Response to stimulus
A methodology built around a single visual stimulus – an image shown to participants before any questioning begins. Rather than asking people what they think about a concept in the abstract, it shows them something concrete and captures their instinctive response to it. Use it to understand how a communication, scenario, or idea is actually received – the immediate impression, the felt meaning, the interpretation – before rational processing takes over.
The qualitative theory
Section titled “The qualitative theory”Image-based stimulus testing is grounded in a simple but important insight: showing people something produces a different kind of response than telling them about it. Abstract descriptions invite considered, intellectual reactions. A concrete image triggers immediate, affective ones – impressions formed before the participant has had time to construct a considered opinion.
This matters because the impressions that form in the first few seconds of exposure to a communication, concept, or creative idea are often the ones that drive behaviour. If a piece of communication creates an unintended first impression – one the participant then rationalises into something more charitable – standard questioning will capture the rationalisation, not the impression. This methodology is designed to capture the impression.
The instruction to respond without describing the image is deliberate. Describing an image is an analytical act – it puts the participant in the role of reporter rather than reactor. By asking instead “what’s the first thing that comes to mind?”, the methodology invites an associative, affective response that reveals what the stimulus means at an intuitive level, not what it depicts at a literal one.
The laddering structure then works outward from that first response. It follows a sequence from free association through emotional response, to interpreted meaning, to behavioural intent. This arc – what it evokes, how it feels, what it communicates, what it makes me do – is a deliberate map of the cognitive and behavioural journey a real person takes when encountering a stimulus in context. The result is a layered account of reception: how the stimulus lands, what it means, and what it prompts.
Configured settings
Section titled “Configured settings”This section documents every setting in the methodology so you can understand the intent behind each choice.
Research framing
Section titled “Research framing”| Field | Value |
|---|---|
| Context & Background | Participants are shown an image as a visual stimulus (e.g. scenario, communication statement, concept/idea) to capture their immediate impressions, feelings, interpretations and intentions. |
| Research Objectives | Understand how people perceive and interpret the stimulus, including meaning, emotional response, clarity and overall impression, to inform messaging or strategy. |
| AI Moderator Tone | Friendly and casual |
The casual tone is essential here. Participants need to feel they can say whatever comes to mind – including reactions that feel vague, unexpected, or hard to articulate. A formal tone signals that careful, considered answers are expected, which is precisely what this methodology is designed to avoid.
Conversation introduction
Section titled “Conversation introduction”Hi, I’m
{{Companion Name}}, your survey companion. I’m going to show you an image and I’d like you to just share your instinctive thoughts, feelings and impressions. There are no right or wrong answers, of course, it’s just how you see it.Please look at this image. Without describing it, what’s the first thing that comes to mind?
The introduction does two things: it establishes the instinctive, no-wrong-answer register, and it delivers the critical instruction – “without describing it”. This constraint is the pivot on which the whole methodology turns. Participants who describe the image are reporting; participants who respond to it are reacting. The question is designed to get reactions.
“It’s just how you see it” is also deliberate – it personalises the response and signals that individual perception is exactly what’s being sought.
Laddering questions
Section titled “Laddering questions”Five questions at maximum, with Randomise on and Paraphrase on:
| # | Question | Purpose |
|---|---|---|
| 1 | What else comes to mind? | Extends the initial free association – surfaces more of the associative landscape |
| 2 | How do you feel about what you see here? | Shifts from cognitive to affective – invites explicit emotional response |
| 3 | What do you understand this to be saying, doing or communicating? | Explores interpreted meaning – what the participant thinks the stimulus intends to communicate |
| 4 | If you encountered this in real life, what might you do? | Surfaces behavioural intent – the downstream consequence of the impression and interpretation |
| 5 | What makes you say that? | A reflective probe that asks for the reasoning or feeling behind the previous response |
Randomise is on so the sequence varies across participants – this reduces order effects and gives a more representative view of initial associations across the sample. It also means not every participant reaches question 4 or 5 in the same state, which adds naturalistic variation to the behavioural intent data.
