To assess conversation partner question load and responsiveness, measure Dev's partner's questions, prompts, comments, acknowledgments, expansions, waits, interruptions, topic follows, repair support, device access, and shared contributions. Link each partner action to the next interaction event. Identify interrogation patterns and missed messages before adding a client target. A partner can increase access by waiting and following Dev's topic, rather than eliciting more speech.

Classify partner contributions

Record open and closed questions, stacked questions, commands, response prompts, comments, acknowledgments, expansions, reflections, waits, topic offers, repairs, interruptions, and closures.

Link partner and client events

Join each partner action to Dev's next response, initiation, repair, pass, pause, or no-opportunity state. Preserve timing and competing speakers.

Measure responsiveness

After Dev communicates, record whether the partner acknowledges content, follows the topic, expands accurately, acts on the referent, asks for repair, redirects, corrects, or misses the message.

Protect opportunity denominators

Count openings where Dev could initiate or continue, rather than treating every minute as an opportunity. Partner speech and task demands can remove openings.

Use this sequence for the audit

Define the interaction goal, classify partner actions, link sequences, quantify question and uptake patterns, ask Dev about fit, change the earliest partner barrier, and reassess cautiously.

Build Dev's conversation-partner behavior audit

Create one versioned conversation-partner behavior audit for the library research tutoring. Include Dev's priority, partners, activity, modes, person-authored turns, topic and timing rules, latency, overlap, initiation, response, repair, script and prompt status, partner behavior, context, health, privacy, client experience, passes, invalidity, missingness, withdrawal, evidence limits, qualified owner, correction, and reassessment trigger. Store only decision-relevant information with role-limited access.

Validate Dev's evidence

Reproduce 40 contributions with 24 questions, eight prompts, five acknowledgments, three comments; Dev responses after 20/24 questions and two initiations; then 12 turns with six questions, three acknowledgments, three comments and four initiations. Verify topic and task changes, timing, device access, repairs, and uptake.

Connect Dev's evidence to a decision

The audit can support fewer stacked questions, longer waits, related comments, explicit topic handoff, AAC access, or better repair support. Partner change is often the first actionable control.

Work through Dev's example

Across 40 partner contributions, the tutor asks 24 questions, gives eight prompts, makes five acknowledgments, and offers three related comments. Dev contributes after 20/24 questions but initiates only twice. In a 12-turn follow-up with six questions, three acknowledgments, and three comments, Dev initiates four times. The sequence is descriptive because topic and task also change. Preserve every planned and valid exchange, partner and setting version, communication mode, authored contribution, topic, timing, prompt, script, repair, partner action, access and health state, pass, withdrawal, invalid event, correction, and unresolved item. This fictional example demonstrates one assessment control. It supplies no diagnosis, universal conversation hierarchy, treatment effect, legal conclusion, coverage decision, payment promise, or outcome guarantee for Dev.

Address Dev's main interpretation risk

A client response rate of 20/24 could look strong while hiding a narrow interview role. Comparing two phases without noting changed task and topic cannot isolate question load. More questions may increase answers while reducing Dev's opportunities to initiate or repair. Review communication access, message authorship, turn rules, topics, timing, partner questions and waits, prompts, scripts, task cues, partner familiarity, context, health, fatigue, privacy, preference, direct experience, and design strength separately. A convincing sequence cannot make an unsafe, inaccessible, unwanted, unauthorized, or irrelevant conversation goal clinically appropriate.

Keep scope and authority clear for Dev

For Dev's conversation-partner behavior audit, the CASP public summary supplies high-level ABA behavioral-health-treatment scope for autistic people, with licensed detail outside this page. The current BACB Ethics Code addresses competence, client involvement, consent and assent when applicable, assessment, medical needs, risk, data, documentation, and evaluation for covered people. The BACB outline is examination content. These sources supply no individualized conversation norm, language diagnosis, legal authority, or requirement that Dev speak, answer, or converse in a typical style.

Treat timing as dyadic and contextual for Dev

The automated latency study analyzed recordings from 210 child-adult pairs and derived non-speech intervals between turns. It also measured adult-partner latency and linked change to a particular intervention dataset. A later turn-taking context study found that latency and overlap varied with activity and partner familiarity in its child samples. These studies support measuring both partners and context for Dev. They supply no universal optimal speed, diagnostic cutoff, or reason to shorten AAC composition time.

Use topic categories descriptively for Dev

The recent topic-shift study distinguished contingent responses, associated and non-associated shifts, reintroductions, explicit and implicit markers, and immediate partner prompts. Its coding system illustrates how one off-topic category can erase meaningful distinctions. It does not establish that topic continuation is better than change for Dev or that a laboratory exchange captures everyday conversational fit.

Keep AAC, repair, and partner findings bounded for Dev

An AAC intervention study involved four minimally speaking autistic children ages six to nine and targeted turn requests, question answers, and comments with peers. A longitudinal repair study compared several neurodevelopmental groups and defined repair as responding to a clarification request. These studies support explicit message, partner, repair, and context records. Their populations and procedures do not create one conversation curriculum, mastery rule, or expected outcome for Dev.

Preserve assent and communication access for Dev

Breaux and Smith propose individualized assent and withdrawal practices while describing an evolving evidence base. Their paper is practice guidance rather than a separate BACB mandate. ASHA's AAC portal says AAC users should always have access to their tools or devices. Give Dev an accessible way to contribute, compose, pass, ask, repair, pause, end, or withdraw. Do not require speech, speed, eye contact, topic conformity, or social endurance as the price of participation.

Choose Dev's next bounded action

Dev asks the tutor to use one question at a time and acknowledge his source before redirecting. The tutor adds a visible topic handoff during open discussion. Record the qualified owner, source evidence, effective date, current partner, activity, mode, and context version, ordinary supports, communication and health state, implementation check, accessible explanation, disagreement route, and reassessment trigger. Preserve the earlier record when partners, vocabulary, devices, modes, topics, settings, outcomes, health, or priorities change. A repaired arrangement creates a dated phase rather than an error in Dev's prior communication.

Close Dev's assessment

Review the conversation-partner behavior audit with Dev, the qualified behavior analyst, relevant partners, access owners, and specialists named in the manifest. Confirm that person-authored turns, partner turns, timing, topic relation, repair, scripts, prompts, question load, context, and client experience remain separate; every denominator is reproducible; AAC, privacy, and basic access remain protected; urgent needs received action; and conclusions stay bounded to sampled exchanges. Keep this page draft and noindex until every required review is complete.

Related resources

Sources