robot as a learning partner for promoting proactive discussion in peer groups: a case study for...

25
Robot as a Learning Partner for Promo1ng Proac1ve Discussion in Peer Groups: A Case Study for Career Development Toshio Mochizuki, Senshu Univ., Japan Yoshitaka Mitate, The University of Kitakyushu, Japan Yoshikazu Tateno, Jun Nakahara, & Naomi Miyake, The University of Tokyo, Japan Takehiro Wakimoto, Aoyama Gakuin University, Japan Yuko Miyata, Freelance Career Consultant 1

Upload: toshio-mochizuki

Post on 03-Jul-2015

104 views

Category:

Education


0 download

DESCRIPTION

This paper describes an experiment on peer groups that had a robot as a learning partner, to examine whether the robot could encourage the participants to talk on their own initiative. The authors measured the number of proactive utterances of each participant during the sessions. The authors compared the experimental groups that had robot facilitators, which were manipulated by professional human facilitators, and the control groups, which were also led by professional human facilitators but without a robot. The result showed that the participants in the experimental sessions talked on their own initiative much more than those in the controlled sessions. Finally, the authors qualitatively examined the characteristics of the proactive utterances in the peer group and found that the utterances contained supportive responses, which encouraged the participants to voluntarily join the dialogue promoting the counseling.

TRANSCRIPT

Page 1: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Robot  as  a  Learning  Partner  for  Promo1ng  Proac1ve  Discussion  in  Peer  Groups:   A  Case  Study  for  Career  Development

Toshio  Mochizuki,  Senshu  Univ.,  Japan  Yoshitaka  Mitate,  The  University  of  Kitakyushu,  Japan  Yoshikazu  Tateno,  Jun  Nakahara,  &  Naomi  Miyake,                     The  University  of  Tokyo,  Japan  Takehiro  Wakimoto,  Aoyama  Gakuin  University,  Japan  Yuko  Miyata,  Freelance  Career  Consultant �1

Page 2: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Background  and  Research  Ques1ons

�2

Page 3: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Robots  as  learning  partner -­‐  an  emerging  agenda  for  CSCL!• Two  reasons  to  apply  robots  as  social  partner  in  learning  (Miyake  &  Okita,  2012;  Okita  et  al.,  2010)  – Robots  have  human-­‐like  appearance  and  behavior  that  can  elicit  social  responses  that  invite  acSve  engagement  • Not  too  human-­‐like  and  sSll  machine-­‐like.  • This  gives  people  room  for  imaginaSon  and  creaSvity  in  social  interacSon  and  elicits  greater  empathy.  !

– A  robot  can  be  an  interface  to  collect  the  process  data  of  collaboraSve  learning.  • In  order  to  analyze  the  mechanisms  and  design  the  principles  of  producSve  learning

�3

Page 4: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Past  research  of  using  robots   as  learning  partners  –  structured  seEngs

• Learning  English  with  children  (Kanda  et  al.,  2004)  – A  certain  social  and  emoSonal  support  was  required  in  order  to  enhance  collaboraSon  between  children  and  robots  !

• Use  of  Honda  ASIMO  in  a  variety  of  se[ngs  (Okita  et  al.,  2010)  – CooperaSve  interacSon  with  more  human-­‐like  voice  and  gestures  was  effecSve  to  make  children  engage.

�4

Page 5: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Past  research  of  using  robots   as  learning  partners  –  structured  seEngs• A  scripted  collaboraSve  learning  based  on  Jigsaw  (Miyake,  2012)  – EffecSve  if  children  recognize  a  robot  as  “just  like  the  other  kid  who  does  not  know  the  answer,  but  sincerely  working  to  know  the  answer”  !

• Jigsaw-­‐based  reciprocal  teaching  (Oshima  &  Oshima,  2013)  – Robots  can  work  as  well  as  human  facilitator.

�5

Page 6: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Agenda

• The  potenSal  of  robot  facilitaSon  in  ill-­‐structured  se[ngs  –With  ore  self-­‐regulated  student  discussions  in  order  to  learn  what  the  students  do  not  know  in  the  absence  of  educaSonal  materials.  !

• To  find  out  whether  it  creates  an  environment  to  encourage  parScipants  to  talk  on  their  own  iniSaSve.  – A  case  study  in  career  development  

• Career  development  is  a  serious  learning  agenda  for  adolescents  in  every  country    

– No  hierarchical  difference  between  robots  and  young  parScipants  

– The  robots’  appearance  is  more  neutral  than  that  of  human  career  counselors

�6

Page 7: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Method  and  Design  (1)

�7

• Design  – Experimental  study  design

student

student

student

student

Career  counselor

robot

Remote  controlExp.

