PreprintPDF Available
Preprints and early-stage research may not have been peer reviewed yet.

Abstract

Background: The development of generative AI systems like ChatGPT has provoked debates about their effective use in educational settings. Aims: The present study explores how university students prompt ChatGPT to solve complex non-routine problems, specifically examining which prompts are associated with higher or lower problem-solving performance. Sample: Seventy-seven university students (53 women; Mage = 22.4 years) participated in the study. Methods: To identify various prompt types employed by students, the study utilized qualitative analysis of interactions with ChatGPT 3.5 during the resolution of the creative problem-solving task. Participants’ performance was measured by the quality, elaboration, and originality of their ideas. Subsequently, two-step clustering was employed to identify groups of low- and high-performing students. Finally, process mining techniques (heuristics miner) were used to analyze the interactions of low- and high-performing students. Results: The findings suggest that including clear evaluation criteria when prompting ChatGPT to generate ideas (rs = .38), providing ChatGPT with an elaborated context for idea generation (rs = .47), and offering specific feedback (rs = .45), enhances the quality, elaboration, and originality of the solutions. Successful problem-solving involves iterative human-AI regulation, with high performers using an overall larger number of prompts (d = 0.82). High performers interacted with ChatGPT through dialogue, where they monitored and regulated the generation of ideas, while low performers used ChatGPT as an information resource. Conclusions: These results emphasize the importance of active and iterative engagement for creative problem-solving and suggest that educational practices should foster metacognitive monitoring and regulation to maximize the benefits of human-AI collaboration.

How Students Prompt ChatGPT for Creative Problem-Solving: Process Mining of Hybrid
Human-AI Regulation
 !"#$%&' ()*+)  , ( -$,./ ( -
0,.10)2* (
%1 1#1)3%',.)()4'.5,.,/*')3, ,%)/+)/%)#$5 67869:
;;4#5,.-#( ,
,#(1')31.*1 ,%/.'% ,%.(% $% 1'+)(%)$ ,,.<6<8:4#
5,.-#( ,
,,#(1')31%.(% $% 1'*==(,.8:>794#5,.-#( ,
/,#(1')3#* 1 %.(% $% 1'1)$:7<8?9:;;4#5,.-#( ,
%1 1#13)%,. ), ()**# ,1 )()$,/*')3, ,%#$%,%1
@96?1 %($()$ 
* First and corresponding author:
*=*=#A4* (=,)*B6:@;9?:6><;;;;;;;;7:<<:799
Co-authors:
 !"#$%&;;;;;;;:;9:::@C
' ()*;;;;;;;?@6?;?6
+)  ;;;;;;;<:76::67
 ( -$,.;;;;;;;9?@79@67
 ( -0,.10);;;;;;;7;6<:@?C
2* (;;;;;;;76?6<@9;6
April 9th, 2025.
This is a preprint of an article.
Contents may change prior to publication.
Acknowledgments: .#1.)%/,(1.11.'.$),)3( ,1)3 1%1=.%,.D%
,)/#,1/,)3)* 1'D 1.E%1. ,(,)/=.1. ,(--)$(D%)1 /3)*
1. ,(,)** 11)31.%1 1#1)3%',.)()4'.5,.,/*')3, ,%#/1.
#*69;8)8:;:7=((-1 , -1%-)$ //1. )DD 11,)%1=.D))
1. %--D%%#--)1/'1.5,., ,)#/1 )FGH-)I,1)=:6??=%
)1 $%-%/# 41.--1 ))31. %D)1.#1.)%#%/.1 )/1)
,%1./ ( 1')31.1J1=31#% 41. %1))(1.#1.)%$ D/// 1/1.
,)11%///13#((%-)%  ( 1'3)1.,)11)31.-#( ,1 )=
 :
How Students Prompt ChatGPT for Creative Problem-Solving: Process Mining of Hybrid
Human-AI Regulation
 4.( 4.1%
K% 4.1%)(('% 3)*1 )(%)#, %%%), 1/D 1.()D-3)*,
K 4.-3)* 4%1#/1%44  11 $/ ()4#D 1.
K)*-1%D 1. ,.,)1J1/,($(#1 ), 1 .,,1 $ 1')3)#1,)*%
K)$ / 4D 1.%-, 3 ,3/,(/%1)*)) 4 (/#%3#(%)(#1 )%
K#,,%%3#(,1 $-)(*%)($ 4( %).' /.#*4#(1 )
 7
How Students Prompt ChatGPT for Creative Problem-Solving: Process Mining of Hybrid
Human-AI Regulation
Background: ./$()-*1)341 $%'%1*%( .1.%-)$)//1%
)#11. 33,1 $#% /#,1 )(%11 4%=
Aims: .-%1%1#/'J-()%.)D# $% 1'%1#/1%-)*-1.11)%)($,)*-(J
))#1 -)(*%%-, 3 ,(('J*  4D. ,.-)*-1%%%), 1/D 1.. 4.)()D
-)(*%)($ 4-3)*,=
Sample: $1'%$# $% 1'%1#/1%F?7D)*L M4 M::=6'%H-1 , -1/1.
%1#/'=
Methods: ) /1 3'$ )#%-)*-11'-%*-()'/'%1#/1%1.%1#/'#1 ( 5/N#( 11 $
('% %)3 1,1 )%D 1..17=?/# 41.%)(#1 ))31.,1 $-)(*%)($ 4
1%=1 , -1%E-3)*,D%*%#/'1.N#( 1'()1 )/) 4 ( 1')31. 
/%=#%N#1(' 1D)%1- ,(#%1 4D% *-()'/ 1) /1 3'4)#-%)3 ()D /. 4.
-3)* 4%1#/1%= (('-),%%*  41,. N#%F.# %1 ,%* HD#%/1)('5
1. 1,1 )%)3()D/. 4.-3)* 4%1#/1%=
Results: .3 / 4%%#44%11.1 ,(#/ 4,($(#1 ), 1 D.-)*-1 4.1
1)41 /%Fr%M=79H-)$ / 4.1D 1.()1/,)1J13) /41 )Fr%
M=6<H/)33 4%-, 3 ,3/,Fr%M=6?H.,%1.N#( 1'()1 )/) 4 ( 1'
)31.%)(#1 )%=#,,%%3#(-)(*%)($ 4 $)($% 11 $.#*4#(1 )D 1.. 4.
-3)*%#% 4)$(((4#*)3-)*-1%Fd M;=9:H= 4.-3)*% 1,1/
D 1..11.)#4./ ()4#D.1.'*) 1)//4#(1/1.41 ))3 /%
D. (()D-3)*%#%/.1% 3)*1 )%)#,=
Conclusions: .% %#(1%*-.% 51. *-)1, )3,1 $ / 11 $44*1 3)
,1 $ -)(*%)($ 4 / %#44%1 1.1 /#,1 )( -,1 ,% %.)#(/ 3)%1 *1,)4 1 $
*) 1) 4/4#(1 )1)*J * 51.3 1%)3.#*,)(()1 )=
KeywordsO41 $1 3 , ( 1(( 4,-)*-1 4  4 *1,)4 1 $% ((% .' /
.#*4#(1 ),1 $-)(*%)($ 4
 6
How Students Prompt ChatGPT for Creative Problem-Solving: Process Mining of Hybrid
Human-AI Regulation
Introduction
1 $ 1'%,..%1/ 1 )(('3),#%/) / $ /#(,)4 1 $-),%%%)31)$()) 4
1.,)(()1 $1#)3,1 $/$)%F($#1(=:;7H=)//%%1. %4-2=
PF:;:7H 1)/#,/1.,),-1%)3,1 $,)4#(1 )/,1 $%), (('
%./4#(1 )%/)%), (('%./4#(1 ))3( 4FH3*D)F/D 
1(=:;9LQ$(Q1(=:;:H=)4#(1 )),,#%D. / $ /#(%%#--)1,.)1.E%
4#(1)'-),%%%1.)#4.%), ( 1,1 )%=33,1 $,)4#(1 )-)*-1%,3, ( 11
%(34#(1 )/(/1)*)-)/#,1 $ %), (('%./ 4#(1 ) F/D 1 (= :;9H=
), (('%./4#(1 )-%1%1.. 4.%1($()3,)(()1 $ 4#(1 )D. ,.
4)#-,),)%1#,1%4)(%%114 %/*) 1) 4-),%%%1),. $%./)#1,)*/
D. 4)#- **% ,)((,1 $(' *4 1. ,)4 1 $ *1,)4 1 $ *)1 )( /
*)1 $1 )(%-,1%)31. I) 1,1 $ 1 %FQ$(QP/D :;7H=
# (/ 4)1.%/3  1 )%,1 $,)4#(1 ) $)($%)-%)4 $ 4 3
4#(1)'-)*-11.1))%1%)1. / $ /#(E%)1.4)#-E% %(34#(1 )= ,)1%1
,1 $%), (('%./4#(1 )*4%D.1.1 4)#-44% ,))/ 1//
11 $3 1# 4)31.,1 $1%3)*I) 1(',)%1#,1 41.-)(*-%11 )1)
%(,1 43  4/$(#1 4 /%F2=P:;:7H=1. %3)*)34#(1 )
1. 1,1 ) % / ,1 )(D 1./ 331**%,)1 #1 41)1.4#(1)'-),%%
(/ 41),1 $%)(#1 )1.1 %411.1.%#*)3 / $ /#(,)1 #1 )%F I%1/1
(= :;@L D' :;;>L 2=  P  :;:7H= )D$ 1. /$1 )3 1 3 , (
1(( 4, FH /#,1 )( /D) 4 ,)1J1% .% 1)/#,/ D/'* ,% 1.
4#(1 ) )3 ,)(()1 $ -),%%% F3 1 (= :;:H= R. ( 1/ 1 )( %,. )
4#(1 ),)(()1 ).%3),#%/),)4#(1 )/%), (('%./4#(1 )D 1. 
.#*)(',)1J1%,1/$,*1%. 4.( 4.11.-)11 ()3.' /.#*4#(1 )
F)(:;::HD. ,.%'%1*%* 4.1%'4 %1 ,(('D)D 1.(%1)/1,1/
%-)/1), 1 ,(4#(1)'1 44%)33 4(1 *%,33)(/ 41.1%#--)1%/J1/%
%), (('%./4#(1 )-),%%%FQ$(Q1(=:;:7H=
 ?
$ 3) 1. 1)/#,1 ) )3 41 $ 1 3 , ( 1(( 4, %,. 1%1
.#*,)(()1 )3),)*-(J-)(*%)($ 4D%- /('4)D 4F% *)$ ,1(=
:;:7H=.,),-1)3.' /.#*1 3 , ( 1(( 4,D. ,.,)* %.#*/1 3 , (
1(( 4,1)%)($,)*-(J-)(*%,%% 11%,)(()1 )1),. $)#1,)*%%#- )
1) 1.)% 1.1  1. ,)#(/,. () F((* 1 (=:;@H= . % ,),-1 D% 3#1.
/$()-/ /#,1 )(-%',.)()4'')(F:;::HD.) 1)/#,/1..' /.#*
4#(1 ) 1.)' F.31 H=  % /3 / % ,)(()1 $ / /-1 $
--),.D. ,.)1..#*%/%.,)1)()$1.( 4-),%%/33,1 $
-)(*%)($ 4=.#(1 *14)( %1)1 %1#/1%1),)*-)3 , 1%(34#(1)%D.)
,*41. )D( 433,1 $('/33 , 1('%#--)1/'1.1/-1%1)1. 
$)($ 4/%/% ((%=,1%1#/''1(=F:;:6H-)$ //%#--)1 $$ /,3)
1.)'=. *-  ,(,%%1#/'3),#%/)-)4** 4,)#%/% 4/3)1. /
'#/4/#1 %1#/1%=. 3 / 4% / ,1/1.1 33,1 $ ,)(()1 ) *-)$/
%1#/1%E*1,)4 1 $/%(34#(1/( 4% ((%=)D$)$ ,%)31%1#44(/D 1.
$ 3' 4%-)%%/ /4 4)D(/44-%-1 ,#((' *),)*-(J1%%=. 
3 / 4%#/%,)1.,%% 1')3%,33)(/ 4/)$% 4.1%#44%1 41.1(%E ( 1'1)
*4 / , 1 ,((' $(#1  )#1-#1 * % %%1 ( 3) -)/#,1 $ .#*S
,)(()1 )F1(=:;:?H=// 1 )(('Q$(Q1(=F:;:7HJ-//)1.
1.)' ' ,)-)1 4 1. / )3 .' / .#* %./ 4#(1 ) ( 4= .'
/*)%11/1.1 %$( .#* #%% , ,)(()1 $(' D) D 1. 1) ., .#*
( 4/-)(*%)($ 44#(1 )=)31. ',)1 #1 )% %1.1 44,),-1
3*D)D. ,./1 (%.)D,/1,1/%-)/1)*)*11)*)*1% 4(%F=4=
%. 31% ,)4 1 $()/)*)1 )()#%(H ,)(()1 $( 41.'.(- 4(%
/1*%,1-)/#,1 $%), (('%./4#(1 )=
. 1,1 )1D.#*%/,$''1./4)3.#* $)($*1=
#* 1.())- %'%1*% $)($ ,1 $ .#* -1 , -1 ) -( 4 J,#1 4 /
$(#1 41%% ()4% / N#  4,)1 #)#% 1,1 )=' / 1(( 4, ,#, (
%-,1)3.#* 1.())-%'%1*%)-1 * 5%1.%'4'1D.#*%/1),. $
%#- ) )#1,)*%= R. ( .' / 1(( 4, *-.% 5% ,)1 #)#% 1,1 ) /
,)(()1 ))1.%'%1*%%#,. %.#*)1.())--(,.#*%%#-$ %)')(
D. 1.' - * (' )$% / $ 3' 1. E% 3 ( )#1-#1F3 1 (= :;:H=#1 1)
 >
,. $1.%)#1,)*% 1 %,#, (1),)% /1.)()3-)*-1 4D. ,.4# /%1.E%
%-)%%/ 3(#,% 1.33,1 $%%)3 ,)(()1 ) F4#'1 (=:;:6H= )*-1 4
3%1) 1. -),%% )3 ,31 4 -#1 %1#,1 )%1.1 %.- 41/ )#1-#1% D 1. 1.
