Multimodal vs. Unimodal Physiological Control in Videogames for Enhanced Realism and Depth

FACULDADE DE ENGENHARIA DA UNIVERSIDADE DO PORTO
Mul imodal s. Unimodal Physiological
Con ol in Videogames o Enhanced
Realism and Dep h
Gonçalo Filipe Lopes Coelho Ama al da Sil a
Mes ado In eg ado em Engenha ia In o má ica e Compu ação
Supe iso s: Rui Rod igues (PhD), Ped o Al es Noguei a (MSc)
Feb ua y 7, 2014
Mul imodal s. Unimodal Physiological Con ol in
Videogames o Enhanced Realism and Dep h
Gonçalo Filipe Lopes Coelho Ama al da Sil a
Mes ado In eg ado em Engenha ia In o má ica e Compu ação
App o ed in o al examina ion by he commi ee:
Chai : José Manuel de Magalhães C uz (PhD)
Ex e nal Examine : Hugo Alexand e Pa edes Guedes da Sil a (PhD)
Supe iso : Rui Ped o Ama al Rod igues (PhD)
Feb ua y 7, 2014
Abs ac
In he las wo decades, ideogames ha e e ol ed in a nea ly explosi e way, om ex -based
ad en u es and pixela ed g aphics o he nea - ealis ic 3D en i onmen s ha a e common in he
supe -p oduc ions o big game companies (AAA i les). Al hough his endency was no mi o ed
in he games’ in e ac ion de ices a ea, ecen s udies in he Human-Compu e In e ac ion ield
(HCI) ha e explo ed bio eedback in e ac ion as an al e na i e o he cu en de ices.
In he con ex o HCI and ideogames, his disse a ion explo es di ec bio eedback: he ex-
plici manipula ion o physiological da a (i.e. one’s own body) in o de o pe o m ac ions inside
he game. Usually, p o o ypes o his bio eedback ype bind o each game mechanic a single phys-
iological senso . Al hough his in eg a ion is easy and quick, we conside ha i has he pi all o
being oo simple - i does no ye jus i y changing om he ac o p essing a bu on o un o a
physical ac ion such as aising one’s oo . As a consequence, his excessi e simplici y was es he
as po en ial ha we belie e o exis in his echnology, o c ea e games o inc easing complexi y
and highe dep h.
To answe o hese limi a ions, in his disse a ion we s i e down an unexplo ed pa h by com-
bining mo e han one senso in a single game mechanic, which we call Mul imodal Di ec Bio eed-
back. Using he Un eal De elopmen Ki , we c ea ed a Fi s -Pe son Shoo e capable o simula ing
eliably eal-li e ac ions, such as a enua ing he ecoil o a gun and unde wa e explo a ion by
using he playe ’s b ea hing da a, among o he mechanisms. To assess p os and cons o his new
bio eedback ype, we de eloped h ee dis inc e sions o he game: Vanilla (Keyboa d/Mouse, o
mode n games), Unimodal Bio eedback (Keyboa d/Mouse + 1 senso pe each game mechanic)
and Mul imodal Bio eedback (Keyboa d/Mouse + 2 senso s pe each game mechanic).
This p o o ype was es ed by 32 pa icipan s whom compa ed he h ee e sions o he game
mechanisms (and each e sion as a whole) in e ms o Fun, Ease o Use, O iginali y, Playabili y
and Fa ou i e Condi ion along wi h ano he ques ionnai e a ge ed a o he pa ame e s (IMI Ques-
ionnai e). Addi ionally, he pa icipan s compa ed he h ee e sions o he mechanisms by using
12 keywo ds which a e ele an o he game design o he game, and p o ided a g ea quan i y o
eedback and commen a ies abou he a ious elemen s o he h ee e sions.
In compliance wi h p e ious s udies, bo h bio eedback e sions we e conside ed mo e un han
he anilla e sion. The unimodal and mul imodal e sions ecei ed simila sco es in pa ame e s
such as "Fun" o "Playabili y", bu bo h we e app ecia ed by he playe s: he unimodal e sion o
i s simplici y o use, and he mul imodal o i s ealism, ac i a ion sa e y and dep h added o he
game. Ou bigges con ibu ion is ha mul imodal bio eedback can ha e a qui e ele an impac
in e ms o added dep h, depending on he way i is used inside he game. On a bounda y case, i
can be used o inc ease he eeling o empowe men on he playe when using ce ain abili ies, o
o in en ionally make in-game ac ions mo e di icul by demanding mo e physical e o om he
playe . We belie e ha bo h bio eedback ypes should be combined simul aneously depending on
he sensa ions ha game designe s wish o con ey o he playe s.
i

ii
Resumo
Nas duas úl imas décadas, a á ea dos jogos de ídeo e oluiu de uma o ma quase explosi a a pa i
das a en u as em ex o e g á icos pixelizados pa a os ambien es 3D quási- o o- ealís icos comuns
nas supe p oduções das g andes emp esas. Apesa des a endência não se e e i icado nos
disposi i os de in e ação com os jogos, es udos ecen es no campo da In e ação Pessoa-Máquina
(IPM) êm explo ado a in e ação a a és de bio eedback como uma al e na i a aos disposi i os
a uais. No con ex o da IPM e jogos de ídeo, es a disse ação explo a unicamen e a e en e do
bio eedback di e o: a manipulação explíci a dos dados isiológicos (ou seja, o p óp io co po) pa a
ealiza ações den o do jogo. Os p o ó ipos adicionais des e ipo ligam a cada mecânica de jogo
um único senso isiológico e embo a es a in eg ação seja ácil e ápida, conside amos que peca
po se demasiado simples e ainda não jus i ica a oca do a o de ca ega num bo ão po uma ação
ísica como le an a um calcanha pa a co e . Po sua ez, es a simplicidade excessi a despe diça
o po encial as o que ac edi amos exis i nes a ecnologia pa a c ia jogos complexos e de maio
p o undidade.
Em espos a a es as limi ações, nes a disse ação en e edámos po um caminho não explo ado
ao combina mais do que um senso em cada mecânica de jogo, a que chamamos Bio eedback Mul-
imodal Di e o. Foi c iado no Un eal De elopmen Ki um Fi s -Pe son Shoo e capaz de simula
ia elmen e ações eais como con e o ecuo de uma a ma, le a a cabo explo ação subaquá ica,
en e ou os. Pa a conhece mos os p ós e con as des e no o ipo de bio eedback, o am desen-
ol idas ês e sões dis in as: S anda d (Teclado/Ra o, pa a os jogos a uais), Bio eedback Uni-
modal (Teclado/Ra o + 1 senso po mecânica de jogo) e Bio eedback Mul imodal (Teclado/Ra o
+ 2 senso es po mecânica).
Es e p o ó ipo oi es ado po 32 olun á ios que compa a am as ês a ian es dos mecanismos
do jogo (e cada a ian e na o alidade) em e mos de Di e são, Facilidade de Uso, O iginalidade,
“Jogabilidade”, Condição P e e ida e u ilizando um ques ioná io di ecionado a ou os pa âme os
(IMI Ques ionnai e). Adicionalmen e, a alia am as ês a ian es dos mecanismos com ecu so a
12 pala as-cha e ele an es ao game design do jogo e o nece am um as o leque de comen á ios
sob e os á ios elemen os das ês a ian es.
Em conco dância com es udos an e io es, ambas as e sões de bio eedback o am mais di e -
idas que do a e são s anda d. As a ian es unimodal e mul imodal ecebe am pon uações semel-
han es em pa âme os como "Di e são" ou "Jogabilidade", mas ambas o am ap eciadas pelos
jogado es: a unimodal pela sua simplicidade de uso, e a mul imodal pelo seu ealismo, segu ança
de a i ação e p o undidade ac escen ada ao jogo. O nosso maio con ibu o é que bio eedback
mul imodal pode e um impac o bas an e ele an e em e mos de p o undidade ac escen ada, de-
pendendo do uso que lhe o dado den o do jogo. Nos casos ex emos, pode ainda se i pa a
aumen a a sensação de pode no jogado ou pa a di icul a in encionalmen e uma ação den o
do jogo ao exigi mais es o ço ísico. Ac edi amos que ambos os ipos de em se combinados
em simul âneo em unção das sensações que os c iado es de ideojogos quei am ansmi i aos
jogado es.
iii
i
Acknowledgemen s
Finally, he longes p ojec o my Mas e ’s deg ee is o e . I was a om imagining all he ha d
wo k ha would come wi h his p ojec , he i s ime I pu my eyes on i . Bu e en a e all he
ough imes and op ing o wo k on i o six ex a mon hs, I would s ill choose o wo k on his
p ojec all o e again. Mo e han my con ibu ion o bio eedback esea ch applied o ideogames,
his wo k is also dedica ed o all he people who make o lo e ideogames. This p ojec wouldn’
exis wi hou you!
To my supe iso s, P o esso Rui Rod igues and Ped o Noguei a, a huge “ hank you” o be-
lie ing ha I was he man o he job, despi e my no -so-long expe ience in games de elopmen ,
and o accep ing my sugges ions ha ook his p ojec “ o he nex le el” (in a manne o speak-
ing).
To Ch is Holden, o allowing us o use and modi y his o iginal "Dungeon Escape" UDK map
— a simple hing as a “yes” made all he di e ence in his p ojec and allowed us o each he
desi ed g aphical quali y o he game. A huge hank you!
I eel obliged o gi e c edi o he au ho o “Game Design Wo kshop” — T acy Fulle on —,
who w o e a wonde ul book on Game Design which inspi ed me o e he cou se o his p ojec .
I ’s in he li le de ails and addi ional e o o his hesis ha his book’s magic is p esen .
Finally, o hose dea es o me: o my pa en s, amily and dogs o keeping me heal hy and
men ally sane du ing all his ime; o my close iends o all he un imes, he e y ea ly pilo
es ing and helping me push o wa d; and a special hanks o my gi l iend, as she pu i nicely,
“ o pu ing up wi h me all he ime”. The e is no be e esea che han a happy one.
Thank you!
