Conceptualisations of the immediate and contextual impetus for information seeking Open Access

Since the early 1980, researchers have developed a number of models for information behaviour (IB) and information seeking (Given et al., 2023, pp. 141–156; Wilson, 2017). The models offer various explanations about why and how people engage in information seeking. For example, it is suggested that information seeking is prompted by information need (Freund, 2015), knowledge need (Choo, 2023) and uncertainty (Kuhlthau, 1993). Due to the diverse approaches such as these, the picture of the factors triggering information seeking is still fragmented and somewhat unclear. Overall, there is a paucity of investigations offering a systematic view on the diverse conceptualisations dealing with the impetus for information seeking. Rare examples include the study in which Savolainen (2017b) examined how researchers have conceptualised information need as a trigger and driver of information seeking. The findings indicate that information need does not purely function as an impetus for information seeking because information need is experienced in a particular context, for example, work task performance. A contextual factor, for example, the urgency of the work task at hand, precedes the experience of information need. In this case, information need occupies the role of an immediate impetus directly setting information seeking in motion, while the urgency of a work task functions as a contextual impetus affecting how the information need is interpreted.

The present study makes use of the above conceptual distinction by examining the ways in which researchers have conceptualised the immediate and contextual impetus for information seeking. In the present study, the term impetus is used interchangeably with the term trigger. Both impetus and trigger are understood as factors that cause something to start or happen. As Savolainen’s (2017b) investigation was limited to a single impetus, that is, the construct of information need, there is a need to take a broader view on the above topic by including triggering factors of other kinds. To achieve this, a conceptual analysis was made to find out how the immediate and contextual triggers are conceptualised in explanatory models for IB and information seeking (e.g. Freund, 2015; Wilson, 1997). Such models offered particularly relevant research material for the study because the relationships between the triggers and information seeking are specified in diagrams and explained in greater detail in texts explicating the content of the diagrams. The research topic is significant because the findings deepen our understanding of what makes people start seeking information. The study pioneers by offering a nuanced picture of the immediate and contextual triggers, as well as their relationships. The findings also provide insights about how to elaborate the conceptualisation of triggering factors in future models for IB.

The rest of the article is structured as follows. First, to give background, the explanatory elements in the models for IB and information seeking are characterised. Thereafter, the research approach, research questions and methodology are specified, followed by the report of the research findings. The last sections discuss the findings and reflect their significance.

A scientific model is a simplified representation of a portion of reality, either an object or a process. Such models highlight what are considered to be key elements or parts of the object or process and the connections among them (Shoemaker et al., 2004, p. 110). Scientific models do not solely describe phenomena but can also explain them by specifying the relationship between cause and effect. In scientific explanations, the notion of cause and effect can be expressed by a number of different words, for example, “bring about”, “make happen” and “catalyse” (Johansson et al., 2024, p 18). If we have reason to believe that two events or states of affairs are related as cause and effect, they are labelled as cause and effect according to their timing; the first occurring is the cause of the other one, the effect (Johansson et al., 2024, p. 47).

Scientific models with explanatory elements can be textual, mathematical, or graphic constructs. For the present study, models combining textual and graphic constructs such as diagrams, are most pertinent. This is because most models for IB or information seeking are presented in the form of diagrams describing, for example, the causes and consequences of an information-seeking activity in a sequential or cyclic fashion (Wilson, 1999, p. 250). Textual constructs include the statements explaining how the components and their relationships presented in a diagram should be interpreted. The components of such models may include, for example, information need as a cause and the modes of information seeking as a consequence or outcome. On the other hand, it is often impossible to depict all causal connections in the diagrams because this would make the graphic constructs unreadable. Nevertheless, as noted by Case and Given (2016, p. 185), it is possible to identify the key factors and their likely sequences and interactions in the process of information seeking. Models make these aspects explicit by means of graphic symbols such as arrows. Usually, they are used to indicate causes and effects, for example, uncertainty → information seeking (Given et al., 2023, p. 144). There are also hybrid types that use both flowchart symbols and boxes to indicate sequences and variables (e.g. Robson and Robinson, 2013).

As Savolainen (2017a, p. 594) has summed up, the 1980s and 1990s were a “golden age” for the modelling of IB because many of the classic frameworks were developed at that time. Highly influential contributions include the model of information-seeking behaviour proposed by Wilson (1981), the Information Search Process (ISP) model (Kuhlthau, 1993), the Comprehensive Model of Information Seeking (Johnson et al., 1995), the model for information-seeking of professionals (Leckie et al., 1996) and the general model of IB (Wilson, 1997). The above models also offer pioneering conceptualisations about the immediate and contextual impetus for information seeking, including, for example, basic needs (Wilson, 1981) and work role (Leckie et al., 1996). Compared to the “golden age”, there has been less interest in the development of general models for IB since the beginning of the 2000s. The interest has shifted to the creation of context-specific models depicting information seeking among diverse groups of people, for example, software engineers (Freund, 2015), as well as information seeking about an issue such as climate change (Choo, 2023). Nevertheless, as the findings of the present study demonstrate, the models developed more recently have enriched the picture of the impetus of information seeking.

