MoMo
A tangible AI interface that keeps you grounded in the real world when your screen pulls you away.
Team
Chloe Ni - Product Designer
Gonzalo Minuto - Product Manager
Xiaoyang Wu - Hardware Designer
Timeline
4 months, 2025 fall
​
Problem
We are losing control of our attention in the media consumption era; with it, we feel miserable.
AI-driven media consumption and AI-assisted production now co-amplify attention fragmentation, creating self-reinforcing patterns of dopamine-driven behavior that undermine critical thinking, agency and cognitive health.​​
Problem 01 - Doom Scrolling
Autoplay, Infinite Scroll, Personalized Feeds, Reward Loops
algorithmic consumption caused time loss, compulsion, mental fatigue
Problem 02 - Multi Tasking
Notifications, Context collapse, Tab overload, Tool blending
fragmented cognition caused by blocked flow state, cognitive overload, stress
Problem 03 - Gen AI overuse
Shortcuts, Auto-complete thinking, Over-assistance, No friction
cognition outsourcing caused dependency, loss of depth, reduced agency
Context
Every existing digital wellbeing tool makes the same mistake: it fights screen addiction with more screen. App timers, usage dashboards, notification blockers — they all occupy the same attentional channel as the problem they're trying to solve. MoMo takes the intervention off-screen entirely, externalizing digital consumption into a tangible, affective presence that you care for. The premise is simple: what if the motivation to put your phone down wasn't compliance with a rule, but empathy for something that needs you?
case study: current market does not have tools that fulfill all of companionship, productivity, and subtleness
Solution
Our solution is to create an intuitive tool that helps you be mindful of your relationship with technology, and change it:
- Only objective is to restore awareness
- Physical and tangible to ground attention in the real world
-No screens, no metrics, no notifications, no settings
-Behavior emerges through interaction rather than instruction
-Uses soft, embodied cues instead of commands or alerts
-Calming and supportive rather than punitive or moralizing
Goal 01
A tangible AI companion that increases users awareness of their attention patterns and reduces compulsive phone use without explicit restriction
Goal 02
Progressive, embodied, and affective cues influence emotional connection and behavioral change toward technology use by creating a gradual, non-abrupt path back to awareness
What is MoMo?
MoMo, stands for Mobile Monster, is a palm-sized, multimodal AI companion that functions as an intuitive productivity tool that helps you be mindful of your relationship with technology. It doesn't have a screen. It doesn't send you notifications. Instead, it breathes, pulses, and reacts, mirroring your digital consumption through biomimetic physical behaviors.
When you're in a healthy usage pattern, MoMo is calm: slow, rhythmic breathing, quiet. As passive scrolling escalates, MoMo's state deteriorates. Its breathing quickens, driven by a micro servo actuating a rack-and-pinion expansion mechanism inside its shell. A coin vibration motor begins pulsing against your palm like an accelerating heartbeat. Audio cues shift. The device is visibly, tangibly distressed.
To soothe MoMo, you have to put the phone down and pet it. A capacitive touch sensor beneath the outer shell detects sustained contact, closes the feedback loop, and triggers a transition back to calm: breathing slows, haptics ease, the device settles.
Design Process
Software
A lightweight iOS companion app handles sensing and communication. Users select the specific apps they find problematic — social media, news, anything — and the app monitors foreground usage in real-time, batching data and transmitting it to MoMo over BLE. Critically, the app displays no metrics, no scores, no dashboards. All feedback lives in the physical device. All data processing stays local, and nothing is stored externally or transmitted to servers. MoMo is deliberately non-surveillant.
Hardware
MoMo is built around an Arduino Nano microcontroller housed inside a 3D-printed internal skeleton, small enough to fit within a 100mm spherical shell. The system coordinates five subsystems: a TI CC2540 BLE module for low-latency phone communication, a micro servo driving the breathing mechanism, a flat coin vibration motor for haptic feedback, a DFPlayer Mini MP3 module with onboard microSD for synchronized audio, and a capacitive touch sensor for petting detection. A custom biomimetic algorithm smooths servo motion to eliminate mechanical jerkiness, producing organic, breath-like movement rather than robotic actuation. The outer shell is furry fabric: it's light enough for the servo to actuate without strain, and it's tactilely rewarding to interact with. Petting it actually feels good, which matters when that's the core interaction.
MoMo’s electromechanical architecture. (a) Mechanism for vertical expansion; (b) BLE module for app connectivity; (c)
Capacitive touch sensor for detecting petting interactions; (d) Arduino Nano microcontroller; (e) Micro servo for actuation; (f) Coinvibration motor for haptic feedback; (g) Speaker for audio output.
Interaction Design
MoMo's behavior is governed by an internal state machine with distinct expressive profiles mapped to phone usage patterns and physical interaction. Crucially, with an integrated AI model, MoMo operates as an affective black box: its internal logic is intentionally opaque. Users aren't optimizing a system; they're responding to something that seems to need them.
​
map of MoMo's reaction stage responding to human input
We validated the multimodal cue design through a pilot study evaluating sound and haptic combinations. Six participants completed pairwise comparisons across stimulus groups, ranking cues by emotional legibility. Consistent patterns emerged, giving us a consolidated set of reactions with clear interpretability across emotional states.
MoMo's behavioral progression
Evaluation and Conclusion
We ran a 10-participant evaluation study in a naturalistic setting — no assigned tasks, just 20 minutes of coexistence with MoMo. Participants beforehand described their tech relationship in terms like "over-attached," "toxic," and "co-dependent." After the session, the framing shifted. Quantitatively: 5/5 for purposefulness, 4/5 for reducing doom-scrolling during the session, 4.17/5 as more effective than standard screen timers. The care-based model generated measurably less psychological resistance than rule-based restriction — reframing disconnection as empathy rather than enforcement worked.
​
​Software platforms are engineered to capture and hold attention. Fighting that with more software is a losing game. MoMo argues for something different: that the most effective interventions might be physical, peripheral, and empathetic, not punitive, not data-driven, not screen-based. A small, breathing thing on your desk that needs you to put the phone down. Sometimes the simplest reframe is the most powerful one.