Gonçalo Filipe Sil a
LIST OF FIGURES
xii

Lis o Tables
5.1 Mapping o Senso s o Game Mechanics . . . . . . . . . . . . . . . . . . . . . . 28
5.2 Ca d So ing - Desc ibing Game Mechanics in Wo ds . . . . . . . . . . . . . . . 29
6.1 Q1 Fun - S a is ical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
6.2 Q1 Ease o Use - S a is ical Analysis . . . . . . . . . . . . . . . . . . . . . . . . 35
6.3 Q1 O iginali y - S a is ical Analysis . . . . . . . . . . . . . . . . . . . . . . . . 37
6.4 Ca d So ing Resul s - Desc ibing Game Mechanics in Wo ds (1) . . . . . . . . . 40
6.5 Ca d So ing Resul s - Desc ibing Game Mechanics in Wo ds (2) . . . . . . . . . 41
6.6 Q2 O iginali y, P e e ence and Playabili y be ween Condi ions - S a is ical Analysis 42
6.7 Q2 IMI Pos Expe ience - S a is ical Analysis . . . . . . . . . . . . . . . . . . . 42
xiii
LIST OF TABLES
xi
Abb e ia ions
AI A i icial In elligence
DLL Dynamic Link Lib a y
EDA Elec ode mal Ac i i y
ECG Elec oca diog aphy
EEG Elec oencelog aphy
EMG Elec omyog aphy
FPS Fi s -Pe son Shoo e
GAZE Eye Gaze T acking Senso
GLOVE Hand T acking Glo e Senso
GSR Gal anic Skin Resis ance
HCI Human Compu e In e ac ion
HR Hea Ra e Senso
NUI Na u al Use In e ace
RESP Respi a ion Senso
SCL Skin Conduc ance Le el
UDK Un eal De elopmen Ki
x
Chap e 1
In oduc ion
1.1 Con ex : E olu ion o Videogames and Bio eedback
Mode n ideogames ha e d as ically e ol ed in he pas wo decades om simple ex -based and
pixela ed g aphics o nea -pho o ealis ic wo lds ea u ing massi e amoun s o a ailable con en
o explo e and in e ac wi h. Wi h he in oduc ion o hese massi e wo lds, ideogames ha e
e ol ed in o complex expe iences iddled wi h in e ac i e and engaging na a i es, ealis ic game-
play mechanics and in ica ely designed a i icial in elligence (AI) sys ems, all o which aim o
close he gap wi h eal li e as much as possible. Ul ima ely, ideogames a e played o he exclu-
si e, highly emo ional and ewa ding expe ience ha hey deli e o playe s [EM07]. A he end
o 2013 we en e ed he eigh h gene a ion o ideogame pla o ms wi h new inpu de ices, bu his
new gene a ion s ill adhe es p ima ily o he gamepad- ype con en ion. Compu e pla o ms s ill
adhe e o he keyboa d/mouse con en ion, as well.
The keyboa d/mouse and gamepad con ol schemes ha e been he mos popula con en ions
o ideogames and o e ime hese de ices ha e e ol ed along wi h he gaming pla o ms. While
he game pla o ms hemsel es ha e e ol ed in e ms o compu a ional and ende ing powe , he
in e ac ion de ices ha e e ol ed mos ly in he di ec ion o usabili y – ei he o accommoda e di -
e en play s yles o o become mo e e gonomic. Hea ily ma ke ed owa ds ha d-co e game s,
he indus y p oduced keyboa ds and mice wi h e gonomic layou s, cus omizable weigh , p o-
g ammable keys o gamepads in many di e en shapes and sizes – mos ly based on he p emise
ha amilia i y wi h a ce ain con olle gua an ees a good pe o mance in he game [B o07,
KT02].
Th oughou he yea s game companies ha e ied o c ea e new echnologies o s ep away
om he adi ional de ices, wi h he mos ecen a emp s ocusing on na u al use in e ac ion
(NUI) de ices: he Nin endo Wiimo e, he Mic oso Kinec and he PlayS a ion Mo e. Wi h
hese new na u al in e ac ion de ices, playe s can now hold a he o’s swo d in an ad en u e game
(“The Legend o Zelda: Skywa d Swo d”, 2011), ide an in la able boa h ough a i e (“Kinec
Ad en u es”, 2010) o mo e a wiza d’s wand o cas spells (“Wonde book: Book o Spells”, 2012).
1

In oduc ion
Figu e 1.1: Diag am depic ing he low o aw and p ocessed (audio isual) physiological da a
be ween he playe and he game.
Recen esea ch in he Human-Compu e In e ac ion (HCI) ield has wo ked on ein oduc-
ing he concep o including eal- ime physiological da a om playe s in eal- ime applica ions
h ough he use o Bio eedback de ices. Bio eedback was c ea ed o iginally in he ield o
medicine in he 1980s as a aining p ocedu e o o e come medical condi ions bu in he las
decade has ound i s way o wo main applica ions in game HCI esea ch (Figu e 1.1). These a e
Di ec and Indi ec Bio eedback:
•The playe in en ionally manipula es physiological da a (i.e. hei body), which is ead by
he sys em in o de o pe o m ac ions. The playe is an ac i e agen in he ans e o
physiological da a.
•The sys em eads he playe ’s physiological da a and uses i o modi y in e nal aspec s o i s
beha iou , bu wi hou him/he being awa e o how is physiological da a a ec ed he game.
The playe is a passi e agen in he bio eedback p ocess.
The second ype is mo e commonly connec ed wi h A ec i e Gaming [GDA05], whe e he
playe ’s emo ional s a e is es ima ed om he collec ed physiological da a and game pa ame e s
a e modi ied in a meaning ul way [BBM+07,CPC+09,HB00,MA07,Nog13,NTR13,NRON13a,
RSL05,To 13,YMJ10]. I can s ill be conside ed a bio eedback sys em as i uses Bio eedback
ins umen a ion o ead physiological da a and p esen i back o he playe , hence closing an in e -
ac ion loop. This ype o bio eedback sys em is commonly e e ed o as A ec i e Bio eedback
2
In oduc ion
and is a e y p omising a ea o A ec i e Compu ing. In his disse a ion, howe e , we will be
ocusing only on di ec bio eedback.
Adap ing adi ional ideogames o make use o di ec bio eedback has been es ed in he pas
wi h p omising esul s. Howe e , hese game adap a ions ha e ocused mos ly on indi ec and
simple game mechanics [NKLM11,DC07,KLT+10]. Addi ionally, we belie e ha un il now no
bio eedback games ha e managed o ake ull ad an age o he physiological de ices’ po en ial.
Rega ding he game design aspec s, physiological senso s in hese games a e in eg a ed in a e y
shallow way and hei con ibu ion o gameplay – wi hou accoun ing o he di e en sensa ion
o ac i a ing an ac ion in-game using ou own body – seems unce ain when compa ed o p essing
a bu on in mode n games. In [Amb11], SCL is used in “Alien Swa m” simply o de ine how
as he game ime coun s down o ze o. In [DC07] and [KLT+10], only indi ec bio eedback is
explo ed hea ily, wi h he playe “ o ei ing game con ol” (i.e. gameplay eedom) in o de o
access o abili ies such as in isibili y, x- ay ision o as e i ing a e. Because hese mechanics
a e ac i a ed using indi ec bio eedback, playe s do no know when hey a e going o ac i a e hose
abili ies and hus ail o ully bene i om hem. We conside ha bio eedback-powe ed abili ies
designed in his manne a e o e simpli ied and end up unbalancing he game (i.e. making i oo
easy), which a e a while will esul in bo edom due o he lack o an app op ia e challenge.
E mi and Mäy a a i m ha “ he essence o a game is oo ed in hei in e ac i e na u e” [EM07].
Since physiological da a con ains li e al in o ma ion on how we a e con olling ou own body, we
eel he e is a g ea un apped po en ial in di ec bio eedback games. Gi en his p emise, ou ques-
ion is whe he in oducing mo e aspec s o ou physiological beha iou in o he game, we can
inc ease his “essence o in e ac i i y” exp essed by E mi and Mäy a. In ou opinion, by inc eas-
ing he bio eedback mechanics’ complexi y we can le e age ha po en ial o c ea e games wi h
a mo e dep h and imp o ed game design.
1.2 Documen S uc u e
Be o e elabo a ing u he on how we ied o inc ease his essence o in e ac i i y, we ake
a glimpse a he cu en s a e o he a o bio eedback esea ch in wo a eas: Medicine and
Games/En e ainmen (Chap e 2). Addi ionally, we p o ide a dedica ed sub-sec ion o in oduce
he physiological senso s used in his disse a ion wi h examples o hei p e ious use in esea ch.
In Chap e 3we app oach he equi emen s o ou game and he c i e ia ollowed in he c e-
a ion o he bio eedback game mechanics, be o e implemen ing he ac ual game. In Chap e 4we
p esen he inal game and i s bio eedback game mechanics, along wi h he sys em’s a chi ec u e
and addi ional e o s done o e ine he quali y o ou wo k. In Chap e s 5and 6, we p esen
ou empi ical s udy wi h 32 egula playe s es ing ou bio eedback game in h ee in e ac ion e -
sions (one o hem wi h s anda d keyboa d/mouse in e ac ion, and he o he wo augmen ed wi h
bio eedback) and p esen hei analysis o he game acco ding o objec i e c i e ia (e.g. a ing
3
In oduc ion
pa ame e s o Fun on a scale o 0 o 5) and subjec i e c i e ia (associa ing a lis o keywo ds o
each game e sion and open-ended pos -game commen a ies).
Finally, we e lec on ou wo k by analysing he collec ed da a and he limi a ions o his s udy
(Chap e 7), p esen ing u u e a enues o esea ch in bio eedback (Chap e 8) and he conclusions
o his disse a ion (Chap e 9).
4
Chap e 2
Rela ed Wo k
A e iew o he cu en s a e o he a on bio eedback and i s closes ela ed ields is p o ided in
he ollowing sec ions. Mos o he p e ious wo k done on physiological in e ac ion has ocused
on es ing how well i applies o medical he apy and in es ablishing guidelines o use o new ields
such as emo ional ecogni ion [MA07,LCCS07,HGSW04,NRON13b,NRON13c] and a ec i e
digi al games [GDA05,YMJ10,KBWA04].
2.1 Field o Medicine
Bio eedback was i s commonly used in medical he apy as a aining ea men o o e come
medical condi ions o pa ien moni o ing [MB+91,BEV+96]. In [DLH+10], a mobile pla o m
o music he apy was p esen ed, whe e use s’ nega i e emo ional s a es we e coun e balanced
h ough music. In a simila app oach, [RBF+08] p esen ed a sys em o body balance ehabili a-
ion by using simple audio equencies o indica e co ec pos u e.
Also in he ield o ehabili a ion ea men s, Huang de eloped an in e ac i e bio eedback
sys em o mo o ehabili a ion inside a 3D i ual wo ld whe e pa ien s y o comple e asks while
ecei ing musical and isual eedback in eal- ime [HIO+05]. Cho, Wang and Mingyu de eloped
EEG-based sys ems ha ha e he po en ial o ea medical condi ions such as he A en ion De ici
Hype ac i e Diso de (ADHD) [WSN10,MJNQ06,CLK+02].
Gi en bio eedback’s seemingly easy in eg a ion wi h ideogames and high po en ial in med-
ical ea men s, a ious se ious games ha e been designed o aiding in he ea men o medical
condi ions, such as: ea men o swallowing dys unc ions using EMG signals, whe e a pa ien is
a ish ying o swallow smalle ish [SBC+11]; ea men o s u e ing o simila speech diso de s
by eading aloud pieces o ex while moni o ing he pa ien ’s GSR le el [LG12]; and Gene al
Anxie y Diso de ea men by combining bio eedback wi h i ual eali y sys ems whe e ce ain
objec s inside he i ual wo ld we e modi ied based on he pa ien ’s hea a e signal [RGP+10].
