FORTH-ICS, TR-458 October 2015

FOUNDATION  FOR  RESEARCH  AND  TECHNOLOGY  -­    HELLAS  (FORTH)  ΙΝSTITUTE  OF  COMPUTER  SCIENCE  

Postal  address:  N.  Plastira  100,    GR-­700  13  Heraklion,  Crete,  Greece  Tel.:  +30  2810391600,  Fax:  +30  2810391601,  Email:  ics@ics.forth.gr,  URL:  www.ics.forth.gr

A virtual reality brainwave entrainment method for human augmentation applications 1

G. Papagiannakis1,2, E. Argento2, E. Baka1,2, M. Maniadakis2, P. Trahanias1,2

2Foundation for Research and Technology – Hellas (FORTH) Institute of Computer Science

100 N. Plastira Str., 70013, Heraklion, Crete http://www.ics.forth.gr

3University of Crete, Computer Science Department,

Voutes Campus,70013, Heraklion, Greece

Technical Report Number 458 October 2015

Keywords: augmented cognition, brainwave entrainment, virtual reality platform, technology-

enhanced learning

Abstract

In order to investigate the brain activity under rhythmic stimulation we developed a novel architecture for Augmented Cognition by integrating latest wearable Virtual Reality (VR) immersive Head-Mounted-Displays with wireless, mobile neuro-biofeedback systems for Real-time, Autonomous Cognitive State Detection, Manipulation and Augmentation. We integrated in our open-source VR framework both latest experimental Oculus Rift HMDs (DK1 and DK2) together with the Emotiv EPOC mobile EEG headset in an innovative, open architecture for user position, orientation and brainwave frequency tracking. We carried out a novel experiment with five groups tracking the brain frequency states of people while relaxing or performing certain sensorimotor tasks in an interactive, immersive 3D environment. The primary aim of this novel research is human cognition augmentation and improvement of task performance via a unique VR-induced entrainment methodology. In most of our experimental groups, we witnessed the transition of the beta-wave to the alpha-wave state. The brainwave entrainment at alpha frequency (10Hz) and relaxation via modern VR, using our open s/w framework, allowed us to verify for the first time enhanced task performance in a fast-learning setup. These first results empower us to plan future studies concerning clinical treatments in VR entrainment using modern wearable devices.

1This work was supported by the FORTH - Institute of Computer Science.

A virtual reality brainwave entrainment method for human augmentation applications

I.   INTRODUCTION From entertainment to industrial design and from education to psychiatry, virtual reality is becoming a powerful new tool for ludic, working, training and treating experiences. Enhancing virtual reality has been identified as a Grand Engineering challenge for the 21st century [1]. One of the cornerstone concepts in such an enhancing endeavor is the advancement of “Presence” in virtual environments. The common view that Presence was the “feeling of being there” [2] in the virtual environment rather than where the participant’s body was actually located, has recently being grounded on the ability of “doing there”, stressing the importance of interactivity [3], [4]. Mixed Reality refers to the suite of technologies that allow the partial (Augmented Reality) or total augmentation (Virtual Reality) of the human senses (virtual, auditory, haptic) with synthetic, computer generated worlds. In this work we aim to integrate modern, latest Virtual Reality (VR) s/w frameworks and h/w [5], [6] with Brainwave Entrainment methods for augmented cognition via enhanced learning during sensorimotor task performance.

Augmented Cognition encompasses methods and designs that harness computation and explicit knowledge about human limitations to open bottlenecks, address biases and deficits in cognition [7]. These ‘Closed Loop’ Systems use modern neuro- Scientific Tools, and other environmental information, to sense an individual’s cognitive state in real time to drive system adaptation and to maximize human information. This is done by measuring cognitive state and removing the burdens of technology to achieve optimal human-computer performance. This creates closed-loop human-computer interaction capability where the computer anticipates, predicts and augments the performance of the user [8]. As already stated in [9] improvement of human performance through integration of ‘convergent technologies: nano-bio-info-cogno’ now becomes feasible.

