PhD Defence: Leveraging sensing-based interaction for supporting reflection at work: the case of...

Leveraging sensing-based interaction for supporting reflection at work: the case of crisis training

Simone Mora

AP Photo/Asahi Shimbun, Noboru Tomura

Outline

• Problem domain

• Research questions

• Theoretical underpinning

• Research methodology

• Technology tools

• Contributions

• Conclusions and future work

2

Image REUTERS/Mainichi Shimbun

During 1992-2012 disasters have affected 4.4 billion people and caused USD 2 trillion in damage worldwide Source: The UN Office for Disaster Risk Reduction

3

Crisis Training approaches

Maximising learning outcomes from experience-based training events

4

Problem:

COST

REALISMlow high

mod

erate

onerious

Serious game

Physical simulation

Tabletop exercise

Protocol training

Research questions

I. How sensing-based interfaces can be designed to enable unobtrusive experience collection during crisis work?

II. How sensing-based interfaces can be designed to trigger and support reflection activities?

III. How sensing-based interfaces for supporting reflection can be rapidly prototyped?

5

What are the opportunities introduced by combining reflective learning theories with sensing-based interfaces for supporting crisis training?

Crisis training

Reflective learning

Sensing-based interfaces

Reflective learning

WORK EXPERIENCE REFLECTIVE PROCESS OUTCOMES

+ New perspectives + Change in

behaviour + Readiness + Commitment

Sharing of work experience + Reflection = New knowledge

6

[Boud, D., Keogh, R., & Walker, D. (1985). Reflection: turning experience into learning.]

Doctoral theses at NTNU, 2015:123

Simone Mora

Leveraging sensing-basedinteraction for supportingreflection at work: the caseof crisis training

NTN

UN

orw

egia

n U

nive

rsity

of

Scie

nce

and

Tech

nolo

gyTh

esis

for

the

degr

ee o

fPh

iloso

phia

e D

octo

rFa

culty

of I

nfor

mat

ion

Tech

nolo

gy,

Mat

hem

atic

s an

d El

ectr

ical

Eng

inee

ring

Dep

artm

ent o

f Com

pute

r an

d In

form

atio

nSc

ienc

eD

octo

ral t

hesi

s

This PhD work

Computer supported reflective learning

7

new reflection cycle

Change

Outcome Fram

e


Plan and do work

Plan workDo work

Monitor work

Initiate reflection Set objectiveInvolve othersPlan session

new re

flectio

n

cycle

Data

Apply outcomeDecide on change to

work (e.g. what, who)Decide how to make the

changeDecide whether further

reflection is needed

Conduct reflection sessionMake related experiences available

Reconstruct or envision work experienceUnderstand meaning Articulate meaningCritique experienceReach a resolution

Check applicability of reflection outcome

Triggering of reflection (new cycle)

Input to next stage in the cycle

StageActivities that may be

involved

Legend:

EU-IST/FP7 2010-2014 Empirical data Validation of theories Technology tools

Reflection Theory Evaluation framework

[Krogstie, B. R., Prilla, M., & Pammer, V. (2013). Understanding and Supporting Reflective Learning Processes in the Workplace: The CSRL Model.]

Sensing-based interaction and tangible user interfaces

8

Tangible interaction denote system that rely on embodied interaction, physical representation of data and embeddedness in the real space

(Hornecker and Buur 2006)

CSCW


Change

Outcome Fram

e


Plan and do work

Plan workDo work

Monitor work

Initiate reflection Set objectiveInvolve othersPlan session

new re

flectio

n

cycle

Data

Apply outcomeDecide on change to

work (e.g. what, who)Decide how to make the

changeDecide whether further

reflection is needed

Conduct reflection sessionMake related experiences available

Reconstruct or envision work experienceUnderstand meaning Articulate meaningCritique experienceReach a resolution

Check applicability of reflection outcome

Triggering of reflection (new cycle)

Input to next stage in the cycle

StageActivities that may be

involved

Legend:

Generate experiences Capture experiences

Re-create experiences

DATA

WORK

REFLECTION

LEARNING

INTUITIVE

PLAYFUL

EMBODIED

EMBEDDED

Methodology overview

9

ENVIRONMENT KNOWLEDGE BASEDESIGN SCIENCE

Hardware and software rapid prototyping

Formative evaluation

ComputerSupportedReflectiveLearning

CrisisTraining

Sensing-based interaction

Relevancecycle

Rigorcycle

Exploratory field studies

Evaluation field studies

Designcycle

[Hevner, Alan R (2007). “A three cycle view of design science research”.]