,)11)3-)*-1%% 4 3 ,1(' 3(#, 41.N#( 1')31J1-)/#,/'41 $1))(%
%#,.%.1F41(=:;::H=(1.)#4.1. % %- /('$)($ 4/D4# /( %
 4 1)/#,/1)%1#/1%F5)$%'1(=:;:6L)(( ,P)(( ,:;:6H/% 4 4
33,1 $-)*-1%* %,.((4-1 ,#(('3)1.)%D 1.)#1,4)#/ 1 3 , (
1(( 4,)*,. ( 4FT*3 %,# 1(=:;:7H=)1.%%)%1.-%1
%1#/'  *% 1) ('5 1,1 )% 1D .#* #%% / .1 1) $%1 41 1.
.$ )%)3. 4./()D-3)* 4%1#/1% %)($ 4,)*-(J,1 $-)(*%)($ 4
1%=
Creative Problem-Solving with the Generative Artificial Intelligence
1 $-)(*%)($ 4 FH % *#(1 3,1/-),%%%%1 ( 3)$ 41 4 ((/3 /
))#1 ,.((4%=1(% / $ /#(%1)%1# N# 4)(%41) 4 ( %)(#1 )%
-(%-, 3 ,%1-%/%1( %.* 43#($(#1 ), 1 F%%1(=:;L)%%
@@<H= % .1(',',( ,N#  4/ $411.  41)41*#(1 -()$( /%/
*1,)4 1 $% ((%1)*) 1)$(#1/%(,11.*)%1#%3#(/) 4 ()%F#*3)/
1(=@@:;@L33 41(=:;;9H=/41 )( %)3(#,'/3(J  ( 1'
/$()- 4/ 331%)(#1 )%D. (*1,)4 1 $% ((%%#--)11.-( 4 *-(*11 )
/$(#1 ))31.)%%)(#1 )%F=P:;:6H= %,)4 5/%,):%1
,1#'% ((F:;@H 4,#, (3),. $ 44()(%#%1  ( 1'4)(% ,(#/ 4
N#( 1'/#,1 )/ )$1 )/ $ /#%1 %F 1/1 )%:;?H=1.%(%),)*
))31.*)%1 /*/D)-(,% ((% 4%1.%,)/*)%1$(#/,)*-1,'
)4 51 )%D)(/D / F 11 %1 1 (= :;:7H= )* /#,1 )( -%-,1 $  %
,)% //1.. 4.%1,)4 1 $4)(1.$ %/())*E%1J))*'F21.D).(:;;:H%
/ $ /#(%*#%1 #/%1/--(' ('5/ $(#1 3)*1 )3)1.' ,,1
)$(%)(#1 )%F == )$1H=%- 1 1% *-)1,3)*(/#,1 ))313 (%1)-)$ /
J-( , 11  4 %114 %FR 44 %:;H='/#,1)%(,/-#/%1/ 4)3
,1 $ 1'')/ / 41 )F211( 1 (= :;9H(/ 41) / %,),1 1D 1.
,)4 5/$(#)3/ 1%%'%1*1 ,/$()-*1/#,1 )(%11 4%F%1(()1
(=:;<H=
 <
)D$ % /#,1 )( -,1 ,% $)($ - /(' .#*/ $ () *' )
()4%#33 , 1= 4 1(/#,1 ))#, (%#$',)%%>,)#1 %3)#/1.19>U)3
# $% 1'%1#/1%#%41 $ 1. %1#/ %D 1.<9U/) 4%)1(%1D('FT /'
:;:6H= * (('%#$'%- 4;@,)#1 %-)1/1.1 %1)* 4D /% %1.
*)%1 ,)**) #% )3 41 $  *)4 %1#/1% F$V(I 1 (= :;:?H= . 4)D 4
141 ) )3 1 3 , ( 1(( 4, 1) -)(*%)($ 4 ,%% 1. ,%% 1' )3 33 , 1
.' /.#*,)(()1 )F31(=:;:H=
1 (= F:;:6H ,)/#,1/ -4 %1/J- *1 $%1 41 4 1. *-,1 )3
.1),1 $-)(*%)($ 4-3)*,*)4# $% 1'%1#/1%=1. %1#/'<<
%1#/1% %)($/ ,)*-(J -)/#,1 *-)$*1 1% #% 4 .1 D. ( >9 %1#/1%
,)*-(1/1.1%D 1.)#1%% %1,=.3 / 4%%.)D/1.11.%1#/1%D.)D)/
D 1..141/%)(#1 )% 1.1D *)) 4 (()1// )3. 4.N#( 1'
,)*-/1)1.)%D.)D)/ /-/1('=// 1 )(('%1#/1%#% 4.1-, $/
1.1%%% /-)1/J1 4(%%*1(33)1%#44%1 41.1%#--)1/-)(*
%)($ 4 *' /#, ,)4 1 $ ()/= #1.*) ./% 1 (= F:;:6H 3)#/ 1.1
41 $  % 4 3 ,1(' 3(#,/ 1. ,1 $ /% 4 -),%% ., 4 1. N#( 1' )3
/% 4)#1,)*%//#, 4,)4 1 $()/=. %1#/')/% 4%1#/1%%.)D/1.11.)%
D.)#%/1))(%/# 41.  1 (/% 4-.%-)/#,/*) )$1 $// $%/% 4
(11 $%,)*-/1)1.)%D.)/ /)1#%41 $1))(%=.%1))(%,1/%,)
,1)% .(- 4 %1#/1% 1) /$()- / 3  )$( ,),-1% 1.)#4. 11 $ -),%%%
%#(1 4 . 4. ,1 $ 1' / 3)%1 4 *)33 , 1 / )$1 $ -)(*%)($ 4
--),.=
)D$ 1 % *-)111). 4.( 4.11.1%1#/1%/1)/$()-% ((%3)33,1 $
,)(()1 ) D 1.  FQ$(Q 1 (= :;:7H= 1 $  1))(% J.  1 D%%%
,)%1#,1 44#*1%%/)1.)1 ,(3*D)%* 1  4,).,*)4/ 331
4#*1%-)$ / 4--)- 13,% /41 4) 4 (,)11F)5 WPX1#(
:;:7L..1(= :;:7H=#1.*)41/)#1-#1%1/1)*).)*)4)#%
,1 4%)(#1 )%1.1*)% * (1),.)1.1.)#1-#1%/$()-/'.#*%F)%.
P#%:;:6LR4P211:;:?H=%- 11.%D%%%$J-1%)313 / 1
/ 33 ,#(1 1) / %1 4# %. 1D 1. 1J1 -)/#,/ ' .1/ 1.1 D 11 ' .#*%
F(,%1 1(=:;:6L11.D%P+)(-:;:7H=1#/1%.)(/. 4.1#%1 41 $
 9
41/ ,)11 1)  ,),1 F 4 1 (= :;:7H / ) J-1% )31 %1#44( 1)
#/%1/D.'41 $1))(%/ /)1-)/#,1.J-,1/%#(1%FT*3 %,# 1
(=:;:7H=
.%,.((4%#/%,)1./3).' /.#*4#(1 )D. 141 4
41/,)11 1).#*D)=J-()1)'%1#/')4( %.%) 44#4
%,)/' %1#/1%E / 41 ) %114 % #% 4 1#( (4#4 41 ) 1))(% 3)
,1 $D 1 4$(/1D)/ %1 ,1%1#/1--),.%O1.)%D.)(/'./ /%/#%/
1)3 )J1/1.*/1.)%D.)( /)1)41D /%3)*%,1,.FR))
1(=:;:7H=1%1 4('%1#/1%)31$(#1/41/ /%%/)1. ,).,
D 1. J %1 4 /% 1. 1. *, 4 )$( ) #J-,1/ %#44%1 )%= #1.*)
%1#/1%1//1)-31))(%1.141/41N#1 1')3 /%( 41.*1)
%(,1/3 ,)111.1( 4/D 1.1. 1 $%=%/)1. 3 / 4%R))1(=
F:;:7H ,)**// 1.1 .#*#%% %.)#(/ ,3#((' $(#1 41/ /% 3)
,)-)1 4 1.* 1) 1.  D) 1) %# ($, / ,).,= )D$ %1#/1%
1,.%/)1.)J-1%*'/J-( , 1 %1#,1 )).)D1)-)*-141 $
1))(%/)$(#1 4 /%-)/#,/'41 $1))(%F*1(=:;:6H=
1)41.,%$%,)4 1 $,)(()1)1.1.*(' /41)=)D$
1. ( 1'1), 1 ,(('44D 1.)#1-#1% %%%1 (3)3)%1 4,1 $-)(*%)($ 4
%(34#(1 )/* 43#(.#*,)(()1 )=
Present Study
.4)()3.' /.#*%'%1*% %1),1%'4 %1.1,- 1( 5)1.%141.%)3
)1. .#* / 1 3 , ( 1(( 4, #(1 *1(' ,. $ 4 *) 33,1 $ )#1,)*%
F((*1(=:;@H=(1.)#4. 1D%()4( $/1.1%'%1*%,)#(/%% %1)(''
#1)*1 4-1 1 $)*#/ ,1 $ 1 %(()D 4.#*%1)3),#%)*),)*-(J/
,1 $%-,1%)$(41 $1))(%%-, (('3 , (D.%)($ 4%),((/ ((
/3 /)))#1 -)(*%F./%1(=:;:6L 1%#D 1(=:;:7L)'P
T.4:;:7L1(=:;:6H=)D$4 $1.)$(1')341 $1))(%1. %
( * 1/)D(/4)#1 .)D#%% 44D 1.1.*=)%-, 3 ,((') - )%1#/' .%
J* /1.1'-%)3-)*-1%%1#/1%#%D. 1,1 4D 1..1=)//%%1. %4-
1.-%1%1#/'%%1.3 %1%,.N#%1 )=
 @
YO ZR.1 1'-% )3 -)*-1% /) %1#/1% *-()' /# 4 ,1 $ -)(*%)($ 4 D 1.
.1[\
))$ 1 %)1'1,(D.11'-%)3-)*-1%3 , ()/1 *1(1)1.
,1 $-)(*%)($ 4-),%%=.3)1.%,)/%,.N#%1 ) *%1) $%1 411.
*-,1)3 / $ /#(-)*-11'-%),1 $-)(*%)($ 4-3)*,=
Y:OZR.11'-%)3-)*-1%#%/ 1,1 )D 1..1%%), 1/D 1.. 4.
,1 $ -)(*%)($ 4 -3)*, / D.1 1'-% )3 -)*-1%  ( / 1) ()D
-3)*,[\
R. (Y/Y:3),#%)1.1'-%)3-)*-1%%1#/1%#%/1.  **/ 1
%%), 1 ) D 1. ,1 $ -3)*, 1.% N#%1 )% ,-1# )(' -1 )3 1. 1,1 )
/'* ,% .' / .#* %'%1*%= 33,1 $ ,)(()1 ) D 1. 41 $  ( ('
$)($%)1I#%1%(,1 41. 4.11'-)3-)*-1#1(%)44 4-),%%)3 11 $
3 *1 ==%1#,1# 41. 1,1 ))$1 *1),. $11)#1,)*%=. %( 4%D 1.
1.,),-1#( 51 ) )3 .' / .#* 4#(1 ) F)(:;::HD. ,. . 4.( 4.1% 1.
*-)1,)3 *#1#(4#(1 )O 1. #%%1%1. D 1. 4#(1)'-)*-1% /1.
%-)/%D 1. 3/, /%#44%1 )%= #--)1 4 1. % /4#'1 (= F:;:6H 1. 
J-()1)' %1#/' D 1. 1 /),1)( %1#/1% 3)#/ 1.1 %1#/1% D.) 44/ 11 $
J,.4%D 1..1/# 4,/* ,D 1 4-3)*/111.1.)%D.)#%/ 1%
%11 , 3)*1 )%)#,=.% % 4.1%%#44%11.11.%1#,1#//$()-*1)3-)*-1
%N#,%*' % *-)11%1. / $ /#(-)*-1% 1.*%($%= ,,)/ 4('1. 1. /
%,. N#%1 ) J-()% 1. #3)(/ 4 )3 1.% 1,1 )% (1 ) 1) -3)*,
)#1,)*%=
Y7OZ)D/)1.-)*-1%N#,%#%/ 1..1 1,1 )/ 331D. 4./
()D-3)* 4%1#/1%[\
 %#**' 1. % %1#/'  *% 1) /-)# #/%1/ 4 )3 .)D 1,1 )% D 1.
.1,.,,1 $-)(*%)($ 4 ,)*-(J ((/3 /1%%=.%1#/'%%1)
/1 3'1.1'-%)3-)*-1%%1#/1%#%/1* D. ,.-)*-1%%%), 1/D 1.. 4.
)()D,1 $-3)*,/J-().)D1.-)*-1 4-),%%/ 33%1D. 4./
()D-3)* 4%1#/1%='('5 4-)*-1 4-11%1. %%1#/'J1/%')/-)*-1
33,1 $%%1)J-().)D. 4.-3)* 4 / $ /#(%44 %1#,1#/ 11 $/ ()4#
 ;
D 1. .1= /%1/ 4 1.% /'* ,% .% *-)11 %1#,1 )( *-( ,1 )%O 3
33,1 $-)*-1%N#,%, /1 3 //#,1)%,-)$ /$ /,%/4# /,)
.)D%1#/1%%.)#(/ 1,1D 1.1))(%3))-1 *(-)(*%)($ 4)#1,)*%=
Methods
Participants
.-%1%1#/'*-()'/%$1'%$# $% 1'%1#/1%3)*(4# $% 1' 4#
5,.-#( ,F:6*?7D)*LM4M::=6'%SDM6=?H=.%*-(,)% %1/)3?<
%), ( %, ,% / .#* 1 % *I)% < ( 3%, ,% *I)% / 7 1,. ,( / %, -( %
*I)%=3 1.% ?7 %1#/1% D #/4/#1% / :6D 4/#1%= . %*-( D%
.)*)4)#% ,/1. , 1'=
./1%1('5/1. %%1#/') 4 1%3)*1.%*-1 , -1 1,1 )%%
1.)%1.J- *1(,)/ 1 ))31(=F:;:6H=)D$D. (1.-$ )#%%1#/'
3),#%/)N#1 11 $-3)*,*1 ,%1.,#1%1#/'J* %-$ )#%('#('5/
N#( 11 $/-),%%/13)*1.-1 , -1%E 1,1 )%=*-)11('1.-%1%1#/'
J-()%/ 331%,.N#%1 )%/3),#%%)/ %1 ,1%-,1%)3-1 , -1%E44*1
D 1..1=
1. ,(--)$(D%)1 /3)*1.1. ,%)** 11)31.3 %1#1.)]% %1 1#1 )
,,)/,D 1.1.1. ,(- , -(%=
Measures
Creative problem-solving performance. )%%%%,1 $-)(*%)($ 4-3)*,1.
