03 · A moment in Japan
The cherry-blossom experiment
The same moment. A different invitation to feel.
One minute of my own footage from Japan: a walk under cherry blossoms in Tokyo, ending on a bronze Buddha at Senso-ji. The same edit, shown two ways. Twenty viewers in four groups of five.
Version A
The haiku
The same footage with natural sound and shamisen, and three haiku on screen. No narration.
Version B
Show and tell
The same footage and sound. The only difference: a voiceover tells you what to feel.
Read the voiceover script
“Spring in Tokyo.” / “The cherry blossoms are in full bloom.” / “Pink petals hang heavy on every branch.” / “Feel your shoulders drop. Feel your breath slow down. You are safe here.” / “We walk slowly beneath the blossoms.” / “Feel the soft light on your skin. Feel how quiet your mind is becoming.” / “The blossoms turn from pink to white.” / “Let yourself feel calm and let yourself feel light.” / “People rest quietly in the park.” / “Everything feels calm and peaceful.” / “The petals remind us that spring is short.” / “Feel gratitude for this moment.” / “A single white blossom, perfect and fragile.” / “Feel tenderness. Feel wonder.” / “At the temple, the Buddha sits in silence.” / “Feel stillness, wanting nothing, needing nothing.” / “A moment of peace.”
The two versions are identical except for the voiceover, so any difference comes from the explaining alone.
Four rooms, five viewers each
Group 1
The ideal viewer
Full attention on the pause; the baseline.
Procedure
- 01
Group drawn from a hat.
- 02
One viewing, alone, on the same laptop with headphones.
- 03
Survey straight away.
- 04
A short debrief.
Four questions, in order
- 01
Recall
“Describe everything you remember seeing and hearing.”
Details remembered, scored out of 5.
- 02
Visual detail
5 multiple-choice questions, e.g. “What colour were the blossoms?”
Questions correct, out of 5.
- 03
How moved
“How moved were you?”
From 1 to 5.
- 04
Found meanings
“In one or two sentences, what was this moment about?”
The viewer's own interpretations in their answer, 0 to 3.
GivenRepeats the voiceover: “A moment of peace.”
FoundThe viewer's own reading: “It felt like the walk was leading somewhere quiet.”
- 05
Eyes on screen
“How much of the clip were you actually looking at the screen?”
Self-reported, 0–100%.
What the viewers showed
| Measure | Haiku · no phone | Haiku · phone | Voiceover · no phone | Voiceover · phone |
|---|---|---|---|---|
| Recall | 4.6/5 (4–5) | 2.4/5 (2–3) | 3.6/5 (3–4) | 3.2/5 (3–4) |
| Visual detail | 4.6/5 (4–5) | 2/5 (1–3) | 2.2/5 (2–3) | 1.2/5 (1–2) |
| How moved | 4.6/5 (4–5) | 2.2/5 (2–3) | 2.6/5 (2–3) | 1.8/5 (1–2) |
| Found meanings | 2.6/3 (2–3) | 0.8/3 (0–1) | 0.8/3 (0–1) | 0.8/3 (0–1) |
| Eyes on screen | 97.8% (95–100) | 52% (45–60) | 89.8% (85–94) | 47.2% (42–53) |
With full attention, the haiku version was remembered best. With a phone, recall leaned toward the voiceover.
Two-way analysis (version × phone), 5 viewers per group. Without a phone, the haiku version scored significantly higher on every measure (p ≤ .04). With a phone in hand, the differences were not significant, and recall leaned toward the voiceover (3.2 vs 2.4, p = .06). The phone lowered every score (p < .001). Effects this large in a 20-person pilot are likely to shrink in a larger study.
What was this moment about?
Group 1 · Haiku · no phone
FoundViewer response
“How fast spring passes. The petals and the slow walk made me think about growing older with my family.”
Group 2 · Haiku · phone
Viewer response
“Taking a quiet walk in spring, though I missed the middle part.”
Group 3 · Voiceover · no phone
GivenViewer response
“How nothing lasts forever, because the voiceover explained that theme right away.”
Group 4 · Voiceover · phone
GivenViewer response
“Impermanence. The voiceover summed up the point of the walk.”
Selected answers, one per group.
Explaining helps viewers follow, but it takes away what they find for themselves.
Results reflect group averages from a 20-person pilot study (5 viewers per group).
04 · What this means for film
What the experiment shows
With full attention, the haiku version was remembered, seen and felt more on every measure. With a phone in hand, those differences disappeared, and the voiceover kept people following the story slightly better. Explanation helps distracted viewers follow. It cannot give back what only watching gives. Ma is the space where the viewer does part of the work, and it only works if someone is looking.
Telling is not seeing
The voiceover named the blossoms and the Buddha, yet its viewers recalled fewer details. Being told where to look did not make people look closer.
Meaning is found
Left alone with the images, viewers built their own readings. That search is where a film becomes personal.
The trade-off
Every feeling said out loud is one the viewer no longer has to feel. Explanation did not buy clarity; it cost memory, feeling and meaning.
For filmmaking, this is the quiet cost of telling instead of showing. A film that explains itself is easier to watch half-distracted, but it asks nothing of us — and what a film asks of us is what makes it stay with us. Ma is not empty time to be filled; it is the space where the viewer does part of the work, and where the work becomes theirs.
This is a 20-person pilot with a one-minute clip. It cannot prove how the whole industry is changing — but the pattern it shows, explanation traded for interpretation, is the same pattern I measured in the Zuko comparison.
Credits
Made by Asli
Filmed in Tokyo, April 2024 · Edited in Adobe Premiere Pro · Haiku by Bashō and Ryōkan, chosen and placed by me · Voiceover script written by me · Website built with Lovable
Ethics for the experiment
Only people who agree take part and anyone can stop; no names are collected. This is an informal pilot.






