Using EEG and Gamiﬁed Neurofeedback Environments to Improve

eSports Performance: Project Neuroprotrainer

Jose L. Soler-Dominguez

1 a

and Carlos Gonzalez

2 b

Instituto de Investigaci

on e Innovaci

on en Bioingenieria (I3B), Universitat Polit

ecnica de Val

encia,

Camino de Vera s/n, 46022, Valencia, Spain

Applied Computing for Learning & Well-being, Unidad TIC, Florida Universit

aria, 46470, Catarroja, Spain

Keywords:

eSports, Gamiﬁcation, Neuroscience, Neurofeedback Training.

Abstract:

Human performance has permanently been an objective for society and, speciﬁcally, for researchers. Tradi-

tional sports, board games, video games... A wide diversity of domains have recall the attention on the reasons

why top players perform better than the rest. eSports can be deﬁned as competitive multiplayer video games.

Nowadays, eSports have a huge impact on society with billions of people playing or consuming related content.

Being a relatively new universe, there is a wide gap when talking about applying scientiﬁc principles to per-

formance analysis and improvement on eSports. This paper tries to establish a new research topic, introducing

Virtual Reality and neuroscience as main frameworks to pursue a double objective: evaluate psycho-cognitive

characteristics of eSports players aiming to proﬁle them and, additionally, using that proﬁle to create custom

psycho-cognitive training plans. Neuroscience and EEG data from players have the ability of explaining the

complex decision making procedures that involve an individual action while playing. Neurofeedback training

(NFT) is a neuro-behavioral technique that will allow using real-time EEG data to drive a gamiﬁed environ-

ment aiming to adapt their brain activity to the optimum performance mode. This project aims, for the ﬁrst

time, to use neurofeedback within a gamiﬁed training environment in order to improve individual performance

on eSports.

1 INTRODUCTION

From chess (De Groot, 2014; Simon and Chase, 1988;

Chase and Simon, 1973) to air-pistol shooting (Cheng

et al., 2017), going through other massive (like foot-

ball (Savelsbergh et al., 2002)) or non-massive sports

(like squash (Abernethy, 1990)) have been object of

research aiming a double goal: (a) Identify and iso-

late psychological, cognitive and motor traits of best

performers (experts) and (b) Try to implement these

characteristics into non-expert through training pro-

grams.

For those researchers following the cognitive ap-

proach, a skilful performance in sport is tradition-

ally linked with the combination of a skilful percep-

tion and a skilful action (Craig and Watson, 2011;

Williams et al., 1999). Additionally other psycholog-

ical perspectives like decision-making (Chamberlain

and Coelho, 1993; Araujo et al., 2006) or risk tak-

ing (Pain and Pain, 2005; Kontos, 2004) have been

https://orcid.org/0000-0002-3819-9022

https://orcid.org/0000-0002-4984-8630

studied as relevant factors related to sports for the last

decades.

With the bloom of neurosciences and EEG acces-

sibility and affordability, several labs started to intro-

duce this neuronal component to the usual psycho-

logical, cognitive and motor dimensions. Brain plays

an important role on sport performance and attending

to this is needed to study its implications on the 21st

century massive sport: competitive video games, the

so called eSports.

In recent years, there has been a growing inter-

est in the research potential of the application of elec-

troencephalography (EEG) to the analysis of cortical

activity in the most relevant phases of action in vari-

ous sports: preparation and execution (Vernon, 2005;

Perry et al., 2011; Park et al., 2015; Xiang et al.,

2018).

The main goal of this project is to implement a

roadmap in order to transfer this previous work on

traditional sports and the potential beneﬁts of neuro-

science and neurotraining on individual performance,

to eSports.

278

Soler-Dominguez, J. and Gonzalez, C.

Using EEG and Gamiﬁed Neurofeedback Environments to Improve eSports Performance: Project Neuroprotrainer.

DOI: 10.5220/0010314502780283

In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 1: GRAPP, pages

278-283

ISBN: 978-989-758-488-6

Aiming to this, it will be needed to:

1. Identify which EEG signals have the ability of de-

scribing an expert eSports player (for different eS-

ports).

2. Design and develop a gamiﬁed environment

where neurofeedback could be implemented, sup-

porting the training of players.

3. Develop a NFT program in order to evaluate the

real outputs over the individual performance.

2 eSports RESEARCH

Social interest on eSports is continuously growing,

even surpassing the popularity of some traditional

sports (BusinessTelegraph, 2020), having for exam-

ple, more audience than the rest of professional

leagues except the National Football League.

