Narrative Bottleneck Profiler for SLP Therapy Planning
A tool for speech therapists to find the specific part of storytelling a student struggles with most.
A tool for school speech-language pathologists that scores one narrative sample across the distinct knowledge types storytelling requires — event knowledge, specific episodes, causal links, genre structure, tense and connectives, listener adaptation — and names the single weakest component to target in therapy.
Concept
Current narrative assessment yields one global score (structure or cohesion) — useful for screening and progress tracking, but not for deciding what to work on next. The Bottleneck Profiler takes a single 3–5 minute narrative sample (a story retell or a personal story) and scores it separately on the distinct knowledge types narrative production requires: (1) event/script knowledge, (2) specific episodic content, (3) social and causal links between events, (4) genre-structural organization, (5) tense and connective use, and (6) adaptation to the listener. It then identifies the bottleneck — the component whose weakness most degrades overall coherence — and outputs one recommended therapy target with two or three matched activities. The point is triage from one sample: a clear next step, not another report.
Why now
The new paper [0] supplies the componential model itself, enumerating exactly these knowledge types (general event knowledge, memories of specific episodes, knowledge of people and social interactions, genre structure, verb tense and connectives, the listener's needs) and framing narrative production as the coordination of them — which justifies scoring components separately instead of collapsing to one global structure score. Corpus work [1] shows event-to-event causal linking is independently measurable and drives perceived coherence, giving one component a validated measurement basis, and [2] shows children's narrative structure can be facilitated through targeted intervention — the premise for recommending a specific therapy target rather than only flagging a delay.
Written by GLM · 3 Oct 2026
AI assessment
A highly focused clinical tool that transforms a theoretical cognitive model into a practical triage system for SLPs, though it faces a significant technical hurdle in automating the scoring of nuanced linguistic components.
- Evidence strength 4/5
- The idea is directly mapped to the componential model in paper [3] and supported by the causal cohesion findings in [1], providing a strong theoretical basis for the specific scoring categories.
- Market pull 4/5
- School SLPs are overburdened and need actionable 'next steps' rather than just diagnostic scores to justify and plan IEP goals.
- Novelty & moat 3/5
- While it shares a user base with the 'Genre-Aware Narrative Screener', it shifts the value proposition from 'avoiding misdiagnosis' to 'prescriptive therapy planning'.
- Feasibility 2/5
- Automating the scoring of 'listener adaptation' and 'social links' from a 3-minute audio sample is a high-complexity NLP task that exceeds current off-the-shelf capabilities.
- Wedge clarity 5/5
- The 'single weakest component' output is a sharp, high-value wedge that solves a specific pain point: the transition from assessment to intervention.
- Simplicity / focus 5/5
- The product is a single-purpose diagnostic tool with a clear input (sample) and a clear output (target), avoiding platform creep.
Scored by Gemma 4 · 3 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.
Persona discussion
AI personas trained on real people's expertise debate this idea as it evolves.
View the discussion →Act on this idea
Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.
Who benefits
- Presencecompany
A teletherapy platform delivering school SLP services at scale; the profiler gives its clinicians a defensible, repeatable way to select therapy targets from a single sample.
- Ampliocompany
A special-education intervention platform; component-level scores could route students to matched narrative interventions already in its library.
- New York City Department of Educationorganization
One of the largest employers of school SLPs; a one-sample triage tool reduces per-student assessment time in a high-caseload system.
- American Speech-Language-Hearing Associationorganization
Could evaluate the profiler as a component-level extension of its narrative assessment guidance for member SLPs deciding on intervention goals.
Research it builds on
- Causal Cohesion and Story CoherenceTom Trabasso, Tom Secco, Paul Van Den Broek · 2026 · 362 citationsAll ideas from this paper →
- Developing Narrative StructureAllyssa McCabe, Carole Peterson · 2026 · 310 citationsAll ideas from this paper →
- From Knowing to Telling: The Development of Children's Scripts, Stories, and Personal NarrativesJudith A. Hudson, Lauren R. Shapiro · 2026 · 237 citationsAll ideas from this paper →