ControlCareer  

counselor student

student

Page 8: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Method  and  Design  (2)• ParScipants  – Undergraduate  students  (sophomores  and  juniors)  in  a  private  university  in  Tokyo  !

– Professional  facilitators  for  career  development  • More  than  5-­‐year  experience  of group  counseling.  

– Desktop  Rovovie-­‐W  robot

�8

Page 9: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Method  and  Design  (3)• Design  – Experimental  study  design

�9

student

student

student

student

Career  counselor

robot

Remote  controlExp.

ControlCareer  

counselor student

student

Page 10: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

�10

Robot  Operated  by  Remote Human    Facilitator

Professional Human    FacilitatorPresent

Page 11: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Method  and  Design  (3)  

• Discussion  theme:  – “So,  please  tell  us  about  your  career  goals  for  the  future,  Mr.  B.”  – “How  do  you  plan  to  use  the  coming  summer  vacaSon  to  prepare  for  your  

career  goals?”    – “Your  plan  is  (summary  of  what  Mr.  B.  says)”;  

“What  do  you  think  of  his  goals,  Mr.  A?”    – “Do  you  all  have  any  suggesSons  for  preparing  for  Mr.  B’s  career  goals?  “ �11

Step Experimental  Group Control  Group1 Pre-­‐quesSonnaire2 Ice-­‐breaking  (10  min.)3 IntroducSon  for  group  counseling  (5  min.)4 Group  Counseling  Session  (30  min.)5 Post  interview  (30min.)6 Post-­‐quesSonnaire

Page 12: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

�12

Hello,  I’m  Robovie.  Nice  to  meet  you  all.  First  of  all,  I  would  like  to  start  thissession  by  introducing  ourselves  to  each  other.  Please  tell  your  name,  the  place  where  you  come  from,  and  your  goal  in  this  session.  We  have  10  minutes  from  now....

Page 13: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Research  Method  and  Design  (3)  

• Discussion  theme:  – “So,  please  tell  us  about  your  career  goals  for  the  future,  Mr.  B.”  – “How  do  you  plan  to  use  the  coming  summer  vacaSon  to  prepare  for  your  

career  goals?”    – “Your  plan  is  (summary  of  what  Mr.  B.  says)”;  

“What  do  you  think  of  his  goal,  Mr.  A?”    – “Do  you  all  have  any  suggesSons  for  preparing  for  Mr.  B’s  career  goal?  “ �13

Step Experimental  Group Control  Group1 Pre-­‐quesSonnaire2 Ice-­‐breaking  (10  min.)3 IntroducSon  for  group  counseling  (5  min.)4 Peer  Group  Counseling  Session  (30  min.)5 Post  interview  (30min.)6 Post-­‐quesSonnaire

Page 14: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Data  collec1on

• QuesSonnaires  (pre  and  post)  – to  examine  the  results  were  significantly  different  – the  General  Self-­‐Efficacy  Scales(GSES)(Sakano  &  Tojo,  1989)    – EffecSveness  of  peer  support  can  be  explained  by  individual’s  improved  self-­‐efficacy  (Bandura,  1997;  Benight  &  Bandura,  2004)  

• Video  of  discussion  during  the  peer  group  acSvity  – to  examine  how  students  parScipated  in  the  discourse  

• Post  group  interview  (video-­‐recorded  &  transcribed)  – To  examine  feelings  about  the  peer  group  experience.

�14

Page 15: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Analysis  and  Findings

Page 16: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Effec1veness  of  the  Peer  Group  for  Career  Counseling  Purposes

�16

2.56% 2.57%2.07% 2.08% 2.13% 2.23%%

2.72% 2.61% 2.36% 2.15% 2.40%% 2.42%

1%

1.5%

2%

2.5%

3%

3.5%

4%

Robot% Human% Robot% Human% Robot% Human%

Agressiveness%of%Ac>on% Anxiety%to%failure% Social%posi>on%of%ability%

Before% AEer%

* **

** *

*p<.05

Page 17: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Coding  rule  of  transcribed  peer  group

�17

F F

FF

P P

P

P

P

P

P

P

prompt

prom

pt

prompt

prompt

Page 18: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Differences  of  Students’  Par1cipa1on  in  the  Peer  

�18

23.5%

21.3%

17.9%

21.9%

79.3%

6.4%

17.2%

10.2%

0% 20% 40% 60% 80% 100%

FP/p%

FP/a%

PP/p%

PP/a%

Experimental% Control%

(S.D.=14.8)

(S.D.=15.1)(S.D.=4.7)