-%1%1#/'*-()'//-1/$% ))31.)/#,1*-)$*1%FH) 4 (('
/1 (/'#1"51(=F:;:H/%/)),F:;;9H=./-1/ $)($/
,)*-(J%, ) D. ,.-1 , -1%./1)41 /%1).,,)**)-)/#,1
1. %,%)/ '%1#33/#'=R 1.1.%% %1,)3.1-1 , -1%D%/1)
/$ % 1. )$1 $ / #%3#( *-)$*1% 1) .(- 11( *I) 1)' *#3,1#
%#-%% 1%,)*-1 1)1.4))#- %(%=R. (1. %1#,1 )%/ /)1J-( , 1('%11
1.1.1 #% D% */1)' (( -1 , -1% 44/D 1. 1 /# 4 1. 1%= 1 (/
%1#,1 )%3)1.1%,3)#/ 1(=F:;:6H=
 
%-)%% D $(#1/ ' 1D) /-/1J-1% /)* 5/ )/= J-1%
%%%%/1.N#( 1'()1 )/) 4 ( 1')3,. /#% 4?-) 1%,(FMD)%1?
M%1H3)(()D 41.*1.)/)()4')3)//1(=F:;@H=Quality D%/3 /%1.( 4*1
)3 1. /% D 1.1. 1% 4)(% F-)/#,1 *-)$*1 / %(% ,%H elaboration D%
*%#/' 1.*)#1 )3/1 (/ originality 3(,1/ 1. # N#%% )3 1. /%F1.
%,) 4*1 J,3)#/ 1(=:;:6H=. 11(  ( 1'D%J,((13)((
/ *% )%D 1.D 4.1/^$(#%)3=993)N#( 1'=@73)()1 )/=9?3)) 4 ( 1'
F/ %P2),.@<<H=.)$(((  ( 1')31.%1.%,)%D%J,((1F,)(/E%
_M=99HD. ,.(()D/3)1.,(,#(1 ))3% 4(*%,)1)-%11.,1 $ 1')3
1.%D%=
Procedure
%1#/'D%,)/#,1/()1)'11.3 %1#1.)]% %1 1#1 )=* (D 1.4(
3)*1 ))#11.J- *1D%%11)-))()3)#/9;;;# $% 1'%1#/1% $ 1 4
1.*1)-1 , -1 %1#/'1 1(/Z1 $.  4=\1 , -1%, $/%*((-)1 ))3
1.,)#%,/ 13)1.  $)($*1=
-) $(-1 , -1%D%% 4/1) / $ /#(D)%11 )% N# 1))*D 1.
#-1)3 $-1 , -1%-%111 *=,.-1 , -1D)/)-%)(,)*-#1D 1.
/1 ,(./D /%)31D ,)3 4#1 )%%# 4,)% %1,'= .D)%11 )% D
4/1)-$1-1 , -1%3)*% 4,.)1./# 41.1%=
3)*/,)%1D%)1 /(,1) ,(('3)1.1%%4=.J- *1(
1%D%,)*-(1/ % 4()-1D 1. D)D%#% 4D3)*D 1.%,)/
13).17=?F(#%H=(1.)#4.-1 , -1%D(()D/1)#%1.D)D%1),,%%
// 1 )( %)#,% ) ,.)% 1) /) %)= (( -1 , -1% -%%/ 1. 111 ) ,.,% /
,)*-(1/1.N# /1%%#(1 4 )J,(#% )%=.1 J- *1(%1/$4
)37>* #1%FSDM@H=
Analytical Procedure
) /1 3' / 331 1'-% )3 -)*-1% #%/ ' -1 , -1% 1.  -)(* %)(#1 ) 1.
,.1)1,)$%1 )1%, -1%D /#,1 $(',)//'1.3 %1/(%1#1.)#% 41(%=1
9=?=  1 (,)/%,.*D%%1( %./,)%%#((')1.% %)3--)J *1(':;U)3
 :
1. %*-(=  %4*1% 1D 1% D %)($/ 1.)#4. / %,#%% ) / *#1#(
4*11)%#,)% %1,'/,,#,' 1.,)/ 4-),%%=#%N#1(')1.J-1%
/-/1(' ,)// 1. * / )3 1. %*-( /)* 5/ )/= . 11
(  ( 1'D%J,((1FD 4.1/^M=9H=
) /1 3'/ 33,%-)*-1#%1D-1 , -1%D 1.()D/. 4.,1 $
-)(*%)($ 4-3)*,1.-%1%1#/'*-()'/1D)%1-,(#%1(4) 1.*
:9F,.1(=:;;6L. #1(=:;;H=. %#1)*1 ,,(#%1 4-),/#D%%/
)1.N#( 1'()1 )/) 4 ( 1')31.%)(#1 )%/%#(1/1D)D((%-1/
,(#%1%FM% (.)#11M=<;HD 1.NM6?()D-3)* 4/NM7:. 4.-3)* 4%1#/1%=.
. 4.-3)* 4%1#/1%%,)/*/('. 4.N#( 1' tF<?HM9=66 p `=;; d M=@?
()1 )tF<?HM;=9;p `=;;d M:=?;/) 4 ( 1'tF<?HM9=?;p `=;;d M=@<=)
%#1.$( / 1')31.%#(1%(11 $*1.)/)3%-1 )D%(%)-3)*/#% 4
*/ %-( 1%/)1.$4,1 $-)(*%)($ 4-3)*,%#(1 4 (*)%1
/1 ,(%-1 )a:F<<HM>:=7>p`=;; VM=@;=)D$4 $1., 1 , %*%%), 1/
D 1.1.1 3 , (1#)3*/ %-( 1%F,((#*1(=:;;:H1.,(#%1**%. - 1.
-%1%1#/' %%/)1.1D)%1-,(#%1*1.)/D. ,.-)$ /%*)1#(%-1 )
)3-1 , -1%F3)(()D 4,)**/1 )%'1*1(=:;::L,)*-(%)-),/# 
#P1:;:7H=
 (('1)('5 1.-)(*%)($ 4-),%%1.()4/1 ,(#/ 41.-1 , -1
#*/-)*-1%N#,%D1 $/3)*1(%=1 9=?=#%N#1-),%%*  4D%
,)/#,1/ )>=7#% 41..# %1 ,%* (4) 1.*F1,1 $1D# %1 ,%
 L ./1 1 (= :;<H= )% %11 D 1. -$ )#% -),%%*  4 %1#/ % F% =4=
4(*P1:;:L1*1(=:;::H1.1.%.)(/3N#,'D%%11)=;L
(41.)())-% 1.%.)(/ 1) =@;L (41.1D)())-% 1.%.)(/ 1) =@;L / (1 $1)%1
1.%.)(/1)=;?=. / 4#(D%%11)=;L,)/ 1 )(/-/, %F).E%^H1)=?;L
//, % ).# %1 ,F%,)H1)=9;=.)('/ 33,3)*-$ )#%%1#/ %D%%11 4
1.,#%(/-/,'1)5)=.,#%(/-/,'%1 *1%1./ ,1 )( 1')3 / $ /#(
%1-%,)% / 4.)D)313)(()D%/3)(()D%= 4./-/,' / ,1%1.1
,)% %11('3)(()D%/$3)(()D%FR I1%1(=:;;>H=)D$. 4./-/,'
%,)1'1)1.1#)3,1 $-)(*%)($ 4D. 11 )F=4=%N#,%( b
 7
b b H % 3#/*1( 1.)1 ,( %%#*-1 )3) %#,,%%3#( -)(* %)(#1 ) F%
#*3)/1(=@@:;@H=
Results
.%#(1%%,1 )3 %1 /1 3 %1. / $ /#(-)*-1%#%/ ,)$%1 )%D 1..1
FYH=J1 1$(#1%1.#%3#(%%)3,.-)*-1'J*  4 1%%%), 1 )D 1.)$((
,1 $-)(*%)($ 4-3)*,FY:H= (('-),%%*  4 %*-()'/1)J* 
1./ 33,% )$((-)*-1 4 %114 %1D. 4./()D-3)* 4 / $ /#(%
FY7H=
Prompts: Useful, Not-So-Useful, and Counterproductive for Creative Problem-Solving
(,)1 %( %1)31. -)*-11'-%1.1-1 , -1%#%/1) %1#,1.1 1. 
1%%)(#1 )=(('FH-1 , -1%-)*-1/.11)41D /%F-)*-1
%6HF:H1.'#%/.1% 3)*1 )%)#,F?H/F7H 1.1.'%/
.11)-)$ /3/,)1.'-)$ //3/,)1. /%41/F%>@H=
 6
Table 1
Descriptive statistics and intercorrelations between prompts and creative problem-solving performance, qualitative description and
examples of individual prompts.
ID Variable Min Max M SD
Correlation
with CPS Description Example
)*-1,)#1 9 7=6; :=>> =6;  #*)3-)*-1% ),)$%1 )= cd
R)/,)#1 7 :6 6;=?> 7<=>@ =7@  $((D)/,)#1)3-)*-1%#%/ )
,)$%1 )=
cd
List of prompts
 Idea generation
F%/)1%
%1#,1 )H
; ;=<? ;=67 =;@ .  1 (-)*-1-.% 4 3)*1 )
3)*1% %1#,1 )=
Z/1),)*#-D 1.1. /%1) *-)$)/ '
-(#%. 11)'=\
: Idea generation
F%-, 3 ,(('3)
,1 $ 1'H
; ? ;=7@ ;=97 =79  -, 3 , %1#,1 )3) /41 )
1)41.D 1.%-, 3 ,$(#1 ), 1 
F) 4 ( 1'/8)#%3#(%%H=
ZR.')#1'1),)% /1.- *'-#-)%)3-(#%.1)'%
,#//( 4,#1%%1,=1'1),)*#-D 1.%)*1. 41.1
11(D (()#1-3)*4)D 1.%)*1. 4# N#=\
7 Idea generation
F%/)-$ )#%
%DH
; 7 ;=9 ;=>> =9 1 , -1%%(,1/) /3)*-$ )#%
%D%/%/.11)41*)
/%%/)1. %(,1 )=
Z')#()1)1.1. /-) 1[\
6 )$ / 4// 1 )(
,)1J1
; 7 ;=77 ;=>: =6<  )$ / 4// 1 )(,)1J1= ZE*)1( * 1/ D.11)/)D 1.1.#'#11.
/I#%1*1%%.)#(/%#,.1.1 1D)#(/%1 ((* #'=\
? 3)*1 )41. 4 ; ? ;=6@ =; =:? e % 4.1% 3)*1 )%)#,
F% * (('1)))4(%,.)R  -/ H=
ZR.1 /)3$41(%/) 1%1[\ZR.1 /%
()) 43) 1)'%1)/'[\
> FeedbackF4(H ; > ;=6; ;=@9 =: /,D 1.)#1'%-, 3 ,/1 (%= Z/)E1(  1=\
< FeedbackF *-)$
,1 $ 1'H
; : ;=:> ;=?< =6?  /,%-, 3 ,(('% 4.11)
*-)$) 4 ( 1')#%3#(%%)3 1%
%D%=
Z(  1=#1D11.#'1)/)%)*1. 4% /%$ /)
-)I,1 )=//)1.J, 1 4/) 4 ((*11.1D ((
()D((1. /%D'=\
9 % 43)feedback ; : ;=:? ;=?6 =;< % 4.11)-)$ /3/,) /%
1.1D,1/=
ZR. ,.)31.% /% %1.*)%1#%3#([\
@ 41 ; ? ;=:? ;=<7 =7 %% 41.Z41\#11) 1.,.1
,)%)(=
Note. -%1%1.*  *($(#)1 /' / $ /#(-1 , -1=J-%1%1.*J *($(#)1 /' / $ /#(-1 , -1=M-%1%1.*$(#
)1 /' / $ /#(-1 , -1%=%)]%r,)33 , 1 %-)1/3)%%), 1 )1D-)*-1,)#1/D)/,)#1/L-*]%.) %-)1/3)%%), 1 )1D
/ $ /#(-)*-1%/=-%1%*,1 $-)(*%)($ 4-3)*,3),.-1 , -1=
ep `=;?#11.%#(1 %)1%11 %1 ,(('% 4 3 ,14 $)(*)3) ,),1 )3)*#(1 -(,)*- %)%Lp `=;;
 ?
.%1)4%1-)% 1 $ %%), 1 )%D 1. ,1 $-)(*%)($ 4-3)*,D3)#/
1.-)*-11'-%= %11.*)%13 , (-)*-13)41 4 /%D%-)$ / 41%
%1#,1 )%1)41.D 1.$(#1 ) , 1 F:L r% M =79 p `=;;H=%%.)D(
-)*-1% 1.1 %-, 3 ,(('  */ 1 ,1 4 /% 1.1 D Z# N#\ ZJ,-1 )(\
ZJ1)/ '\)Z%#- % 4\D *)33,1 $1.-)*-1%% *-('-.% 41. 1%
%1#,1 )%FLr%M=;@pM=66H=('% %)31%, -1%$(/1.1D.-1 , -1%%/
3) /41 )%)((')1% %1#,1 )%1.'-.%/1. %1#,1 )% 1. )DD)/%
/)* 11/1.$(#1 ), 1 =) %1,-1 , -1%3 (/1),-1#1.11. *-)$/
1)'%.)#(/Z*)3#3),. (/1)-('D 1.\D. ((%).(- 4Z11(1) ,%1. 