In parallel, the research production on eSports is

also acquiring some relevance, as shows the last sur-

vey paper (Reitman et al., 2020) on eSports research.

Since the ﬁrst work on eSports was written on 2002

(Bryce and Rutter, 2002), Google Scholar currently

returns more than 30.000 articles with the key word

eSport.

Even when the research approach to eSports could

come from different domains like Sports Science,

Cognitive Science, Law, Sociology, Business, Media

Studies or Informatics (Reitman et al., 2020), this pa-

per tries to introduce a new approach, using the neu-

roscience lens.

Following the Guttman’s taxonomy of modern

sports (Guttman, 1992), as it is visible on Fig.1, eS-

ports will fall inside the Intellectual Contests class.

Intellectual here could be understood as ”leaded by

mind” but it is impossible to separate cognition from

psycho-motor relevance in eSports.

There is a reduced amount of papers focused on

understanding the characteristics that distinguish and

the ones which make similar eSports and traditional

sports (Railsback and Caporusso, 2018).

By the same way, most eSports are collective

games where teamwork and global strategy and tac-

tics play a relevant role in the ﬁnal result. This project

is focused on improving the individual contribution of

each player. More accurately, is focused on improv-

ing performance of each single action executed within

a match.

Attending to the different types of eSports, fol-

lowing a taxonomy genre-leaded, it is possible to es-

tablish some main categories such as MOBA (Multi-

player Online Battle Arenas) like League of Legends

or Dota 2, FPS (First-Person Shooters) like Valorant

or Rainbow Six, RTS (Real Time Strategy) like Star-

craft 2 or Age of Empires, CCG (Collectible Card

Games) like Legends of Runeterra or Hearthstone or

Sports games like NBA, NHL or FIFA.

Even considering the particular differences be-

tween each main genre, all of them have some com-

mon traits: they are collective competitive games

(mainly) and they divide the gameplay in two dimen-

sions, micro (each individual action, short-term con-

sequences) and macro (mid/long-term strategy, team-

work). This project aims to improve micro perfor-

mance using NFT.

3 NEUROFEEDBACK

APPLICATIONS IN SPORTS

PERFORMANCE

As Xiang et al. (2018) summarized on their meta-

analysis about neurofeedback training for sport per-

formance, expertise on a certain sport is linked to a

superior cognitive-motor processing and this, with the

reduction of task-irrelevant processes and the increase

of task-relevant processes.

Some authors analyzed the importance of physi-

cal abilities and demostrated that motor skills are a

intrinsical part of eSports (Hilvoorde and Pot, 2016).

There is a relevant corpus of works where these

ﬁndings are explained (Xiang et al., 2018):

• Expert practitioners of different visuomotor based

activities like golf, archery, and target shooting

evidenced lower cerebral cortical activation when

compared with beginners (Hung et al., 2008).

• This lower cortical activity has also been found

jointly with a lesser activation in the language

zones of the left hemisphere (Hatﬁeld et al.,

2004).

• In elite athletes, preparation of precise visuo-

motor performance is related to a low cortical ac-

tivation (Del Percio et al., 2009).

The neuroProTrainer project aims to establish a

translation from this ﬁndings within the sports do-

main to the eSports, focusing on the micro-gameplay.

Micro-gameplay represents the individual perfor-

mance in single actions in front of macro-gameplay,

a concept that deﬁnes the more strategic, long-term

performance of a team, normally.

Using EEG and Gamiﬁed Neurofeedback Environments to Improve eSports Performance: Project Neuroprotrainer

279

Figure 1: Guttman’s modern sports taxonomy.

4 METHOD

This research is based on a basic but with a high level

of ecology BCI (Brain Computer Interface) which is

the Emotiv Insight (Fig. 2). It is a 5-channel dry sen-

sors EEG device that measures activity from all corti-

cal lobes of the brain, providing in depth information.

This BCI delivers six performance metrics that

will help to establish a neural proﬁle of players, both

resting and playing. The set of metrics consists of

(Emotive, 2019):

• Stress (FRU) represents a measure of comfort

with a certain activity. Higher stress levels are

related to negative implications, feeling over-

whelmed or simply not being able to make a par-

ticular action due to adverse consequences of fail-

ing. Low level of stress could foster performance

but a higher level could have terrible effects on

individuals.