5
P elude: Requi emen s and Design Choices
12

Chap e 4
Implemen a ion: A Di ec Bio eedback
Game
To compa e how bo h mul imodal and unimodal bio eedback-augmen ed ideogames impac game
design and use expe ience in mode n Fi s -Pe son Shoo e ideogames, we a e in e es ed in an-
swe ing he ollowing ques ions:
1. Wha a e he playe s’ esponses o physiologically augmen ed gameplay mechanics in mod-
e n Fi s -Pe son Shoo e games?
2. How do Mul imodal and Unimodal mechanisms compa e o each o he in e ms o in-game
ac ions, si ua ion con ex and playe expe ience?
3. Can we enhance game ealism and dep h using Mul imodal and Unimodal Bio eedback? I
so, how a ?
As a means o answe he p e ious ques ions, h ee iden ical e sions o he game wi h di e en
in e ac ion me hods we e de eloped:
•No mal (o Vanilla) — Mouse and keyboa d only.
•Unimodal — Mouse/Keyboa d plus one physiological senso pe augmen ed gameplay me-
chanic.
•Mul imodal — Mouse/Keyboa d plus wo physiological senso s pe augmen ed gameplay
mechanic.
13
Implemen a ion: A Di ec Bio eedback Game
4.1 “Gene icShoo e 3000” Game Desc ip ion
In ou game he playe has h ee guns a his/he disposal: a Physics Gun o in e ac wi h physics-
enabled objec s and emo e obs acles ou o he way when necessa y; a Link Gun and a Shock
Ri le om he “Un eal Tou namen ” games which unc ion simila ly o a machine gun and an
o dina y i le, espec i ely.
Besides basic locomo ion, he main cha ac e can also pe o m o he ypical ac ions such as:
sp in , jump, c ouch, pu ou o ches o sol e puzzles and in e ac wi h bu ons/handles on walls
and loo s. Las ly, he e a e wo special abili ies:
1. Possession — The playe en e s he body o an enemy and con ols i in a i s -pe son iew.
This lea es he main cha ac e ’s body immobile and ulne able o nea by enemies o he
du a ion o he possession. The playe can use he possessed body o ac i a e doo s and
swi ches o o igh o he enemies – on ce ain key poin s, his abili y is he only way o
p og ess in he game. The possession echnique has no ime limi and ends when he pos-
sessed body dies (e.g. igh s e sus o he enemies, d owning, suicidal), he playe s ops he
abili y o when he main cha ac e ’s body is killed (in his case he playe loses he game).
2. In isibili y — The playe becomes in isible o all enemies and may eely oam he le el.
This echnique ends whene e he playe in e up s i o a e i s powe le el (indica ed on
he lowe le co ne o he sc een) ends.
Enemies un he same AI algo i hm in he en i e game and ha e ei he he Link Gun o he
Shock Ri le. The e a e no boss encoun e s a his cu en s age. Fo ou es scena io, he game
akes place in a looded p ison whe e he main cha ac e is inca ce a ed and has o escape. To
his end he/she mus ind h ee keys o exi he p ison, which o ou es pu poses is he “winning
condi ion” - ha is, playe s ha e o keep playing un il hey manage o escape he p ison. The le el
is based on an exis ing UDK map called “Dungeon Escape”1, which we modi ied as necessa y o
change he le el’s in e ac ion logic (e.g. beha iou o ga es, handles, swi ches, spawning enemies)
and o include an unde wa e sec ion ha did no exis in he o iginal le el. Enemies and obs acles
we e placed o o ce usage o all he main cha ac e ’s abili ies a some poin in he le el.
4.2 Game Mechanics Subjec o Bio eedback Con ol
The e a e a o al o eigh game mechanisms ha a e ac i a ed di e en ly depending on he game
e sion being played.
4.2.1 Gun Recoil A enua ion
Based on popula FPS i les such as “Coun e -S ike” o “Call o Du y”, he Link Gun and he
Shock Ri le mo e he playe ’s iew whene e a sho is i ed. In he anilla e sion, no a enua ion
1C ea ed by Ch is Holden, who gen ly au ho ized us o use he map on ou p ojec . The o iginal map is a ailable a
h p://ch isholden.ne /07.h m .
14
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.1: In isibili y iew mode wi h an enemy in on . In isibili y’s pos -p ocessing e ec s
occlude he sc een bo de s and change he colo scheme.
is a ailable and he ecoil alues a e app oxima ed as ai h ully as possible o he a e age ecoil
exhibi ed in exis ing FPS i les.
The eal-wo ld logic behind his mechanism is ha when someone i es a gun, con ac ing
he i ing a m and sus aining he b ea hing cycle will esul in s eadie aim and educed weapon
kickback e ec s – in eali y his does no necessa ily apply, bu common popula wisdom dic a es
o he wise and hus makes he mechanism eel eal in his way.
In he unimodal e sion, playe s can signi ican ly a enua e ecoil by con ac ing hei igh a m
(EMG-A m). In he mul imodal e sion, playe s mus bo h con ac hei igh a m (EMG-A m)
and inc ease hei ches olume (RESP) o apply he e ec o he same in ensi y. Ac i a ing he
mechanism only pa ially (by ei he con ac ing hei a m o inc easing hei ches olume) esul s
in a pa ial educ ion e ec . Bo h bio eedback e sions ha e hei ecoil alues sligh ly inc eased
o o ce he usage o he physiological senso s while a he same ime ying o app oxima e he
beha iou o a eal-li e gun.
4.2.2 In isibili y
The ime In isibili y can emain ac i e is he same o he h ee e sions – a du a ion ba is dis-
played on he lowe le co ne o he sc een (Figu e 4.1).
As he e is no eal-wo ld analogue o becoming in isible, we le e aged a common playe
beha iou in s eal h games - becoming ense and ha ing e y shallow, slow b ea hing (pe haps
15
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.2: Unde wa e pos -p ocessing e ec s wi h blue in ing and dis ance-based blu ing.
in a subconscious a emp a no making noise in-game, al hough his beha iou o making oo
much noise o b ea hing loudly using he senso s was no implemen ed in he game). Howe e , o
balance he game’s design we decided o ea i as a special abili y (based on popula ic ion wo ks
such as “D agon Ball”,“E agon” o “Ha y Po e ”) which equi es a eeling o empowe men
be o e ac i a ing powe ul abili ies.
In he anilla e sion, playe s p ess he “Q” key o become in isible. In he unimodal e sion,
playe s a e equi ed o b ea he in (RESP) o ac i a e he powe , and hold hei b ea h in o de
o main ain in isibili y ac i e. I hey b ea he ou , he powe is deac i a ed. In he mul imodal
e sion, playe s mus i s make a closed is pose (GLOVE) - mimicking he magic seal o a spell
o o concen a e powe - and hen b ea he in (RESP) o ac i a e he powe , subjec o he same no
b ea hing cons ain s as in he unimodal e sion.
4.2.3 Unde wa e B ea hing
Simila ly o wha is done in o he i les such as he old “Tomb Raide ” games, he playe has
limi ed ime o s ay unde wa e and ha e hei oxygen le el displayed on he sc een (Figu e 4.2).
In he anilla e sion, oxygen dec eases linea ly o e ime wi hou any in e en ion om he
playe .
In bo h bio eedback condi ions he eal-wo ld analogies a e ob ious: playe s a e equi ed o
physically hold hei b ea h using only he RESP senso while hey a e unde wa e . T ying o
b ea he in while unde wa e esul s on illing he lungs wi h wa e and d own ins an ly as a esul .
16
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.3: Possession pos -p ocess e ec s include bo de o e lays wi h ed colo (meaning “dan-
ge ”, o be associa ed wi h enemies). Keybinds a e also communica ed on he bo om o he sc een
(on he Vanilla e sion only!) as a good design p ac ice ollowed by ideogames.
While his is no wha would no mally happen in eal-li e - ha is, people would ill hei lungs
wi h wa e and eel a s ong need o come o he su ace –, ea lie play es ing showed ha playe s
con inued di ing igno ing heal h losses in he cha ac e . Igno ing he RESP senso and heal h
losses was a beha iou ha we did no desi e in he game, hus we ook a mo e puni i e app oach
o his mechanism.
RESP le els while di ing a e also connec ed o he on-sc een oxygen ba (Figu e 4.2), eplac-
ing he anilla e sion’s dec easing oxygen beha iou .
4.2.4 Possession
In he anilla e sion, playe s ake o e he enemy’s body by p essing he “V” key (Figu e 4.3).
Simila o o he popula i les (in his case “Dishono ed”), o e u n o hei o iginal body, hey
can pe o m a Release ac ion using he “V” key, o a Suicide ac ion using he “H” key.
In he unimodal e sion, playe s Possess enemies by blowing ho ai on he TEMP senso ,
which mimics he playe ’s soul lea ing he body as i en e s he possession’s a ge . In he mul i-
modal e sion, playe s i s pe o m he hand pose displayed (Figu e 4.4 on he le ) and hen blow
ho ai on o he TEMP senso . To main ain consis ency wi h he In isibili y echnique, his pose
esembles he magic seal o a magical abili y.
17

Implemen a ion: A Di ec Bio eedback Game
Figu e 4.4: Possession hand ges u es o bo h bio eedback e sions: ini ia e and e e se (le ),
suicide ( igh ).
Fo bo h e sions, o e u n o he o iginal body hey ha e o pe o m one o he hand poses
( e e se o suicide) displayed on Figu e 4.4.
4.2.5 Fi e Blow: In e ac ion wi h Fi e Objec s
This ac ion was designed o be used as a puzzle sol ing abili y inside he game in ol ing o ches
o o he i e objec s. In u u e wo k we would like o i o also be a s eal h mechanic, whe e
blown-ou o ches dec ease he isibili y o an AI enemy - i allows o mo e s a egic hinking
om he playe .
In he anilla e sion, playe s pu ou a o ch o ligh up a i eplace by p essing he “B” bu on
(Figu e 4.5). In he unimodal e sion, playe s blow on he TEMP senso .
In he mul imodal e sion playe s mus i s inhale hea ily o inc ease hei ches olume
h ough he RESP senso and inish he ac ion by blowing on he TEMP senso . Since he in-
game puzzles we e in he size o o ches o i eplaces, hey equi ed a la ge b ea hing e o – he
mul imodal e sion po ays his in e ac ion mo e ealis ically.
4.2.6 Sp in ing
On he lowe co ne o he sc een is a s amina ba ha shows o how much longe he playe
can un be o e being o ced o eco e his s eng h (Figu e 4.6). S amina dec ease o e ime and
maximum speed a e he same on he h ee condi ions.
In he anilla e sion, Sp in ing is ac i a ed by holding down he “Le Shi ” key. In he
unimodal e sion, playe s can li he le leg’s heel o oo ip (EMG-Leg). In he mul imodal
e sion, playe s a e equi ed o use bo h ee (2x EMG-Leg).
4.2.7 I em Use: Using Objec s o Equipping Them
This abili y is used o pick up objec s inside he game, o open locked doo s o o in e ac wi h
o he “usable” game objec s. In he anilla e sion, playe s use he “E” key.