%(%,)*-/1)1.4))#-\J-( , 1('*-.% 5 41./3))#1,)*%1))1.Z%
) 4 (%-)%% (\/Z%#%3#(%-)%% (\=
,)/ -)$ / 4 // 1 )( ,)1J1 F6H 3) / 41 ) J.  1/ 1. %1)4
(1 )%. -D 1.)$((-3)*,Fr% M=6<p `=;;H=%,%(1. %
-)*-1-)$ /%/1 (/,4)#/ 3)*1 )1.1)33%// 1 )(,#%3)41 4 /%=
 (('-)$ / 4%-, 3 ,3/,F<H1.1 */1 *-)$ 4 1.1.) 4 ( 1'
)#%3#(%%D%%1)4('%%), 1/D 1.-3)*,Fr% M=6?p `=;;H=%%.)D 
(  %-, 3 , 3/, )33//1 (/ J-(1 )% ) .)D 1)., 1. %D
%#%N#1 11 )%=. %,)1%1%D 1.4(3/,F>H%#,.%Z/)E1(  11' 1
4 =\
1. %#- % 4(' 1. D 1. -)*-1 1'-% 1.1 / / )1 %.)D % 4 3 ,1
(1 )%. -D 1.,1 $-)(*%)($ 4FH-3)*,= %1 /41 )%/)
-$ )#%%DF7HD.-1 , -1%%/.11)()1))%(,1/%D
/ /)1%.)D% 4 3 ,1%%), 1 )Fr% M =9 p M=:H=,)/% 43)3/,F9H
D. ,. $)($/ %1#,1 4.11)$(#1 1.1.) 4 ( 1')#%3#(%%)3-$ )#%('
41/%D% / 1)%(,1 1. / 1,)% //*)%1 -)* % 4 (%) / /)1 %.)D
% 4 3 ,1(1 )%. -Fr%M=;<pM=?7H= (('-%% 41.Z41\#11)F@H/ /)1
-)$ /// 1 )(3 1%3)-3)*,Fr%M=7pM=:?H=
,)1%11)#%3#(F/)1%)#%3#(H%114 %1.)('41 $%%), 1 )D%
3)#/1D-3)*,/1.-)*-1% 4.11)%$% 3)*1 )
%)#,F?L r% M =:? p M=;:>H=. 3)*1 )41. 4%114 % #%/'-1 , -1%
 >
D% * (1)-3)* 4))4(%,.),)%#(1 4R  -/ -4%=))$%D ((
/1 (/ (1 1. -)(*%)($ 4 -),%% %%), 1/ D 1. 3)*1 ) 41. 4 D%
N#( 11 $('/ 3313)*1.-),%%%#%/'-1 , -1%*-()' 4*)33,1 $-)*-1%
1. ,)$%1 )%=
Differences between High- and Low-Performing Individuals
%,% (:. 4.-3)* 4-1 , -1%#%/% 4 3 ,1('*)-)*-1% 1. 
,)$%1 )%D 1..1tF<?HM7=?>p`=;;dM;=9:,)*-/1)1. ()D-3)* 4
,)#1-1%=// 1 )(('1.-)*-1%#%/'. 4.-3)* 4-1 , -1%D% 4 3 ,1('
()4tF<?HM7=:?p`=;;dM;=9:,)*-/1)1.)%#%/'()D-3)* 4-1 , -1%=
,)*- %))3 / $ /#(-)*-1%%.)D/1.1. 4.-3)* 4-1 , -1%D*)
( ('1)*-()' /41 )%-, 3 ,((' * 43),1 $)#1,)*%F:Ha:F<<HM:;=@@
p `=;;V M=?:-)$ / 4// 1 )(,)1J1F6Ha:F<<HM:;=@@ p `=;; V M=?:/
)33 4%-, 3 ,3/,F<Ha:F<<HM?=> p `=;; V M=6?=1.,)1'1.()D
-3)* 44)#-#%/.1*))31% 3)*1 )%)#,F?Ha:F<<HM?=<p M
=;:7V M=:>=
Table 2
Comparison of prompts used in high- and low-performing groups.
ID Variable Low CPS (N = 45) High CPS (N = 32) Comparison
M (SD) M (SD) ).E%d
)*-1,)#1 :=?>F=@H 6=?@F7=H ;=9: 
R)/,)#1 :9=?>F:?=6H ?<=66F6?=>6H ;=9: 
,,#,FUH ,,#,FUH *E%V
 Idea generationF%/)1% %1#,1 )H <=U 9=7U =:
: Idea generationF%-, 3 ,(('3),1 $ 1'H >=<U ?7=U =?: 
7 Idea generationF%/)-$ )#%%DH 6=6U :=?U =?
6 )$ / 4// 1 )(,)1J1 >=<U ?7=U =?: 
? 3)*1 )41. 4 7?=>U :=?U =:> e
> FeedbackF4(H <=9U 7=7U =>
< FeedbackF *-)$,1 $ 1'H 6=6U 6;=>U =6? 
9 % 43)feedback <=9U :=@U =;?
@ 41 ?=>U 9=9U =;6
Note= )D-%1% -1 , -1%D 1.()D ,1 $-)(*%)($ 4-3)*,L . 4.-%1% -1 , -1%
D 1.. 4.,1 $-)(*%)($ 4-3)*,=,,#,FUH-%1%1.-,14)3-1 , -1% 4)#-1.1
#%/4 $-)*-11(%1),=
ep`=;?#11.%#(1 %)1%11 %1 ,(('% 4 3 ,14 $)(*)3) ,),1 )3)*#(1 -(,)*- %)%L p`
 <
=;;
. / ,1('3)(()D% 4-.%  4# %.)D $ %#( 51 )% )3 1. ,1 $ -)(*%)($ 4
-),%% ()D / . 4.-3)* 4 -1 , -1%= % *-.% 5/ '1* 1 (= F:;::H
1-1 4-),%%*  44-.%N# %,#1 )/#1)$ )#%1.%.)(/, 1 =( D 1.
-$ )#%%,.F%4(*P1:;:L1*1(=:;::H 4#$ %#( 5%
%N#,%),,# 41(%1;U)31.1 *=
Figure 1
Visualization of the creative problem-solving process of a) high- and b) low-performing
individuals.
 9
Note. . / ,1('3)(()D% 4-.% $ %#(('-%1 1. %N#1 (3()D)3 -)*-1% D 1. 
-),%%%.)D 41.J,1#*)3 %1,% ,.,1 $ 1'/ ,1('3)(()D 4)1.%/
)1.$1()4/1=.#*/1. ,%%)31.)D%-%1 1.3N#,' D 1.
D. ,.),1 $ 1'3)(()D%)1.=
% ,  % 1. -),%% 1. . 4.-3)* 4 4)#- )33% $ )#% D'% D. ,.
-1 , -1% -)*-1/ .1 3) ,1 $ -)(*%)($ 4=  1. ()4%1 %N#,
-1 , -1%41 /%FL:H-)$ /%-, 3 ,3/,F<H//// 1 )(,)1J1
F6H/,),(#/D 1.4(3/,F>H=))$ 1 % *-)111))11.1. 4.
-3)*% 1. %%, ),)**# ,1/D 1..1 11 $('=.D())-%1D
/41 )3),1 $ 1'F:H/-)$ / 4// 1 )(,)1J1F6H/%#%N#1('
1D -)$ / 4 ,)1J1 F6H / -)$ / 4 3/, F>H=  )1. D)/% 1.
,)$%1 )%1. %%, )-%1 / ,1 )( 1,1 )1D.#*#%%/
.1=
,)1%11.-),%%1.()D-3)* 44)#-*)%1(',)% %1%)3% , /
41 )FH) 3)*1 )41. 4F?H=3)*1 )41. 4 %#%/% * (('1)
3)*1 )%,.)))4()R  -/ L -1 , -1%%/N#%1 )% 1.1/ / )1 3(,1
%D%41/ -$ )#%%1-%=
(7$(#1% 1.3 1%%)3)1.-),%%*)/(%= 1%%-%1%1./41)
D. ,. -),%% *)/( ,,#1(' 3(,1% 1. ,1#( .$ ) ,)// 1. $1 ()4%=
-, 3 ,((' 1*%#%.)DD((1.%N#,%)3,1 $ 1 % 1.$1()4%,-)/#,/
'1.-),%%*)/(=%,%1.-),%%*)/(* /3)*/1 1.()D4)#-
J-( %96U)3$1% 1.()D4)#-/67U 1.. 4.4)#-=)$%('1.
*)/(* /3)*/1 1.. 4.4)#-J-( %<:U)3$1% 1.. 4.4)#-/
69U 1. ()D-3)* 4 4)#-= . )$(- 1D 4)#-% % #/%1/( ,#%
-1 , -1% )1. 4)#-% *-()'/ 3 $ )#1 )3   -)*-1% % * ((' F=4= % 4 3)
3/,9)41@H=)D$1.%#%N#1%11 %1 ,(('% %%.)D/1.11.
-),%%% ()D/. 4.-3)* 44)#-% //% 4 3 ,1('/ %1 ,1 $a:F<<HM9=?:
pM=;;6=
Table 3
Fitness of process models in low- and high-performing groups of participants.
 @
)#-
)D  4.
),%%*)/(F)DH 96U 67U
),%%*)/(F 4.H 69U <:U
Note. )D-%1% ()D,1 $-)(*%)($ 4 -3)*,L. 4. -%1%. 4.,1 $ -)(*%)($ 4
-3)*,=
%3 (%1-D,)%1#,1/1. /( 5/-)*-1 4%N#,%/)1.-),%%)3
1.. 4.-3)* 44)#- 1.3)*)31 1=. %%N#,%11/D 1. /41 )
,(#/ 4 1.$(#1 ) , 1  F:H-)$ / 4 %-, 3 , 3/,F<H // 4 // 1 )(
,)1J1F6H/,),(#/ 4D 1.4(3/,F>H=. % /( 5/%N#,,,)#1/
3) 7@U )3 1. $1 ()4% 1. . 4.-3)* 4 4)#- / @U )3 $1 ()4% 1. ()D
-3)* 44)#-=.%3 / 4% / ,11.1. 4.-3)* 4 / $ /#(%D*)( ('1)
44%1#,1#// 11 $-)*-1 4%N#,%%#44%1 41.11.%-11%*'
*-)113)33,1 $,1 $-)(*%)($ 4D 1.41 $=
) ,),(#/ . 4. ,1 $ -)(*%)($ 4 -3)*, D% %%), 1/ D 1. *)
3N#1/()4-)*-1%1.1-)$ //$(#1 ), 1 D.41 4 /%/1 (/
,)1J1/-, %3/,=#1.*). 4.-3)*% 1,1/D 1.41 $ 
*#1#(/ 11 $*D. (()D-3)*%#%/1.41 $1))(*)3N#1('
%) 11 $%)#,3) 3)*1 )41. 4=
Discussion
.4)()31.-%1%1#/'D%1)FH /1 3'1.-)*-1%1.1# $% 1'%1#/1%#%D.
%)($ 4,)*-(J))#1 1%FHJ* 1.-)*-1%%%), 1/D 1.. 4./()D
-)(*%)($ 4 -3)*, / F,H J-() 1. -)*-1 4 -),%% . 4. / ()D
-3)* 4 / $ /#(%=
#3 / 4%$(/1D)/ %1 ,14)#-%)3%1#/1%D.)/ 33/)1. 1.,)*-(J 1'
)31. -)*-1%/ 1.,1 $ 1')31. %)(#1 )%=.)%D 1.. 4.,1 $-)(*
%)($ 4-3)*,,)% %11('*-()'/-)*-1%1.1 ,)-)1/,($(#1 ), 1 
D.41 4 /%%#,. %Z1 /% 3) -)/#,11.1*#%1 *11. , 1  )3
) 4 ( 1'3%  ( 1'/*1 ( 1'\)33/ ,.,)1J1#(/1 (%%#,.%Z*4 1.1
')#  1)' /% 4 1%/ D 1. *-)$ 4 -)/#,1=  % (( D )D )#1 1.
% 1#1 )f\/-)$ //%-, 3 ,3/,3) 11 $ *-)$*1%#,.%Z. % //)%
 :;
)1*11.) 4 ( 1', 1 )L1'* 4 1*)#,)$1 )(\=,)1%11.)%D 1.
()D-3)*,1//1)#%4(3/,-)*-1%%#,.%Z/)E1(  1\)% *-('
% 4.11)%(3$(#1 1%%-)%%% ZR. ,.)3')# /% %1.%1[\= (('
#% 4.1*('% 3)*1 )%)#,% 4N#%1 )%( ZR.1/) 1%1[\
D%(%)%%), 1/D 1.()D-)(*%)($ 4-3)*,=
#1.*)1.-%1%1#/'%.)D/1.11. 1,1 )%)3. 4.-3)* 4%1#/1%
D)$((()4/,)1 /*)-)*-1%=#%N#1-),%%*  4)3,)$%1 )%
1D%1#/1%/.13)#/1.11. 1,1 )%1'(1D%1#/1%/41 $
-('/% 4 3 ,1)(3)33 , 1,1 $-)(*%)($ 4= 4.-3)* 4 / $ /#(%
1// 1) 44 11 $ -)*-1 4 %N#,%  4 ( 4/ D 1. 1.)1 ,(
,),-1#( 51 )% )3 .' / .#*1 3 , ( 1(( 4, F3 1 (= :;:H / .' /
.#* 4#(1 ) F)( :;::H=  1.% -1 , -1% # (1 #-)1. E% %-)%%
1.)#4.,)1 #)#%3/,())-%D 1.,. 11 )., 41.N#( 1')31.%)(#1 )=.
1## (1).#* -#1(()D 4)1.1.#%/1.1),. $*)1)41.