• Engagement (ENG) quantiﬁes how immersed is

the player on the tasks he is involved. ENG

could be expressed as a combination of atten-

tion and concentration. It’s the opposite of bore-

dom. This metric presents an increased physio-

logical arousal, beta waves presence and low al-

pha waves.

• Interest (VAL) is the degree of attraction or aver-

sion to the current stimuli, environment or activ-

ity. In the psychological domain is called Valence.

The interest is related with the enjoyment of the

current task, normally throwing higher values of

interest when the user is enjoying the task and

lower values when he/she is disliking the activ-

ity. Medium Interest values usually show a neutral

feeling about the task.

• Excitement (EXC) is an awareness or feeling of

physiological arousal with a positive value. The

detection of this performance metric is designed

to detect short-term changes on excitement, usu-

ally within a few seconds.

• Focus (FOC) is a measure of ﬁxed attention to

one speciﬁc task. A high level of attention is usu-

ally accompanied by a low level of task switching

which correlates positively with low level of focus

and distraction.

• Relaxation (MED) quantiﬁes the ability to achieve

a calm state of mind. It also represents a metric

of the ability to reduce activation levels, rest and

recover from intense concentration.

Establishing a neuro-proﬁle of players will pro-

vide relevant information about individual character-

istics of those who better perform on a video game

and the differences with those of lower performance.

With this information, a customized training plan

could be designed.

Additionally, following the works listed previ-

ously, NFT will be applied in order to get perfor-

mance metrics to their optimum level. NFT could be

deﬁned as a speciﬁc paradigm based on operative con-

ditioning, in which the individual learns how to inﬂu-

ence the electrical activity of his brain through differ-

Figure 2: Emotiv Insight.

GRAPP 2021 - 16th International Conference on Computer Graphics Theory and Applications

280

ent mental states (Kerick et al., 2001; Pop-Jordanova

and Demerdzieva, 2010; Hammond, 2011).

NFT could be based on different frequencies,

coming from different cortical areas (Fajardo and

Guzm

an, 2016; Xiang et al., 2018). Different corti-

cal areas of the brain produce different rhythms:

1. Delta (1Hz-4Hz): It is a wave associated with

sleep that is present in states of muscle relaxation,

so it is very useful in stretching and sports warm-

ups (Behncke, 2004).

2. Theta (4Hz-8Hz): This frequency has been linked

to pre-competition stress. Additionally it corre-

lates also with positive moods. (Thompson et al.,

2008).

3. Alpha (8Hz-12Hz): This wave promotes focus-

ing, diminishing emotions, toughts and distrac-

tions (Beauchamp et al., 2012).

4. SMR (Sensorimotor Rhythm, 12Hz-15Hz): This

rhythm is related to body movement and concen-

tration. A higher sensorimotor signal during the

last instant before an action initiation has been

demonstrated to be linked to better performance

(Gruzelier et al., 2010; Cheng et al., 2017).

5. Beta (13Hz-30Hz): Is associated with the muscle

contractions that happen in isotonic movements

and are suppressed prior to and during movement

changes (Baker, 2007). Also related to states of

wakefulness and mental activity, states of alert-

ness and active concentration (Beauchamp et al.,

2012).

Relevance and contribution of each wave have

to be evaluated aiming to create a subset of cortical

rhythms that have inﬂuence on eSports players per-

formance. Jointly, this subset of waves and the per-

formance metrics given by the BCI, will compose the

NeuroScorecard (NS) for each individual. Attending

to their current level and the pre-ﬁxed goal for each

parameter of the NS, a customized NFT plan will be

designed.

NFT will be delivered by a gamiﬁed training en-

vironment (GTE), not deﬁned yet, with a simple de-

sign premise: Correct feedback (contributes to the

pre-deﬁned goal) will be translated into a positive ac-

tion within the gamiﬁed environment and wrong feed-

back (that one on the contrary sense of the pre-deﬁned

goal) will represent a negative action in the gamiﬁed

environment.

As an example, if the GTE is similar to a ”Flappy

Bird” game (Fig. 3), when an individual successfully

tunes its cortical activity as expected, the main char-

acter should elude the obstacles of the level. But, if

the tuning is wrong, the avatar will collide with an

obstacle, loosing some health or one life.

Figure 3: Flappy Bird, by .GEARS Studios, as a reference

for the GTE.

This GTE should be responsive to users’ activity,

changing values like timing, difﬁculty or sounds at-

tending to current performance. Different levels of

intensity of training could be conﬁgured. Introducing

machine learning elements at design level could fos-

ter a procedural GTE (PGTE) that will be dinamically

set up for each user, taking into consideration his/her

starting point and evolution.