The above review suggests that IB models incorporate components that are pertinent for the analysis of the causes of information seeking. Following Wilson (2000, p. 49), the scope of IB models is defined by understanding information behaviour as “the totality of human behaviour in relation to sources and channels of information, including both active and passive information seeking, and information use”. Information seeking is approached as a subdomain of IB “encompassing the range of activities dealing with identifying, discovering and accessing information resources (humans and systems) in response to goals and intentions” (Wilson, 2000, p. 49). Moreover, information-searching behaviour is understood as a “subset of information seeking, referring to the actions involved in interacting with an information search system” (Wilson, 2000, p. 49). On this basis, the models examined in the present study include frameworks which conceptualise information seeking as an integral part of IB (e.g. Wilson, 1997), models focused on information-seeking behaviour (e.g. Byström and Järvelin, 1995), and models for information search(ing) (e.g. Kuhlthau, 1993). In addition, models conceptualising information seeking about particular topics such as climate change (e.g. Choo, 2023) are included. However, to sharpen the focus of the study, models primarily characterising information retrieval (e.g. Ingwersen and Järvelin, 2005) were excluded. For similar reasons, models focusing on collaborative information seeking, searching and retrieval were excluded (e.g. Karunakaran et al., 2013). As the present study concentrates on a limited set of explanatory models that explicitly conceptualise the immediate and contextual triggers for information seeking, the selection of the frameworks is potentially biased. The sample of models excludes explanatory frameworks that conceptualise the immediate and contextual triggers specified in studies in which information seeking, information search and information retrieval are approached in an integrated way (e.g. Chowdhury et al., 2014; Ingwersen and Järvelin, 2005; Pharo, 2004). As triggers of this kind also deal with information search and retrieval, the analysis of models such as these would require a separate investigation. The exclusion of models for collaborative information seeking is another source of potential bias. So far, however, the repertoire of such models is very limited. In fact, the framework proposed by Karunakaran et al. (2013) is probably the only explanatory model conceptualising collaborative information seeking. As their model would represent an “outlier” in the sample, it was excluded from the study.

To elaborate on the picture of the impetus for information seeking, the present investigation was oriented by the conclusion drawn by Savolainen (2017b). As reviewed in the introductory section, the impetus for information seeking may appear at two stages. A contextual impetus, such as a requirement of a work task at hand, precedes an immediate impetus such as an information need. The contextual trigger affects how the immediate impetus is interpreted and how the latter finally prompts the information-seeking process. Drawing on the above assumptions, the present study seeks answers to the following research questions:

The research material was identified by searching seven major databases: Academic Search Ultimate (Ebsco), Google Scholar, Library and Information Science Abstracts (LISA), Sage Journals Online, Science Direct, Scopus and Wiley Online. The search terms included information behaviour, information seeking, information searching and model. It appeared that the searches extensively identified the literature relevant to the research topic. Diverse databases such as LISA, Science Direct and Scopus retrieved the same items, thus suggesting that the sample of studies is saturated and that additional searches from other databases would not result in the identification of new material relevant to the research topic. In addition, to check the coverage of the research material, major reviews of IB research (Given et al., 2023; Wilson, 2017) were examined. The searches resulted in the identification of 42 potentially relevant investigations. A closer reading revealed that 22 of these studies are marginally relevant because they mainly repeated the descriptions of models presented in earlier investigations. Similarly, studies that did not explicitly conceptualise the triggers of information seeking were excluded. Examples include the model of information seeking characteristics proposed by Ellis (1989), the nonlinear model of information seeking developed by Foster (2004) and the model of information practices proposed by McKenzie (2003). Thus, the final sample includes 20 studies published within the period of 1981–2023. These investigations propose altogether ten models with diagrams and accompanying textual explanations, focusing on the immediate and/or contextual impetus for information seeking. Most of the ten models date back to the 1980 and 1980s (for example, Kuhlthau, 1993; Krikelas, 1983; Wilson, 1981, 1997). The focus on older (classic) models such as Johnson et al. (1995) and Wilson (1981, 1997) is simply due to the fact that they represent the best examples of frameworks explicitly conceptualising the immediate and contextual triggers of information seeking. By means of diagrams, these models also specify the components of the immediate and contextual triggers, as well as their relationships. Since the beginning of the 2000s, there has been a paucity of models of this kind. This is reflected in the fact that the sample of ten models examined in the present investigation includes only two frameworks developed in the 2000s (Choo, 2023; Freund, 2015).

The research material was examined by means of conceptual analysis. It is an approach that treats the components of the study objects as classes of objects, events, properties or relationships (Furner, 2004). The analysis involves defining the meaning of a concept, for example, immediate impetus and its attributes, for instance, information need and uncertainty, by identifying and specifying the contexts in which it is classified under the concept in question. In the present study, the context is a model. To conduct the conceptual analysis, relevant text portions (paragraphs and sentences) characterising the factors indicative of the immediate and contextual impetus for information seeking were first identified from the research material. More specifically, to answer RQ1 and RQ2, the characterisations of such factors were identified by scrutinising the diagrams presenting models for IB or information seeking, as well as their textual explanations. Relevant terms indicative of the immediate impetus include, for example, need (Krikelas, 1983) and cognitive uncertainty (Kuhlthau, 1993), while contextual impetus is exemplified by the constructs of work role (Leckie et al., 1996) and subjective task (Byström and Järvelin, 1995). Immediate and contextual impetus were identified by looking at the (temporal) order of the triggering factors presented in the diagrams. As a cause of action, immediate impetus, for example, uncertainty is placed closest to the effect, that is, information seeking, while contextual impetus, for example, work task, is placed before the immediate impetus. This setting can be illustrated graphically as contextual impetus → immediate impetus → information seeking. The analysis was continued by scrutinising how the textual explications of the diagrams substantiate the characterisations of the immediate and contextual triggers. Finally, the explications of individual triggers, for example, information need and uncertainty, presented in diverse models were compared by seeking similarities and differences between the characterisations. The comparative analysis was continued until no new similarities or differences were found.