18
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.5: A puzzle whe e he playe has o ligh up he i eplace combining a o ch and “Fi e
Blow” o eed powe o a non-wo king ga e.
Figu e 4.6: The S amina ba shows on he sc een when he playe p esses Shi o he EMG-Leg
senso s a e used. A blu e ec is applied o he sc een o enhance he sensa ion o unning – his
is a ea u e ha al eady exis s in he UDK engine which we le unchanged.
19
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.7: “I em Use” wo ks on handles o open ce ain doo s, o he in e ac i e objec s o o
equip key i ems like he P ison Keys (which a e equi ed o inish he le el).
In he unimodal e sion, playe s close he hand equipped wi h he GLOVE de ice (Figu e 4.7).
In he mul imodal e sion, playe s a e equi ed o con ac he igh a m (o al e na i ely o mo e
i in any di ec ion) while g abbing he same objec wi h he GLOVE.
4.2.8 G abbing Objec s (Physics Gun)
To use his ac ion playe s ha e o equip he Physics Gun. In he anilla e sion, his is ac i a ed
by d agging he Le Mouse Bu on. In he unimodal e sion, playe s close hei hand (GLOVE)
o g ab he objec and open i o elease he objec - o mo e i while g abbing om one place o
ano he he playe can use he WASD keys o mo e o use he mouse o o a e he came a.
Un o una ely, he GLOVE de ice did no ha e an accele ome e and his was an abili y ha
we wan ed o pe o m pu ely on he GLOVE – playe s who wished o o a e he came a we e
o ced o do so by mo ing he mouse wi h he ip o hei closed hand. This is a usabili y aspec
ha we wish o imp o e in u u e wo k.
In he mul imodal e sion, playe s i s con ac he igh a m (EMG-A m) as in he “I em
Use” abili y un il he “G ab” wo d on he sc een (Figu e 4.8) is li g een, and hen close hei
hand (GLOVE) o g ab he objec - a his s age he a m can be elaxed as he objec is being held
success ully.
20
Implemen a ion: A Di ec Bio eedback Game
Figu e 4.8: A physics-enabled objec is co e ing a hole whe e he playe has o pass h ough by
clea ing he way wi h he Physics Gun.
4.3 Sys em A chi ec u e
To c ea e addi ional unc ionali y, UDK has he abili y o in eg a e cus om de eloped Windows
Dynamic-link Lib a ies (DLLs) wi h unc ions ha can be called om inside he UDK’s sc ip ing
language (Un ealSc ip ). We w o e a cus om DLL in C++ which is esponsible o e ie ing
physiological da a om ou side UDK, esul ing in he o e all sys em a chi ec u e depic ed on
Figu e 4.9.
The NeXus-10 de ice is esponsible o cap u ing physiological da a o he EMG, RESP and
TEMP signals. Signal p ocessing he e is handled by he Bio ace+ so wa e sui , which also
pe o ms he signal acquisi ion om he NeXus-10 de ice ia Blue oo h. The GLOVE’s da a
s eam is accessed by a C++ lib a y p o ided by he glo e’s manu ac u e .
The Bio eedbackDLL p o ides no malized da a in o he he game a e i is p ocessed by he
Bio ace+ So wa e sui e, as we desc ibe now in mo e de ail.
4.4 Calib a ion o Physiological De ices
Towa ds he end o de elopmen we quickly el he need o no malize he physiological de ices’
da a eed o he game – his p o ided an easy abs ac ion o wo k o e as we we e no wo king on
each pe son’s unique physiological alues any mo e, bu a he on he expec ed beha iou o each
o he physiological de ices ac oss a gene al popula ion. Fo his pu pose, we c ea ed a e y simple
21
An Empi ical S udy wi h Playe s
Table 5.1: Mapping o Senso s o Game Mechanics
Mechanism Unimodal Mul imodal
Gun Recoil EMG-A m EMG-A m + RESP
In isibili y RESP GLOVE + RESP
Unde wa e — RESP - same in bo h e sions —
Possession TEMP GLOVE + TEMP
Fi e Blow TEMP RESP + TEMP
Sp in ing EMG-Leg 2x EMG-Leg
I em Use GLOVE EMG-A m + GLOVE
G ab Objec GLOVE EMG-A m + GLOVE
cap u e di e ences in posi i e and nega i e aspec s o gameplay – allowing us o gain a signi ican
insigh on hei game expe ience wi h and wi hou he physiological de ices.
In he second es ound, pa icipan s played he P ison Le el h ee imes o each condi ion
un il hey managed o escape he p ison (app oxima ely 15-20 minu es in he ini ial play h oughs
and 5-15 minu es on subsequen uns). A he end o he hi d P ison Le el, playe s compa ed
each e sion o he game (Vanilla s Unimodal s Mul imodal) acco ding o a ings o O iginali y,
P e e ence and Playabili y. Playabili y was inspi ed in B own’s imme sion heo y whe e game
con ols we e an elemen o he i s ba ie o imme sion, “Accessibili y” [BC04]. Being a a he
ambiguous e m, in he con ex o ou s udy we p esen ed i o ou pa icipan s as “ he deg ee
o which hey el he game con ols we e an obs acle in each condi ion” (in simple e ms, he
quali y o game con ols). They we e also equi ed o commen hei a ing choices o jus i y hei
opinion and a oid andomly- illed ques ions.
Las ly, hey illed an In insic Mo i a ion In en o y (IMI) Pos Expe ience Ques ionnai e wi h
each ques ion epea ed h ee imes o compa e he game e sions. Playe s we e ul ima ely ac-
knowledged again o hei pa icipa ion and ecei ed a chocola e ba as a ewa d.
5.3 Tes ing Appa a us
The game was played on a 64-bi desk op compu e a a 1680x1050 esolu ion, unning Windows
7 En e p ise SP 1 wi h he ollowing ha dwa e speci ica ions:
•In el R
Co eTM2 Quad Q9550 @ 2.83 GHz
•4 GB RAM
•NVIDIA R
GeFo ce R
9800 GTX
•Moni o ASUS VW222S wi h a size o 22 inches
•2 desk op speake s o small size
Physiological da a was cap u ed using he NeXus-10 de ice and he 5DT Glo e, and in eg a ed
in he game using ou cus om DLL as desc ibed in he p e ious sec ion. The machine’s ame
28

An Empi ical S udy wi h Playe s
Table 5.2: Ca d So ing - Desc ibing Game Mechanics in Wo ds
Useless
in-game Simple Realis ic Exhaus ing
Comple e
(“don’ mess
wi h i ”)
Imagina y Co e in-game Complex
Con using Relaxing In ui i e
Incomple e
(“some hing’s
missing”)
a es we e logged o all h ee condi ions o moni o pe o mance issues ha migh de ac playe
expe ience. Luckily, no signi ican (>5 ames pe second) ame a e di e ences we e obse ed.
5.4 Pa icipan s’ Demog aphic Da a
Thi y- wo pa icipan s (29 male, 3 emale) aged om 18 o 27 (M=21.28, SD=2.56) pa icipa ed
in he s udy – be o e pa icipa ing in he s udy sessions hey we e equi ed o ill a demog aphics
ques ionnai e. Figu e 5.1 depic s he a e age numbe o hou s pe week ou pa icipan s spen
playing ideogames.
Playe s ha e also epo ed which game gen es hey usually play (Figu e 5.2) - FPS, Ad en u e
and Ac ion gen es we e consis en ly men ioned among playe s, so we can assume ou playe s o
be ai ly amilia (e en i no p o icien ) wi h he gen e.
29
An Empi ical S udy wi h Playe s
Figu e 5.1: The pa icipan s’ a e age ime spen playing ideogames.
Figu e 5.2: Game gen es usually played by ou pa icipan s.
30
Chap e 6
Empi ic S udy Resul s
This sec ion is di ided in wo main sub-sec ions. The i s o hese analyses how pa icipan s a ed
each o he indi idual gameplay mechanics and how hey di e ed be ween hei a ian s. Ra ings
ega ding he Fun, Ease o Use and O iginali y aspec s we e p ocessed using One-way Analysis
o Va iance (ANOVA) es s wi h he Vanilla, Unimodal and Mul imodal e sions o he game as
he wi hin-subjec s ac o . Pos -hoc Tukey es s we e pe o med when s a is ical signi icance was
me .
Fo he lis o e alua ion keywo ds desc ibing he gameplay mechanics, we p esen a compa -
ison o he mos equen keywo ds be ween he di e en mechanisms. This is mean as a mo e
subjec i e and p elimina y insigh on ou pa icipan s’ game expe ience.
In sub-sec ion 6.3 we ocus ou analysis on how pa icipan s pe cei ed he o e all gameplay
expe ience o be a ec ed ac oss he h ee gameplay condi ions in espec o he componen s o
O iginali y, P e e ence and Playabili y. Finally, o a mo e objec i e analysis o playe expe ience
p o ided by each o he game condi ions, we eso ed o he In insic Mo i a ion In en o y (IMI)
ques ionnai e. He e, he measu ed componen s we e: In e es /Enjoymen , Pe cei ed Compe ence,
E o /Impo ance, P essu e/Tension, Pe cei ed Choice, Value/Use ulness and Rela edness.
6.1 Ques ionnai e 1: Indi idual Mechanics
6.1.1 Fun Ra ings
In he i s ques ionnai e we asked playe s o a e he Fun o each mechanism (Figu e 6.1) on a
scale o 0 (No un) o 5(Ve y un). Fo he mechanisms Gun Recoil (χ2(2) = 0.856,p>0.01);
In isibili y (χ2(2) = 0.805,p>0.01); Unde wa e (i only has wo e sions, by de ini ion i does
no iola e he sphe ici y assump ion); Fi e Blow (χ2(2) = 0.950,p>0.01); I em Use (χ2(2) =
0.996,p>0.01) and G ab (χ2(2) = 0.756,p>0.01), Mauchly’s sphe ici y assump ion had been
me . Fo Possess (χ2(2) = 0.883,p<0.01) and Sp in (χ2(2) = 0.644,p<0.01) i was iola ed.
31
Empi ic S udy Resul s
Figu e 6.1: Mean Fun alues be ween condi ions wi h co esponding S anda d E o measu es.
As a esul , deg ees o eedom o bo h mechanisms we e co ec ed using G eenhouse-Geisse
es ima es o sphe ici y (ε=0.71562 and ε=0.73746, espec i ely).
S a is ical signi icance was achie ed o all mechanisms: Gun Recoil F(2,62) = 13.584,p<
0.01; In isibili y F(2,62) = 61.076,p<0.01; Unde wa e F(1,31) = 114.33,p<0.01; Pos-
session F(2,62) = 64.807,p<0.01; Fi e Blow F(2,62) = 77.196,p<0.01; Sp in F(2,62) =
74.975,p<0.01; I em Use F(2,62) = 96.38,p<0.01; G ab F(2,62) = 56.564,p<0.01.
