A review of MR companions for navigation

CASE STUDY

UX/UI-XR-EMERCIVE

PROJECT BY - yASH ANIL bHAGAT

RESEARCH · SYSTEMATIC LITERATURE REVIEW

MR Companions for Navigation

A systematic review of an emerging field — and the four gaps it found.

Springer ICWIHI 2025 · First Author · Sep 2024 – Apr 2025 · Lead Researcher · Supervisor: Prof. Delwyn Remedios, IIT Hyderabad

AT A GLANCE

Lonelier, and more connected, than any generation before

I wanted to know if a Mixed Reality companion could help people navigate the world and their own social anxiety without replacing real human contact. Before designing anything, I read the field systematically: 35 papers, 8 databases, four themes and mapped exactly where the research stops.

Lonelier, and more connected, than any generation before

Lonelier, and more connected, than any generation before

35+

papers read in full, screened from 8 databases

themes mapped across the field’s research

ICWIHI 25

published · Springer AISC · first author

“The internet offers immense possibilities for encounter and solidarity. Yet the desire for digital connectivity can isolate us from those closest to us.” — Pope Francis

03 · TIMELINE

Seven months, from a hunch to a published paper

Sep 2024 – Apr 2025 · Systematic Literature Review

Topic & scoping

Sep – Oct ‘24

• MR companions + loneliness framing

• Research question set

Search · 8 databases

Oct – Nov ‘24

• 8 databases queried

• Boolean search strings

Screening → 35+ papers

Nov ‘24 – Jan ‘25

• Title + abstract screening

• 35+ papers read in full

Thematic analysis

Dec ‘24 – Feb ‘25

• Coded the full corpus

• Four themes mapped

Findings & writing

Jan – Apr ‘25

• Four gaps identified

• Full paper drafted

ICWIHI 2025 published

Apr 2025

• Springer ICWIHI 2025

• First author

03 · TIMELINE

Seven months, from a hunch to a published paper

Sep ‘24

Oct

Nov

Dec

Jan ‘25

Feb

Mar

Apr

Topic & scoping

Search · 8 databases

Screening → 35+ papers

Thematic analysis

Findings & writing

Published · Apr 2025

TIMELINE

Seven months, from a hunch to a published paper

Sep ‘24

Oct

Nov

Dec

Jan ‘25

Feb

Mar

Apr

Topic & scoping

Search · 8 databases

Screening → 35+ papers

Thematic analysis

Findings & writing

Published · Apr 2025

SCENARIO

◇ DISCOVER

You’d rather stay lost than ask

New city. Phone almost dead. You’ve turned the same corner three times. The person at the chai stall would know the way in a heartbeat and you’d sooner keep walking than say a word to them.

Asking a stranger for help in public can be daunting, especially for young adults. We’re the most connected generation, yet often the loneliest.

So the question that started this project: could the thing already in your pocket help you move through the world and toward people instead of away from them?

You’d rather stay lost than ask

New city. Phone almost dead. You’ve turned the same corner three times. The person at the chai stall would know the way in a heartbeat and you’d sooner keep walking than say a word to them.

Asking a stranger for help in public can be daunting, especially for young adults. We’re the most connected generation, yet often the loneliest.

So the question that started this project: could the thing already in your pocket help you move through the world and toward people instead of away from them?

THE PROBLEM, IN DATA

◇ DISCOVER

Our World in Data (2024)

35%

Of young adults reported insufficient social connections, feelings of loneliness in Asia.

32%

Of respondents noted lack of social ties, social bonding in India.

Our World in Data (2024)

35%

Of young adults reported insufficient social connections, feelings of loneliness in Asia.

32%

Of respondents noted lack of social ties, social bonding in India.

Our World in Data (2024)

35%

Of young adults reported insufficient social connections, feelings of loneliness in Asia.

32%

Of respondents noted lack of social ties, social bonding in India.