The analysis of the ten models indicated that researchers have taken six major approaches to the conceptualisation of the immediate impetus, while the characterisations of the contextual triggers were less consistent. Therefore, for the sake of simplicity, the findings are reported by taking the content of an immediate impetus, for example, information need or uncertainty, as a point of departure. This approach is also preferable because the immediate impetus occupies a central role in the initiation of the information-seeking process. As the contextual triggers function as “background catalysts”, they are reviewed in relation to the immediate triggers.

The pioneering models of information-seeking behaviour were presented in the early 1980s by Wilson (1981) and Krikelas (1983). Common to their contributions is the idea that information seeking is triggered by human needs. Since the 1960s, it has been commonly assumed that people engage in information seeking in order to meet their information needs. The popularity of this construct may be partly due to the impactful review articles on “information needs and uses” published in the volumes of the Annual Review of Information Science and Technology (ARIST) since the mid-1960s (e.g. Menzel, 1966). The ARIST articles reflected the consensus among researchers about the centrality of the concept of information need, particularly in the domain of Library and Information Science. There were also impactful articles such as Taylor’s (1968) study, where he identified four levels of need for information (or information need) articulated by library users in the context of reference interview. Taylor proposed that a need providing an immediate impetus to an information-seeking process may first exist at a pre-unconscious “visceral” level and that such needs cannot always be expressed in detail.

Wilson (1981) offered an alternative perspective on the triggers for information seeking by arguing that information need is not a primary need, but a secondary need that arises out of needs of a more basic kind. More specifically, he suggested that information seeking is prompted by an interrelated complex of basic needs, that is, physiological needs (e.g. the need for food), affective needs (e.g. the need for attainment) and cognitive needs (e.g. the need to plan) (Wilson, 1981, pp. 7–8). The complex of basic needs was posited as an immediate trigger of information seeking. Wilson proposed that basic needs can be interrelated so that physiological needs can trigger affective and/or cognitive needs. Moreover, affective needs may give rise to cognitive needs, while cognitive needs may generate affective needs. On the other hand, the constellation of basic needs may appear differently in various contexts. Of them, the work context is particularly central. Therefore, work-related factors function as significant contextual triggers of information seeking. Wilson’s (1981) model is presented in Figure 1.

The nested boxes in Figure 1 depict the relationships between immediate and contextual triggers of information seeking. We may think that the combination of basic needs may appear differently, so that in certain contexts, cognitive needs may prevail when the person is immersed in writing at her desk, for example. At the same time, physiological and affective needs may have a marginal role as an immediate impetus for information seeking. The contextual triggers appear at diverse levels of generality. The most concrete level is the work role and the requirements of work task performance, while work environment, socio-cultural environment, politico-economic environment and physical environment function as general level background factors indirectly affecting an individual's willingness to engage in information seeking. The model also suggests that needs per se do not necessarily result in information-seeking behaviour. For example, a cognitive need, such as the need to learn more about an issue, does not always trigger the response of information seeking. This is because personal, interpersonal (or role-related) and environmental factors (labelled as “barriers”) may inhibit information seeking.

Taken together, Wilson’s (1981) early model exemplifies a strong contribution to the analysis of the triggers of information seeking. The diagram identifies clearly the components of the immediate and contextual triggers and their causal relationships, indicated by arrows. Moreover, the textual explanation of the model is presented clearly and in sufficient detail while characterising the qualities of basic needs and contextual factors of information seeking. On the other hand, the model suggests that the identification of the ultimate impetus for information seeking is a demanding task. This is because the elements of the contextual impetus are embedded in the interplay of numerous factors at diverse levels of generality.

Another pioneering model positing the construct of need as an immediate impetus for information seeking was proposed by Krikelas (1983). His model is presented in Figure 2.

The diagram suggests that information seeking is most directly triggered by immediate or deferred needs experienced by the information user. On the other hand, the textual explanation of the model reveals that the diagram does not tell all about the factors setting information seeking in motion. Krikelas (1983, p. 8) argued that need is a nebulous concept that is difficult to specify; therefore, it is more useful to define need as “recognition of the existence of uncertainty”. Although need is ultimately defined as a derivative of the concept of uncertainty, Krikelas preferred the term need, while the component of uncertainty was not included in the diagram. More specifically, need was defined as a “function of extrinsic uncertainty produced by a perceived discrepancy between the individual's current level of certainty about important environmental objects and criterion state he seeks to achieve” (Krikelas, 1983, pp. 6–7). Said otherwise, information seeking is initiated when someone perceives that the current state of possessed knowledge is less than needed to deal with some issue or problem. From this perspective, need (immediate or deferred) is the direct trigger of information seeking, while uncertainty is an integral element of this impetus.

