Pediatric Eye Exam Robot

Pediatric Eye Exam Robot

Designing Emotionally Supportive Interaction for Children in Clinical Robotics

The Problem

OCT eye exams are intimidating for children. The clinical environment, unfamiliar robotic equipment, and lack of emotional scaffolding leads to fear, low cooperation, and disrupted workflows.

The Solution

A character-driven interaction system, including a named robot persona, pre-exposure animated media, and evidence-backed storyboard designed to reduce anxiety and build trust before the exam begins.

Who this is for

Children aged 2-12 undergoing OCT eye exams, and their caregivers. Young patients experience fear and low cooperation in clinical settings due to unfamiliar robotic equipment and lack of emotional scaffolding.

What we learned

Through 30 caregiver surveys at Kellogg Eye Center, 78% of emotional responses to character-driven interaction were curious or comfortable, compared to stress and anxiety reported with standard OCT exams.

Role
UX Design Research Specialist
Team
UARTS Faculty Engineering/Arts Student Teams (FEAST) · Kellogg Eye Center
Timeline
Nov 2024 – Dec 2025
Tools
Figma · Miro · Procreate · R · Excel · ACM LaTeX · Canva · Promiere Pro

TL;DR

Interdisciplinary HRI research at UMich designing emotionally intelligent interaction for pediatric clinical robotics

Challenge

Children fear the robotic OCT device. The mechanical design was finalized but the robot had no personality, no interaction behavior, and no emotional scaffolding for young patients.

Approach

Character system design grounded in survey data. Four concepts tested across age groups, refined through medical staff feedback, and narrowed to one final persona: Sam the Ranger.

Solution

A full interaction system consisting of robot persona, pre-exposure animated video, narrative storyboard, and data-backed design rationale designed to reduce first-contact anxiety.

Impact

Findings being synthesized into a co-authored ACM paper. Physical robot prototype in progress. Research contributes to the growing field of child-robot interaction in clinical settings.

Key Design Decisions

Three choices that shaped the interaction system with each grounded in research, not assumption.

01

Character over function

The robot's mechanical capability was already defined. The missing layer was personality. A named, story-driven character (Sam) gave children a relationship to form before the exam to shift the experience from clinical procedure to familiar encounter.

02

Age-differentiated design

Ages 3–6 responded to gentle, animal-inspired characters emphasizing comfort. Older children preferred detective-style characters with interactive "missions." One character can't serve both so the system accounts for the developmental stage.

03

Pre-exposure before first contact

Pediatric media literature shows that familiarity before an event significantly reduces anxiety. The animated pre-exposure video that is designed for home and waiting room viewing creates that familiarity before the child ever sees the robot.

The Robot

↳ The mechanical arm was finalized as our job was giving it a soul.

robot arm video demo

Research & Strategy

Four methods, one goal to find what actually makes children less afraid.

It's a super cool project. Has the potential to make the situation so much easier for children.

Caregiver, Kellogg Eye Center survey

The premise is good. More narrative and early introduction would make it fully effective.

Caregiver, Kellogg Eye Center survey
Contextual Inquiry

Shadowed real OCT procedures at Kellogg Eye Center by observing wait times, emotional friction points, and how children responded to the equipment in the moment.

Character System Design

Generated multiple character concepts targeting age-based responses. Four finalists were refined through medical staff feedback for hygiene and clinical compatibility.

Survey Analysis

Analyzed post-visit survey responses from children and families. Created data visualizations revealing character preferences, emotional states, and media response patterns.

One observation led to two completely different design problems.

01

Observe real exams

Shadowing OCT procedures revealed two very different kids in the same room — a 6-year-old gripping their parent's hand in fear, and a 10-year-old slumped in the chair, visibly bored. Same exam, completely different emotional states.

02

Two personas, two distinct problems

That observation became two personas and journey maps — each revealing a fundamentally different design need.

Age 6 😨 Scared

"What is that thing? Is it going to hurt me?"

Needs comfort and familiarity before anything else.

Age 10 😑 Bored

"How long is this going to take?"

Needs engagement and a sense of mission to stay cooperative.

03

One solution cannot serve both

A gentle animal character calms a 6-year-old but feels babyish to a 10-year-old. A mission-based character engages older kids but overwhelms younger ones. This is why we developed 4 distinct concepts.

Persona: Age 6, Scared
Persona: Age 10, Bored
personas
Journey Map: Age 6
Journey Map: Age 10
journey maps

Character Design

↳ Four concepts and Sam the Ranger, selected for cross-age appeal and clinical compatibility.

Each character was designed to target a different emotional register and age group. Survey data and medical staff review narrowed four designs to Sam — a ranger-style character with a mission-based narrative that works across developmental stages.

Character Design Board
character design board created with team
Optic Character
SAM the Ranger
A-Eye Character
Cheeze Character
four character finalists with Optic · Sam the Ranger · A-Eye · Cheeze

Storyboard & Pre-Exposure Media

↳ Narrative arc designed to mirror a child's emotional journey.

Narrative Structure

Two-part storyboard following exposition → conflict → rising action → climax → resolution. Part 1: Sam's introduction (trust + excitement). Part 2: Sam's farewell (closure + positive memory).

Pre-Exposure Video

Calming animated video introducing Sam played at home and in clinic waiting areas. Based on pediatric media literature showing pre-familiarity significantly reduces first-contact anxiety.

Academic Paper

Co-first authoring a forthcoming ACM paper synthesizing findings across media design, child–robot interaction, and clinical UX research. Written in ACM LaTeX.

SAM the Ranger Storyboard
SAM the Ranger storyboard

Data Analysis & Deliverables

↳ Quantitative survey data + qualitative emotional feedback synthesized into design decisions.

Survey Analysis & Visualizations Character Design System Narrative Storyboard ACM Paper Pre-Exposure Video Physical Robot Prototype — continued by team
Survey data analysis
survey data analysis

Survey Results

30 caregivers evaluated character designs after viewing pre-exposure videos at Kellogg Eye Center.

93%

preferred robotic OCT over standard exam

78%

of emotional responses were curious or comfortable

30

caregivers surveyed at Kellogg Eye Center

4

character designs evaluated across age groups

Reflection

1
Emotional design is clinical design

In pediatric healthcare, reducing anxiety is a functional requirement.

2
Data and story work together

Neither alone was enough to make the right design decision.

3
IRB constraints are design constraints

Collecting data from children under ethical protocols required the same careful framing as any sensitive UX research context.