Gallup Global Emotions Report

27%

Of individuals aged 19-29 reported feeling “very lonely” or “fairly lonely” Worldwide.

21%

Of the general U.S. young adults population feels lonely (Harvard Report).

Gallup Global Emotions Report

27%

Of individuals aged 19-29 reported feeling “very lonely” or “fairly lonely” Worldwide.

21%

Of the general U.S. young adults population feels lonely (Harvard Report).

Our World in Data (2024)

35%

Of young adults reported insufficient social connections, feelings of loneliness in Asia.

32%

Of respondents noted lack of social ties, social bonding in India.

Gallup Global Emotions Report

27%

Of individuals aged 19-29 reported feeling “very lonely” or “fairly lonely” Worldwide.

21%

Of the general U.S. young adults population feels lonely (Harvard Report).

THE QUESTION

◇ DEFINE

What does the field actually know?

Before proposing a companion, I wanted the honest state of the evidence not my hunch about it. So the review set out to answer one question:

How can Mixed Reality companions be designed to enhance the navigation experience and what does existing literature reveal about the current state, limits, and future of MR companion design?

How can Mixed Reality companions be designed to enhance the navigation experience and what does existing literature reveal about the current state, limits, and future of MR companion design?

Navigation was the lens on purpose: it’s where spatial cognition, real-time help, and social context collide the exact place where isolation and the need for support meet.

MY ROLE

Sole researcher, first author

I defined the scope, designed the review method, searched eight databases, screened and selected every paper, ran the thematic analysis, and wrote the publication. My supervisor, Prof. Delwyn Remedios, guided methodology and database choice — the research decisions were mine.

Research scope

SLR methodology

Screening & selection

Thematic analysis

Academic writing

WHY A REVIEW

◇ DEFINE

Map the territory before you build

The honest first move wasn’t to prototype, it was to find out what the field had already proven. A systematic review forces that honesty: explicit search terms, defined inclusion rules, and a traceable decision at every stage. For a topic scattered across HCI, psychology, education and medicine, it was the only way to build one coherent picture.

THE DECISION

I chose a systematic review over a quick literature survey. A survey lets you cherry-pick what supports your idea; an SLR makes you account for everything you find. I wanted a method that could prove me wrong.

WHY A REVIEW

◇ DEFINE

Map the territory before you build

THE DECISION

WHY A REVIEW

◇ DEFINE

Map the territory before you build

THE DECISION

THE SEARCH

◇ DISCOVER

Eight databases, one honest net

The same five search terms ran across all eight sources ‘Mixed Reality personal assistants,’ ‘user experience in MR,’ ‘social isolation and MR,’ ‘empathy in MR interactions,’ and ‘cognitive load in MR assistants.’

Google Scholar

IEEE Xplore

SpringerLink

ScienceDirect

Scopus

PubMed

Elsevier

ResearchGate

THE DECISION

I deliberately pulled in medical databases like PubMed. Social isolation isn’t only an interaction problem it’s clinical. Pure design databases would have missed half the evidence for why any of this matters.

SELECTION

◇ DEFINE

From a wide net to thirty-five + papers

Inclusion: 2007–2024, English, MR personal assistants touching user experience, empathy, cognitive load, or social isolation. Exclusion: VR-only or non-MR work, no agent present, purely theoretical papers. What passed, I read in full.

Wide search across 8 databases

hundreds of candidate papers

Screened on inclusion / exclusion criteria

2007–2024 · MR + UX / empathy / cognitive load / isolation

35 papers read in full

35+

Papers Reviewed

database

themes

months

published paper

35+

Papers Reviewed

Databases

Themes

Months

Published Paper

THE DECISION

I kept the net wide, then cut hard on explicit criteria rather than on gut feel. Thirty-five was where the evidence was strong enough to categorise honestly — not a number I set out to hit.