In order to sharpen the picture of the immediate impetus, Krikelas (1983, p. 8) posited that the type of need is also affected by the nature of the environment in which information is sought. To this end, Krikelas (1983, p. 11) referred to “need-creating event/environment”, which can produce both immediate and deferred needs. Thus, such environments occupy the role of a contextual trigger. Environments of this kind may include “a wide range of personal and job-related situations; they may also respond to some specific (perhaps dramatic) event”, as Krikelas (1983, p. 8) put it. In such environments, activities associated with satisfying immediate needs were postulated as “information-seeking behaviour”, while those associated with deferred needs were defined as “information gathering behaviour” (Krikelas, 1983, p. 8). Deferred needs lead to storing information in memory and/or physical media, while immediate needs are dealt with as though they were urgent (Case and Given, 2016, p. 149). Deferred needs prompt information gathering from internal sources, such as one's memory, while immediate needs give rise to information seeking from internal, as well as external sources.

Krikelas's model exemplifies an early attempt to conceptualise the triggers of information seeking. The diagram depicted in Figure 2 describes causal processes generally flowing downward, suggesting that deferred and immediate needs prompt information gathering or information seeking, while the “need-creating event/environment” was assumed to function “in the background” as a contextual trigger. On the other hand, Krikelas explained the nature of immediate and deferred need at a general level only. In this regard, the construct of immediate need has remained vague because it is also associated with information giving (or information sharing). The diagram suggests that information giving affects the formation of an immediate need; however, the nature of this connection is not explicated in detail. All in all, as Savolainen (2017b, p. 8) concluded, Krikelas's early conceptualisation of the triggers of information seeking remained somewhat rudimentary.

Different from the above approaches advocating needs as immediate triggers of information seeking, researchers have traditionally proposed that information need is the factor that in the final end makes people seek access to information sources. One of the impactful models drawing on this assumption is the framework developed by Byström and Järvelin (1995) (see Figure 3).

The diagram depicts the relationship between (work) task complexity and information-related activities in the workplace context using iterative feedback loops and paying attention to both personal and situational elements. The arrows depicted in Figure 3 suggest causal relationships between the components since the qualities of a subjective (work) task – a contextual trigger – is assumed to affect the ways in which information need is experienced by the information seeker. The requirements of the task, most notably its perceived complexity affects the formation of the information need. During this process, other contextual factors such as the situation in which the work task is performed, and an individual's qualifications (education and experience) affect the ways in which the information need is interpreted. On the other hand, the textual explication of the model reveals that the characterisation of the immediate impetus, that is, information need, has remained at a general level. Information need is qualified regarding the types of information required in work task performance, that is, the need for problem information, domain information, and problem-solving information (Byström and Järvelin, 1995, pp. 195–196). This approach was preferred because the information need was understood as an act to determine how to handle the information requirements for the task at hand. Given the general-level approach to information need, its nature as the immediate impetus remains somewhat vague. This is due to the fact that Byström and Järvelin devoted their main attention to the elaboration of the contextual trigger, that is, the subjective task with regard to its complexity.

Similar ideas of the central role of information needs as an immediate impetus and work task-related factors as contextual triggers are presented in the framework developed by Leckie et al. (1996). They named it a general model of information-seeking behaviour and suggested that it would be applicable to all professionals. The model is depicted in Figure 4.

Different from the model proposed by Byström and Järvelin (1995), work role is posited as an ultimate contextual impetus which affects work task as a more concrete level contextual trigger. It, in turn, influences how the information need is interpreted. The textual explication of the model specifies that information needs arise out of situations pertaining to a specific task that is associated with one or more of the work roles played by the professional in the course of daily practice (Leckie et al., 1996, p. 182). The ways in which such requirements are interpreted give rise to the experience of information need, which triggers attempts to seek information from diverse sources. However, this occurs only if certain contextual conditions are met; for example, a work role such as project manager requires that an individual engage in information seeking.

Again, however, a specifying question arises about how the contextual triggers would causally influence the initiation of information seeking? Work roles and tasks, per se, may not prompt information seeking, but they affect the ways in which the information need is interpreted regarding the importance and urgency of a work task at hand. As critiqued by Savolainen (2017b, pp. 13–14), the nature of information need as an immediate impetus is not reflected in depth in the textual explication of the above model. However, it is acknowledged that information need is not constant because its content is context sensitive. The ways in which information need is interpreted is affected by a variety of intervening factors (Leckie et al., 1996, pp. 182–183). They include, for example, individual demographics (e.g. career stage), frequency (recurring need or new), predictability (anticipated need or unexpected), importance (degrees of urgency) and complexity (easily resolved or difficult). Taken together, the level of complexity, the degree of importance and urgency, and whether the information need is anticipated or unexpected together will affect when and how strongly an information need will trigger the information-seeking activity.

More recently, the constellation in which context-sensitive information needs functions as an immediate impetus is conceptualised in the model proposed by Freund (2015). Similar to the framework developed by Leckie et al. (1996), her model deals with information-seeking behaviour of professionals, more specifically, of software engineers. Freund's model is depicted in Figure 5.

Freund's model further refines the nature of contextual triggers affecting the ways in which information need is interpreted as an immediate impetus. First, two types of conditioning variables, that is, information requirements, for example, level of detail and information seeking constraints, e.g. findability of information, affect how information need is interpreted (Freund, 2015, p. 1601). Second, the interpretation of information need is affected by a variety of contextual triggers. In this regard, the person's work role, expertise and familiarity with a work task constitute a general level contextual impetus. In addition, the nature of the work project, for example, its stage and length, as well as the type of work task and the information-seeking task at hand affect more directly the ways in which the information need is experienced and interpreted.