(S.D.=16.7)(S.D.=19.7)

(S.D.=20.1)(S.D.=6.6)

(S.D.=34.9)

***

***

*

**p<.001,  *p<.05

Page 19: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Differences  of  Students’  Par1cipa1on  in  the  Peer  

�19

23.5%

21.3%

17.9%

21.9%

79.3%

6.4%

17.2%

10.2%

0% 20% 40% 60% 80% 100%

FP/p%

FP/a%

PP/p%

PP/a%

Experimental% Control%

(S.D.=14.8)

(S.D.=15.1)(S.D.=4.7)

(S.D.=16.7)(S.D.=19.7)

(S.D.=20.1)(S.D.=6.6)

(S.D.=34.9)

***

***

*

**p<.001,  *p<.05

Human  facilitators  could  speak  fluently  and  use  their  acSve  

listening  strategy  

The  robots  lacked  fluency  because  the  operators  could  only  input  messages  using  the  keyboard  aser  understanding  the  parScipants’  uterances  

F P

prompt

Page 20: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Differences  of  Students’  Par1cipa1on  in  the  Peer  

�20

23.5%

21.3%

17.9%

21.9%

79.3%

6.4%

17.2%

10.2%

0% 20% 40% 60% 80% 100%

FP/p%

FP/a%

PP/p%

PP/a%

Experimental% Control%

(S.D.=14.8)

(S.D.=15.1)(S.D.=4.7)

(S.D.=16.7)(S.D.=19.7)

(S.D.=20.1)(S.D.=6.6)

(S.D.=34.9)

***

***

*

**p<.001,  *p<.05

F P

The  parScipants  spoke  to  the  robot  facilitator  much  more  than  they  did  to  the  human  facilitator.    

Page 21: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Differences  of  Students’  Par1cipa1on  in  the  Peer  

�21

23.5%

21.3%

17.9%

21.9%

79.3%

6.4%

17.2%

10.2%

0% 20% 40% 60% 80% 100%

FP/p%

FP/a%

PP/p%

PP/a%

Experimental% Control%

(S.D.=14.8)

(S.D.=15.1)(S.D.=4.7)

(S.D.=16.7)(S.D.=19.7)

(S.D.=20.1)(S.D.=6.6)

(S.D.=34.9)

*

**p<.001,  *p<.05

The  parScipants  in  the  experimental  groups  parScipated  in  the  discussion  much  more  proacSvely  than  in  the  control  groups,  even  with  fewer  prompts  from  the  robot  facilitators.  

**p<.001,  *p<.05

F P

P

P

Page 22: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Characteris1cs  of  Proac1ve  USerances   in  the  Peer  Groups  • To  clarify  how  the  parScipants  proacSvely  spoke  in  the  experimental  groups    –Whether  the  peer  groups  produced  a  posiSve  mood  that  encouraged  the  members  to  disclose  what  they  thought  (Paine  et  al.,  1989)  

• AcSve  listening  skills  for  proacSve  discussion  – clarifying,  paraphrasing,  and  summarizing  • given  in  the  scenario  in  both  group  

– SupporSve  responses  • But  difficult  for  the  robot  operator  to  react  smoothly  by  using  supporSng  response.   �22

Page 23: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Characteris1cs  of  Proac1ve  USerances   in  the  Peer  Groups  

�23

The  facilitator  should  try  to  elicit  supporSve  responses  from  other   members  during  the  self-­‐help  group  session  (Paine  et  al.,  1989)

Page 24: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Characteris1cs  of  Proac1ve  USerances   in  the  Peer  Groups  

�24

There  were  significantly  many  supporSve  responses  from  the  parScipants  in  the  experimental  groups  and  from  the  facilitators  in  the  control  groups.  (χ2(1)  =  69.664,  p  <  .01),  

The  facilitator  should  try  to  elicit  supporSve  responses  from  other   members  during  the  self-­‐help  group  session  (Paine  et  al.,  1989)

Uh huh

Page 25: Robot as a Learning Partner for Promoting Proactive Discussion in Peer Groups:  A Case Study for Career Development

Conclusion• Both  experimental  and  control  peer  groups  produced  similar  results  with  regard  to  self-­‐efficacy  improvement  

• The  parScipants  in  the  experimental  groups  talked  on  their  own  iniSaSve  during  the  sessions.    

• The  parScipants  in  the  experimental  groups  used  supporSve  responses  much  more  than  those  in  the  control  groups  did.

�25

The  robot  has  the  potenSal  to  create  a  similar  effecSve  peer  discussion,  and  a  more  parScipant-­‐centered  proacSve  discussion