Tukey pos -huc es s e ealed ha pa icipan s ound bo h he unimodal and mul imodal bio eed-
back e sions o he eigh mechanics we e mo e un han he anilla e sion o he game (p<
0.01). Possession’s mul imodal e sion was he only mechanism ha playe s conside ed mo e
un han he unimodal e sion (p<0.01). No u he di e ences we e de ec ed in he emaining
mechanisms (p>0.01).
The Unde wa e mechanism did no equi e a Tukey es no a Mauchly’s sphe ici y es as i
only exis ed in wo e sions: anilla and bio eedback con ol. I was conside ed mo e Fun in he
bio eedback con ol e sion han he anilla e sion.
6.1.2 Ease o Use Ra ings
Pa icipan s we e asked o a e each mechanism also ega ding hei Ease o Use (Figu e 6.2) om
0 (Ve y ha d) o 5 (Ve y easy). Fo all eigh mechanisms Mauchly’s sphe ici y assump ion was
me : Gun Recoil (χ2(2) = 0.955,p>0.01); In isibili y (χ2(2) = 0.870,p>0.01); Unde wa-
e ; Possess (χ2(2) = 0.883,p>0.01); Fi e Blow (χ2(2) = 0.830,p>0.01); Sp in (χ2(2) =
0.994,p>0.01); I em Use (χ2(2) = 0.814,p>0.01) and G ab (χ2(2) = 0.874,p>0.01).
S a is ical signi icance was achie ed as well o all eigh mechanisms: Gun Recoil F(2,62) =
7.281,p<0.01; In isibili y F(2,62) = 24.363,p<0.01; Unde wa e F(1,31) = 32.137,p<
0.01; Possess F(2,62) = 28.981,p<0.01; Fi e Blow F(2,62) = 28.257,p<0.01; Sp in F(2,62) =
20.851,p<0.01; I em Use F(2,62) = 40.374,p<0.01; G ab F(2,62) = 46.669,p<0.01.
Tukey pos -hoc es s showed ha playe s:
1. Found he anilla e sions easie o use han he unimodal e sions o In isibili y, Possess,
Sp in and G ab (p<0.01);
32
Empi ic S udy Resul s
Table 6.1: Q1 Fun - S a is ical Analysis
One-Way ANOVA Tukey Pos -Hoc
Fun F p V-UBF V-MBF UBF-MBF
Gun Recoil 13.584 < 0.001 < 0.001 < 0.001 0.794
In isibili y 61.076 < 0.001 < 0.001 < 0.001 0.131
Unde wa e B ea hing 111.330 < 0.001 —
Possession 64.807 < 0.001 < 0.001 < 0.001 < 0.001
Fi e Blow 77.196 < 0.001 < 0.001 < 0.001 0.846
Sp in 74.975 < 0.001 < 0.001 < 0.001 0.784
I em Use 96.380 < 0.001 < 0.001 < 0.001 0.612
G ab 56.564 < 0.001 < 0.001 < 0.001 0.698
2. Found all mechanisms easie o use in he anilla e sion han he mul imodal e sion (p<
0.01)
3. Found Fi e Blow, Use and G ab easie o use in he unimodal e sion han he mul imodal
e sion (p<0.01).
No di e ences we e de ec ed o he emaining combina ions (p>0.01).
The Unde wa e B ea hing mechanic did no equi e a Tukey es no a Mauchly’s sphe ici y
es as i only exis ed in wo e sions: anilla and bio eedback con ol. I was conside ed easie o
use in he anilla e sion han he bio eedback e sion.
6.1.3 O iginali y Ra ings
O iginali y was a ed on a scale o 0 (No o iginali y) o 5 (Ve y o iginal) – he esul s a e sum-
ma ized on Figu e 6.3. Fo he mechanisms In isibili y (χ2(2) = 0.746,p>0.01) and Unde wa-
e Mauchly’s sphe ici y assump ion was me . Fo Gun Recoil (χ2(2) = 0.724,p<0.01); Pos-
sess (χ2(2) = 0.471,p<0.01); Fi e Blow (χ2(2) = 0.414,p<0.01); Sp in (χ2(2) = 0.473,p<
0.01); I em Use (χ2(2) = 0.636,p<0.01) and G ab (χ2(2) = 0.504,p<0.01) i was iola ed.
G eenhouse-Geisse co ec ions we e applied in hose cases (ε=0.78344; 0.65385; 0.63047;
0.65496; 0.73294 and 0.66852, espec i ely).
S a is ical signi icance was achie ed o all eigh mechanisms: Gun Recoil F(2,62) = 203.19,p<
0.01; In isibili y F(2,62) = 183.73,p<0.01; Unde wa e F(1,31) = 317.05,p<0.01; Pos-
sess F(2,62) = 127.76,p<0.01; Fi e Blow F(2,62) = 154.90,p<0.01; Sp in F(2,62) =
236.54,p<0.01; I em Use F(2,62) = 339.10,p<0.01; G ab F(2,62) = 160.68,p<0.01.
Tukey pos -hoc es s e ealed ha playe s ound bo h bio eedback e sions mo e o iginal han
he anilla e sion (p<0.01). No di e ences we e de ec ed be ween he unimodal and mul imodal
e sions (p>0.01).
The Unde wa e B ea hing mechanic did no equi e a Tukey es no a Mauchly’s sphe ici y
es as i only exis ed in wo e sions: anilla and bio eedback con ol. Playe s ound ha he
Unde wa e bio eedback mechanic was clea ly mo e o iginal han i s anilla coun e pa .
33

Empi ic S udy Resul s
Figu e 6.2: Mean Ease o Use alues be ween condi ions wi h co esponding S anda d E o mea-
su es.
6.2 Gameplay Keywo ds
Tables 6.4-6.5 show he esul s o he pa icipan s’ e alua ion o each game mechanic (in each
e sion) using he p esen ed lis o keywo ds. We will be discussing only ele an esul s o he
able as he amoun o in o ma ion is oo ex ensi e o a comple e discussion. Howe e , he esul s
a e p esen ed as a whole so hey can be used as a e e ence and compa ison basis o u u e wo k
on new di ec bio eedback games and game design expe imen s. To iew he ga he ed da a in a
mo e iendly o ma (column cha s), please e e o Appendix A.
The Fi e Blow mechanic egis e ed mode a e “Useless” alues mos likely because i was only
used as a puzzle-sol ing abili y and did no ha e much in luence in he main ac ion sec ions o he
game. Howe e , we a e ce ain ha i s e olu ion o a s eal h- ela ed mechanic in he game (see
Sec ion 4.2.5) would d op his alue o 0% as i would be mo e ele an o he game.
The “Co e in-game” keywo d egis e ed an inc easing end in he bio eedback e sions, al-
hough no on a e y signi ican scale – su p isingly, he mul imodal e sion o Sp in ecei ed
his adjec i e 44% o he imes (agains 28% in he anilla e sion). This p o ides a hin ha his
mechanic eels mo e like a co e aspec o gameplay on he mul imodal e sion o he game.
Fo he “Simple/Complex” pai – and wi hin ou expec a ions – he bio eedback condi ions
su e ed ele an dec eases owa ds he anilla e sion (especially in he mul imodal e sion) and
inc eases in complexi y owa ds he mul imodal e sion. Howe e , “Complex” did no always
ansla e o some hing necessa ily bad acco ding o playe s’ eedback:
•“The bio eedback sec ions don’ make he game ha d a all, bu hey make i less luid com-
pa ed o he anilla condi ion. While on he la e you can walk and open doo s qui e as ,
on he bio eedback e sions you ha e o s op and use he glo e” (P20)
•“i ’s complex in a good way, i has dep h” (P15 on mul imodal’s Gun Recoil)
•“I liked he unimodal e sion because we ha e o use mo e elemen s o ou body o ge
imme sed in he game i sel and i makes i mo e in e es ing. The mul imodal e sion is one
34
Empi ic S udy Resul s
Table 6.2: Q1 Ease o Use - S a is ical Analysis
One-Way ANOVA Tukey Pos -Hoc
Ease o Use F p V-UBF V-MBF UBF-MBF
Gun Recoil 7.281 < 0.01 0.210 < 0.001 0.098
In isibili y 24.363 < 0.001 < 0.001 < 0.001 0.106
Unde wa e B ea hing 32.137 < 0.001 —
Possession 28.981 < 0.001 < 0.001 < 0.001 0.249
Fi e Blow 28.257 < 0.001 0.031 < 0.001 < 0.001
Sp in 20.851 < 0.001 < 0.001 < 0.001 0.115
I em Use 40.374 < 0.001 0.156 < 0.001 < 0.001
G ab 46.669 < 0.001 < 0.001 < 0.001 < 0.001
s ep u he compa ed o he unimodal e sion because i b ings a li le mo e complexi y
(dep h) o he ges u es execu ion” (P25)
•“I ’s posi i e because i gi es mo e dep h o he game (and mo e imme sion), bu o he
people migh p e e some hing mo e simple” (P4 on mul imodal’s Gun Recoil)
In e ms o Realism, bo h bio eedback e sions had he expec ed esul s and egis e ed la ge
inc eases owa ds he anilla e sion. Du ing de elopmen phase, unde wa e b ea hing was one
o he mos challenging ac ions in he bio eedback-con olled e sions and we we e e y cu ious
abou he playe s’ eac ion o i in he inal e sion:
•“I wish he di ing sec ion was bigge , i was e y sho .” (P4)
•“Some imes I hink i would be nice o eel in a mo ie o game how would i be wi hou
b ea hing” (P15)
•“The di ing sys em is e y c ea i e and he hand seals sys em [o he mul imodal e sion]
is as well” (P2)
•“On he b ea hing sec ion i was e y exhaus ing – i i we e a game wi h e y s ong un-
de wa e componen s o we had o epea i se e al imes i would be e y exhaus ing.”
(P16)
•“I liked all mechanisms in gene al excep o he unde wa e b ea hing one.” (P30)
Rega ding he “Imagina y” keywo d, i was a sibling concep o Realism whe e a game me-
chanic could ail o achie e ealis ic ac ion (because i does no exis in eal-li e) bu was able
o ep oduce qui e well how playe s would go abou pe o ming ac ions ha hey could imagine
hemsel es doing (e.g. cas ing a spell o becoming in isible). This explains why he special abil-
i ies In isibili y and Possess ob ained he highes esul s in he “Imagina y” pa ame e – playe s
seem o iden i y mo e wi h hose ac ions as he hey y o ap in he playe s’ imagina ion.
35
Empi ic S udy Resul s
Figu e 6.3: Mean O iginali y alues be ween condi ions wi h co esponding S anda d E o mea-
su es.
Exhaus i eness was also expec ed o inc ease in he bio eedback condi ions, al hough i seems
some pa icipan s we e able o cope wi h i du ing gameplay. I emains o be seen in u u e s udies
wi h longe es ing pe iods whe he he ype o bio eedback used (unimodal o mul imodal) o e en
bio eedback i sel ha e an impo an in luence on he playe s’ willingness o play.