Taken together, Freund's model demonstrates in a nuanced way how the contextual triggers are based on the interplay of dynamically changing factors relevant to work task performance. On the other hand, the complex setting makes it even more difficult to trace how an individual contextual trigger would influence the immediate impetus before the person decides to seek access to an information source. All that can be said is that this may vary situationally and case by case. From this perspective, an individual contextual impetus per se is not important. Instead, the interpretation of the information needed depends on how a certain combination of the contextual triggers appears during work task performance.

As reviewed above, the model proposed by Krikelas (1983) suggests that uncertainty is an integral element of the construct of need, which functions as the immediate impetus for information seeking. Kuhlthau (1993) ISP model exemplifies frameworks in which uncertainty occupies the major role as a trigger that immediately sets information seeking in motion. The model is presented in Figure 6.

The conceptualisation of the impetus for information seeking is affected by how Kuhlthau (1993) approached it in the particular context of learning task performance, for example, writing an essay. As the diagram indicates, the stage of Initiation is particularly important for the conceptualisation of uncertainty as the immediate impetus. In a later explication of her model, Kuhlthau (2004, p. 25) emphasised that in this regard, the interplay of cognitive and affective experiences is crucially important. The nature of such an interplay is crystallised in the Uncertainty Principle formulated by Kuhlthau (1993, p. 347): “Uncertainty is a cognitive state that commonly causes affective symptoms of anxiety and lack of confidence … the affective symptoms of uncertainty, confusion and frustration are associated with vague, unclear thoughts about a topic or question … uncertainty due to a lack of understanding, a gap in meaning, or a limited construct initiates the process of information seeking”.

The diagram presented in Figure 6 confirms the assumption that uncertainty is primarily posited as an affective factor. This is because it manifests itself in the feelings experienced by an information seeker at the initiation stage of the ISP. On the other hand, the textual explication of the model suggests that cognitive uncertainty is more fundamental regarding the immediate impetus of information seeking. As explained by Kuhlthau (1993, p. 343), uncertainty first manifests itself as cognitive uncertainty, appearing as unclear thoughts or gaps in meaning. Uncertainty of this kind then gives rise to affective uncertainty appearing in negatively coloured feelings such as apprehension, doubt and irritation (Kuhlthau, 1993, p. 347). Similar to Krikelas (1983), Kuhlthau (1993, p. 343) suggests that uncertainty precedes the recognition of the need for information. Different from Krikelas's model, however, Kuhlthau’s (1993) diagram does not include the component of “need” (for information). As to the conceptualisation of the immediate impetus, that is, uncertainty, the diagram offers a clear answer. On the other hand, the nature of the contextual impetus remains somewhat unclear. As cognitive and affective uncertainty are experienced in the context of a (learning) task at hand, we may think that the ultimate impetus can be traced to the requirements of the task at hand, for example, its urgency or importance. However, while explicating her model, Kuhlthau did not much characterise the nature of the task; the emphasis was laid on the specification of the nature of cognitive and affective uncertainty at various stages of the ISP.

Another example of models in which uncertainty occupies a central role is the framework proposed by Wilson (1999). He departed from the assumption that uncertainty is a basic notion in IB and that much information seeking is occasioned by uncertainty (Wilson, 1999, p. 265). Different from Kuhlthau (1993), however, Wilson conceptualised uncertainty as an impetus of information seeking in the context of problem solving. It is assumed that in this particular context, the goal of uncertainty resolution is basic to the use of any information resource. Wilson’s (1999) model is depicted in Figure 7.

The diagram suggests that information seeking is goal-directed behaviour serving the ends of the resolution of the problem. The model identifies stages that an individual passes through in his or her information-seeking behaviour: problem identification (“What kind of problem do I have?”), problem definition (“Exactly what is the nature of my problem?”), problem resolution (“How do I find the answer to my problem?”), and, potentially, a solution statement (“This is the answer to the problem”). The model hypothesises that each stage sees the successive resolution of more and more uncertainty and that where uncertainty fails to be resolved at any one stage, it may result in a feedback loop to the previous stage for further resolution (Wilson, 1999, p. 266).

The model depicted in Figure 7 is a generic framework that can be applied to any information problem. Different from the models proposed by Krikelas (1983) and Kuhlthau (1993), no assumptions are made about the existence of information need because it is assumed that the requirement of uncertainty resolution functions as an immediate impetus for information seeking. Even though uncertainty triggers information seeking at various stages of the problem-solving process, it is most likely that the immediate impetus appears when the individual, for the first time, identifies the existence of a problem. Compared to Kuhlthau’s (1993) framework, Wilson’s (1999) model is less detailed in that it does not differentiate between cognitive and affective uncertainty. However, as Wilson (1999) primarily approaches problem-solving as a rational process, we may think that cognitive uncertainty would occupy a central role as an immediate impetus of information seeking. Wilson’s (1999) model does not offer a direct answer to the question about the nature of the contextual impetus. Nevertheless, it can be traced to the nature of the problem at hand, for example, its urgency, because it may prompt an individual to ask herself: “what kind of problem do I have?”

The problem-solving framework reviewed above was preceded by Wilson (1997) perhaps most impactful contribution to IB research. His model for IB depicted in Figure 8 incorporates information-seeking behaviour, information processing and use. Nevertheless, the model is highly relevant for the examination of the triggers of information seeking because Wilson (1997) offered a new perspective on their complex nature.