Exploratory studies

10

• OBSERVATIONS • SHADOWING • VIDEO • INTERVIEWS

METHODS:

Prototypes

11

2010 2011 2012 2013 2014

Exploratory field studies

Prototypes

F2F1

F6F5

G1 G2 G3 G4

W1 W4W2

D1 D2

W3

C1 C2

Research abroad

Papers

CITY MIT

P7 P1 P4

P6

P5

P2

P4

Field evaluations

F3 F4

Formative Evaluations

REL

EVA

NC

EC

YCLE

DES

IGN

CYC

LER

IGO

RC

YCLE

W4W3

W2W1

C2C1

D2D1

Evaluation studies

12

• QUESTIONNAIRES • INTERVIEWS • OBSERVATIONS

METHODS:

13

0" 1" 2" 3" 4" 5"

To interact with W. I had to shift my sight away from the rescue work

To interact with W. I had to shift my hands away from the rescue work

To interact with W. I had to shift my attention away from the rescue work

I believe that after some usages I will interact with W. without shifting attention

W. empowered me to capture data without interupting the rescue work

W. Interfered with the rescue work

Novices"

Experts"

Error! No text of specified style in document. Page 5

Version 1.0

collected data. In addition all the teams have filled in a pre and post questionnaire. All the participants did it on a voluntary basis, without any organizational pressure.

The respondents were overall satisfied with the use of the system (SAT01, M=4.20, SD=0.58) and perceived it as a useful tool for training (SAT02, M=4.26, SD=0,61). The respondents also agreed that the system helped them to reflect on their work (CA2, 4.06, SD=0,59). The information collected with WATCHiT was perceived as accurate (GAE17 M=3.97, SD= 0,57), relevant (GAE19, 4,11, SD=0,53), and collection of data was effortless (GAE18, 3,91, SD=0,56).

1.4.2 Level 2: Learning

1.4.2.1 Learning Process

Respondents agree that the system has provided them with support for reflection in terms of relevant contents, provided help in capturing and reconstructing work experiences and guidance throughout the reflection process (CAs, M=4.13, SD=0.42). The acceptance rate is higher in the population of participants with more than 5 years of experience (M=4.26, SD=0.41), than among novices (M=3.96, SD=0.37). A higher acceptance rate is found among experts across all the app-specific questions asked (see table X), the performed (two-tailed) t-test shows that the difference is statistically significant (p=0.036).

Table X. Means and standard deviations for app specific questions

ID Question All (N=35)*two users didn’t report yrs of experience

M (SD)

At least 5 years of experience (N=18)

Less than 5 years of experience

(N=15)

CA2 WATCHiT helped me to reflect on experiences

4.06 (0.59) 4.28 (0.46) 3.73 (0.59)

CA6 WATCHiT helped me to reconstruct a work experience

4.29 (0.52) 4.39 (0.50) 4.13 (0.52)

CA7 WATCHiT helped me by capturing my reflection outcomes

4.15 (0.56) 4.18 (0.53) 4.07 (0.59)

CA12 WATCHiT helped me by providing accurate information about my work

4.03 (0.51) 4.22 (0.55) 3.80 (0.41)

CA40 WATCHiT provided relevant content for reflection

4.29 (0.52) 4.44 (0.62) 4.07 (0.27)

CA41 WATCHiT guided me through the reflection process

3.91 (0.61) 4.00 (0.59) 3.80 (0.68)

In both experiments we have added in the pre-questionnaire the short reflection scale (CRs). Results show that participants have a positive inclination towards reflection both on an individual level (M=4.38, SD=0.30) and as a team (M=4.22, SD=0.38). As the evaluations

Evaluation studies

Outline

• Problem domain





• Contributions


14

WATCHiT

P2 P3 P6 P7

RQ1 RQ2

C2 C4C1

CAPTURE EXPERIENCE

15

17

TAGHAPTIC FEEDBACK

You're in: New York City, U.S.CroMAR 8:24 AM

MoreMapAR NotesLayers

Face Time

Share

Timeline

"I'm trapped in my car#panic!"

@marco via twitter

Next tweet containing #panic (1 of 10+)

0.7 Km5" later

MMS from +3934667xxx

Tags: car, obstacle

Noise level

95 dB CCTV Video

"Agent Dokes in sector 5 needs an ambulance"

Yesterday (Oct, 27th)

14:00 16:00 18:00 20:00 22:00 24:00

CroMAR

19

P1 P2 P6 P7

RQ2 RQ3

C3 C4C1

RE-CREATE EXP

20

PHYSICAL EXPLORATION

RADAR & MAPSMULTIPLE LAYERS

TIMELINE NAVIGATION

REFLECTION OUTCOMES EDITOR

P4 P5 P7

RQ2 RQ3

C3 C4

Don’t Panic

C1

GENERATE EXP

22

23

NAVIGATE

CALM, MOVE OR RESTRAIN CITIZENS

MANAGE INFORMATION UNDER STRESS

Outline

• Problem domain





• Contributions


25

Papers

26

P1: Mora, S., Boron, A., & Divitini, M. (2012). CroMAR: Mobile Augmented Reality for Supporting Reflection on Crowd Management. International Journal of Mobile Human Computer Interaction.