Brainwave Entrainment refers to the affection of brain areas and their response to rhythmic sensory inputs, such as sound or light. There are a lot of studies that indicate the effects of such a procedure applied in several groups of people, suggesting that a state of Brain Entrainment can affect behavior and cognition in different ways, e.g the enhancement of creativity and even of the technique of contemporary dancers, and the decrease of their anxiety state [10]. The primary aim of this novel research is to answer the following research question: is human cognition augmentation and improvement of task performance achievable via a modern, VR-induced brainwave entrainment methodology?

In most of our experimental groups that were carried out to answer the above question, we witnessed the transition of the beta-wave state to the alpha-wave state. The achieved brainwave entrainment at alpha frequency (10Hz) in modern VR, using our open s/w framework, allowed us to verify for the first time, enhanced task performance in a fast-learning setup. These first results empower us to plan future studies concerning clinical treatments in VR entrainment. The figure below illustrates the overview of our VR architecture [6] and an example of a subject undergoing the described experiment, using the mobile wearable HMD and EEG tracking headsets .

Fig. 1.  Overview of our novel, open VR platform (top) with the associated wearable h/w devices (Oculus Rift HMD and Emotiv EEG headset) woren by a subject of ou experimens (bottom). A detailed description of the class diagram of the VR architecture is provided in [6]

In the following Section II of this paper we describe in detail the relevant state-of-the-art. In Section III our method for our experiments and in Section IV our results for each group of the experiment. We conclude with the final Section V with the discussion of our results and future work.

II.   STATE OF THE ART Augmentation of human intellect: "Intelligence Amplification" (IA) as it was initially described in [11] is still

not within easy grasp in its entirety. However, recent advances in Virtual & augmented Reality [2], [3] , [12], [13], Brain-wave entrainment [14], Computational Neuroscience and Neurofeedback [15], [16], Computational creativity and psychology [17] open new avenues of disruptive research for future emerging technologies for Augmented Cognition and IA.

In the area of Mixed Reality [20] we have developed over the last decade several frameworks for interactive simulation of virtual characters [5],[19] featuring basic presence and interaction with virtual characters [3], [12], [6]. Similarly, other researchers have extensively been engaged in the usage of virtual characters in Mixed Reality environments [42] where a more complete list is provided in [21].

Brainwave Entrainment, achieved by photic or audio stimulation, has an extended historical path. In fact, the first known clinical application belongs to the French psychologist Pierre Janet, at the turn of the 20th century, in the late 1800s. A lot of studies followed afterwards by many important researchers such as [33], [34], [35] and [40]. However, they have not considered the possible effects of a photic stimulation. Therefore, after the initial demonstration of Berger in 1934 on electrical activity recorded from the human brain, Adrian and Mathews [36] showed that the Berger rhythm, which means alpha rhythm, could be further amplified by photic stimulation at the same frequency. In 1942, Dempsey and Morison [37] found that Brainwave Entrainment could also be induced by tactile stimulus and in 1959 Chaitran [38] reported entrainment effects with an auditory stimulus. Meanwhile, in 1956, W. Gray Walter [39] published the results of his researches on the emotional and psychological effects of a flickering light stimulation. Flickering light produced frequency-dependent sensations of pattern, movement and colour. The development of Brain Entrainment tools increased greatly from 1973, with Oster’s result [30] which evidenced the properties of binaural beats. Studies on the effects of Brain Entrainment on situations like pain, headaches, migraines, anxiety and stress followed in the 1980s, as well as the incorporation of specific devices which directly allowed the achievement of an entrainment, such as David Siever’s device [32]. This kind of research expanded in the 1990s to include learning and memory, learning disabilities, attention deficit hyperactivity disorder, behavioral problems and PMS. [31], [32].