Different from his prior model (Wilson, 1981), no reference is made to basic needs as immediate triggers of information seeking. Thereby, the question about the nature of such factors remains somewhat open. Instead, the model devotes more attention to diverse factors functioning as contextual impetus for information seeking. As suggested by the arrows presented in the diagram, the contextual impetus of information seeking can be located in two factors, that is, person-in-context and the context of information need. In the textual explication of the new model, the above components are characterised at a general level only (Wilson, 1997, p. 568). Person-in-context is primarily associated with a situation in which an individual decides to seek information, while the construct of the context of information need is not depicted in greater detail. The same applies to the construct of information need, which was terminologically “rehabilitated” and acknowledged as a relevant construct of IB research.

As Figure 8 suggests, information seeking is also propelled by factors constitutive of the activating mechanisms. Most notably, such factors include stress, the weighing of risks against rewards and the level of self-efficacy. On the other hand, the model does not offer a conclusive answer to the question about the immediate impetus for information seeking. The diagram suggests that the weighing of risks against rewards, and the level of self-efficacy, would be most closely related to the component of information-seeking behaviour. However, as Wilson (1999, p. 265) noted later on, the experience of stress may even implicate a causal nature in that it can bring people to seek information that is useful for problem-solving in particular. From this perspective, the experience of stress can serve as an immediate impetus for information seeking. We may think that a person is likely to start seeking information, particularly if he or she experiences stress-related information need in a personal context. In addition, he or she believes that the outcome (reward) of information seeking outweighs the risk of obtaining information that increases the stress and is convinced of his or her capabilities to find relevant information. Given this assumption, however, the question about the identification of the trigger most directly prompting the information-seeking process remains open since the immediate impetus may be formed by a situation-specific combination of the three activating factors.

Over 2 decades later, Wilson (2020) elaborated his model by replacing the term information-seeking behaviour with the construct of information discovery. It was considered a more appropriate term because purposive information seeking is only one of the activities through which people discover information (Wilson, 2020, p. 40). On the other hand, the construct of “context of information need” is retained in the new framework. Wilson (2020, p. 41) noted that, as to factors triggering information discovery, there is an information need, but it is not dominant in consciousness at the time relevant information is received. This is because the information received brings the need into consciousness: the person has not been actively seeking information to satisfy that need at this point. The notion of serendipitous discovery or information encountering applies here, since the acquisition of relevant information is more or less accidental. On the other hand, one of the modes of information discovery, that is, monitoring, does not necessarily satisfy any immediate information need, but simply to keep abreast of developments in areas of interest. Taken together, Wilson’s (2020) model does not offer anything radically new to the conceptualisation of the triggers of information seeking. As previously, the context of information need, related to person-in-context, is assumed to be a major factor that affects how an individual interprets a stressful situation, weighs the risks or costs of information seeking against rewards or outcomes, and assesses his or her capabilities to discover relevant information.

Different from models in which need/information need, uncertainty or diverse activating mechanisms function as immediate impetus for information seeking, the Comprehensive Model of Information Seeking (CMIS) posits that such triggers can be located in the perceived qualities of information sources, while diverse antecedent factors precede them as contextual impetus (Johnson et al., 1995). CMIS was originally developed in the context of health-related information seeking, but the model can also be applied in other domains, for example, organisational information seeking (Johnson, 1996). CMIS is presented in Figure 9.

CMIS is explicitly a causal model because the antecedents, that is, contextual triggers, “provide the underlying imperatives to seek information” (Johnson et al., 1995, p. 277). On the other hand, the connection between immediate and contextual triggers is characterised at a general level only in the textual explanation of the model. The antecedent of demographics refers to the information seeker's age, sex, ethnicity, education and socioeconomic status. As explained by Yang et al. (2023, p. 361), the component of direct experience stands for one’s personal and social history with the topic of interest. Therefore, CMIS predicts that a person is more likely to seek information on a topic that affects them or people they know than one that does not. The salience of information refers to the personal significance of health information for an individual. Both direct experience and salience address the relevance of a topic to an individual; the more important a topic is to an individual, the more likely the person is to seek information about that topic. Finally, beliefs are essentially a person’s perceptions of self-efficacy and response. If a person believes that he or she is able to engage in information seeking related to a topic, that person is more likely to seek out information related to that topic.

As to the immediate impetus for information seeking, CMIS specifies two factors constitutive of information carriers (or information sources). In considering the characteristics of information carriers, Johnson et al. (1995, pp. 281–282) referred to the user's perception of their credibility and authority, and the accuracy and comprehensibility of the information. The concept of the utility of an information carrier relates to the relevance, topicality and interest of the information and its usefulness and importance for achieving the user's goals. The above explication suggests that the characteristics of information sources and the qualities of utilities, for example, their accuracy and topicality, represent a potential which may be realised if the person decides to engage in information seeking. Therefore, a positive quality of a source per se does not necessarily trigger an information-seeking action. Similarly, the antecedents per se do not indicate how information seeking is prompted in a real-life situation. It is evident that sociodemographic characteristics such as a person's age and education level can only indirectly affect the decision to seek information. Experience about the topic of information seeking may also have an indirect effect on the decision to seek information, similar to the salience of information and an individual's self-efficacy beliefs. Taken together, CMIS does not specify which of the information carrier factors is the most powerful immediate trigger of information seeking. This can only be confirmed in empirical studies by examining whether an information carrier characteristic, for example, credibility or the perceived usefulness of a source, is most strongly correlated with information-seeking actions.