P2: Mora, S., & Divitini, M. (2014). Supporting Debriefing with Sensor Data: A Reflective Approach to Crisis Training. In Proceedings of Information Systems for Crisis Response and Management in Mediterranean Countries, ISCRAM-MED.

P3: Mora, S., & Divitini, M. (2014). WATCHiT: a modular and wearable tool for data collection in crisis management and training. In Proceedings of the European Conference in Ambient Intelligence, AMI.

P4: Di Loreto, I., Mora, S., & Divitini, M. (2012). Don’t Panic: Enhancing Soft Skills for Civil Protection Workers. In Proceedings of International Conference on Serious Games Development Applications, SGDA.

P5: Mora, S., Di Loreto, I., & Divitini, M. The interactive-token approach to board games. Ready for submission.

P6: Müller, L., Divitini, M., Mora, S., Rivera-Pelayo, V., & Stork, W. (2014). Context Becomes Content: Sensor Data for Computer Supported Reflective Learning. IEEE Transactions on Learning Technologies.

P7: Mora, S., & Farshchian, B. A. (2010). A Unified Architecture for Supporting Direct Tag-Based and Indirect Network-Based Resource Discovery. In Proceedings of the International Conference on Ambient Intelligence, AMI.

27

C1C2

C3 C4

P2ISCRAM-

med´14

P3AMI‘14

P4SGDA

‘12

P6IEEETLT

‘15

P7AMI‘10

P1IJMHCI

‘12P5‘15

Instantiation of CSRL model

Design of tools forexperience-capture

Novel sensing-basedinteraction techniques

Challenges in buildingprototypes

Conference Journal Submitted Contribution

Theo

ry

Tech

nolo

gy

TEL ISCRAM TEI

Contribution I

28

Implementation and evaluation of MIRROR Computer Supported Reflective Learning (CSRL) theory

C2

C3 C4

P3AMI‘14

P7AMI‘10

P1IJMHCI‘12

P5‘15

C1 P2ISCRAM-

med´14

P4SGDA‘12

P6IEEETLT

‘15

Contribution II

29

Knowledge about designing experience-capturing tools for crisis workers

1. Mobility of work and sensing

2. Different crises, different relevant data

3. Different types of data

4. Sensor data and user-submitted data

5. Different use, different sharing

6. Intuitive, hands-free interaction

7. Automate and discrete capturing

C1P4SGDA‘12

P6IEEETLT

‘15

C3 C4P7AMI‘10

P1IJMHCI‘12

P5‘15

C2P3AMI‘14

P2ISCRAM-

med´14

Contribution III

30

Novel sensing-based interaction techniques to support crisis training

CAPTURE EXPERIENCES RE-CREATE EXPERIENCES GENERATE EXPERIENCES

C2P2

ISCRAM-med´14

C1P4SGDA‘12

P6IEEETLT

‘15

C4P7AMI‘10

C3P1IJMHCI‘12

P5‘15

P3AMI‘14

Contribution IV

31

Knowledge about implementing prototypes to be deployed into the wildC3P1

IJMHCI‘12

C2P2

ISCRAM-med´14

C1P4SGDA‘12

P6IEEETLT

‘15

C4P7AMI‘10

P5‘15

P3AMI‘14

32

C2

C3 C4

C1The CSRL model can be instantiated into an ecology of sensing-based interfaces for the capture, re-creation and generation of crisis work experiences

Future work will further map with technology the activities described in the model

Conclusions and future work

Theory in the field of tangible, embodied and embedded computer can drive the design of user interfaces for supporting reflection in crisis training

Future work will generalise findings to other application domains

Building sensing-based interfaces require a wide skill set. Further technology to be tested during (simulated) crisis work has to be build for high resilience

Future work will develop tools to ease the development of sensing-based systems

Unpredictably of relevance for the data and highly varying crisis scenarios make hard to design experience-capturing tools. A design space is provided in a set of design choices

Future work will explore further implementation of the design space

Thanks!

The work here presented is co-funded by EU-ICT 7FP MIRROR project (http://www.mirror-project.eu). Photos by ANPAS Piemonte and Simone Mora.

PhD Defence: Leveraging sensing-based interaction for supporting reflection at work: the case of...

Technology

Transcript of PhD Defence: Leveraging sensing-based interaction for supporting reflection at work: the case of...