rather than just an intellectual activity. Initially, searching commonly increases uncertainty rather than reducing uncertainty. Uncertainty, common in the early stages of any constructive process, has not been adequately addressed in the systems and services designed for information seeking. User's uncertainty is traditionally considered a negative to be reduced as quickly as possible. Most information retrieval systems work fairly well when a problem is well defined in the collection stage of the ISP. But few are designed for a person who is learning in the exploration stage. An Uncertainty Principle has been developed to provide a theoretical framework for library and information services to accommodate uncertainty and lack of constructs in the early stages of the ISP. The following principle is based on the findings of the research into the user's perspective of information seeking described above. Uncertainty Principle: Uncertainty is a cognitive state that commonly causes affective symptoms of anxiety and lack of confidence. Uncertainty and anxiety can be expected in the early stages of the ISP. The affective symptoms of uncertainty, confusion and frustration are associated with vague, unclear thoughts about a topic or question. As knowledge states shift to more clearly focused thoughts, a parallel shift occurs in feelings of increased confidence. Uncertainty due to a lack of understanding, a gap in meaning or a limited construction initiates the process of information seeking. The Uncertainty Principle is expanded by six corollaries: process, formulation, redundancy, mood, prediction and interest. These are further explained and discussed in Seeking Meaning. Can this principle be applied to the design of information retrieval systems? If we think of uncertainty as a sign of the beginning of innovation and creativity the goal of IR design shifts considerably. Can systems be designed that do not close the person down too quickly – that are sufficiently open to accommodate exploring and formulating? These research findings have had important implications for library and information services and have had considerable impact on development of user-centered services. Many librarians are aware of process, and the term process is common in library vocabulary. Library services that formerly concentrated solely on the physical attributes of information seeking, such as locating and circulating materials, are attending to the more cognitive and affective attributes of using information for solving problems, for learning and for seeking meaning. For example, reference librarians who have become aware of stages in the ISP describe important changes in the way they view students and faculty who approach the reference desk. They now listen for an indication of the stage in the process of the user and particularly note when someone is "in the dip" and needs some extra help to explore for learning in order to formulate a research focus. They are especially careful not to give too much too soon and to assist in pacing the use of resources by suggesting strategies for exploring information to form a focus for research. Librarians planning instructional sessions describe being more cautious about offering one-shot sessions where students are expected to learn everything at once. Instead they are accommodating the user's constructive process by giving a series of instructional sessions spread over a period of time aimed at different tasks in the stages of the ISP. Once aware of the ISP, teachers also change the way they design assignments to give more time for exploring and formulating. They are acknowledging the learning process and finding new ways to access and evaluate the construction process of students. Once librarians become aware of the process of information seeking, they find new ways of intervening with users. The notion of a zone of intervention, introduced in Seeking Meaning, describes a time when or an area where the user needs the help of another to move ahead, a concept based on Vygotsky's zone of proximal development. Identifying when intervention is needed and determining what mediation and instruction are appropriate is the librarian's professional task and is important for IR system designers to consider as well. Intervention into areas where an individual is self-sufficient is not only unnecessary but may be intrusive and annoying. However, mediation into areas where individuals cannot proceed on their own, or can advance only with great difficulty, is enabling and enriching. The area where a person can do with assistance what he or she cannot do alone is the zone of intervention. The zone of intervention varies from person to person and from time to time. For some it simply involves pointing out one source in response to a single factual request. For others it encompasses extensive collaboration in the ISP. Five zones of intervention are described in Seeking Meaning along with corresponding levels of library service. The diagnostic task is an important professional skill for designing library and information services. Can the concept of zones of intervention and levels of mediation be applied to designing IR systems? Are IR systems presently designed to work at all levels of intervention or just on simple levels? Although this research has had considerable impact on library and information services, there has been little impact thus far on design of IR systems. In fact the concept of process in the holistic sense that the user experiences seems to be somewhat outside of the paradigm of IR. User-centered information systems call for a broad vision of users as people actively engaged in tasks rather than the narrow view of information searchers at the time of interface with the system. The following concepts emerging from this research may be useful for designing information retrieval systems that support innovation, creativity and learning. Concept of Process. The first and foremost concept emerging from this work has been the concept of process. In many cases, information seeking is more than source location and fact-finding or even question answering and problem solving. Ongoing studies have revealed a complex inquiry process that involves learning