Just-in-Time Prompt¶
Contextual prompt — instantiates Cognitive Load Reduction
Surfaces the exact piece of guidance at the moment and place of need, so the user neither recalls it nor breaks flow to look it up.
A Just-in-Time Prompt delivers one small, fixed piece of guidance at the instant a task step needs it and in the place the user is already looking — a tooltip, a contextual hint, a spoken cue, an inline note that appears only when its step comes up. Its defining property is that it is triggered and in-flow: the prompt is pushed by the user reaching a particular point, not pulled by the user going to find it, and it arrives without making the user leave what they are doing to consult a manual, switch to another tool, or hold the fact in memory until it is relevant. It carries no computation and offers no depth to expand — it is a static cue whose entire value is arriving at the right second in the right spot.
Example¶
A field technician is servicing a rooftop HVAC unit she has not touched before. The full service manual is a 200-page PDF; opening it means climbing down to the van, losing her place, and finding the one relevant torque spec among hundreds — a context switch that costs minutes and breaks her concentration each time.
Instead she works from a handheld app that tracks which step she is on. As she reaches the compressor-bolt step, the screen surfaces exactly one line — "torque to 24 N·m, star pattern" — with nothing else competing for the screen. She never recalled the figure, never left the task to search for it, and never saw the 199 pages that don't apply to this step. When she advances, that cue disappears and the next one takes its place. The guidance met her at the bolt, at the moment the bolt was in her hand; the manual's knowledge was delivered without the manual's interruption.
How it works¶
- Detect the moment. A trigger — task position, sensor state, screen focus, or an explicit "next" — determines when the cue fires; timing is the whole game.
- Surface one thing, in place. The prompt shows the single relevant fact where the user is already attending, so no tool-switch or page-hunt is needed.
- Offload the recall. Because the fact appears exactly when needed, the user never has to carry it forward in working memory.
- Vanish on advance. The cue clears when its step passes, keeping only the currently-relevant guidance present so nothing accumulates on screen.
Tuning parameters¶
- Trigger precision — how tightly the cue is bound to the exact moment of need. Loose triggers fire too early (ignored) or too late (useless); tight triggers require reliable context detection.
- Interruptiveness — a quiet inline hint vs. a modal that demands acknowledgment. More salience guarantees the cue is seen but risks becoming the very interruption it was meant to prevent.
- Content length — how much the prompt says. A single fact preserves flow; a paragraph turns a cue into a detour.
- Dismissal and recurrence — whether a prompt can be silenced and whether it returns next time, trading against nagging an expert who already knows.
When it helps, and when it misleads¶
A just-in-time prompt is strongest when the knowledge is thin but positionally critical — a spec, a warning, a next action that is trivial to state but expensive to look up or easy to forget at the one moment it matters. By meeting the user in place, it removes both the recall burden and the context-switch cost that make an otherwise-available manual practically useless mid-task — the animating idea behind performance-support systems, which put help at the moment of need rather than in a separate document.[1]
Its failure modes are interruption and dependency. Prompts that fire too often, too loudly, or off-target become noise; users learn to swat them away and miss the one that mattered — the boy-who-cried-wolf failure of over-prompting. And permanent prompting can prevent skill from forming, keeping a user forever reliant on being told the next step. The guarding discipline is ruthless relevance and restraint: fire only on a precise trigger, say only the one needed thing, and let experts turn cues down so the channel stays trusted rather than tuned out.
How it implements the components¶
context_switch_boundary— its core act: delivering guidance inside the user's current flow so they never break away to a manual or another tool to get it.sequencing_rule— the cue is bound to when its step occurs, releasing the right fact at the right moment rather than all at once.external_memory_support— by appearing exactly when needed, it stands in for a fact the user would otherwise have to hold in working memory.
It does not implement progressive_detail_boundary — a JIT prompt pushes one fixed cue at the moment of need and offers no depth to expand on demand; user-pulled drill-down is Progressive Disclosure for Load Reduction. Nor does it implement comprehension_or_performance_check — it surfaces a static fact and computes nothing; validating a recommendation is Decision Support Tool.
Related¶
- Instantiates: Cognitive Load Reduction — the Just-in-Time Prompt supplies the archetype's in-flow, moment-of-need delivery machinery.
- Sibling mechanisms: Chunked Instructions · Template · Worked Example · Visual Aid · Progressive Disclosure for Load Reduction · Simplified Interface · Decision Support Tool · Checklist
Editorial Notes¶
Form Classification¶
Form family: Interface, Display & Cue
Rationale: Just-in-Time Prompt operates as a user-facing prompt, display, template, or perceptual cue that shapes attention and action at the point of use because it surfaces the exact piece of guidance at the moment and place of need, so the user neither recalls it nor breaks flow to look it up
Independent corroboration: The frozen evidence defines Just-in-Time Prompt as 'Surfaces the exact piece of guidance at the moment and place of need, so the user neither recalls it nor breaks flow to look it up', so its operative form is Interface, Display & Cue.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Human-Computer Interaction
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Universal
Rationale: HCI developed contextual prompts embedded in the user's workflow and triggered by interface state.
Related originating lineages:
- Cognitive Science — External-memory and cognitive-load research materially shaped prompts that reduce recall demands.
- Education & Pedagogy — Instructional scaffolding supplied point-of-need guidance and fading support.
- Psychology — Cognitive-load research materially shaped the externalization of memory at the moment of action.
Review resolution: Both independent reviews place the primary lineage in human_computer_interaction. The queued differences (alternate_origin_disagreement, domain_reach_disagreement) concern secondary metadata rather than primary provenance. The final retains education_pedagogy, psychology, cognitive_science only where a reviewer supplied a formative-lineage rationale; downstream application by itself is not treated as origin. origin_mode=cross_disciplinary_synthesis records the relationship among origin traditions, while domain_reach=universal records application breadth separately. encyclopedia_synthesis=false reflects whether either reviewer identified a corpus-specific synthesis, and confidence=high preserves the more cautious evidence assessment.
Review outcome: Reconciled after independent review; high confidence.
References¶
[1] Electronic Performance Support Systems (EPSS), a concept associated with Gloria Gery's 1991 work, embed help, guidance, and reference directly into the flow of a task so that support arrives at the moment of need rather than requiring the worker to stop and consult separate training or documentation. The just-in-time prompt is the atomic unit of that idea. withdrawn registry ↩