> For the complete documentation index, see [llms.txt](https://help.dscout.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.dscout.com/study-basics/question-types/open-ended-questions/talk-out-loud-questions.md). # Talk out loud questions
Study typeAvailability
Usability teststrue
Intercept studiestrue
AI moderated studiestrue
Talk out loud questions ask participants to respond to a prompt in their own words—speaking aloud—so you capture the reasoning and reactions behind their answers, not just the answer itself. How the question is moderated depends on your study type: * In **usability tests** and **intercept studies**, participants think aloud while their screen and session are recorded, producing a video response and a transcript. * In **AI moderated studies**, participants have a spoken exchange with the AI moderator, which can ask dynamic follow-ups to probe deeper.
### Talk out loud question settings The settings available on a Talk out loud question depend on your study type. Select your study type below.
{% tabs %} {% tab title="Usability tests" %} In a usability test, Talk out loud is an open-ended question that prompts the participant to respond aloud while their screen and session are recorded. You'll configure the following: | Setting | Description | | --------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Prompt** | The question or instructions the participant responds to. | | **Privacy blur** (optional) | Blurs the participant's screen to protect their personally identifiable information (PII). Audio is still recorded. For more information, see [Privacy blur](/study-basics/testing-prototypes/privacy-blur.md). | | **Stimuli** (optional) | Attach images or media for the participant to react to as they respond. For more information, see [Question stimuli](/study-basics/question-types/question-settings-and-organization/question-stimuli.md). | | **Skip logic** (optional) | Route participants to a later point in your study based on this question. For more information, see [Skip logic](/study-basics/design-a-study/skip-logic-in-usability-tests-media-surveys-and-intercept-studies.md). | | {% endtab %} | | {% tab title="Intercept studies" %} In an intercept study, Talk out loud is an open-ended question that prompts the participant to respond aloud while their session is recorded. You'll configure the following: | Setting | Description | | --------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Prompt** | The question or instructions the participant responds to. | | **Privacy blur** (optional) | Blurs the participant's screen to protect their personally identifiable information (PII). Audio is still recorded. For more information, see [Privacy blur](/study-basics/testing-prototypes/privacy-blur.md). | | **Stimuli** (optional) | Attach images or media for the participant to react to as they respond. For more information, see [Question stimuli](/study-basics/question-types/question-settings-and-organization/question-stimuli.md). | | **Skip logic** (optional) | Route participants to a later point in your study based on this question. For more information, see [Skip logic](/study-basics/design-a-study/skip-logic-in-usability-tests-media-surveys-and-intercept-studies.md). | | {% endtab %} | | {% tab title="AI moderated studies" %} In an AI moderated study, Talk out loud is a spoken exchange between the participant and the AI moderator, which can probe with dynamic follow-ups. You'll configure the following: | Setting | Description | | --------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Prompt** | The question the AI moderator asks the participant. | | **Read verbatim** | When on, the AI moderator asks your prompt exactly as written. When off, it may rephrase the prompt naturally while staying on topic. | | **Follow-up questions** | How many follow-ups the AI moderator asks in response to the participant's answer: **Only if response is short**, **1 follow-up**, or **2–3 follow-ups**. | | **Follow-up guidance** (optional) | Guide the kind of follow-ups the moderator asks. For example, to focus on a competitor the participant mentions or ask the reasoning behind a choice. | | **Visual intelligence** | Captures expressions in participant responses. For more information, see [Visual intelligence](/ai-moderated-studies/analyze-an-ai-moderated-study/visual-intelligence.md). | | **Stimuli** (optional) | Attach images or media for the participant to react to as they respond. For more information, see [Question stimuli](/study-basics/question-types/question-settings-and-organization/question-stimuli.md). | {% hint style="info" %} **Note:** AI moderated studies support up to 30 Talk out loud and Task questions combined. Multiple choice and Scale questions don't count toward this limit. For more information, see [Question types in AI moderated studies](/ai-moderated-studies/what-are-ai-moderated-studies/question-types-in-ai-moderated-studies.md). {% endhint %} {% endtab %} {% endtabs %} ### Add a Talk out loud question to a study {% stepper %} {% step %} Navigate to the study tab of your builder—the **Usability test**, **Intercept**, or **AI moderator** tab, depending on your study type. {% endstep %} {% step %} Click **+ Add** in the center of the builder or in the left sidebar. {% endstep %} {% step %} Select **Talk out loud**. (In usability tests and intercept studies, you'll find it under the **Usability** category.)
{% endstep %} {% step %} Enter your **Prompt** and configure the available settings. For details, see [Talk out loud question settings](#talk-out-loud-question-settings) above. {% endstep %} {% step %} Click **Done**. {% endstep %} {% endstepper %} ### Analyze a Talk out loud question Every Talk out loud response is captured in the participant's session recording and transcript, so you can watch or read exactly what they said. * In **usability tests** and **intercept studies**, responses are transcribed verbatim and available across all analysis views. * In **AI moderated studies**, responses are transcribed alongside the session video. Dscout AI also generates summaries for each Talk out loud question and surfaces emerging themes across participants. For more information, see [View AI moderated study responses](/ai-moderated-studies/analyze-an-ai-moderated-study/view-ai-moderated-study-responses.md). ### Frequently asked questions
How should I write a Talk out loud prompt? Ask one open question at a time, and write it the way you'd say it out loud. In AI moderated studies, **Review your study** flags Talk out loud prompts that are likely to get weaker responses: * **Not genuinely open:** A single word or a yes/no answer fully answers the question. * **Awkward read aloud:** The moderator reads your prompt word for word, so embedded dates and times, list-like structure, or a length no one would say in one turn don't work well when spoken. * **Compound or double-barreled:** The prompt stacks two questions into one, so you can't tell which one the participant is answering. Ask the second as its own question. The same principles apply to Talk out loud questions in usability tests and intercept studies. For more information, see [Review your study](/dscout-ai/built-in-ai-features/review-your-study.md).
Do participants see my prompt as text in an AI moderated study? Not as a block of written text. Participants hear the AI moderator speak your prompt, and captions of the spoken words appear on screen as the moderator says them. If you turn off **Read verbatim**, the captions show the moderator's rephrased wording. Because prompts are heard rather than read, keep them short enough to follow by ear. If participants need to look at something, such as a concept description or a design, attach it to the question as a stimulus instead. For more information, see [Question stimuli](/study-basics/question-types/question-settings-and-organization/question-stimuli.md).
Can I save a Talk out loud question to my question bank? Yes. In the builder, open the question's **More options** menu and select **Save question**. When you add a saved Talk out loud question to a usability test or intercept study, it stays a Talk out loud question. If you add it to a study type that doesn't support Talk out loud, it's added as a typed response question instead. For more information, see [Question banks](/study-basics/question-types/question-settings-and-organization/question-banks.md).
--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.dscout.com/study-basics/question-types/open-ended-questions/talk-out-loud-questions.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.