from a variety of inconsistent and even incompatible sources. This view of information seeking as a process in which a person is actively constructing a new understanding from the information encountered, has some important implications for the way systems are designed. Exploration and formulation within the information search process are essential considerations for developing systems that enable the user's constructive process. For example, browsing traditionally considered a haphazard and insignificant approach to information seeking is being reconsidered for its potential as a strategy in the early stages. Participants in the study of the ISP in the workplace explain that they are looking for a new angle to present rather than just reporting on facts they are finding. Uncertainty. Uncertainty is another concept that emerges from this research. There is a marked contrast between information systems that are built on a principle of order and certainty, and the individual's profound experience of uncertainty in the early stages of the information search process. These findings indicate the need for considering uncertainty as a natural, essential characteristic of information seeking rather than regarding the reduction of uncertainty as the primary objective of information seeking. Uncertainty is a concept for information retrieval design that offers insight into the user's quest for a personal perspective within the process of information seeking, what I have called "formulating a focus." When uncertainty is accepted as an important concept many other related corollary concepts become apparent. One is the notion of the relationship between uniqueness and redundancy of information. At the beginning of an information seeking process we can expect that the likelihood of encountering uniqueness (new information) will be high and redundancy (familiar information) to be low. As the process progresses and the person learns more about the problem there is likely to be more of a balance between the two types of information. Finally, at the close of the process we can expect that the ratio may be reversed with uniqueness low and redundancy high. Therefore, uncertainty may be associated with high uniqueness and confidence with high redundancy. This concept may lead to new ways of thinking about relevance judgments. Another concept related to uncertainty is the notion of the mood or stance of an individual in the process of seeking information. At the beginning of an information seeking process an invitational mood is most appropriate for opening the individual to new ideas. At the close of the process, an indicative mood might be more productive for leading the individual to summarize, organize and present existing ideas. However, this may be counterintuitive for the more uncertain one feels the more likely he or she is to take an indicative approach with prescriptive steps for guidance. Can information retrieval systems be designed to incorporate the concept of stance to assist users through the ISP? Interest is another concept associated with uncertainty. At the beginning of the process, when uncertainty decreases, interest frequently increases along with personal knowledge. These critical shifts in interest and the relation to motivation and incentive may be important factors to consider in designing user-centered information retrieval systems. Complexity. The concept of complexity is important for understanding the experience of uncertainty in the information search process. Findings suggest that it is an individual's perception of the complexity of a task that determines his or her experience of process and degree of uncertainty. Since it is the perception of complexity, and not the complexity inherent in the task, tasks cannot be labeled in advance as complex or simple. A task in which considerable construction is required is likely to be considered to have a higher degree of complexity than a task that is considered routine, regardless of the inherent complexity of the task. This more subjective approach to determining complexity may prove productive for developing systems that respond to a wide range of users in a variety of tasks. Task complexity is emerging as an important concept for understanding why and when the stages of the ISP are experienced by users in contrast to information seeking that is a more straightforward source-location and question-answering endeavor. In the workplace complex tasks that lead to innovation are often seen as time consuming and disruptive to routine. Can task complexity inform designers of IR systems? Enough. The concept of enough involves the deceptively simple question of, "What is enough?" What is enough may have been a fairly straightforward notion when a person could gather all there was to know on a problem or topic in a contained collection. The concept of enough is quite a different matter in the present information environment. Understanding "what is enough" is essential for making sense of the information available to us. Enough relates to seeking meaning in a quantity of information by determining what one needs to know and by formulating a perspective on which to build. The ISP treats the concept of enough as what is enough to make sense for oneself within a context and to accomplish the task at hand. By applying the concept of enough in each stage of the process, systems may be developed that accommodate the ability to recognize an information need, explore information on a general topic, formulate a specific focus, gather information pertaining to a specific focus, and prepare to share what has been solved, learned or created. This research addresses information use as well as information seeking. It does not stop at locating information, but is concerned with interpreting and using it. These ideas challenge some of the underlying principles of IR system as they the of information in libraries. This research becomes more pertinent as and system designers begin to their to the of more user-centered User-centered systems will need to accommodate users the as people seeking information to learn and in the context of their is with the of Information and Library at The of New in New can be by at by at or by at