Paraphrase is on – unlike the Keyhole technique, these questions don’t depend on a specific sequential arc, so natural variation in phrasing is acceptable and keeps the conversation from feeling scripted.
Default probe
Section titled “Default probe”Tell me more about
{{topic}}? What makes you say that?
Paraphrase on. The two-line probe follows the same pattern as the Top-of-mind methodology – the first line opens up the topic, the second pushes toward the reasoning or feeling behind it. Together they draw out both the content of the response and its affective basis.
Default prompt
Section titled “Default prompt”Thinking now about
{{topic}}, tell me about that?
Paraphrase on. Kept deliberately open – the prompt introduces a topic without signalling what kind of response is expected.
Probe & prompt limit
Section titled “Probe & prompt limit”9 – calibrated to the number of active topics without making the conversation feel exhaustive. Adjust upward if you add more topics with probing or prompting enabled.
Conversation ending
Section titled “Conversation ending”- Ask a final question before wrapping up: On
The final question is generated dynamically by the AI Moderator based on what was discussed, which suits this methodology’s open-ended, exploratory nature. Because responses to stimulus are highly variable – different participants notice different things – a fixed closing question would often feel disconnected from what was actually said.
Adapting this methodology
Section titled “Adapting this methodology”When you apply this methodology to a new conversation, these are the fields you’ll typically need to change:
- Add your stimulus image using the Add Media section – the conversation introduction references it directly.
- Context & Background – describe what the stimulus is, what it’s intended to communicate, and any relevant context about the audience or category.
- Research Objectives – update to reflect what you specifically need to understand: is this primarily about emotional response, message clarity, brand fit, or behavioural intent?
- Laddering questions – the current set covers the full arc from association through to behavioural intent. If your study has a narrower focus (e.g. purely communication clarity), you may not need all five. Remove questions that aren’t relevant rather than leaving them active – unused laddering questions still affect conversational pacing.
- Probe & Prompt Limit – scale to match the number of topics you configure with active probing or prompting.
The conversation introduction – and particularly the “without describing it” instruction – is the most important thing to preserve. It’s what separates a stimulus response study from a description task. Change the framing carefully if at all.
Frequently asked questions
Section titled “Frequently asked questions”Why does the introduction say “without describing it”?
Describing an image is an analytical act that puts participants in a reporting mindset rather than a reacting one. “What do you see?” produces a literal inventory; “what comes to mind?” produces an associative response. The instruction is specifically designed to prevent participants from treating the exercise as a description task – which would produce surface-level observation rather than the impressions, feelings, and meanings the methodology is designed to capture.
Why is Randomise on for the laddering questions?
It reduces order effects across participants. If every participant gets the emotional question (Q2) immediately after the initial response, the data for that question is always collected at the same point in the conversation. Randomising means the emotional question sometimes arrives after participants have already explored associations – which produces more varied and representative data across the sample.
Why is Paraphrase on here but off in the Keyhole technique?
The Keyhole laddering questions follow a deliberate narrative arc – their specific phrasing matters because they’re guiding participants through a sequential experience from start to finish. The questions here don’t depend on each other in the same way. They’re exploring different dimensions of response (association, emotion, interpretation, intent) that can be approached in any order. Paraphrase variation keeps the conversation natural without undermining a sequential structure that doesn’t exist here.
Can I use multiple images as stimuli?
The methodology is configured for a single stimulus shown at the start of the conversation. If you need to test multiple stimuli, create a separate conversation for each one – this ensures each image gets its own clean, uncontaminated set of first impressions. Showing stimuli sequentially within a single conversation introduces order effects and priming that make it difficult to isolate responses to each one.
What kinds of stimulus work best with this methodology?
The methodology is designed for any image-based stimulus: communication concepts, advertising visuals, pack designs, scenario descriptions, mood boards, or concept statements presented as images. The key requirement is that the stimulus is a single, self-contained image participants can respond to instinctively. Stimuli that require significant reading or explanation before a response is possible are less suited – the first-impression dynamic depends on an immediate, pre-verbal reaction to something visible.