FOUR THEMES

◇ DEFINE

Four ways the field talks about companions

The 35 papers didn’t sit in one conversation. They clustered into four — and the split itself showed where the attention was, and where it wasn’t.

User Experience & Engagement

How MR assistants shape immersion, engagement and satisfaction.

Social Isolation & Well-being

Whether MR companions can genuinely reduce loneliness — especially in young adults.

Cognitive Load & Task Performance

How real-time MR guidance changes mental strain during tasks.

Technological Frameworks

The architectures behind MR assistants, and how platforms compare.

User Experience & Engagement

How MR assistants shape immersion, engagement and satisfaction.

Social Isolation & Well-being

Whether MR companions can genuinely reduce loneliness — especially in young adults.

Cognitive Load & Task Performance

How real-time MR guidance changes mental strain during tasks.

Technological Frameworks

The architectures behind MR assistants, and how platforms compare.

KEY FINDINGS

◇ DEFINE

Four things the research made clear

Across the 35 papers, four findings held up consistently and two of them quietly rewrote how I’d design a companion later.

FINDING 01

Give it a face but not a human one

Human-like features boost engagement, yet hyper-realism triggers the Uncanny Valley. Clean, abstract characters consistently beat photorealism.

FINDING 02

Social Isolation & Well-being

Whether MR companions can genuinely reduce loneliness especially in young adults.

FINDING 03

Guidance lowers the mental tax

In complex tasks, real-time MR guidance cuts cognitive load turning flat 2D map logic into 3D, lived direction.

FINDING 04

The loneliness evidence is thin

MR companions can ease isolation, but the studies are short-term, small-scale and headset-bound. The long-term effect is essentially unstudied.

11 · KEY FINDINGS

◇ DEFINE

Four things the research made clear

Across the 35 papers, four findings held up consistently and two of them quietly rewrote how I’d design a companion later.

FINDING 01

Give it a face but not a human one

Human-like features boost engagement, yet hyper-realism triggers the Uncanny Valley. Clean, abstract characters consistently beat photorealism.

FINDING 02

The body says more than the voice

Gesture, expression and posture drive social presence more than dialogue does but most systems still lean almost entirely on voice.

FINDING 03

Guidance lowers the mental tax

In complex tasks, real-time MR guidance cuts cognitive load turning flat 2D map logic into 3D, lived direction.

FINDING 04

The loneliness evidence is thin

MR companions can ease isolation, but the studies are short-term, small-scale and headset-bound. The long-term effect is essentially unstudied.

KEY FINDINGS

◇ DEFINE

Four things the research made clear

Across the 35 papers, four findings held up consistently and two of them quietly rewrote how I’d design a companion later.

FINDING 01

Give it a face but not a human one

Human-like features boost engagement, yet hyper-realism triggers the Uncanny Valley. Clean, abstract characters consistently beat photorealism.

FINDING 02

The body says more than the voice

Gesture, expression and posture drive social presence more than dialogue does but most systems still lean almost entirely on voice.

FINDING 03

Guidance lowers the mental tax

In complex tasks, real-time MR guidance cuts cognitive load turning flat 2D map logic into 3D, lived direction.

FINDING 04

The loneliness evidence is thin

MR companions can ease isolation, but the studies are short-term, small-scale and headset-bound. The long-term effect is essentially unstudied.

TO BE CONTINUED

The review found the gaps. Designing into them became ARiA.

Part 2 takes these four findings and builds into them — a mobile, chibi-faced MR companion, tested with real users. Research becomes a design response.

PUBLICATION

Peer-reviewed, and on the record

Bhagat, Y. A., & Remedios, D. J. (2025). The Potential of Mixed Reality (MR) Companions for Navigation — A Systematic Literature Review. In Proceedings of ICWIHI 2025, Springer AISC Series. First Author.

Springer AISC

ICWIHI 2025

First Author

Peer-reviewed

Yash A. Bhagat

Design Researcher · HCI · Systems

yashbhagat.4199@gmail.com