Digital games are unique in their ability to motivate and stimulate people. Indeed, gamers often surprise themselves and are capable of much more than expected. The concept of Serious Games (SGs) is based on tacking advantage of these game assets to motivate students and ease their learning process [43]. Games also promote active pedagogy by placing learners in the position of central actor, thus providing them with a sense of power and control over their learning activities and curriculum [44]. Game mechanics also trigger emotions, such as happiness, surprise, impatience or even frustration. These emotions have proven to increase concentration and memorizing skills [45]. SGs have the advantage of offering virtual environments, ideal for simulating various infrastructures and situations in which learners can practice and develop their skills [46]. Interactive Learning Events (ILEs) is a modern term that is used to describe all different aspects of serious games, simulations and gamification that are employed for learning purposes [46] via modern digital computing platforms.

In the following figure we illustrate the position of our proposed architecture for human augmentation based on a novel, open VR platform that allows for brainwave entrainment at alpha frequency for enhanced learning, with respect to other mostly relevant associated technologies and methodologies described in the state-of-the-art.

Fig. 2.  Overview of the associated human augmentation technologies and methodologies with our new approach at the centre of the figure

I.   OUR METHOD

A.  Group Participants of the experiment In order to verify our premise, we conducted several experiments. Twenty people, sixteen to forty four years

old, both male and female, were fully informed about the procedure and participated in our study, which was held a the Foundation for Research and Technology in Crete with the collaboration of the Computer Science Department, of the University of Crete. Subjects were randomly placed in one of the five following groups: a) Control (C), b) Basic Entrainment (B), c) Simple Learning (SL), d) Entrainment Learning (EL), e) Entrainment Fast Learning (EFL). The duration of the experiment for each participant depended on the objectives of each group. The first and the last task of the experiment were the same for all groups. The first one was a VR familiarization task, based on an example demo of the Oculus Rift Head Mounted Display (HMD), so that participants would familiarize themselves with the 3D environment, orientation and navigation metaphors. The second task was a VR meditation application and the last one was a VR maze from which the participants had to find their way out in the shorted possible time.

B.  Procedure followed In order to study the influence of brainwave entrainment, combined with the active exploration of an

environment in VR, with the improvement of task performance and cognition augmentation, we organized the five groups mentioned above.

In each group the procedure began with the familiarization task for default duration of three to four minutes, depending also on the previous gaming experience of each participant. This familiarization task consists of the navigation through a large virtual house, located in Tuscany. At first, the Control group (C) after having the experience of this task they tried to solve the way-finding problem of the VR maze. The maze was divided in two parts, separated by two sets of ring-like marker structures, which were located at the one third and at the end of the correct route to the exit. Every time a participant met a ring, we noted the time as the purpose of the task was not only to get out of the maze but as quickly as possible. The time factor was relevant for being able to determine any possible improvements of task performance.

Augmented  Cognition

Brainwave  entrainment  &  neurofeedback

VR  entrainment  learning  platform  for  augmetned  

cognition

Gamified  VR  framework  and  

Presence

The Basic entrainment group (B) had one more task between familiarization and the maze, the one of entrainment, called relaxation.

The most commonly used methods of Brain Entrainment are to stimulate the brain at the desired frequency via flashing lights, different kinds of auditory tones, or a combination of both. The most common forms of auditory stimulation are based on isochronic, monaural or binaural beats [22]. In our experiment, we used an isochronic beat tuned in 10 Hz, which is an evenly spaced tone that simply turns on and off. While they were listening to this beat, the participants were transported into a virtual garden where they had almost 15 minutes to relax and concentrate. There was no task required from them to perform, except of navigating themselves in the garden only by changing movement direction with their head. The purpose of the task though, wasn’t only navigation but relaxation as well, which would lead to brainwave entrainment at alpha state. After having completed the relaxation task, the four participants of the group underwent the test of the maze, as it is the standard last task for all groups.

The third group was the one of Simple Learning (SL). The purpose of this group was to solve the problem of the maze after having seen the solution path towards the exit via an automated, prerecorded fly-through of the virtual camera in the 3D environment. The participants, after the familiarization task, navigated into the maze for five times so that they had enough time for planning the solution.

The Entrainment Learning group (EL) had to accomplish the same task as SL group except that, after familiarization task, they had the relaxation one in order for them to achieve brain entrainment and also during their entire challenge within the maze they were subject to the 10hz isochronic audio stimulus.