1. 1.,)#(/()F((*1(=:;@H=#3 / 4%( 4D 1.1.)%)34#'1(=
F:;:6HD.)3)#/1.11.Z1#,1#//-1 $ 1'\%1#/1%-)*-1//3 /41/
1J1 * ) 4 )# . 4. -3)*% D.) -1/('-)$ // /1 (/ ,)1J1 / %-, 3 ,
3/,= )$%(' 1.  Z%1#,1#/ 1*( \ %1#/1% - * (' ,)-'-%1/ 
)#1-#1%% * ( 1) )# ()D-3)* 4 %1#/1% D.) #%/ .1 *(' % )D'
3)*1 )%)#,=. % %#44%1%1.1/ ()4 ,3(,1 $--),.g,.,1 5/',(
4# /, 1)  / ,)1 #)#% $ % )g' (/% 11 )#1,)*% 1. (  #J* /
/)-1 ))3%-)%%=
#1.*) 1. 1, /1 )# %1#/' / ,1/ 1.1 . 4.-3)* 4 -1 , -1%
44/ *1,)4 1 $*) 1) 4)3 / $ /#(%-)%%F=4=Z%1. %) 4 (/8)#%3#(
)#4.[  -),/[\H / %#%N#1(' 4#(1/ 1. -),%% ' -)$ / 4 // 1 )(
,)1J1 ) 3/,= .% 3 / 4% ( 4 D 1. ,1 ,),-1#( 51 )% )3 1. )( )3
*1,)4 1 ),1 $ 1'F/P#/:;:6L=P:;:7H/,1 $
-)(*%)($ 4 F4)$ P  13(/ :;?L 2=  P  :;:7H= % %#44%1/ '
R F@@?:;;H33 , 1( 4-),%%N# %,)%11*1,)4 1 $*) 1) 4
D.%1#/1%,)*-,.%1-)31. -)4%%4 %1%1)3-/1* /%1//%=
1. %%1#/'1.$(#1 ), 1 D) 4 ( 1'/#%3#(%%* 41.1%1#/1%./1)
 :
%%%% .141/ %-)%% 4 %1 1.  -%)( %1//% 3) 1.% / *% )%=
#1#%1#/ %%.)#(/1.3)J* 1.%-, 3 ,%1//%%1#/1%--('D.$(#1 4
41/ /%=-1 ,#(%,.%%.)#(/ $%1 411., 1 1.1%1#/1%#%1)
/1* D.1.41/)#1,)* %Z4))/)#4.\1)-),/D 1.=
High-Performing StudentsGenerate Ideas with Constraints
. -%1 %1#/' %.)D/ 1.1 1. *)%1 33 , 1 -)*-1% 3) / 41 ) ,(#// 1.
$(#1 ) , 1  )3 1. 3 ( )#1,)*=  4.-3)* 4 %1#/1% D *) ( (' 1) %
.11)41 /%1.1./1)Z) 4 (\)Z#%3#(\D 1. 1.,)1J1)31.1%
D. ( ()D-3)* 4 %1#/1% )(' %/ .1 1) 41 /% D 1.)#1 ' 3#1.
%-, 3 ,1 )%= #%N#1(' . 4.-3)* 4 %1#/1% D ( 1) -)$ / 1. %-, 3 ,
3/,1.14# //1. /41 )-),%%=
% %#,. 1. -%1 %1#/' %#44%1% 1.1 -)$ / 4.1 D 1. $(#1 ), 1 
/ ,1('/# 4 /41 )()4D 1.%-, 3 ,3/,1.)#4.)#11.-),%%*'%$
% * (3#,1 )1)-)$ / 4,)%1 1%3).#*,1 $ 1'=,1 $ 1'%,.,)%1 1%
%%%%1 ((*1%1.1,%1 *#(1,1 $1.  41.1. .  1 1F#4.1
)*-:;>H=)%1 1%-)$ /3*D)D 1. D. ,. / $ /#(%*#%1)-1-#%. 4
1.*1)J-())$(%)(#1 )%/1. )#1% /1.)J=(1.)#4.1.3(#,'F#*)3 /%
41/H % /#,/ 1. /%  . 4. ) 4 ( 1'= * ((' )*- / F:;::H
/%, ,)%1 1%%*,. %*%1.13),#%,1 $33)1%// ,1 1.-)(*%)($ 4
-),%%= ' %11 4 ,( )#/ % / ( * 11 )% ,)%1 1% ,)#4 / $ /#(% 1)
)$1 D 1.  /3 / -*1% (/ 4 1) *) 33,1 $ / ) 4 ( )#1,)*%=
#1.*)/ )%1(=F:;9H%.)D/1.1,)%1 / /41 )(%)(/%1)*)
,,#1 /$(#1 )=,. %*%%%), 1/D 1.1.%),((/,)%1 1% 4N#%1 )%
*'4# /1)-)/#,*) )$1 $%-)%%F5)$%'1(=:;:6H=.-%1%1#/'
%#44%1% 1.1 $(#1 ) , 1  / %-, 3 , 3/, ,)#(/ %$ % ,)%1 1% 3) /
41 )D 1.41 $=)#/%1/1.33,1 $%%)31. %--),. 1 %%%1 (1)
#/%1/.)D(4(4#4*)/(%,1#(('41 /%=
Providing Generative AI with Rich Context for More Creative Outcomes
 ::
#3 / 4% / ,11.1. 4.-3)* 4%1#/1%D.)%)#4.111%)(#1 )%3)*.1
1// 1) -)$ / *) ()1/ ,)1J1 1.  -)*-1%= ) %,. %#44%1% 1.1
/1 (/ -)*-1% *-)$ -3)*, ' *J * 5 4 )D(/4 #1 ( 51 ) / ., 4
,)1J1#( 3)*1 )1 $(F)D1(=:;:;H=. %33,1,J-( /'1.D'
(4 (4#4 *)/(% -),%% 3)*1 )= .% *)/(%  1 / ) $%1 / / $%
/1%1%*// 4J1% $)D(/4,)%%$ )#%/)* %=R.4 $*)/1 (/
-)*-11.*)/(,1 $/( 4($1 3)*1 )*)33,1 $('(/ 41)*)
-, % / ,)1J1#((' --)- 1 %-)%%= . % -),%% % 3, ( 11/ ' %(3111 )
*,. %*%D. ,.(()D1.*)/(1)D 4.1.% 4 3 ,,)3,.1)(1 $1))1.%
/ ,-1# #,/ (1 )%. -% D 1.  1. -)*-1 F+%D 1 (= :;<H= . %%#(1%
%-)%%1.1)1)(',,#1#1(%),)*-.% $%1.*)/(,1/$14)3
1.// 1 )(,)1J11)-)$ /*)#,/%D%=)D$%D/ %,#%%(11. %%*
*,. %* *' J-(  D.' % *-( 3)*1 )41. 4 %114 % ,  /1 *1( 1)
,1 $-)(*%)($ 4=R. (()1/-)*-1%,)#41.*)/(1)1 $/ $%/
($1/1 (%)$(')/)4 , -)*-1%*'(/ 1)%#-3 , ( %-)%% 1.1(,
) 4 ( 1'=
Why Does Information Gathering with ChatGPT Fail?
.)('-)*-1 1.-%1%1#/'%%), 1/D 1.()D-)(*%)($ 4-3)*,D%
3),#%/ ) 3)*1 ) 1 $( 3)* .1= . 3)(()D 4 J,-1 /*)%11% 1.
3)*1 )41. 4%114'*-()'/'()D-3)* 4 / $ /#(%O
- 4-)*-1OR.1/)1.# %1[
)(()D 4-)*-1OR./)1.# %( $[
 (-)*-1OR.'/),. (/( %1#33/1)'%[
.%1#/1/ /)1-)$ /$(#1 ), 1 F ==1.%D%%.)#(/%) 4 (%-)%% (
/#%3#(3) ,% 4%(%H,)1J1F ==')#,)%#(11,)*-'1.1%%1)
)#1-3)*1.,)*-1 1 )'/$()- 4D$% ))31)'H)3/,=%1/1.
.#*#%/$ %/1. )D1.%)(#1 )%%/)1.%D%-)$ //O4 $1.#'
%1)3$41(%# (/%*((3)%1.)*/*1.#'E%3#%)31%)1.1,. (/D ((
( 1),#//( 1=.%-%1 1.*)%1 ,)**) /%% * ( 1) 1.)%(/' ( %1/ '
 /41% F,)*- 2=  P  :;:6H= .1 41/ 1. *)%1 ,)**)
 :7
%D%,#%1.*)%1,)**)/1.#%(%1) 4 (%D%(%)1.*)%1-)(=
 ,(4(4#4*)/(%%% 4-) ( 1 %1)/ 331-)%% (,)1 #1 )%)34 $
1J1%/)1.-11%1.'.$(/3)*1  4/1F+%D 1(=:;<H1.'D ((
-)/#,1.*)%1,)**)%D%#(%%1.#%-)$ /%%-, 3 , ,)%1 1%%/ %,#%%/
)$=  )1. D)/% .1 D (( 411. *)%1,)**) %D% #(%% -)*-1/
)1.D %=
Reassessing The Role of AI-Generated Feedback
1.,)1J1)3-$ )#%%,.1.)('%#- % 43 / 4)31.-%1%1#/' %%%), 1/
D 1. 1. 33,1 )3 1. 3/, 41/ ' .1= )1. ()D / . 4.-3)* 4
-1 , -1%%/.13)3/,F=4=ZR.1 /')#.$41/ %1.%1[\H#1
1. %-)*-1D%)1,)(1/D 1.-3)*,=
)D$1.%$(%1#/ %%#44%1 41.13/,41/'41 $ %
-)* % 4$#F% 1(=:;:6L'1(=:;:6L 1 %%1(=:;:6H='1(=
F:;:6H3)#/1.1%1#/1%, $ 441 $3/, ,%/1. $ % )-3)*,
/1. 33,1 $/*)1 $1 )(%11% *-)$/=))$41 $ %113/,
/#,/J 1'/($1/3()D %1#/1%F. 1(=:;:6H=)D$%1(=F:;:6H
3)#/1.1%1#/1%, $ 4J-13/,*-()'/1. %3/,3)$ % )*))31
1. %1#/1% , $/ 41/ 3/, F99U1) ?@UH=$ 1.)#4. J-141/
3/,D%1/. 4. )$((N#( 1'/#%3#(%%,)*-/1)41/3/,
1.%1#/' %#44%1%1.1 41/3/,,%1 (( 3 , ( -1 ,#((' ,)1J1%
D.)3/,D)#(/)1.D %-)$ //=.1(P4 %, F:;:7H3)#/1.1
-1 , -1% D.) , $/ #1)*1/ (1 * 3/, )#1-3)*/ 1.)% , $ 4 )
3/,=))$1.'(%)3)#/*)1 $1 )(33,1%)33/,/-)-)%/1,.)()4'
%% %1/3/, 1$1 )%3)-)*)1 4,1 $ 1'=
.'/ 33,1D1.3/,1.3)*1 )/%1#/ %/1.-%1
%1#/' %1.1%1#/1%1.3)*, $/3/,%'%1*1 ,(('D. ( 1. %%1#/'1.'
N#%1/3/,/( 1('=#1#%,.%.)#(/ *1)$(#11.%'%1*1 ,)()3
41/3/,)((3 ()#1,)*%%#* 11/'%1#/1%=
 :6
Limitation%
%- 11. % 4.1%4 /3)*1. %%1#/'%$(( * 11 )%*#%1,)D(/4/= ,1. %
%1#/'3),#%/%-, 3 ,((')1.)()3-)*-1#% ,1 $-)(*%)($ 4 D/ /)1
J*  .)D / $ /#( / 33,% - ) .1 J- , -)(*%)($ 4 ( 1 %
,)4 1 $/*1,)4 1 $% ((%)*)1 $1 ) 1,1/D 1..1#%=#1#%,.
%.)#(/J-( , 1(' $%1 411.% 1,1 )%=
#1.*)1.#%)3-),%%*  41)('5-)(*%)($ 4-),%%%-%1%
%$(( * 11 )%=(1.)#4.1..# %1 ,%* (4) 1.* %33,1 $3)%*((%*-(% 5%
1'- ,( ,(%%))* %11 4% FR I1% 1 (= :;;>H 1. -),%% *)/(% * / 3)* %*((
%*-(% (D'%  % ,),% )#1 1. 4( 5 ( 1' )3 3 / 4% 1) (4 -)-#(1 )%
F4#'1(=:;:6H=))$) %./( 4 %, 1 ,(% ($1,1 $ 1 %, %1.
J1,1/*)/(/$'3N#1.$ )%*'(/1))$%1 *1/(1 )%. -%1D
,1 $ 1 %F1*1(=:;::H= ,1.1%, -1%D.#*,)//1.($()3/1 (1.1
D%--)- 13)1.-%1%1#/'* 4.1)1,,#1('3(,11.* ,))*,)%1#,1#
)31.-)(*%)($ 4-),%%%=// 1 )(('1.4#( 1')3,)//,1 $ 1 %* 4.1* %%
#,// 33,%%1#/1%114 %,)*-( ,1 41. 1-11 ))3%#(1%=.3)
3#1# %1#/ % %.)#(/  * 1) $( /1 1.-%1/ *)/(% ,)%% (4 / *) / $%
%*-(%=
 (('1. %%1#/'D%,)/#,1/#% 4.17=?--)J *1('% J*)1.%31 1%
(%=11.1 *1.4(#%%E#/%1/ 4)341 $D%%1 ((*4 4/
1.*)/(%1.*%($%N# /*)J-( , 1,3#(('%1#,1#/-)*-1 41)-)/#,. 4.
N#( 1' %-)%%=  , 1. D$% )% %#,. %6) ) (#/7=< )1 .$
1)/#,/% 4 3 ,1(' *-)$/,- ( 1 % (4#4*)/( 4 ,(#/ 411,)1J1#(
1-11 ),).,/ %1#,1 )3)(()D 4=.% *-)$*1%.$*/ 1,1 )%
*))#%11)#%)%(()D 4$(1 $('D)$4#-)*-1%1)' (/#%(
)#1-#1%= % %#(1 1. 1.%.)(/ 3) )1  4 3#,1 )( %-)%% .%  ()D/
-)11 ((',.4 41.1#)3-)*-1 4%114 %//3))#1 1%%=)D$1. %
%. 31 /)% )1 / *  %. 1. $(# )3 33,1 $ -)*-1 4 ,)*-(J ) %,. 1% $
,)1J1%=)$(%) 4*)/(%%#,.%)/)7)%$ ,%( Z/-%,.\-3)*
D(()('D.-)$ // D 1.J1% $ ,4)#/ 3)*1 )*#(1 %1-4)(% /()4
 :?