6.3 Ques ionnai e 2: Game Condi ions
In he second ques ionnai e pa icipan s had o compa e he h ee game condi ions on hei en i e y
ins ead o e alua ing game mechanics indi idually. The ques ionnai e was illed a e all h ee
P ison Le els we e comple ed o limi subjec i i y - ha is, by o cing playe s o compa e all
h ee gaming condi ions a he same ime, i o ced hem o e lec on each o hem and weigh
hem agains one ano he . The compa ison o mean sco es o each measu ed aspec is shown on
Figu es 6.4 and 6.5.
6.3.1 O iginali y, P e e ence, Playabili y
O iginali y was a ed on a scale o 0 (No o iginali y) o 5 (Ve y o iginal) and P e e ence on a scale
o 1 (Ve y low) o 5 (Ve y high). Playe s we e also equi ed o a e he sen ence “I eel ha he
game con ols we e an obs acle in my way o play well...” [BC04] o each game condi ion using
a se en-le el Like scale in o de o quan i y Playabili y (1 – comple ely disag ee; 4 – neu al;
7 – comple ely ag ee). Mauchly’s es s a is ic e ealed ha he sphe ici y assump ion was me
o P e e ence (χ2(2) = 0.933,p>0.01) and Playabili y (χ2(2) = 0.955,p>0.01) bu no o
O iginali y (χ2(2) = 0.359,p<0.01). The e o e, O iginali y’s deg ees o eedom we e co ec ed
using a G eenhouse-Geisse es ima e o sphe ici y (ε=0.60937).
S a is ical signi icance was encoun e ed o all h ee pa ame e s: O iginali y F(2,62) = 107.82,p<
0.01; P e e ence F(2,62) = 24.905,p<0.01; Playabili y F(2,62) = 23.188,p<0.01.
Tukey pos -hoc es s e ealed ha :
36
Empi ic S udy Resul s
Table 6.3: Q1 O iginali y - S a is ical Analysis
One-Way ANOVA Tukey Pos -Hoc
O iginali y F p V-UBF V-MBF UBF-MBF
Gun Recoil 203.19 < 0.001 < 0.001 < 0.001 0.046
In isibili y 183.73 < 0.001 < 0.001 < 0.001 0.054
Unde wa e B ea hing 317.05 < 0.001 —
Possession 127.76 < 0.001 < 0.001 < 0.001 0.020
Fi e Blow 154.90 < 0.001 < 0.001 < 0.001 0.111
Sp in 236.54 < 0.001 < 0.001 < 0.001 0.234
I em Use 339.10 < 0.001 < 0.001 < 0.001 0.028
G ab 160.68 < 0.001 < 0.001 < 0.001 0.463
1. Playe s ound bo h bio eedback e sions mo e o iginal han i s anilla coun e pa (p<
0.01), wi h no di e ence de ec ed be ween he unimodal and mul imodal e sions (p>
0.01). This esul co obo a es wha was ob ained in he i s ques ionnai e ega ding he
a ious game mechanics.
2. Playe s p e e ed bo h bio eedback e sions when compa ed o he anilla e sion (p<
0.01). No signi ican di e ence was obse ed be ween he unimodal and mul imodal e -
sions (p>0.01).
3. Playe s disag eed wi h he s a emen ha game con ols did no s op hem om played well
– in o he wo ds, i means ha he playabili y o ou game con ols eached he desi ed le el.
Howe e , Tukey es s indica ed ha as con ols inc eased in complexi y ( anilla being he
leas complex condi ion, ollowed by unimodal and hen mul imodal), playe s ag eed less
wi h he s a emen (p<0.01).
6.4 IMI Pos Expe ience Ques ionnai e
The IMI ques ionnai e ha we used has se en pa ame e s: In e es /Enjoymen , Pe cei ed Com-
pe ence, E o /Impo ance, P essu e/Tension, Pe cei ed Choice, Value/Use ulness and Rela ed-
ness. Fo he pa ame e s: In e es /Enjoymen (χ2(2) = 0.753,p>0.01); Pe cei ed Compe-
ence (χ2(2) = 0.958,p>0.01); P essu e/Tension (χ2(2) = 0.737,p>0.01) and Rela edness
(χ2(2) = 0.805,p>0.01), Mauchly’s sphe ici y assump ion was me . Fo E o /Impo ance
(χ2(2) = 0.533,p<0.01); Pe cei ed Choice (χ2(2) = 0.300,p<0.01) and Value/Use ulness
(χ2(2) = 0.633,p<0.01) i was iola ed. The e o e, deg ees o eedom we e co ec ed using
G eenhouse-Geisse es ima es o sphe ici y (ε=0.68157, 0.58816 and 0.73176, espec i ely).
S a is ical signi icance was encoun e ed o : In e es /Enjoymen F(2,62) = 58.903,p<0.01;
E o /Impo ance F(2,62) = 40.743,p<0.01; P essu e/Tension F(2,62) = 25.658,p<0.01;
Pe cei ed Choice F(2,62) = 10.320,p<0.01; Value/Use ulness F(2,62) = 92.141,p<0.01;
37
Empi ic S udy Resul s
44

Chap e 7
Resul s’ Analysis and Discussion
We in oduced a new a ie y o Di ec Bio eedback in games ha had no been explo ed be o e
– Mul imodal Di ec Bio eedback – and heo ize on he possible di e ences ha i can b ing o
bio eedback games. We also analyse empi ically ou es game o p o ide a comp ehensi e case
s udy on bio eedback games e sus he cu en mouse/keyboa d con ol schemes.
7.1 Fun Le els in Bio eedback Con ol
The Fun and In e es /Enjoymen me ics show s ong e idence ha playe s enjoyed playing he
game mo e using bio eedback con ol han wi h a s anda d keyboa d/mouse scheme. In ou opin-
ion, he e a e a leas wo ac o s ha seem o con ibu e owa ds his ou come.
The i s is he no el y ac o o he echnology: less han a hand ul o pa icipan s hea d abou
bio eedback in ideogames, and none o hem e e had he chance o y and play ideogames wi h
i . The ela i e success o ou implemen a ion (and e en p e ious wo ks such as [NKLM11,DC07]
can imp in an implici biasing e ec owa ds highe Fun a ings. The only way o assess whe he
he answe would be di e en – on a consume -le el se ing such as a home use scena io – is o
make a long e m s udy in ol ing se e al games ( o mi iga e game design issues and eplica e
accu a ely he li e o he a e age game ) wi h a bio eedback g oup and a con ol g oup. How-
e e , his s udy would equi e a sample popula ion well o e 100 subjec s o each g oup and an
ex ended ime ame o o e wo weeks o e en mon hs. Un o una ely he numbe o exis ing
bio eedback games is e y li le, which makes he idea in easible and e y ha d o execu e. The
only way o uly assess he impac o bio eedback is o in oduce i in a comme cial con ex and
see how game s espond o i and adop i (o no ) o e ime.
One o ou main conce ns is ha Di ec Bio eedback ideogames end up being mo e exhaus -
ing han oday’s ideogames, due o hei s ong physical componen . Some o ou pa icipan s
sha e his conce n as well (as shown in Tables 6.4-6.5) and his could nega i ely impac he a e -
age play ime o game sessions:
45
Resul s’ Analysis and Discussion
•“I hink i could become i ing on he long e m and i wouldn’ be so un (in he si ua ions
whe e you play 10 hou s pe day)” (P10)
•“Du ing he game i would be cool o swi ch be ween unimodal and mul imodal acco ding
o he playe ’s i edness” (P14)
•“I ’s exhaus ing a e a while...” (P27)
•“In his game i wouldn’ be exhaus ing because you don’ un o oo long, bu i i we e
manda o y i would be i ing” (P30 on Bio eedback Sp in )
On he o he hand, some cope well wi h i and e en emb ace i as a necessa y consequence o
his new gameplay s yle:
•‘I ’s i ing, bu i ’s a pleasan i ing” (P2 abou bo h Bio eedback e sions o Sp in )
•“I also liked o con ac he a m in o de o ac i a e Use and G ab – i was sligh ly uncom-
o able bu i was ewa ding o he achie ed ealism’ (...)’ (P5)
•“In e ms o p ecision poin ing a gun and holding b ea h is mo e i ing, bu o app ecia e a
game i becomes mo e in e es ing – howe e on a compe i i e le el i ’s less use ul” (P8 on
Gun Recoil)
The second ac o ha we eel i migh impac he Fun a ings is ela ed wi h he wo Imme sion
heo ies desc ibed by B own [BC04] and E mi [EM07]. B own’s “Engagemen ” s a es ha playe s
need o in es ime, e o and a en ion in o de o each ha s age o imme sion, once he game
con ols and he gen e ma ch he playe ’s p e e ences; E mi’s “Challenge-based imme sion” is
“ he eeling o imme sion when one is able o achie e a sa is ying balance o challenges and
abili ies” and “can be ela ed o mo o skills o men al skills” [EM07].
•“In bo h bio eedback e sions we ha e o imagine wha we would no mally do o ac i a e
he ac ions, i gi es a li le mo e o imme sion and p o ides some di e ences in gameplay.
I ’s also mo e challenging ha ing o ind ou in hese wo wha we ha e o do.” (P24)
The implici e o ha he bio eedback condi ions demanded om playe s is a hin ha playe s
should eel mo e imme sed in he bio eedback condi ions, and Fun and Enjoymen a e indica o s
ha a e ela ed wi h he b oad concep o Imme sion.
One pa icipan made an in e es ing ema k in ou game abou imme sion and bio eedback
– we conside ha bio eedback can be posi i ely combined wi h o he echnologies o enhance
u he he sensa ion o imme sion:
46
Resul s’ Analysis and Discussion
•“In he bio eedback condi ions I hough he in e ac i i y was qui e o iginal [c ea i e], bo h
in he su ounding objec s ( he boxes and handles) as well in he ac ions ha ou cha ac e
made like b ea hing o gun ecoil con ol – hey help p o iding he game a bigge eeling o
imme sion. The eal eeling o imme sion would be o accoun o head mo emen , like he
Oculus Ri 1de ice.” (P11)
Many use s complained abou he lack o an accele ome e (o a simila de ice o ack spa ial
o ien a ion) in ou GLOVE de ice because hey wan ed o g ab objec s using only he glo e:
•“[in he bio eedback e sions] you need o add an accele ome e on he glo e” (P5)
•“The g ab mechanic in bo h condi ions does no wo k well because we ha e o use ou hand
closed (while g abbing he objec ) o mo e he mouse.” (P8)
•“G abbing he objec and hen e u ning o he mouse... i eally makes me wan o mo e i
wi h he glo e” (P12)
•“I don’ like ha ing o g ab and hen no being able o use he mouse” (P21)
The mouse solu ion was no he bes in e ms o p ope game design, bu i was he only one
in ou each in e ms o a ailable ime and human esou ces. We hypo hesize ha Bio eedback
can each i s ull po en ial when combined wi h o he echnologies such as o ien a ion acke s
(e.g. accele ome e ), i ual eali y de ices such as he Oculus Ri , o e en wi h he exis ing
mo ion-de ices o games: he Mic oso Kinec , Nin endo Wii Remo e o PlayS a ion Mo e.