The difference between the EL group and the Entrainment Fast Learning group is the speed of the shown solution, which practically refers to the speed of the navigation through the maze. Minutely the tasks were familiarization, relaxation, fast learning and maze. This is a subtle difference that yield very promising results, since the EFL group provided the shortest timings and faster performance in way-finding in the maze.

Throughout this procedure we recorded the EEG data through the mobile, Emotiv EPOC mobile EEG headset, as mentioned before. Although EEG is at a disadvantage compared to other, more expensive methods, like MEG of fMRI, it offers the proper temporal resolution and, due to its equipment portability, allows a range and a freedom of movements that almost cannot be achieved with any other neuroimaging techniques [27] while can be worn in addition to the Oculus Rift HMD. EEG measurements entail the attachment of the electrodes to standardized locations of the scalp, so, as we used the headset mentioned above, we satisfied this condition. However, to facilitate this attachment we used liquid lenses or even we asked from certain participants to have their hair at a wet state, in order to improve the headset sensor conductivity. The following table describes the task performed by each group in our experiment.

Table I. Tasks performed by each group

Group name

Task Familiarization Relaxation Learning Fast

Learning Maze

C ✔ ✔

B ✔ ✔ ✔ ✔

SL ✔ ✔

EL ✔ ✔ ✔ ✔

EFL ✔ ✔ ✔ ✔  

C.  Signal processing of the tracked EEG data After having obtained the recorded EEG data, we used EEGlab and NBT for analyzing the data and signal

processing. EEGlab , as well as NBT, is a toolbox and graphic user interface, running on top of the MATLAB environment, developed for processing collections of single – trial and/or averaged EEG data of any number of channels [23].

To filter the data as well as clean them from any possible artifacts, we eployed the preprocessing tools, offered by both EEGlab and NBT [23] since sampling of the data lead to aliasing. In general, for being able to eliminate this aliasing process, it is suggested in the bibliography that the data should be low-pass filtered at frequencies at

least one third lower than the sampling rate [24], [25]. Given that in our case the sampling rate was 128 Hz, we used a basic FIR filter with the lower edge of the frequency at 45 Hz and the higher one at 55 Hz.

To clean the data, we removed further artifacts induced by eye blinking and abrupt head or body movements throughout the entire process. The Emotiv headset allowed every participant to move freely in every direction but this caused most of the times unavoidable artifacts. Given the purpose of our study, we could not eliminate those movements at their entirety. For an efficient and a more promising cleaning, we, sometimes, had also to reject a number of erroneous channels, marked as such by the toolkits.

In addition to all these, we run ICA for data decomposition, which helped us with the further EEG processing. In this case, EEGlab offered us the most reliable results composed of 14 components, as the exact number of the channels used. Subsequently, examining components activation and their maps, we removed the erroneous components.

Through this procedure we were able to determine the frequencies that corresponded to the microvolts we recorded wth the Emotiv headset throughout the experiment. At the end, it allowed us to plot either the spectrum of each channel or the one of the whole signal and the activity of each channel in relation to the proportional brain activity area.

II.   RESULTS Throughout the procedure we came up against a few people with different level of motion sickness asociated

with the VR 3D environment. Most of the times though, these participants were those who had limited 3D gaming experience/exposure.

A.  C group (Control) The results obtained from the C group were within a range of frequencies from about 11Hz to 17Hz. There are not big differences between the familiarization and maze tasks, which means that the participants almost maintained their primary frequency throughout the procedure. If, for example, the power spectrum of a participant presenteda peak at about 17Hz in the familiarization task, then, in the maze task, the frequency will peak at almost 16 Hz. We noticed though that at certain ocassions, in the familiarization task, there were more than one peak for the same spectrum, but usually concerning the same brainwave state, which in our case was Alpha or lower Beta. Fig. 3.  Power spectrum for familiarization task in C group. The several colored lines show the respective channels. Tha scalp map indicates the activity of the specific frequency in different brain areas. The red color represents the maximum intense activity. (C group – Familiarization task)

 

B.  B group (Basic entrainment) In the Basic Entrainment group (BE) we found almost the same results concerning the familiarization task. The power spectrum presented an average peak at about 15Hz. After having applied brainwave entrainment through the use of the isochronic beat mentioned before, we noticed changes in the associated EEG data. The spectrum peaked at around 11 – 12 Hz and we were able to verify this result through the spectrum plot given by EEGlab and NBT, but also through the activity of each channel individually.