. 4.(',)1J1#( 5/-)*-1%=3,11.%*)/(%-1 ,#(('%% 1 $1)-)*-1N#( 1'
)31N#  4%#%11 (,)1J1/-)(*3* 41)/( $#%3#()#1,)*%=./-
1. 1% %-, ((' /)* %N#  4 ,)*-(J %'1.% % ) %#%1 / N# ' 1. *)
, 1 ,(1. %,)1J1,)*%=. %$)($ 4(/%,-%#44%1%1.1-)*-1 4%114 %*#%1
1 ()/)1)('1)1.1'-)31%#1(%)1)1.,- ( 1 %/J-,11 )%)3%-, 3 ,
*)/(%=R. (4(-#-)%1))(%( 6)*'-3)*D((D 1.(%%#%33)11.3#((
-)11 ( )3 %) 4 *)/(% % #(),/ )(' 1.)#4. 1.)#4.13#( ()1 / D((
%N#,/-)*-1 4= #1# %,. %.)#(/ 1.3) $%1 41 .)D#%%114 % /-1
,)%% / 331 *)/( 1'-% / D.1. %1#/1% ,  1 / 1) *1,. 1.  -)*-1 4
--),.1)1.*)/(/1.1%1./=.% % 4.1%D ((  %%1 ( 3) %#--)1 4
-)/#,1 $.#*S,)(()1 )  ,% 4('/ $%-)(*%)($ 4$ )*1%=
Future Directions
 . % *)%1 ,1 D) R  F:;:7H / %,#%%/ .)D $ )#% 1, /1 3(,1 ,)4 1 $
*1,)4 1 $)*)1 $1 )(-),%%%D.(%D)/D 1.,)*-#1 13,%=1. %
%% 3#1# %,. %.)#(/ J-() .)D 1.%-),%%% 3(,1/ 1, /1 D.
%1#/1%D)D 1. 41 $1))(%=.-%1%1#/'%.)D/1.1$ )#%-)*-11'-%
* 4.1.$3(,1/ 1.*1,)4 1 $*) 1) 4)4#(1 )F =='-)$ / 43/,
)*),)1J1#( 3)*1 )1).1H=)D$ 1* /#,(D.1%1//%)
, 1 %1#/1% #%/ 1) $(#1 .1E% %-)%%= /%1/ 4J,1('.)D %1#/1%
*1,)4 1 $(' *) 1)/ 1.  ,)$%1 )% D 1. .1 1) ,. $ /% / / *)
*-)11('( ()#1,)*D)#(/-)$ /$(#( % 4.1% 1)1. -)(*%)($ 4=.#%
3#1.%,. 1. % %//=)J*-( 1 %%'*-1)*1 ,1.1-)-(D) 4D 1.
41 $.$. 4.%(333 ,,'1)%)($1.-)(*F)'PT.4:;:7H#111.
%*1 *1. 1% 1%1 %1.%*%3)%1#/1%%)($ 41.1%D 1.)#141 $
F 1 (= :;:6H= . %  %% 1. N#%1 )O R.1 1, /1 ,)%-)/ 1) 1.% 1D)
*)1 $1 )(,)%1#,1% ,%/%(333 ,,'#1#,.4/1% 1%1[
) )/(' 3#1# %1#/ % %.)#(/ 141 ,)4 1 $ *1,)4 1 $ 33,1 $ /
*)1 $1 )( $ (% 1) ,)*- 1. -)(*%)($ 4 -),%% D 1. 41 $  1) 1.
-)(*%)($ 4-),%%D 1.)1.1))(%F5$/)1(=:;9H%#,.%/ 4 1(%,33)(/%/
-%)( 5//%.)/% F *1 (=:;:7L+ 3 1 (=:;:7H= )(( , P )(( ,
 :>
F:;:6H -/ ,1 1.1 41 $  1))(% *' #(1 *1(' -(, *) 1/ 1 )(  ( 4
$ )*1%#11. %D (()1-)%% (D 1.)#1-)-%,. 1)1.-)11 (/)D% /%
%%), 1/D 1.41 $1))(%=
 -)11 ( /)D% / % 1.1 .1 -%1% 1%(3 % . 4.(' ( ( %)#,
F))-:;:7HD. ,.*'(/%1#/1%1))$('1#%1 1%J-1 %/(') 1%4# /,
1.' / %,)#4 4 1.* 3)* J-() 4 (11 $ -%-,1 $% F 4 1(= :;:7H= . %
1/,'*'%-, (('.*3#(1)%1#/1%D.)*-()'1. 3)*1 )% 4%114 %
/1 3 / 1.-%1%1#/'/1)1.)%D.)/))1-)%%%%J-1/)* %-, 3 ,%1//%
3) / $(#1 ) FR  :;::H= .3) 1 % ,#, ( 3) 3#1# %,. )1 )(' 1)
$%1 41 1. 3 1% )3 41 $  1))(% #1 (%) 1) , 1 ,((' %%%% 1.  -)11 (
/D,%//$()-%114 %1)* 1 411.% %%%# 41.1%#,.1,.)()4 %1#('
., 1. /#,1 )( J- , D 1.)#1 #/*  4 , 1 ,( 1.  4 / /-/1
-)(*%)($ 4% ((%=
Educational Implications
.3 / 4%)31. %%1#/'.$%$( *-( ,1 )%3)( 4/ %1#,1 )(-,1 ,%
%11 4%D.%1#/1%#%41 $1))(%%#,.%.11)%)($,)*-(J-)(*%=
 %1)#%#(1%%#44%11.133,1 $,)(()1 )D 1.41 $N# %J-( , 11  4
-)*-1 /% 4= -, 3 ,((' %1#/1% %.)#(/ ( .)D1) -)$ / ,(,)%1 1%/
%1#,1 )%g%#,.%%-, 3' 4) 4 ( 1')#%3#(%%%,($(#1 ), 1 g1)4# /
1.  1)D/ . 4.N#( 1' )#1-#1%= . % % -1 ,#((' *-)11 4 $ - ) %,.
%#44%1 41.141/)#1-#1%1/1))$('% * (F)%. P#%:;:6LR4
P211:;:?H-)11 (('(/ 41)/$()-*1)3%),((/Z*)),#(1#%\)3)D(/4
F%% P),11:;:6H=
,)/)#%#(1%,#1 )/#,1)%1.111 4.1- * ('% 3)*1 )%)#,
*'(/1)(%%33,1 $-)(*%)($ 4=.Z))4(( \--),.g% 4)/3,1#(
N#%1 )%g, ' (/ -/ ,1( ) %#-3 , ( /%=  $ 1.1 41 $  % -) 1)
.((#, 1 4 ,),1 3)*1 ) F.(( 1 (= :;:6H 1,.% %.)#(/ ,)#4 Z,)
,1 )\ * /%1 D. %1#/1%44 D 1. 41 $  % ,)(()1) 1.1 N# %
%114 , -#1%/,)%11$(#1 )F31(=:;:7L1(=:;:?H=
 :<
 ((' 141 4 -)*-1 4 / 4#(1)' % ((% 1) ,(%%))* ,1 $ 1 % *' 3)%1
( 4,#(1#,1/ )3(,1 ) $ % )/J-()1 )=' ,)#4 4%1#/1% 1)
44D 1. .1 %  1,1 $ -1 1. 1. %11 , %)#, )3 3)*1 )
/#,1)%,-)*)1, 1 ,(1.  4/-)(*%)($ 4=. %--),..,%%1#/1%]
 ( 1'1)3 1.  /% 11 $('$(#141/%-)%%33,1 $('//$()-
*1,)4 1 $%114 %3)$ 41 4.#*,)(()1 )=
References
5$/)=#=P#/ ,=+=F:;9H=/%1/ 4/%) 4)#1(1 *
,)4 1 $33,1 $/*1,)4 1 $-),%%%1)3)%1%(34#(1 )D 1./$,/
( 41,.)()4 %===,.#P==F/%=HHandbook of self-regulation
of learning and performanceF://=--=:?6S:<;H=)#1(/48'()P, %)#-=
.11-%O88/) =)48;=67:68@<97?>@<;69<
,.=R5 42=P+)4(=F:;;6H=D)1-(#%13 %1$(#1 )=Arbeits-
Und Diskussionspapiere Des Lehrstuhls Für Soziologie, 2F:H=
*= =D%)=)#/=PIID =F:;:6H=$()- 4$(#1 $
I#/4*13)1 *)341 $1 3 , ( 1(( 4,=Assessment & Evaluation in
Higher Education. /$,)( -#( ,1 )=
.11-%O88/) =)48;=;9;8;:>;:@79=:;:6=:77?7:
/=P#/=F:;:6H=4 4)##%OJ-() 41.($(%)3
*1,)4 1 $,)1)( ,1 $ /1 )=Creativity Research Journal=/$,)( 
-#( ,1 )=.11-%O88/) =)48;=;9;8;6;;6@=:;:6=:76<<<;
)D==='/=# .=2-(=. D(=(1=
.'*=%1'=%((=4D(=1+)%%=2#4= 4.
=. (/=*%.=T 4(==R#=R 1====*)/ =F:;:;H=
4#4)/(%D.)1%=Neural Information Processing Systems, 33
9<<S@;=
%1(()=$ /%)=P2.((/ =F:;>H=1 $ 1'1,. N#%1).,)D(/4
1%3D 1. 4()($ 1#(1*% 1.,)1J1)3)D(/4 1% $1- %%=.
 :9
Journal of Technology Transfer, 42F:H:?7S:>>=.11-%O88/) =)48;=;;<8%;@>;>
@?;@;
./%)%%  TP=F:;:6H=1 3 , ( 1(( 4,%#--)1,1 $ 1' 
('/% 4-),%%%[International Journal of Architectural Computing=/$,)( 
-#( ,1 )=.11-%O88/) =)48;=<<86<9;<<:6:?6>7<
.(( =%,*-%=$)#h+=)I =5=,1=' =()D5
=) (#=P#1%,..(( =F:;:6H=((#, 1 )1%/3,,,#,')3
.1//3)%'%1*1 ,$ D%O)*-1 $('% %=Journal of Medical
Internet Research, 26?7>6=.11-%O88/) =)48;=:@>8?7>6
. ==.==P)#==F:;:6H= 4'-(' 4D 1.41 $O/% 4
/$(#1 ))3)(-(' 4/#,1 )(4*D 1.41 $%%,33)(/ 43)
%113/, 1,1 )=Journal of Research on Technology in Education=/$,
)( -#( ,1 )=.11-%O88/) =)48;=;9;8?7@?:7=:;:6=:779;9?
. #=4=.=R4=P %=F:;;H=)#%1/%,((,(#%1 4
(4) 1.*3)* J/1'-11 #1% (4/1%$ )*1=KDD '01:
Proceedings of the seventh ACM SIGKDD international conference on Knowledge
discovery and data mining==.11-%O88/) =)48;=6?8?;:?:=?;:?6@
))-=F:;:7H=J*  4%, ,/#,1 ) .1OJ-()1)'%1#/')3
41 $1 3 , ( 1(( 4,=Journal of Science Education and Technology7:F666S
6?:H=.11-%O88/) =)48;=;;<8%;@?>;:7;;7@'
4=,=.*=)((=P#%,.=F:;::';H=)D1)-)*-1[
--)1# 1 %/,.((4%)35)/3D%.)1( 43).#* 1,1 ) 
,1 $--( ,1 )%)341 $*)/(%=Generative AICHI: Generative AI and
Computer Human InteractioniR)%.)-1]::jD(% 1/11%=
.11-%O88/) =)48;=69??;8C $=::;@=;7@;
/.1(==P4 %, =F:;:7H=#1)*1//,/1 $ 1'O1.)()3
1,)4 1 $) 1) 4  $41.  4=Psychology of Aesthetics, Creativity,
and the Arts=/$,)( -#( ,1 )=.11-%O88/J=/) =)48;=;7<8,;;;;?@:
 :@
((*=(=k((=P * %1==F:;@H=' /1(( 4,=Business
& Information Systems Engineering,61>7<S>67=.11-%O88/) =)48;=;;<8%:?@@;@
;;?@?:
 11 %1='( 4=%%(=)-)(/= ='=PT. / =F:;:7H=
Future of Jobs Report 2023.R)(/,))* ,)#*=
 4= = 4=P-#/=F:;:7H=1#/1%]-,-1 )%)3#% 4.1
-.'% ,%,(%%%$ 1#(1#1)=International Journal of Educational Technology in Higher
Education, 201 ,(#*>7=.11-%O88/) =)48;=9>8%6:7@;:7;;676
)%. ==P#%==F:;:6H=1 $.,% / $ /#(,1 $ 1'#1/#,%
1. ,)((,1 $ / $% 1' )3 )$( ,)11= Science Advances, 10F:9H=
.11-%O88/) =)48;=:>8%, /$=/?:@;
4(*2=P1=F:;:H=('5 41*-)(/13)#/%1/ 41.( 4
-),%% /#,/ ' *1,)4 1 $ -)*-1%= Learning and Instruction, 72 ;:;?=
.11-%O88/) =)48;=;>8I=( %1#,=:;@=;?=;;:
(,%1 ='=%=2((==2k((=Pk((=F:;:6H=)1,.%
%-)1 [ $(#1 4 1. /1,1 ( 1' )3 41/ 1J1% *)4 %1#/1 %%'%=
Computers and Education: Artificial Intelligence, 6 1 ,( #* ;;:;@=
.11-%O88/) =)48;=;>8I=, =:;:6=;;:;@
($# += #1 = )/( = )1(( = /   %  == % 11. 