Aiming to determine if NFT is effective, the train-

ing program will be delivered to three different groups

with three different feedbacks:

• Right feedback.

• Wrong feedback.

• No feedback.

Following this experimental design it will be pos-

sible to isolate the inﬂuence of NFT on individual per-

formance.

5 CONCLUSIONS AND FURTHER

WORK

The purpose of this paper is to provide a solid founda-

tion and methodology for an innovative ﬁeld of study:

the application of the NFT to eSports.

Starting from the extensive previous work apply-

ing EEG and NFT on sports and due to the strong sim-

ilarities between certain actions on traditional sports

and other actions on competitive video games, the

working hypothesis of this project is that NFT could

help players to improve their micro (individual per-

formance on direct actions).

Using a mix of cortical signals: (i) Performance-

metrics delivered by the BCI; (ii) Direct waves inter-

pretation it will be possible to establish a cortical ac-

tivity reference for expert players that, additionally,

will serve as road-map for aspirants.

Once the pro-players cortical proﬁle has been de-

ﬁned, a PGTE will be developed. It will be used to

deliver a NFT program in order to develop their neu-

ral control in order to increase performance within the

Using EEG and Gamiﬁed Neurofeedback Environments to Improve eSports Performance: Project Neuroprotrainer

281

game, trying to mimic pro-players cortical activity.

If both the PGTE and the NFT obtain positive re-

sults, this tools could be used by professional and am-

ateurs eSports teams as a valuable development re-

source.

Next steps of this project will be related to estab-

lishing a link between complementary physiological

signals and performance, using like eye tracking or

electro-dermal activity (EDA).

6 FUNDING

This project has been funded (2020 and 2021) by

the Regional Government, Generalitat Valenciana,

through the Conselleria de Innovaci

on, Universi-

dades, Ciencia y Sociedad Digital department, under

the Emergent Research Groups (ERG) program.

7 AUTHOR CONTRIBUTIONS

All authors listed have made a substantial, direct and

intellectual contribution to the work, and approved it

for publication.

REFERENCES

Abernethy, B. (1990). Expertise, visual search, and infor-

mation pick-up in squash. Perception, 19(1):63–77.

Araujo, D., Davids, K., and Hristovski, R. (2006). The eco-

logical dynamics of decision making in sport. Psy-

chology of sport and exercise, 7(6):653–676.

Baker, S. N. (2007). Oscillatory interactions between sen-

sorimotor cortex and the periphery. Current opinion

in neurobiology, 17(6):649–655.

Beauchamp, M. K., Harvey, R. H., and Beauchamp, P. H.

(2012). An integrated biofeedback and psychologi-

cal skills training program for canada’s olympic short-

track speedskating team. Journal of clinical sport psy-

chology, 6(1):67–84.

Behncke, L. (2004). Mental skills training for sports: A

brief review. Online J Sport Psychol, 6(1).

Bryce, J. and Rutter, J. (2002). Killing like a girl: Gendered

gaming and girl gamers’ visibility. In CGDC Conf.

BusinessTelegraph (2020). Esports viewership vs tradi-

tional sports: Revenue and statistics for investors.

Chamberlain, C. J. and Coelho, A. J. (1993). The perceptual

side of action: Decision-making in sport. In Advances

in psychology, volume 102, pages 135–157. Elsevier.

Chase, W. G. and Simon, H. A. (1973). The mind’s eye in

chess. In Visual information processing, pages 215–

281. Elsevier.

Cheng, M.-Y., Wang, K.-P., Hung, C.-L., Tu, Y.-L., Huang,

C.-J., Koester, D., Schack, T., and Hung, T.-M.

(2017). Higher power of sensorimotor rhythm is as-

sociated with better performance in skilled air-pistol

shooters. Psychology of Sport and Exercise, 32:47–

53.

Craig, C. and Watson, G. (2011). An affordance based ap-

proach to decision making in sport: discussing a novel

methodological framework. Revista de psicolog

ıa del

deporte, 20(2):689–708.

De Groot, A. D. (2014). Thought and choice in chess, vol-

ume 4. Walter de Gruyter GmbH & Co KG.

Del Percio, C., Babiloni, C., Bertollo, M., Marzano, N.,

Iacoboni, M., Infarinato, F., Lizio, R., Stocchi, M.,

Robazza, C., Cibelli, G., et al. (2009). Visuo-

attentional and sensorimotor alpha rhythms are related

to visuo-motor performance in athletes. Human brain

mapping, 30(11):3527–3540.