7.2 Unimodal s. Mul imodal Game Design
Apa om Possess and Unde wa e , we we e no able o unco e signi ican di e ences in Fun
be ween unimodal and mul imodal mechanisms. Playe s ended o use he “Realism”, “Exhaus-
i e” and “Comple e” keywo ds o desc ibe mechanics in he mul imodal e sion – su p isingly,
he unimodal Gun Recoil sco ed lowe in “Comple e” han he anilla and mul imodal e sions.
Game con ols can be conside ed o be sligh ly ha de o use on an inc easing scale om anilla o
unimodal and mul imodal – al hough some unimodal mechanisms can s and in pa wi h he anilla
e sion (Gun Recoil, Fi e Blow and Use) acco ding o ou s a is ical esul s.
Some playe s eached he consensus ha he mul imodal was he one wi h highe ealism, bu
su e ed om unnecessa y le els o exhaus i eness. They alued he simplici y o he unimodal
design, which was less also i ing:
•“Ha ing o con ac he a m always o open doo s is bad, i ’s be e o do ha only when
i ing he gun” (P1 abou he mul imodal e sion o he G ab mechanic)
1h p://www.oculus .com/ - “The Oculus Ri is a nex -gene a ion i ual eali y headse designed o imme si e
gaming.”
47
Resul s’ Analysis and Discussion
•“In he unimodal e sion I also hough he expe ience was ewa ding because i was qui e
in e ac i e, simple and i s a game ype ha is mo e ookie and/o lazy – i s ill o ces us o
pu e o bu in a ’ligh e ’ way... ’in e ac i i y ligh ’.” (P5)
•“I hink ha he mul imodal e sion ies o be mo e imme si e, bu i ’s oo complex. The
unimodal e sion can be imme si e and simple. Bo h ypes b ing added alue when com-
pa ed o he anilla e sion.” (P10)
•“I was con using ha ing o coo dina e bo h hings [ he a m and b ea hing] (...) I hink
i wouldn’ be he bes combina ion and ha ing o b ea h dis u bs he ac ion” (P29 abou
Mul imodal Gun Recoil)
O he s alued he asse i eness and sa e y o he mul imodal design, which added he highes
ealism, mo e sense o con ol o a oided senso aul s – ei he unin ended ac i a ions o he
senso s on hei own (i.e. alse posi i es) o he ou -o -con ex ac i a ion o he playe s’ abili ies:
•“I p e e he complexi y o he mul imodal e sion o ha ing mo e de ails ha make he
game mo e ealis ic.” (P4)
•“I hink i makes sense ha ing dis inc ions be ween he ac ions o di ing and using a powe
[In isibili y] (...) he unimodal e sion wasn’ e y in ui i e. (. . . ) The mul imodal e -
sion was mo e comple e o dis inguishing o ch blowing and Possess” (P10 – In isibili y
and Unde wa e B ea hing sha e he RESP senso ; Ai Blow and Possess sha e he TEMP
senso )
•“Closing he hand has mo e eedback o he game o know ha i ’s wha I wan ” (P12 on
In isibili y)
•“I would choose he unimodal e sion only due o he de ail o a m con ac ion in Use and
G ab. Dis ega ding ha I would use he mul imodal e sion, since i uses ways ha a e mo e
explici and di ec o indica e he ac ions ha we wan o pe o m – Possess, In isibili y,
unning wi h bo h legs, e c. (...) B ea hing in and ac i a ing he In isibili y mechanism by
acciden doesn’ seem good o me” (P14)
•“In he mul imodal e sion when I used Possess, I el mo e con ol by ha ing o pe o m
a ges u e be o e ac i a ing he powe han by ha ing o blow on he enemies [in he mul i-
modal e sion]. (...) In he unimodal e sion, ha ing o b ea he bo h o s ay in isible and o
b ea he unde wa e was e y dis ac ing.” (P22)
Las ly, some pa icipan s epo ed ha bio eedback b oke game low mo e o en, mos likely
due o he ime spen execu ing physical ac ions in compa ison o p essing a bu on. Ou “Playa-
bili y” esul s in he second ques ionnai e seem o con i m his as well, bu we hink his is a
48
Resul s’ Analysis and Discussion
by-p oduc o exposing playe s o h ee di e en combina ions o game con ols in such a sho
ime pe iod. Addi ionally, playe s ha e yea s and yea s o amilia i y wi h he keyboa d/mouse
con ol scheme.
•“The mul imodal e sion is a li le oo complex o a game whe e you ha e o be as ,
wi h comba and ac ion. The unimodal e sion has he be e quan i y o in e ac ion when
compa ed o he o he wo.” (P18)
•“The bio eedback sec ions don’ make he game ha d in any way, bu hey make i less luid
in compa ison wi h he anilla e sion. While in he la e i ’s possible o walk and ac i a e
handles (I em Use) qui e as , when using bio eedback you need o s op and use he glo e.”
(P20)
•“The mul imodal e sion is mo e complex and b eaks a li le he pace o he game. The
unimodal one allows us o play in a mo e luid way.” (P26)
•“In he mul imodal e sion, while he ac ions we e mo e ealis ic, asks ook longe o be
execu ed. In he unimodal mode, ac ions had mo e ealism han he anilla e sion and he
di e ence in ime execu ion o ac ions was almos unno iceable.” (P29)
Based on ou playe s’ eedback and ou expe ience, we hink ha a mixed app oach should be
aken ins ead o a s ic implemen a ion o one o hese bio eedback ypes. Long- e m exhaus i e-
ness, “na u alness” o he physical ac ion and eliabili y o mechanism ac i a ion a e he guidelines
o be ollowed. As in mode n games, books o mo ies, i depends on wha eelings and sensa ions
he game designe is ying o con ey o he a ge audience.
Fo example, doo s ha open easily should no need o e powe ing a m con ac ion in o de o
open hem: closing he hand and mo ing he a m wi hou any con ac ion is mo e han enough.
Objec s ha a e ligh and hea y all unde his ca ego y as well. Pu ing ou a ma chs ick o a
candle should only need he TEMP senso ins ead o b ea hing in all he ai you can hold i s .
Conside ing he panoply o c ea i e ways playe s can in e ac wi h a game, he e may be scena ios
whe e e en simple ac ions should be di icul o execu e. Fo example, conside a game sec ion
whe e he game cha ac e was sho in a gun igh and ba icade him o he sel un il ein o cemen s
a i e. Howe e , being hea ily inju ed, he cha ac e does no ha e much s eng h le o mo e
he hea y objec s needed o ba icade he oom. In his scena io i migh be a good op ion o use
mul imodal bio eedback o pu posely make his simple in e ac ion mo e challenging and ealis ic,
ins ead o eso ing o, o example, quick- ime e en s.
Common gameplay si ua ions wi h ha de obs acles (e.g. doo s ha will no budge) o mo e
complex ac ions (e.g. manipula e o h ow objec s) can jus i y he use o Mul imodal Bio eedback.
The eeling o empowe men o ac i a e a special powe as in ou game (In isibili y, Possess) is
some hing ha mos pa icipan s el i wo ked pe ec ly. This is ano he aspec whe e he po en ial
49

Resul s’ Analysis and Discussion
o Mul imodal Bio eedback is app ecia ed, which some o ou pa icipan s dubbed as he “hand-
sign sys em”:
•“Since I like Na u o i ends up being un... ” (P3 abou Mul imodal Possession)
•“I liked he pa o making a ges u e in Possess o con ol he enemies.” (P25)
•“In he Possessing powe I liked mo e he mul imodal e sion because we needed o make
hand signs in compa ison wi h he unimodal e sion whe e you only had o blow ai – he
mul imodal e sion was mo e ealis ic and in e ac i e.” (P13)
•“[excep Sp in ] In all o he mechanisms, I p e e he mul imodal e sion – I liked using a
hand symbol combined wi h ano he ac ion o ac i a e he mechanisms.” (P5)
7.3 Possible Changes in Gameplay
The success ul in eg a ion o Bio eedback gameplay mechanics in ideogames will ha dly ollow
a plug and play pa adigm – no a leas un il a e y ma u e, s anda dized pe iod. As such, he
design o bio eedback games is an impo an ace o hei de elopmen . Thus, i is impo an o
examine how ou bio eedback game design choices impac ed playe expe ience. Following his
hesis, Pa icipan 19 p o ided he ollowing di e en ial analysis o Bio eedback and adi ional
ideogames:
•“The g ea di e ence be ween games ha jus use mouse and keyboa d and hose whe e
senso s a e used, o example, is ha in games wi h senso s a highe impo ance is gi en o
gameplay ins ead o mo e p ope ly he game objec i e – o example in a shoo e , by playing
wi h mouse and keyboa d he enemies a e s onge and in bigge numbe s, while when
playing wi h senso s mo e impo ance is gi en o con ols and mo emen s o he playe ”
(P19)
Senso s can change he ocus and gameplay o he game, which may equi e ce ain adap a-
ions on he game logic o i he use o he new de ices. Some o he mechanics in Tables 6.4-6.5
we e conside ed sligh ly mo e ele an o gameplay (“Co e in-game”) in he bio eedback e sions.
Rega ding Pe cei ed Choice di e ences in he IMI Ques ionnai e, Pa icipan 22 said:
•“[in bio eedback] Al hough i was some hing ha was s o ed in he compu e , I el ha he
choices inside he game we e mine” (P22)
This was he only explici s a emen ega ding Pe cei ed Choice eco ded in he en i e s udy.
While pa icipan s we en’ e y exp essi e on his opic, he ANOVA es s showed ha playe s
expe ienced a sligh ly highe sense o Pe cei ed Choice in he bio eedback condi ions agains
he anilla e sion o he game, whe e i was mo e neu al. I could be because some o he
senso s could be ac i a ed in mo e han one way (e.g. EMG-A m can be ac i a ed ei he by
50
Resul s’ Analysis and Discussion
muscle con ac ion o by simply s e ching he a m in any di ec ion), while bu on-based ac ions
ypically a e ac i a ed always he same way.
The ac ha Pa icipan 22 conside ed ha he choices we e he own could be because he ac
ha he game uses he playe ’s own physiological signals o igge ac ions inside he game could
c ea e a subconscious bond be ween he playe ’s physical and i ual, in-game body. I s ops being
a ma e o manipula ing he cha ac e inside he game – which does wha we ell him/he o do
ia keyboa d/mouse commands – and becomes a si ua ion whe e playe s can mani es hei will
by using hei body ac i ely o in e ac wi h he game wo ld.
7.4 Limi a ions
Ou in es iga ion is limi ed in he sense ha we buil a game p o o ype on one gen e om all
o hose ha cu en ly exis oday (e.g. Racing games, Pla o me s, e c.). In o de o assess he
po en ial o Bio eedback, mo e p o o ypes need o be buil in o he game gen es. Nacke e al. also
e e his in hei p e ious wo k and while ou wo k builds on hei s’, i is conce ned wi h a as ly
di e en game gen e (2D Pla o me s. 3D Fi s -Pe son Shoo e wi h supe na u al abili ies).