As a causal model, CMIS has similarities with the Risk Information Seeking Model (RISP) originally developed by Griffin et al. (1999). Since the late 1990s, RISP has been used in slightly modified versions. RISP identifies diverse antecedent contextual triggers related to demographic factors and relevant hazard experience, for example (Yang et al., 2014). Most importantly, however, RISP suggests that information insufficiency functions as an immediate impetus of information seeking. Information insufficiency refers to “a person's perception that he or she needs more information to deal with the risk, that is, that his or her current knowledge is less than sufficient” (Jin and Lane, 2024, p. 1589). As information insufficiency refers to the amount of information needed to reach judgmental confidence, it is closely related to the concept of information need since both are concerned with the perceived deficit between current and desired levels of information and knowledge (Choo, 2023, p. 1088).

Choo (2023) made use of the ideas of RISP by developing a model of climate change information seeking. He also tested his model using survey data gathered in Canada and the United States. Different from the original RISP model, Choo prefers the construct of knowledge need over information insufficiency, indicative of an immediate impetus for information seeking. The model is presented in Figure 10.

The diagram suggests a causal setting in which demographic characteristics, a person's experience with information seeking about climate change and his or her values dealing with it function as contextual triggers. Information seeking is prompted more directly by the cognitive precursors standing for immediate triggers. Of them, perceived knowledge need is particularly important (Choo, 2023, p. 1089). The model posits that people try to reach subjectively satisfactory levels of judgmental confidence in the information that they hold about a given risk or hazard in order to form their attitudes, beliefs and behavioural intentions. Therefore, knowledge need is a direct motivator of intention to actively seek information on climate change. Moreover, as explained by Choo (2023, p. 1090), affective precursors, that is, affective response to the risk motivates information seeking. It is assumed that the higher the level of dread, the higher the perceived risk, the more people would want to seek information about the risk. While affective response is a moderating variable in the RISP model, in the framework proposed by Choo (2023, p. 1090), affective response was approached as a direct motivator of intention to actively seek information about climate change. In addition, situational precursors, most notably social norms, may have an effect on how an individual engages in information seeking about climate change. Social norms refer to individuals’ perceptions of other people expecting them to learn about the risk, and their inclination to respond to social pressure or expectations that they should acquire sufficient information to deal with the risk. Finally, social trust can influence whether an individual starts seeking information. In general, social trust can be defined as “the willingness to rely on those who have the responsibility for making decisions and taking actions related to the management of technology, the environment, medicine, or other realms of public health and safety” (Siegrist et al., 2000, p. 354).

Drawing on the findings of his empirical study, Choo (2023, p. 1093) demonstrated that knowledge need, affective response, social norms and social trust can significantly predict intention to seek information on climate change. On the other hand, it should be noted that the above factors predicted information-seeking intention, not actual (real) activities dealing with identifying and accessing information sources. Therefore, in real-life situations, it remains open whether knowledge need, affective responses, social norms or social trust functions as the most powerful immediate impetus, or whether a combination of these factors sets information seeking in motion. As their relative significance in the initiation of information seeking can vary in empirical studies, the model in itself may not offer a final word about the most impactful immediate impetus.

The present investigation elaborated the picture of the motivators for information seeking by concentrating on the conceptualisations of the immediate and contextual impetus for information seeking. The analysis of ten pertinent models revealed six main approaches to the initial impetus and a variety of conceptualisations of the contextual triggers. The main findings are summarised in Figure 11.

Figure 11 specifies six main approaches to the immediate impetus, while the summary of the contextual triggers is structured into five thematic groups. In Figure 11, the arrows with dashed lines illustrate that contextual triggers affect the ways in which immediate impetus finally “pushes” information seeking in motion. This final effect is illustrated by the arrows with solid lines.

All in all, the findings indicate that the immediate triggers identified in the classic models in the 1980s and 1990s are continually highly relevant. In this regard, the construct of information need occupies a central role (Byström and Järvelin, 1995; Freund, 2015; Leckie et al., 1996; Wilson, 1997, 2020). The repertoire of the immediate triggers is complemented by factors such as uncertainty, stress and knowledge need. However, the approach to needs as immediate triggers proposed by Wilson (1981) and Krikelas (1983) has not gained popularity in IB research. This may be due to the fundamental ambiguity of the concept of need (Savolainen, 2017b, p. 5). The use of some triggers, such as information carrier characteristics and information insufficiency, has been limited because they are strongly dependent on the premises of a particular model, that is, CMIS and RISP. The findings also indicate that recent models, in particular, have brought to the fore the complex nature of the contextual triggers. The downside of this development is that it has become increasingly difficult to identify the specific roles of individual contextual triggers because they form a complex and dynamic interplay. Nevertheless, as Figure 11 suggests, contextual triggers arising from the requirements of (work) task performance are particularly popular in the models for IB and information seeking, independent of the nature of the immediate impetus.

The analysis of the ten models identified a few weaknesses in the conceptualisations of the triggers. This is mainly due to the general nature of certain constructs, such as information need or unspecified ways in which the diagrams are explained textually. For example, the construct of information need is sometimes left as a “black-boxed” category by assuming that it is self-explanatory (e.g. Leckie et al., 1996). This problem originates from the models’ tendency to simplify and abstract away complex behaviours. Overall, the findings demonstrate that the identification of an ultimate impetus for information seeking is difficult. This is particularly the case while looking at models in which the triggers are constituted by an interplay of diverse contextual factors (e.g. Freund, 2015; Choo, 2023). From the perspective of the complex models, the reliable identification of the temporally first “cause” is a demanding or even an impossible task. Moreover, in the case of a long-time information seeking process when an individual passively monitors his or her environment, it is difficult to trace the point when a trigger for the first time made an individual access an information source.

