Image Theatre

Image Theatre – Evaluation Method 01
About This Project

Days in the Bay – Tiger’s Bay Men’s Group

Men from Tiger’s Bay, a working-class Protestant neighbourhood in North Belfast, devised and performed a short piece about their own streets. Using Image Theatre, they externalised what their community meant to them – without having to articulate it in words first. The project was led by David Grant, Senior Lecturer in Drama at Queen’s University Belfast, and represents his most developed application of image-based methods in a community evaluation context.

Location Tiger’s Bay, North Belfast
Participants Adult men from the local community
Delivered by Queen’s University Belfast
Evaluation Method: 01

Image Theatre

When words fail or deceive, ask people to show you. The body reveals what the interview misses.

Phase 0: Logic Model & Topic Mapping

Identify the “Evaluation Gap” this method fills. Use it when direct questioning produces safe, predictable answers – or when your participants find written and verbal formats alienating. It works best for post-experience reflection and groups with mixed communication styles or abilities where standard methods create barriers before you’ve even started.

Primary Objective

Surface what participants genuinely felt – including ambivalence, contradiction, and the things they couldn’t find words for – by letting the body make the argument instead.

Key Stakeholders

Community group members, arts project participants, people with disabilities or neurodivergent needs, anyone for whom a questionnaire or interview creates more barrier than insight.

Part 1: The DIY Step-by-Step Method

1

Frame the Task

Tell participants exactly what they’re doing and why there’s no right answer. Most people relax the moment they hear this isn’t a test. Keep the brief short – over-explaining kills spontaneity, and spontaneity is the whole point.

The Details:
  • Action: Tell the group: “We’re making a frozen image – like a photograph – to show [the experience / place / question]. No words while you build it. Don’t plan, just arrange yourselves.” Two minutes maximum briefing.
  • Tip: Never describe what a good image looks like. The moment you do, you’ve already contaminated the data.
Say Exactly This: “Make me a frozen photograph that shows [the experience / this place / how you feel about X]. Don’t discuss it first. Trust your instincts. You’ve got two minutes.”
2

Build the Image

Participants arrange themselves in the space and freeze. No words during the build. Where they stand, who they face, who they turn away from, how close they get – all of it is data, including the choices they don’t know they’re making.

The Details:
  • Action: Step back completely. If the group stalls, repeat the prompt once – don’t demonstrate or suggest. The struggle to find the image is part of the process.
  • Tip: Photograph or film the moment it settles, before anyone moves. This is your primary evidence artefact. Everything else is commentary on it.
3

Read the Room

Open the image to interpretation – starting with observers, not image-makers. Ask what people see, not what was intended. Proximity, orientation, who’s hidden behind whom: the spatial detail is where the meaning lives. Hold multiple readings simultaneously. The moment one interpretation wins, the richness is gone.

The Details:
  • Action: Ask observers first. Then invite image-makers to respond – but never insist on a definitive reading. Record the discussion verbatim if you can, or take dense notes.
  • Tip: Your job is to keep multiple interpretations alive, not to resolve them. Sit with the ambiguity. That tension is your most useful output.
Facilitator Prompts: “What do you notice first?” / “What’s the relationship between these two people?” / “Is anything hidden in this image?” / “Who reads it differently – and why?”
4

Map the Voices (Optional)

When the topic involves internal conflict – identity, belonging, difficult choices – go one level deeper. Boal called this the Rainbow of Desire. Identify all the competing impulses the protagonist holds at once, embody each one, and let them speak simultaneously. Then ask the protagonist to arrange them in the room by strength of pull. The resulting spatial map is a richer account of a complex inner state than any interview could produce.

The Details:
  • Action: Ask the group: “How many different things does this person want at once?” Name each impulse, assign a person to it, run the simultaneous voices, then have the protagonist arrange them spatially. Photograph the final arrangement.
  • Tip: Only attempt this after Steps 2 and 3 have warmed the group up. It requires enough trust and physical ease that it falls flat if introduced too early.
5

Harvest the Data

The image, the discussion, and your observations are your dataset. You don’t need to analyse everything on the day – artefacts can be reviewed later by people with deeper subject knowledge who may spot things you missed. Write up immediately while it’s live in your memory.