Secondly, he app op ia e ha dwa e, obus amewo ks and echnology o p ocess he da a
e u ned by he physiological de ices o Bio eedback games a e no ye a ailable as o his ime.
Calib a ion o he physiological signals be o e gameplay was done manually. The ac ha playe s
would need o pe o m calib a ion manually and equip he bio eedback de ices on hei own – a
p ocess ha ook us app oxima ely 15-20 minu es – makes Bio eedback in easible o he e e yday
consump ion o ideogames a he cu en ime.
Addi ionally, some o ou pa icipan s expe ienced undesi ed ac i a ions in he TEMP and
RESP senso s which we could no sol e du ing he de elopmen phase. The TEMP senso some-
imes picked up e y small ac i a ions ha should be conside ed alse posi i es. The RESP senso
did no ha e he expec ed a ia ion due o a e y small change in he ches olume while b ea hing.
This mos ly happened on pa icipan s wi h a mo e muscula ches a ea.
E e ybody has a di e en physiology: his is an aspec ha wasn’ epo ed in o he bio eed-
back games and should be ca e ully obse ed in u u e bio eedback p o o ypes.
51
Resul s’ Analysis and Discussion
52
Chap e 8
Fu u e Wo k
Ou ini ial explo a ion on his no el ype o Di ec Bio eedback opens se e al in e es ing esea ch
a enues o u u e p ojec s. To he bes o ou knowledge, ou p o o ype succeeded in combin-
ing, o he i s ime, mo e han one di ec physiological senso inside one game mechanic. I
will be in e es ing o see how new p o o ypes build on ou inding o u he imp o e use expe-
ience and inno a e on he design o bio eedback augmen ed gameplay mechanics. I would also
be s imula ing o see mo e heo e ical s udies on how le el design guidelines should be adap ed
o accommoda e o he idiosync asies ha bio eedback in oduces in he game design p ocess.
Las ly, i would be e en mo e exci ing o see hese same s udies applied o o he game gen es (e.g.
Ad en u e, Explo a ion and Role Playing Games).
One pa icula de ail we no iced is ha one o ou pa icipan s (P8) did no seem as pleased as
he o he pa icipan s ega ding he bio eedback in e ac ion. In his wo ds, his ype o in e ac ion
is no good o hose who dedica e hemsel es o highly compe i i e game en i onmen s (e.g.
online games like “S a c a II”,“Diablo III” o “League o Legends”). This pa icipan belie ed
ha his echnology was ideal o hose who like o app ecia e he s o y o he 3D en i onmen
o a game. In ou opinion, a s udy o compa e he impac o bio eedback echniques on epo ed
Fun be ween playe s wi h di e en playing philosophies would po en ially yield aluable esul s.
Playe s could, o example, be segmen ed as “na a i e app ecia o ”, “compe i i e”, “social co-
ope a o ” o “en i onmen ancie ”.1
We we e unable o d aw meaning ul conclusions on whe he con ex ual bio eedback mechan-
ics (e.g. closing a hand has one e ec i we a e holding a doo knob and ano he when g abbing
an enemy by he neck) had any e ec on he gameplay expe ience when compa ed wi h o he
ypes o bio eedback. In ou unimodal p o o ype, some playe s a i med hey we e con used when
1Au ho ’s no e: a he ime o he w i ing o his documen we did no know abou he exis ence o heo ies on
pe sonali y and playe ypes, and me ely ansc ibed (li e ally) he playe ypes sugges ed by P8. A e ecen ly aking
knowledge o Ba le’s “Hea s, Clubs, Diamonds, Spades” playe model [Ba 96], he ou play s yles desc ibed by
Ba le (Kille s, Achie e s, Explo e s, Socialize s) show an in e es ing simila i y o he concep s exposed by P8. In
conclusion, Ba le’s playe model and ela ed heo ies a e a possible s a ing poin o explo e he connec ion be ween
play s yles and he e ec i eness o Bio eedback.
53
REFERENCES
[NPI05] A u o Nakasone, Helmu P endinge , and Mi su u Ishizuka. Emo ion ecogni ion
om elec omyog aphy and skin conduc ance. In P oc. o he 5 h In e na ional
Wo kshop on Biosignal In e p e a ion, pages 219–222. Ci esee , 2005.
[NRON13a] Ped o A Noguei a, Rui Rod igues, Eugénio Oli ei a, and Lenna E Nacke. Guided
emo ional s a e egula ion: Unde s anding and shaping playe s’ a ec i e expe i-
ences in digi al games. In P oceedings o he 9 h Annual AAAI Con e ence on
A i icial In elligence and In e ac i e Digi al En e ainmen (AIIDE). Bos on, Mas-
sachuse s, USA, 2013.
[NRON13b] Ped o A Noguei a, Rui Rod igues, Eugénio Oli ei a, and Lenna E Nacke. A hy-
b id app oach a emo ional s a e de ec ion: Me ging heo e ical models o emo ion
wi h da a-d i en s a is ical classi ie s. In Web In elligence (WI) and In elligen Agen
Technologies (IAT), 2013 IEEE/WIC/ACM In e na ional Join Con e ences on, ol-
ume 2, pages 253–260. IEEE, 2013.
[NRON13c] Ped o A Noguei a, Rui Rod igues, Eugénio Oli ei a, and Lenna E Nacke. A
eg ession-based me hod o ligh weigh emo ional s a e de ec ion in in e ac i e en-
i onmen s. In P oceedings o he XVI Po uguese Con e ence on A i icial In elli-
gence (EPIA), Ang a do He oísmo, Aço es, Po ugal ( o appea , 2013), 2013.
[NTR13] Ped o A Noguei a, Vasco To es, and Rui Rod igues. Au oma ic emo ional eac ions
iden i ica ion: A so wa e ool o o line use expe ience esea ch. In En e ainmen
Compu ing–ICEC 2013, pages 164–167. Sp inge , 2013.
[PCH+11] Ma hew KXJ Pan, Jih-Shiang Chang, Gokhan H Himme oglu, AJung Moon,
Thomas W Hazel on, Ka on E MacLean, and Elizabe h A C o . Now whe e was i?:
physiologically- igge ed bookma king. In P oceedings o he SIGCHI Con e ence
on Human Fac o s in Compu ing Sys ems, pages 363–372. ACM, 2011.
[PSGO13] A inash Pa nandi, Youngpyo Son, and Rica do Gu ie ez-Osuna. A con ol-
heo e ic app oach o adap i e physiological games. In A ec i e Compu ing and
In elligen In e ac ion (ACII), 2013 Humaine Associa ion Con e ence on, pages 7–
12. IEEE, 2013.
[RBF+08] Lau a Rocchi, Ma co Benocci, Elisabe a Fa ella, Luca Benini, and Lo enzo Chia i.
Valida ion o a wi eless po able bio eedback sys em o balance con ol: p elimi-
na y esul s. In Pe asi e Compu ing Technologies o Heal hca e, 2008. Pe asi e-
Heal h 2008. Second In e na ional Con e ence on, pages 254–257. IEEE, 2008.
[RGP+10] Giuseppe Ri a, And ea Gaggioli, Fede ica Palla icini, Da ide Alge i, Alessand a
Go ini, and Claudia Repe o. Ubiqui ous heal h o he ea men o gene alized
anxie y diso de s. 2010.
[RSL05] P amila Rani, Nilanjan Sa ka , and Changchun Liu. Main aining op imal challenge
in compu e games h ough eal- ime physiological eedback. In P oceedings o he
11 h In e na ional Con e ence on Human Compu e In e ac ion, pages 184–192,
2005.
[SBC+11] Ca a E S epp, Deanna B i on, Chelsey Chang, Albe L Me a i, and Yoky Ma -
suoka. Feasibili y o game-based elec omyog aphic bio eedback o dysphagia
ehabili a ion. In Neu al Enginee ing (NER), 2011 5 h In e na ional IEEE/EMBS
Con e ence on, pages 233–236. IEEE, 2011.
60

REFERENCES
[SG06] J Da id Smi h and TC G aham. Use o eye mo emen s o ideo game con ol.
In P oceedings o he 2006 ACM SIGCHI in e na ional con e ence on Ad ances in
compu e en e ainmen echnology, page 20. ACM, 2006.
[SLR+11] E an A Suma, Belinda Lange, Albe Rizzo, Da id M K um, and Ma k Bolas. Faas :
The lexible ac ion and a icula ed skele on oolki . In Vi ual Reali y Con e ence
(VR), 2011 IEEE, pages 247–248. IEEE, 2011.
[TNS13] Luís Filipe Teó ilo, Ped o Al es Noguei a, and Ped o B andão Sil a. Gemini: A
gene ic mul i-modal na u al in e ace amewo k o ideogames. In Ad ances in
In o ma ion Sys ems and Technologies, pages 873–884. Sp inge , 2013.
[To 13] Vasco Pe ei a To es. De elopmen o bio eedback mechanisms in a p ocedu al
en i onmen using biome ic senso s. Mas e ’s hesis, 2013.
[WSN10] Qiang Wang, Olga Sou ina, and Minh Khoa Nguyen. Eeg-based "se ious" games
design o medical applica ions. In Cybe wo lds (CW), 2010 In e na ional Con e -
ence on, pages 270–276. IEEE, 2010.
[YMJ10] Geo gios N Yannakakis, Héc o P Ma ínez, and A na Jhala. Towa ds a ec i e
came a con ol in games. Use Modeling and Use -Adap ed In e ac ion, 20(4):313–
340, 2010.
61
REFERENCES
62
Appendix A
Ca d So ing - Plo ed Cha s
Fo ease o in e p e a ion o he da a p esen ed on Tables 6.4-6.5, we plo ed he esul s o he ca d
so ing p ocess. The esul s a e g ouped in he same manne hey a e p esen ed on Tables 6.4-6.5:
•Useless in-game / Co e in-game;
•Incomple e / Comple e;
•Imagina y / Realis ic;
•Relaxed / Exhaus ing;
•Simple / Complex;
•Con using / In ui i e.
No ice, howe e , ha he p esen ed compa ison pai s a e me ely one way o in e p e he col-
lec ed da a. O he pai s could be es ablished o ex ac use ul in o ma ion - o example: “Complex
s. Con using”, “Simple s. In ui i e” o “Realis ic s. Exhaus ing”.
63
Ca d So ing - Plo ed Cha s
Figu e A.1: Useless in-game s. Co e in-game.
64
Ca d So ing - Plo ed Cha s
Figu e A.2: Incomple e s. Comple e.
65

Ca d So ing - Plo ed Cha s
Figu e A.3: Imagina y s. Realis ic.
66
Ca d So ing - Plo ed Cha s
Figu e A.4: Relaxed s. Exhaus ing.
67
Ca d So ing - Plo ed Cha s
Figu e A.5: Simple s. Complex.
68
Ca d So ing - Plo ed Cha s
Figu e A.6: Con using s. In ui i e.
69