The novelty value of the findings can be reflected by presenting a few comparative notions. Drawing on conceptual analysis, Savolainen (2017b) examined information need as a trigger and driver of information seeking. The findings indicate that since the 1990s, the factors triggering and driving information seeking have increasingly been approached as context-sensitive phenomena. The results of the present investigation confirm this conclusion. Further support for this finding can be obtained from Agarwal’s (2022) recent investigation, where he proposed a unified model of information need and seeking behaviour. Similar to Wilson (1997), Agarwal concluded that information seeking is propelled by a multitude of factors that are contextual in nature. These factors include the person's physical environment, factors relating to the person themselves (thoughts, value system, cognition and experience), the role, task or situation, as well as the sources or system that the person has access to. As a synthesizing contribution to IB research, Agarwal (2022) unified model suggests that the classic conceptualisations of the impetus for information seeking, particularly those proposed by Wilson (1997), are continually significant.

Even though information need is an established construct to understand the nature of triggers of purposive information seeking in particular, it is evident that, as an immediate impetus, information need alone cannot explain why people initiate information seeking. Triggers of information seeking also include affective factors such as emotions (Savolainen, 2014, p. 63). Both negative emotions (e.g. anxiety, confusion and doubt) and positively coloured emotions such as joy and thrill can lead people to identify and access information sources in work-related and leisure contexts. Looking at the ten models examined in the present study, only Choo’s (2023) framework devotes explicit attention to the significance of affective responses. Moreover, as noted above, there are modes of IB whose initiation is difficult to explain merely by referring to the propelling effect of information need. Information can be obtained passively or encountered in an unplanned way (Erdelez, 1999; Wilson, 2020, p. 41). Therefore, the question about the identification of the impetus for information seeking of this kind becomes more demanding. For example, Harati and Isfandyari-Moghaddam (2023) have critiqued the ways in which many models simply approach information-seeking behaviour as a step-by-step linear process, starting from the need for information, progressing to the decision-making process for seeking information and ending with the ultimate selection of information resources. However, unplanned behaviour is often an integral part of information seeking. This means that users do not always decide to use resources or services in advance; rather, they tend to use them due to possible stimuli in the environment and often without the help of intermediaries such as librarians. Information seeking of this kind is shaped by a “hidden or uncertain information need”, as Harati and Isfandyari-Moghaddam (2023) put it. As “hidden information need” is related to the “visceral need” identified by Taylor (1968), this suggests that classic conceptualisations of the triggers of information seeking are continually relevant in today's information environments.

The present investigation focused on models suggesting causal connections between contextual triggers, the immediate impetus and information seeking. However, the information-seeking process may not necessarily be initiated by a set of “pushing” causal factors. There are models proposing that diverse “pulling” factors can also explain why people engage in information seeking. For example, the expectancy-value model developed by Eccles and Wigfield (2002) suggests that people's choices and performance can be explained by referring to their beliefs and expectations about how well they will do on the activity in the future (Savolainen, 2012). Such future-oriented beliefs also explain why people are motivated to engage in information seeking and continue it until the outcome expectations are sufficiently met. Similarly, the model of everyday information practices proposed by (Savolainen, 2008, pp. 64–66) suggests that many of the motivating factors of information seeking can be explained by referring to people's future goals dealing with their everyday projects. This is because an individual's goals, expectations and interests may rather “pull from the future” than causally “push from the past” to seek information in order to further leisure-related projects, for example. In the ten models examined in the present study, components indicative of people's future expectations were largely omitted. Rare examples include the components of “belief” (Johnson et al., 1995), “risk/reward” (Wilson, 1997), “information task: goal” (Freund, 2015) and “social norms” (Choo, 2023). This suggests that there is a need to enhance the modelling of the triggers by including factors indicative of people's beliefs and expectations about why it is important to seek information.

Although the present investigation concentrates on the conceptual issues of IB research, the findings also have practical implications. The findings suggest that information needs experienced by the users of information systems and services are particularly important immediate triggers of information seeking. This suggests that in the development of information services, particular attention should be directed to the analysis of information needs of potential users. As tasks at hand occupy a central role among contextual triggers, the designers of information systems and services should examine the typical requirements of tasks performed by potential users in work-related contexts in particular.

Based on the conceptual analysis of ten pertinent models, the present study elaborated the picture of factors that make people start seeking information. The unique contribution of the investigation is the distinction between immediate and contextual impetus for information seeking. The findings highlight that the classic construct of information need continually occupies a central role as an immediate impetus for information seeking. As to the contextual triggers, (work) task-related factors are most prevalent in the models for IB and information seeking. As the findings are based on the analysis of a limited number of models, there is a need to examine a broader set of frameworks in order to complement the picture of immediate and contextual triggers. The findings of conceptual investigations can be substantiated by analysing how the triggering factors are approached in empirical studies on information seeking. More specifically, to this end, future investigations may compare the similarities and differences in diverse empirical approaches to immediate and contextual triggers of information seeking. It is particularly important to examine the extent to which the findings of empirical investigations support the assumptions presented in the models for IB and information seeking. As such models necessarily offer a simplified picture of the impetus for information seeking, empirical findings can offer insights about how the modelling of the triggers of information seeking could be elaborated in the future.