== = =T%  = F:;7H= 1 $ 1'%,1 )O 3 / 4% 3)*3 $ ,1 $ /)* %=
Frontiers in Psychology, 4<>=.11-%O88/) =)48;=779@83-%'4=:;7=;;<>
/D  = Q$(Q = P  (( = F:;9H= (34#(1 ) ,)4#(1 ) / %./
4#(1 ),)(()1 $( 4$ )*1%===,.#P==F/%=H
Handbook of self-regulation of learning and performanceF--=97S;>H=)#1(/48'()
P, %)#-=.11-%O88/) =)48;=67:68@<97?>@<;69>
1*=#**(=P1=F:;::H=% 4# %1 ,% 1)('5)(*
)($ 4 -),%%%= Journal of Learning Analytics, 9F:H >>S9>=
.11-%O88/) =)48;=9>;98I(=:;::=<7>7
 7;
#4.1)*-=F:;>H=1 $ 1'/,)%1 1O /%)13)%===2#3*P=
F/%=H Creativity and reason in cognitive development F--=:S::?H=* /4
 $% 1'%%=.11-%O88/) =)48;=;<8,)@<97@@6@>@=;
#2=P1=F:;:7H=$%1 41 4( 4-),%%%1.)#4.('% %)3$ 41 )
.$ ) #% 4 ()4 3 (%= Journal of Computing in Higher Education=
.11-%O88/) =)48;=;;<8%:?:9;:7;@7<:7
 1%#D =/=#R=P)*#=F:;:7H=)%.#*S,)(()1 )(/1)
*),1 $1[ %1.1 ,$(#1 ) )3.#**//41/. #-)1'=
Computers in Human Behavior 1 ,( #* 7@=
.11-%O88/) =)48;=;>8I=,.=:;::=;<?;:
%%==)$(2==P33 4==F:;H= Creative approaches to problem-
solving: A framework for innovation and changeF7//=H=4=
% = k31 = . =(,%1  =k(( = 2k(( = P ' = F:;:6H=
*-  %,. 1O)*- 41 $/J-1/,1)1#/1%]R 1 4O
% 4.1%3)*1#/1,.%=Psychologie in Erziehung und Unterricht, 71F:H9;S@:=
.11-O88/J=/) =)48;=:7<98-#:;:6=1;9/
Q$(Q=P/D =F:;7H= D3)1 %O 4#(1 4 ( 4 . Educational
Psychologist, 48FH:?S7@=.11-%O88/) =)48;=;9;8;;6>?:;=:;:=<69;;>
Q$(Q = (*4 = )), %  = P 2 %,. == F:;:H= 1,)4 1 ) 
)(()1 $  4=  = %% = %. * == )%h P == R % F/%.),
International handbook of computer-supported collaborative learning F--= :9S:@6H=
- 4=
Q$(Q=4#'=P/D =F:;:7H=#*/1 3 , ( 1(( 4,,)(()1 )3)
%), (('%./4#(1 )( 4=British Journal of Educational Technology, 54F?H
;?<S;<>=.11-%O88/) =)48;=8I1=77:?
)% *)$ ,=31.(=)=1=P *%=F:;:7H=--)1# 1 %)3
1 3 , ( 1(( 4, 3)%#--)1 4 ,)*-(J-)(*%)($ 4O / 4%3)* %,)- 4
$ D= Computers and Education: Artificial Intelligence, 4 1 ,( #* ;;79=
.11-%O88/) =)48;=;>8I=, =:;:7=;;79
 7
)%% = = F@@<H= %1#,1 )( /% 4 *)/(% 3) D((%1#,1#/ / %1#,1#/
-)(*%)($ 4( 4)#1,)*%=Educational Technology Research and Development,
45>?S@6=.11-%O88/) =)48;=;;<8;::@@>7
211( = * 2= = R ((%) =P #((1= = F:;9H= ,.%E -,-1 )% )3
,1 $ 1' 1. ,(%%))*= Creativity Research Journal, 30F:H >6S<=
.11-%O88/) =)48;=;9;8;6;;6@=:;9=66>?;7
21.D).(==F:;;:H=$ % ))3())*E%J))*'O$$ D=Theory Into Practice,
41F6H::S:9=.11-%O88/) =)48;=:;<8%?67;6:1 -6;6l:
5)$%'==5=P211==F:;:6H=.1)3,1 $ N# 'g)*N#%1 )
% 41)-)*-14  4=Journal of Creative Behavior=/$,)( -#( ,1 )=
.11-%O88/) =)48;=;;:8I),=><
 *= 1 = +  3 = =)$ , =  4. = )( = P
V$ W=F:;:7H=)D/)%1#/1%(D 1.(m1 *-%)( 5/%,33)(/%[ British
Journal of Educational Technology= /$, )(  -#( ,1 )=
.11-%O88/) =)48;=8I1=766
)5 W=PX1#(=F:;:7H=(#1#1)13,1#(O,)*-1 $('% %)3.1/
)1.,.1)1%]-)3 , ,'/) 4 ( 1' %, 1 3 ,D 1 43).#* 1 %= Future
Internet, 15F;H77>=.11-%O88/) =)48;=77@;83 ?;;77>
,((#*==T.4=,.2==P#,==F:;;:H=1.-,1 ,)3
/ ,.)1)* 51 ) )3 N#1 11 $ $ (%= Psychological Methods 7FH @S6;=
.11-%O88/) =)48;=;7<8;9:@9@J=<==@
./1=/) =P I%==F:;<H=# %1 ,*  4$*-/O 1,1 $
/1D/,)3)*,D* ==( %n=)-)(/=/( 4R=$
/(%1 = 2#* = 1(/ P = R% F/%=H Proceedings of the BPM Demo
Track and BPM Dissertation AwardF--=?H=R=)4=
11.D%= P +)(-== F:;:7H= ,/* ,%]-,-1 )% )3.141/ D 11
)#1-#1%O -,1 ,( --( ,1 ) )3 # 4]% * 11 ) *= Australasian Journal of
Educational Technology, 39F?H9:S;;=.11-%O88/) =)48;=6<6:8I1=99@>
 7:
/ )%2==1(==R11%==P#*3)/==F:;9H= * 4 %$'1. 4O
J*  4 1. )( )3 ,)%1 1% 1.)#4.)#1 1. ,1 $ -),%%= Psychology of
Aesthetics, Creativity, and the Arts, 12F6H6<S699=.11-%O88/) =)48;=;7<8,;;;;69
%% =P),11==F:;:6H=1 3 , ( 1(( 4,/ ((#% )%)3#/%1/ 4
%, 1 3 ,%,.= Nature, 6176@S?9=.11-%O88/) =)48;=;798%6?9>;:6;<6>;
'=%=,. ((= )D=R=1 ,.=),.=P(,%1 
=F:;:6H=% 4%1) 4$ /,%/3/, 1)1.,(%%))*O41/
3/, ,%%%,)/' %1#/1%]1J1 $ % )*)1 $1 )/-)% 1 $ *)1 )%=
Computers and Education: Artificial Intelligence, 6 1 ,( #* ;;@@=
.11-%O88/) =)48;=;>8I=, =:;:7=;;@@
)(=F:;::H=.,),-1)3.' /.#*4#(1 )OJ*-( 3' 4.)D1)%#--)1
')#4(%]%(34#(1/( 4= Computers and Education: Artificial Intelligence
31 ,(#*;;;<;=.11-%O88/) =)48;=;>8I=, =:;::=;;;<;
)(( ,=P)(( ,=F:;:6H=%1#,1)%% )$1)%O3#1#3),#%/--),.1)D
( 4)--)1# 1 %D 1.-)*-1%=SSRN=.11-%O88/J=/) =)48;=:7@8%%=69;:6>7
#*3)/ = = 1  = (( )11 = = F:;@H= 1 $ .  4 ),%%%O 4 4
)$1 $ 33)1%=  Oxford Research Encyclopedias= J3)/  $% 1' %%=
.11-%O88/) =)48;=;@78,3)8@<9;@;::69?=;7=<:
#*3)/==)('== 1m(*)=.(*==P)%==F@@H=
),%%('1 ,*)/(%)3,1 $,-, 1 %=Creativity Research Journal, 4F:H@S::=
.11-%O88/) =)48;=;9;8;6;;6@;@?7679;
4#'=)4=4=P#4C=F:;:6H=#*,)(()1 )-11% 
%% %1/ ,/* , D 1 4= Studies in Higher Education= /$, )(  -#( ,1 )=
.11-%O88/) =)48;=;9;8;7;<?;<@=:;:6=:7:7?@7
 I%1/==,.1)(/1 =P.) ==F:;@H=3)*1 )-),%% 4*)1 $1 )/
4)#-,1 $ 1'===#(#%P== I%1/F/%=H The Oxford handbook of group
creativity and innovation F--=9<S;:H=J3)/ $% 1'%%=
 77
)'=PT.4R=F:;:7H=J- *1($ /,)1.-)/#,1 $ 1'33,1%)341 $
1 3 , ( 1(( 4,= Science, 381F>>?6H 9<S@:=
.11-%O88/) =)48;=:>8%, ,=/.:?9>
=F:;@H=OECD Future of Education and Skills 2030: OECD learning Compass==
-=F:;:6H=OpenAI o1 System Card.-=
3=((* =I)1.=+%51n=2*-3 =,' R=P.%)=
F:;:H=% (( 4#-% (( 4/% (( 4O ,%3) .' / 1(( 4,= Morals &
Machines, 1F:H:67@=.11-%O88/) =)48;=?<<8:<6<?<6:;:::6
3=1'==2#3*==#1=P.%)=F:;:7H=1 $ 1' 1.4)3
41 $ = Nature Human Behaviour, 7FH 97>S979=
.11-%O88/) =)48;=;798%6?>:;:7;<?
$V(I=2o p =)*o$ p=*=5)$ =./ ==1(=F:;:?H=
 4./#,1 ) %1#/1%E -,-1 )%)3.1O 4()(%1#/' )3(' ,1 )%=
PLOS ONE, 20F:H;7?;=.11-%O88/) =)48;=7<8I)#(=-)=;7?;
D'=2=F:;;>H=)#-,1 $ 1'O#% ,(-3)*,/,)(()1 ). Psychology of
Music, 34F:H69S>?=.11-%O88;=<<8;7;?<7?>;>;>9?;
1 %% = 1 = .* = #5 = 1 = R4 = )) = %4 R=
R%,.#=P(%)==F:;:6H=)*- 41.N#( 1')3.#*/.1
3/, ) %1#/1%] D 1 4= Learning and Instruction, 91 1 ,( ;9@6=
.11-%O88/) =)48;=;>8I=( %1#,=:;:6=;9@6
..==+)===.1==.*===#4==PC#==F:;:7H=
,D 1.1.*,. %O%%%% 41.,- ( 1')341 $%)($ 4#1.1 ,
%%%%*1%= Australasian Journal of Educational Technology, 39F?H ?@S9=
.11-%O88/) =)48;=6<6:8I1=9@;:
)//== 44%==P#*3)/==F:;@H= %/ %*/ 1 ) 1 $
)(*)($ 4O 4 4  %% 1.)#4. ),%1 4= Creativity Research Journal,
31FH6=.11-%O88/) =)48;=;9;8;6;;6@=:;9=?7::>9
 76
33 4==('==P%%==F:;;9H=/%1/ 4 / $ /#(-)(*
%)($ 4 %1'(O ' 1)( 4 / --(' 4 ,1 $ -)(* %)($ 4= Learning and
Individual Differences, 18F6H7@;S6;=.11-%O88/) =)48;=;>8I=( / 3=:;;<==;;<
)*-=P=F:;::H=1 $ 1'3)*,)%1 1%O.)'/--( ,1 )%1)/#,1 )=
Thinking Skills and Creativity, 46 1 ,( #* ;96=
.11-%O88/) =)48;=;>8I=1%,=:;::=;96
 1/1 )%F:;?H=Transforming our world: The 2030 agenda for sustainable development=
 1/1 )%=
2=P=F:;:7H=)]11 1q1 $)4#(1 )/), (('./
4#(1 ) %1#/' )3 1. 1(%O 1 ,= Journal of Creative Behavior=
/$,)( -#( ,1 )=.11-%O88/) =)48;=;;:8I),=>:>
 2= P  = F:;:7H= )D , D *%# *1,)4 1 ) ,1 $ -)(*
%)($ 4[1// 51 ))31.%,(=Thinking Skills and Creativity, 49FH1 ,(
#*;76?=.11-%O88/) =)48;=;>8I=1%,=:;:7=;76?
 2= P  = F:;:6H= Z )D *' / % ) 4 (q\ 1 $ 1,)4 1 $
) 1) 4 / 4#(1 ) 2 /41 . (/= Thinking Skills and Creativity
1 ,(#*;?6=.11-%O88/) =)48;=;>8I=1%,=:;:6=;?6
 = P 2= F:;:7H= ) R / 1,)4 1 ) 3) 1 $ 1'[ ,%%'
)/ 1 )('% %)31 $1,)4 1 )= Psychology of Aesthetics, Creativity, and
Arts=/$,)( -#( ,1 )=.11-%O88/) =)48;=;7<8,;;;;>6<
=P2=F:;:6H=)%*1,)4 1 )*11,1 $-)(*%)($ 4[
* J/*1.)/%('% % )3 D 1 4= The Journal of Creative Behavior= /$, )( 
-#( ,1 )=.11-%O88/) =)48;=;;:8I),=>7;
=,.1)=#$%'=()$r+=$,.=)*=P2=
F:;:6H=.1*-)$%1 $)(*)($ 43)*,  $% 1'1#/1%O
 J- *1( 1#/'= Computers & Education 1 ,( #* ;?;7=
.11-%O88/) =)48;=;>8I=,)*-/#=:;:6=;?;7
 7?
