Information Need¶

Represent the situated gap between what an actor currently knows and what the actor must know to interpret, decide, or act, allowing the gap to evolve from unarticulated unease into a formal question and system-facing query.

Version: v2 · 2026-09-06 · History
Domain-specific #: 2067
Origin domain: information science
Subdomain: information behavior
Aliases: Information needs, Need for information, User information need

Core Idea¶

An information need is a situated gap between an actor's current knowledge and the knowledge required to interpret a situation, make a decision, perform a task, or resolve a concern. The need is not identical to the words eventually typed into a system. Taylor distinguished four levels: a visceral but unexpressed need, a conscious description, a formalized question, and the compromised question actually presented to an information system.^[1]

Because articulation changes the actor's understanding, an information need is dynamic. Searching, encountering evidence, and receiving partial answers can narrow, redirect, or reveal the gap rather than simply fill a fixed slot.

Structural Signature¶

An actor situated in a task, role, problem, or lived context.
A current knowledge state, whether fully conscious or not.
A demanded knowledge state sufficient for some purpose.
A consequential gap between the two states.
A trigger such as uncertainty, anomaly, decision, breakdown, or curiosity.
A degree of recognition by the actor.
Progressive articulation into descriptions, questions, or queries.
Constraints imposed by language, channel, system, time, cost, and social setting.
A relevance relation between candidate information and the situated gap.
Information-seeking, consultation, observation, or deliberate nonseeking as possible responses.
Feedback by which encountered information changes the need itself.
A stopping judgment based on sufficiency for the purpose, not exhaustive knowledge.
Possible mismatch between expressed request and underlying need.

What It Is Not¶

It is not the query, document request, search terms, or answer. It is not an information requirement imposed by a specification unless that requirement corresponds to the actor's situated knowledge gap. It is not information seeking, which is a possible downstream behavior, and it is not generic ignorance detached from a purpose or context.

Scope of Application¶

Information needs are studied in library reference work, search and recommendation, professional information behavior, health communication, intelligence analysis, community information services, product research, and human–computer interaction. Wilson argued that “information need” should be understood through the more fundamental human needs and contexts that give information value, rather than treated as an autonomous mental object.^[2]

Belkin, Oddy, and Brooks modeled retrieval as engagement with an anomalous state of knowledge: users often cannot specify precisely what would resolve the anomaly because the missing knowledge is what would enable the specification.^[3]

Clarity¶

Name the actor, task or situation, decision horizon, present state, required state, articulation level, constraints, and criterion for sufficient resolution. Distinguish observed behavior from inferred need. A search log reveals a system-facing query, not the full need, and nonuse does not prove absence of need.

Manages Complexity¶

The abstraction separates four things systems often collapse: the underlying gap, the actor's conscious account, the negotiated or formal question, and the machine query. That separation explains why syntactically perfect query matching can return irrelevant results and why interviews, clarification, reformulation, and exploratory interfaces can outperform one-shot retrieval.

Abstract Reasoning¶

Identify the actor and situated purpose.
Reconstruct the current knowledge state without assuming the query is complete.
Specify what action or interpretation is blocked.
Infer the knowledge that would remove or reduce the blockage.
Determine the need's articulation level.
Translate it into a formal question while preserving context.
Select sources and channels appropriate to the evidence required.
Evaluate candidate information by situational relevance and credibility.
Update the need as learning changes the problem representation.
Stop when the actor's purpose is sufficiently supported, recording unresolved residuals.

Dervin and Nilan's user-centered turn emphasized situation, sense-making, and use rather than evaluating users solely against system-defined requests.^[4]

Knowledge Transfer¶

The portable pattern is treat a request as a lossy expression of a deeper situated gap; diagnose the gap, preserve the actor's purpose through translation, and expect the target to move as understanding improves. The proposed immediate parent is Uncertainty.

Examples¶

A clinician asks for one drug interaction but actually needs a safe treatment choice given renal function, other medicines, and urgency. A student types a broad topic because the assignment's comparison criterion is not yet understood. A maintenance technician reports “the manual” but needs one tolerance value tied to a particular machine revision.

In each case, returning documents that match the literal words can fail while a clarification process succeeds.

Structural Tensions¶

Felt need versus expressible question.
User language versus system vocabulary.
Stable task versus evolving understanding.
Recall breadth versus relevance precision.
Immediate answer versus deeper problem reframing.
Observed query versus inferred underlying purpose.
Information abundance versus usable sufficiency.

Structural–Framed Character¶

Gap-relative inquiry is structural. User state, situated task, question articulation, relevance, retrieval interfaces, and information use are constitutive. The abstraction is domain-specific.

Structural Core vs. Domain Accent¶

The structural core is current knowledge < purpose-required knowledge -> articulated request -> evidence-mediated update. The domain accent is information behavior and retrieval-system interaction.

Uncertainty is the proposed immediate parent. Curiosity, Relevance, Search, Sensemaking, Feedback, and Need–Solution Alignment are related primes. Information Seeking is the closest domain-specific downstream process.

The prospective queue contains one strict edge to prime:uncertainty. No live DAG mutation is authorized.

Relationships to Other Abstractions¶

Current abstraction Information Need Domain-specific

Parents (1) — more general patterns this builds on

Information Need is a kind of Uncertainty Prime

Uncertainty is the proposed immediate parent.

Hierarchy path (1) — routes to 1 parentless root

Information Need → Uncertainty

Neighborhood in Abstraction Space¶

Information Need sits in a sparse region of the domain-specific corpus (98^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Peer Learning — 0.75
Hypocrisy — 0.75
Information Seeking — 0.75
Use Case — 0.74
Semantic gap — 0.74

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

A typed search query.
An information requirement in a system specification.
Information seeking behavior.
Relevance as a property of a candidate result relative to the need.
General ignorance with no situated consequence.
Demand or recorded use treated as a complete proxy for need.

References¶

[1] Robert S. Taylor, “The Process of Asking Questions,” American Documentation 13, no. 4 (1962): 391–396, doi:10.1002/asi.5090130405. registry ↩

[2] T. D. Wilson, “On User Studies and Information Needs,” Journal of Documentation 37, no. 1 (1981): 3–15, doi:10.1108/eb026702. registry ↩

[3] Nicholas J. Belkin, Robert N. Oddy, and Helen M. Brooks, “ASK for Information Retrieval: Part I. Background and Theory,” Journal of Documentation 38, no. 2 (1982): 61–71, doi:10.1108/eb026722. registry ↩

[4] Brenda Dervin and Michael Nilan, “Information Needs and Uses,” Annual Review of Information Science and Technology 21 (1986): 3–33. registry ↩