06 / Individual · Interactive installation
A Cactus’s Gaze
An interactive installation that lets a cactus answer touch through sound.
A Cactus’s Gaze began with photographs of names carved into cacti at a botanical garden. I translated that observation into a multimedia installation: a living cactus responds to touch with changing scream sounds, while a short collage animation imagines a world where plants observe, display and mark humans.

- Role
- Individual designer, researcher and interaction builder
- Year
- 2024
- Format
- Multimedia interactive installation
- Tools
- Photoshop, After Effects, Premiere Pro, Arduino and Max
Current outcome
The completed work was exhibited as a working interaction. Visitors approached the cactus, tested different kinds of touch and compared its sound response with Roy’s animated story. The project brings field observation, narrative development, motion design, electronics and exhibition making into one experience.
01 / Context and field research
Names carved into living cacti made the power relationship visible.
At a botanical garden, I photographed cacti covered with names and drawings scratched into their surfaces. The marks showed a contradiction: the plants were kept inside a protected space and still treated as surfaces available for visitors to use.
My first concept asked people to write on a cactus while a screen displayed changing plant data. I removed that action because it repeated the damage I wanted to question and could confuse the meaning of the work. Touch offered a safer, clearer starting point for the interaction.


02 / Narrative reframing
Roy enters a world where humans occupy the role of plants.
The animation follows Roy, a cactus visiting a park where humans are planted in pots and displayed for other species. At the end of the visit, Roy writes his name on a human arm. The reversal carries the botanical-garden observation into a fictional world and gives the audience another position from which to read the installation.
An early drawn prototype did not communicate that this was a plant-led world. I moved to digital collage and stop-motion-inspired movement, building scenes in Photoshop, animating them in After Effects and assembling the final sequence in Premiere Pro. The uneven scale and cut-out movement support the story’s uncomfortable tone.
03 / Interaction prototype
Touch data drives a changing sound response.
I used Arduino to monitor changing electrical readings from the plant and sent the serial data to Max. When the readings moved through different ranges, the background track lowered and different scream samples played. A light touch and a firmer hold therefore produced different responses.
Two failures shaped the final setup. The cactus initially produced no usable data, so I corrected the Arduino code and rebuilt the resistor and cable connections. Max later stopped receiving the signal when I changed devices; checking and matching the serial port restored the link. These tests became a practical checklist for the exhibition build.


04 / Exhibition experience
A living cactus, moving image and sound create one encounter.
The exhibition joined three elements: the cactus as the physical interface, the animated plant world on screen and the sound response produced through Max. I installed the display, concealed the electronics and cables, positioned the speaker and calibrated the input in the gallery.
Visitors approached in different ways. Some touched one stem carefully, some held the cactus with both hands, and others gathered around to compare the sound with the scene on screen. The documentation shows that the interaction attracted participation, while the range of gestures exposed the need for clearer guidance around safe contact.


05 / Reflection and next test
The response was readable; the invitation still needs care.
The working installation gave the concept a clear cause and effect: touch the cactus, hear a response, then read that action against Roy’s story. It also revealed an ethical tension. The work asks people to touch a living plant while questioning the way people treat plants.
A next version would move more of the sensing away from the plant’s body, calibrate the input for light contact and add a concise instruction before visitors reach for it. I would also use short post-interaction prompts to record what people understood, since exhibition behaviour alone cannot show whether the intended critique was clear.