=)*=#$%'= ,.1)=   +=$,. = 2*) ,)$=
2=F:;:?H=Z.1,** %1%=., *-)11 3)=\- %1* ,( 3%
/1,)4 1 $,,#,' 1#/1%E141 ))3.1)11 1),/* ,
R 1 4=British Journal of Educational Technology=.11-%O88/) =)48;=8I1=7?@
+  3 = )$ W =  =  * =  4. = 1 = V$ W = P
)(=F:;:7H=)D1)/% 4/$(#1-%)( 5/%,33)(/%3)%(34#(1/
( 4= Metacognition and Learning, 18F7H <97S9;= .11-%O88/) =)48;=;;<8%6;@
;:7;@7>'
+%D =.5= *= %5) 1=)%= )*5= =2 %s= P
)()%#. =F:;<H=111 ) %((')#/=Proceedings of the 31st International
Conference on Neural Information Processing Systems (NIPS'17)F--=>;;;S>;;H=#(
3)*1 )),%% 4'%1*%=.11-%O88/) =)48;=????87:@?:::=7:@?76@
R I1%====(%1$/R===P($%/ )%=2=F:;;>H= Process
mining with the Heuristics Miner algorithm=,. %,. $% 1 1 /.)$=
R4=P211=F:;:?H=R]/ 331D]1.%*O1 $.)*)4 1',)%%
%=arXiv=.11-%O88/) =)48;=69??;8C $=:?;=@7>
R ==F@@?H=.1/1 (% %(34#(1/( 4= Educational Psychologist, 30F6H
<7S9<=.11-%O88/) =)48;=:;<8%?7:>@9?-7;;6l:
R ==F:;;H=(34#(1/( 4$ D/3)**)/(%)3 3)*1 )-),%% 4=
= = T *** P = = ,.# F/%=H Self-regulated learning and academic
achievement: Theoretical perspectives F:/ / --= ?7S9@H= )#1(/4=
.11-%O88/) =)48;=67:68@<96;>;;7:
R ==F:;::H=)/( 4%(34#(1/( 4%(%/) 4( 4%, ,O)D
1, /1 / ( 4 ('1 ,% .(- /$()- % ((% 3) %(34#(1/ ( 4=
Metacognition and Learning, 17F7H<<7S<@=.11-%O88/) =)48;=;;<8%6;@;::;@7;?'
R ==F:;:7H=)(%3) 3)*1 ) 1,/1#%/1)*)/((34#(1/ 4=
+=2)$)$ ,=5$/)== %)P=31.(F/%=HAdvances in analytics
for learning and teachingF--=<?S@>H=.11-%O88/) =)48;=;;<8@<97;77;@@::l
 7>
R))==R4=#%1) =P #) 2=F:;:7H= /%1/ 4 4( %.% ) 4
4#4%1#/1%] / 41 )%114 % 3),1 $D 1 4D 1.1#((4#4
41 ) 1))(%= Journal of Educational Computing Research, 61F<H 6>669:=
.11-%O88/) =)48;=<<8;<7?>77:7<?@@@
R=I *=PD/=F:;:6H=3 1%/,.((4%)3,)(()1 )1D
%1#/1%/,)$%1 )(41 $1 3 , ( 1(( 4, -)4** 4( 4O
*-  ,( ,% %1#/'= Education Sciences, 14F6H 677=
.11-%O88/) =)48;=77@;8/#,%, 6;6;677
T /'==F:;:6H=. *-,1)341 $)%1#/144*1/1. ,%. 4.
/#,1 )=Journal of Information Technology, Cybersecurity, and Artificial Intelligence,
1FH7;S79=.11-%O88/) =)48;=<;<?8I 1, =:;:6=$= =;;6
T*3 %,#  = = R)4 = 1* = P 4 Y= F:;:7H= R.' ).' ,]1
-)*-1O)D)J-1%1'F/3 (H1)/% 4-)*-1%=Proceedings of the
2023 CHI Conference on Human Factors in Computing Systems (CHI '23)1 ,(#*
67< - ( :7S:9 :;:7 *#4 *' F--= :H=
.11-%O88/) =)48;=6?87?66?69=7?9799
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The paper presents the most comprehensive and large-scale global study to date on how higher education students perceived the use of ChatGPT in early 2024. With a sample of 23,218 students from 109 countries and territories, the study reveals that students primarily used ChatGPT for brainstorming, summarizing texts, and finding research articles, with a few using it for professional and creative writing. They found it useful for simplifying complex information and summarizing content, but less reliable for providing information and supporting classroom learning, though some considered its information clearer than that from peers and teachers. Moreover, students agreed on the need for AI regulations at all levels due to concerns about ChatGPT promoting cheating, plagiarism, and social isolation. However, they believed ChatGPT could potentially enhance their access to knowledge and improve their learning experience, study efficiency, and chances of achieving good grades. While ChatGPT was perceived as effective in potentially improving AI literacy, digital communication, and content creation skills, it was less useful for interpersonal communication, decision-making, numeracy, native language proficiency, and the development of critical thinking skills. Students also felt that ChatGPT would boost demand for AI-related skills and facilitate remote work without significantly impacting unemployment. Emotionally, students mostly felt positive using ChatGPT, with curiosity and calmness being the most common emotions. Further examinations reveal variations in students’ perceptions across different socio-demographic and geographic factors, with key factors influencing students’ use of ChatGPT also being identified. Higher education institutions’ managers and teachers may benefit from these findings while formulating the curricula and instructions/regulations for ChatGPT use, as well as when designing the teaching methods and assessment tools. Moreover, policymakers may also consider the findings when formulating strategies for secondary and higher education system development, especially in light of changing labor market needs and related digital skills development.
Preprint
Full-text available
Numerous powerful large language models (LLMs) are now available for use as writing support tools, idea generators, and beyond. Although these LLMs are marketed as helpful creative assistants, several works have shown that using an LLM as a creative partner results in a narrower set of creative outputs. However, these studies only consider the effects of interacting with a single LLM, begging the question of whether such narrowed creativity stems from using a particular LLM -- which arguably has a limited range of outputs -- or from using LLMs in general as creative assistants. To study this question, we elicit creative responses from humans and a broad set of LLMs using standardized creativity tests and compare the population-level diversity of responses. We find that LLM responses are much more similar to other LLM responses than human responses are to each other, even after controlling for response structure and other key variables. This finding of significant homogeneity in creative outputs across the LLMs we evaluate adds a new dimension to the ongoing conversation about creativity and LLMs. If today's LLMs behave similarly, using them as a creative partners -- regardless of the model used -- may drive all users towards a limited set of "creative" outputs.
Preprint
Full-text available
Recent studies have identified ChatGPT as an epistemic authority; however, no research has yet examined how epistemic beliefs and metacognitive accuracy impact actual work with ChatGPT-generated content, which is often prone to factual inaccuracies. Therefore, the present experimental study aimed to examine how university students integrate correct and incorrect information from expert-written and ChatGPT-generated articles when writing independently (N = 49) or with ChatGPT assistance (N = 49). Students working with ChatGPT-4o integrated more correct information from both expert-written (d = 0.64) and ChatGPT-generated articles (d = 0.95), but ChatGPT-assisted writing did not affect the amount of incorrect information sourced from the ChatGPT-generated article. Regardless of the condition, hierarchical regressions revealed that lower metacognitive bias was moderately associated with increased inclusion of correct information from the expert-written article (R2 = 12%). Conversely, a higher metacognitive bias (R2 = 10%) and epistemic beliefs (R2 = 12%) were moderately related to the inclusion of incorrect information from ChatGPT-generated articles. These findings suggest that while ChatGPT assistance enhances the integration of correct human- and AI-generated content, metacognitive skills remain essential to mitigate the risks of incorporating incorrect AI-generated information.
Article
Full-text available
The rapid adoption of Artificial Intelligence (AI) in higher education is reshaping students’ learning experiences, with tools such as ChatGPT, Grammarly, and Microsoft Copilot becoming integral to academic work. This study, informed by data from the Digital Education Council Global AI Student Survey 2024, examines the impact of AI on students, focusing on usage patterns, trust in AI-generated content, ethical awareness, and expectations for institutional support. Findings indicate that 86% of students use AI for various academic tasks, with a majority expressing concerns about trust, fairness, and over-reliance on AI. While students value AI’s benefits, only 5% are fully aware of institutional guidelines on AI use, and 72% desire more AI literacy courses, reflecting a significant need for comprehensive support in navigating AI responsibly. The study underscores the importance of clear ethical guidelines, faculty training, and student involvement in AI policy formation to foster responsible AI use and preserve academic integrity. These insights offer valuable guidance for educators and policymakers seeking to integrate AI ethically and effectively into higher education.
Article
Full-text available
As artificial intelligence and natural language processing methods rapidly develop, communication plays a pivotal role in everyday interactions. In this theoretical paper, we explore the overlap and commonalities between question-asking and prompt engineering. While seemingly distinct, these processes share a common foundation in essential skills like creativity, critical thinking, and cognitive flexibility. We contend that prompt engineering, the art of crafting cues for language models, and question-asking, the skill of formulating inquiries, form a symbiotic relationship. Delving into question complexity through Bloom's taxonomy and diverse types of questions, we propose strategies for not only efficient but also engaging prompt design. Our theoretical contribution emphasizes the dynamic role of creativity in both processes, offering intriguing perspectives on human-machine interactions and advancing our understanding of language models and communication skills.
Article
Full-text available
This study focuses on Generative Artificial Intelligence (AI) and its transformative impact on design ideation. Generative AI, recognized for its ability to produce a wide array of design alternatives, has become an important tool in design, reshaping traditional methodologies. It facilitates the generation of novel and diverse design forms, acting as a co-creator in the design process. This technology, through machine learning and pattern recognition, analyzes extensive design datasets, enabling the production of innovative solutions. The utilization of generative AI extends beyond replicating AI-provided solutions; it aids in developing and influencing novel concepts, thus fostering original design solutions. This aligns with the concept of ‘reflective practice’ in design, where designers iteratively refine concepts through a dialogue between thought and action. The study employed a quasi-experimental design with 40 design students, randomly assigned to two groups of 20 each. Conducted in two phases, each phase involved a distinct urban furniture design task. In Phase 1, Group A was provided with a text-to-image generating AI tool, while Group B was not. In Phase 2, both groups undertook a similar task without AI assistance. This design exercise allowed for examining the influence of AI on creativity and cognitive load. Design outcomes from both tasks were anonymized and evaluated by experienced professionals using the Creative Product Semantic Scale (CPSS), which measures Novelty, Resolution, and Elaboration and Synthesis. Additionally, the NASA Task Load Index (NASA TLX) questionnaire assessed cognitive load aspects such as mental demand and effort. Findings suggest that generative AI significantly influences the creative design process, enhancing the quality of design outcomes and reducing cognitive load. The AI group demonstrated better performance in both tasks, indicating the impact of AI tools on design skills. This study underscores the potential of AI tools in design education, balancing cognitive load management with creativity enhancement.
Article
Full-text available
Background Large language models (LLMs) have raised both interest and concern in the academic community. They offer the potential for automating literature search and synthesis for systematic reviews but raise concerns regarding their reliability, as the tendency to generate unsupported (hallucinated) content persist. Objective The aim of the study is to assess the performance of LLMs such as ChatGPT and Bard (subsequently rebranded Gemini) to produce references in the context of scientific writing. Methods The performance of ChatGPT and Bard in replicating the results of human-conducted systematic reviews was assessed. Using systematic reviews pertaining to shoulder rotator cuff pathology, these LLMs were tested by providing the same inclusion criteria and comparing the results with original systematic review references, serving as gold standards. The study used 3 key performance metrics: recall, precision, and F1-score, alongside the hallucination rate. Papers were considered “hallucinated” if any 2 of the following information were wrong: title, first author, or year of publication. Results In total, 11 systematic reviews across 4 fields yielded 33 prompts to LLMs (3 LLMs×11 reviews), with 471 references analyzed. Precision rates for GPT-3.5, GPT-4, and Bard were 9.4% (13/139), 13.4% (16/119), and 0% (0/104) respectively (P<.001). Recall rates were 11.9% (13/109) for GPT-3.5 and 13.7% (15/109) for GPT-4, with Bard failing to retrieve any relevant papers (P<.001). Hallucination rates stood at 39.6% (55/139) for GPT-3.5, 28.6% (34/119) for GPT-4, and 91.4% (95/104) for Bard (P<.001). Further analysis of nonhallucinated papers retrieved by GPT models revealed significant differences in identifying various criteria, such as randomized studies, participant criteria, and intervention criteria. The study also noted the geographical and open-access biases in the papers retrieved by the LLMs. Conclusions Given their current performance, it is not recommended for LLMs to be deployed as the primary or exclusive tool for conducting systematic reviews. Any references generated by such models warrant thorough validation by researchers. The high occurrence of hallucinations in LLMs highlights the necessity for refining their training and functionality before confidently using them for rigorous academic purposes.
Article
Full-text available
The present study examined the relationship between kindergarteners’ creative performance, global metacognitive monitoring (i.e., the ability to self-evaluate their overall creative performance) and metacognitive regulation (i.e., the ability to select their most original idea). Prior research established a positive correlation between metacognitive accuracy in global self-evaluation and creativity in adults, but little is known about how these relate to idea selection in younger populations. The study involved fifty 4- to 6-year-olds who completed two creativity tasks (an Unusual Uses Task and a Product Improvement Task) and selected their most original idea for each. The results showed a moderate correlation between the children's ability to select their most original ideas for the two tasks, which suggests that metacognitive skills are transferable within the domain of creativity already at this age. Additionally, there was a moderate-to-strong trend in children capable of accurately identifying original ideas and the ability to perform more creatively, which is in line with previous findings in adults. Furthermore, children who displayed higher metacognitive accuracy in global self-evaluation were also better at metacognitive regulation, i.e., they were better capable of selecting their most original idea. However, even the children who performed best in idea selection tended to overestimate their overall creative performance. The study provides insights into the early development of creative metacognitive monitoring and regulation, and suggests potential implications for fostering creativity in education.
Chapter
Much creative work takes place in groups or teams, but also individual creative efforts cannot be seen as separate from a social context. In recent decades, the questions “What makes groups and teams creative?” and “How is creativity shaped by the social context?” have therefore received increasing research attention. This book provides a comprehensive overview of this work and is organized into five sections. After an introductory section, a second section (individuals and groups) discusses issues of group composition, diversity, newcomers, and conflict. The third section, on basic processes and theoretical approaches, discusses cognitive, motivational, and affective processes in groups as they relate to group creativity and provides theoretical approaches to group creativity based on information-processing theory, social identity theory, network theories, and decision-making theories. The fourth section focuses on the (social) context in which group creativity takes place and examines the role of norms and culture, the organizational context, and technology. The final section offers practical applications in terms of effective brainstorming, the role of leadership, and how group creativity plays a role in industry, science, and the arts. This Handbook of Group Creativity not only summarizes the state-of-the-science in group creativity research but also offers many suggestions on how this blossoming field may further develop and on how group creativity may be stimulated in practice.