Emotive (2019). Performance metrics.

Fajardo, A. and Guzm

an, A. L. (2016). Neurofeedback,

aplicaciones y eﬁcacia. Interdisciplinaria, 33(1):81–

93.

Gruzelier, J., Inoue, A., Smart, R., Steed, A., and Steffert, T.

(2010). Acting performance and ﬂow state enhanced

with sensory-motor rhythm neurofeedback comparing

ecologically valid immersive vr and training screen

scenarios. Neuroscience letters, 480(2):112–116.

Guttman, A. (1992). From ritual to record. Sport and reli-

gion, pages 143–152.

Hammond, D. C. (2011). What is neurofeedback: An up-

date. Journal of Neurotherapy, 15(4):305–336.

Hatﬁeld, B. D., Hauﬂer, A. J., Hung, T.-M., and Spald-

ing, T. W. (2004). Electroencephalographic studies of

skilled psychomotor performance. Journal of Clinical

Neurophysiology, 21(3):144–156.

Hilvoorde, I. v. and Pot, N. (2016). Embodiment and fun-

damental motor skills in esports. Sport, Ethics and

Philosophy, 10(1):14–27.

Hung, T.-M., Hauﬂer, A. J., Lo, L.-C., Mayer-Kress, G.,

and Hatﬁeld, B. D. (2008). Visuomotor expertise and

dimensional complexity of cerebral cortical activity.

Medicine & Science in Sports & Exercise, 40(4):752–

759.

Kerick, S. E., McDowell, K., Hung, T.-M., Santa Maria,

D. L., Spalding, T. W., and Hatﬁeld, B. D. (2001).

The role of the left temporal region under the cogni-

tive motor demands of shooting in skilled marksmen.

Biological psychology, 58(3):263–277.

Kontos, A. P. (2004). Perceived risk, risk taking, estimation

of ability and injury among adolescent sport partic-

ipants. Journal of pediatric psychology, 29(6):447–

455.

Pain, M. T. and Pain, M. A. (2005). Essay: risk taking in

sport. The lancet, 366:S33–S34.

Park, J. L., Fairweather, M. M., and Donaldson, D. I.

(2015). Making the case for mobile cognition: Eeg

and sports performance. Neuroscience & Biobehav-

ioral Reviews, 52:117–130.

GRAPP 2021 - 16th International Conference on Computer Graphics Theory and Applications

282

Perry, F. D., Shaw, L., and Zaichkowsky, L. (2011).

Biofeedback and neurofeedback in sports. Biofeed-

back, 39(3):95–100.

Pop-Jordanova, N. and Demerdzieva, A. (2010). Biofeed-

back training for peak performance in sport-case

study. Macedonian journal of medical sciences,

3(2):113–118.

Railsback, D. and Caporusso, N. (2018). Investigating the

human factors in esports performance. In Interna-

tional Conference on Applied Human Factors and Er-

gonomics, pages 325–334. Springer.

Reitman, J. G., Anderson-Coto, M. J., Wu, M., Lee, J. S.,

and Steinkuehler, C. (2020). Esports research: A lit-

erature review. Games and Culture, 15(1):32–50.

Savelsbergh, G. J., Williams, A. M., Kamp, J. V. D., and

Ward, P. (2002). Visual search, anticipation and exper-

tise in soccer goalkeepers. Journal of sports sciences,

20(3):279–287.

Simon, H. and Chase, W. (1988). Skill in chess. In Com-

puter chess compendium, pages 175–188. Springer.

Thompson, T., Steffert, T., Ros, T., Leach, J., and Gruzelier,

J. (2008). Eeg applications for sport and performance.

Methods, 45(4):279–288.

Vernon, D. J. (2005). Can neurofeedback training enhance

performance? an evaluation of the evidence with im-

plications for future research. Applied psychophysiol-

ogy and biofeedback, 30(4):347.

Williams, A. M., Davids, K., and Williams, J. G. P. (1999).

Visual perception and action in sport. Taylor & Fran-

cis.

Xiang, M.-Q., Hou, X.-H., Liao, B.-G., Liao, J.-W., and Hu,

M. (2018). The effect of neurofeedback training for

sport performance in athletes: A meta-analysis. Psy-

chology of Sport and Exercise, 36:114–122.

Using EEG and Gamiﬁed Neurofeedback Environments to Improve eSports Performance: Project Neuroprotrainer

283