The Details:
  • Action: Document what was shown, not just what was said. Record specific spatial choices, moments of surprise, silence, laughter. These are as evidentially significant as direct quotes.
  • Tip: Run the transcript through a free word-cloud tool (Wordcloud.com, Mentimeter). The most frequent words appear largest – a fast, visual way to identify dominant themes across multiple sessions, and one that reads as data to funders who expect numbers.

Part 2: The Realist Pathway

Inputs
What you need: Four or more participants, open floor space, a phone for photos or video, a facilitator comfortable with stepping back, and optionally a word-cloud tool for transcript analysis afterwards.
Mechanism
Why it works: Physical arrangement in space bypasses the cognitive filter that edits spoken and written responses. Drawing on embodied cognition – the understanding that the body thinks, not just the brain – participants make spatial choices that reflect felt knowledge they haven’t yet rationalised or censored. The image arrives before the defence does. This is especially powerful on sensitive topics and with groups who have learned to give “safe” answers in formal settings.
Outcomes
What you get: Unguarded, unedited responses. Insights that surprise participants themselves. Reusable visual artefacts for reporting. And conversations that go somewhere a survey or interview never reaches.

Evidence Manifest

How the mechanism has been verified – specific cases where Image Theatre produced evidence that other methods missed.

Evidence Type What it Proves Data Source
Embodied Insight
Unplanned Spatial Meaning
When participants create an image they believe represents one thing, the spatial arrangement often reveals something else entirely – a truth the group hadn’t consciously planned to show. This is the core mechanism in action: the body externalises knowledge the words hadn’t reached. Peer-reviewed research on Image Theatre in community settings.
Memory Retrieval
Physical Recall of Past Experience
When participants are asked to physically reconstruct an experience from months earlier, specific and detailed recollections emerge that direct questioning fails to surface. Physical re-enactment unlocks stored experience that language alone cannot access – making this method particularly valuable for post-project evaluation. Peer-reviewed research on embodied recall and community evaluation.
Quantified Output
Word Cloud Analysis
Discussion transcripts from Image Theatre sessions, processed through word-cloud software, reveal consistent cross-group themes and meaningful differences between audiences. Qualitative discussion data becomes a visual, comparable, reportable output – bridging the gap between rich description and the numerical evidence funders typically expect. Established practice in qualitative data visualisation. Free tools include Wordcloud.com and Mentimeter.
Practitioner Evidence
Embodied Field Notes
Practitioners who record their post-session observations through movement or drawing, rather than written notes alone, often retrieve more accurate and detailed recall of what happened. The embodied mechanism is available to facilitators as a data-gathering tool, not just to participants. Documented in applied and participatory arts practice.
Reporting Hacks: Turning Images into Proof

The artefacts you made in the room are your evidence. Use them directly – don’t bury them in prose.

The Annotated Photo

A photograph of the image is not illustration – it’s primary data. Add two or three call-out labels naming what the group noticed in the spatial arrangement. It becomes an analytical exhibit, not decoration. Treat it like a figure in a paper: give it a number, a caption, and a place in your findings.

THE HACK: Annotate the photo in Canva or Google Slides, export as a single-page exhibit, and include it as a numbered appendix figure.
The Word Cloud as a Finding

Paste your discussion transcript into a free word-cloud tool. Frequency drives scale – the words that came up most appear largest. Stack one cloud per session to show what was consistent across groups and what shifted. It converts a qualitative discussion into a visual that reads as quantitative evidence without distorting what was actually said.

THE HACK: Generate one word cloud per session, align them side by side, and label them by group or date – instant cross-session comparison.
The Before and After

If you ran Image Theatre at the start and end of a project, print both photographs together. The change in spatial relationships – who moved closer, who turned to face forward, what disappeared – shows shift more directly than any summary paragraph. Let the images make the argument.

THE HACK: Place both images side by side, date-labelled only, with the single question “What changed?” below – and let the reader draw the conclusion.

The “One-Sentence” Report

“Because we used Image Theatre, participants in our community felt safe enough to show what they couldn’t say – revealing evidence about their experience that no survey or interview would have reached.”

Your Voice Matters|Built by Practitioners|For Practitioners

Help us improve this tool.

Tell us who you are, what you used, and what would make it better. It takes two minutes.

Give Feedback →