In the figure below we are able to see the power spectrum of an individual channel and, also, its activity. The colors on top present this activity based on each color, as we mentioned before for the scalp map. The red one witnesses the highest possible activity of the specific brain area in the indicated frequency. Next to this box of colors there is a scale for deciphering the mean of each color. The graph below presents the power spectrum of the specific channel and we are clearly able to see that , in this example, frequency peaks at almost 12 Hz. Finally, the dots in the right present the position of each channel on the scalp. Every time we choose a channel to represent , the respective dot become red and the name of the channel is marked with the blue line shown in the middle box.

Fig. 4.  Example of an individual channel from the right hemisphere of the brain. The colors on the top represent the activity of the specific brain area where the electrode was placed. The graph below witnesses the power spectrum and the dots in the right show the position of each electrode on the scalp. (B group – Relaxation task)

Fig. 4 indicates the spectrum peak at around 12 Hz. We can observe that, except from the dominant frequency,

which we are every time able to identify, there is always, also a high activity at lower frequencies, which constitute delta state.

Throughout the maze task, the participants were able to maintain the obtained frequency, and hence the alpha state in which they were brain entrained though the 15 minutes of the relaxation task.

C.  SL group (Simple Learning) From this group on, the participants had to deal with an extra task, this of learning, in which, they had to watch

the solution, before they had the chance to solve the exit problem of the maze by themselves. This gave them the opportunity to plan a possible solution so that they would gain time by not making useless mistakes.

In the familiarization task, we had almost the same results as above, with an average spectrum peak at around 12 Hz. In this group, there is no relaxation task, so the procedure proceeds with the simply learning task. All the participants, throughout this task, presented a peak at around 11.5 Hz, which is clearly an alpha state, since they all were subject to the isochronic 10hz auditory stimulus.

Fig. 5.  Power spectrum plot for Simple Learning task in SL group using EEGlab. We witness that frequency peaks at around 11.5 Hz and we can see the activity of this frequency in the brain via the colors appeared in the scalp map.. (SL group – Simple Learning task EEGlab)

Fig. 6.  Power spectrum plot for Simple Learning task in SL group using NBT. We can witness that frequency peaks at around 11 – 12 Hz. The different colored lines show the respective channel (SL group – Simple Learning task NBT

D.  EL group (Entrainment Learning) The familiarization task of EL group gave us the expected result of a peak at around 11 – 12 Hz. Therefore,

in the relaxation task, the participants managed to find themselves slightly in a lower range of frequency peaking at about 10 Hz.

Fig. 7.  Example of a specific channel and its activity using NBT The colors on the top represent the activity of the specific brain area where the electrode was placed. The graph below witnesses the power spectrum and the dots in the right show the position of each electrode on the scalp.. (EL group – Relaxation task)

After this task, the participants were allowed to watch the solution of the maze problem in a kind of slow rhythm. During this task, the peak of the frequency was almost maintained at 10 to 11 Hz.

Fig. 8.  Power spectrum plot for Learning task in EL group using NBT. We can witness the frequency peak at almost 11 Hz..The colored lines indicate the diferrent channels. (EL group – Learning task NBT)

The same brain state with a peak of 10 to 11 Hz was observed in the maze task as well. The alpha state though,

as the dominant state, didn’t exclude the existence of other states as well.

E.  EFL group (Entrainment Fast Learning) In the familiarization task we noticed the expected result of the frequency peaking at around 10 Hz.

Fig. 9.  Power spectrum plot for Relaxation task in EFL group using EEGlab.We witness that frequency peaks at around 10 Hz. The activity in the brain during this frequency is shown via the scalp map above the graph. (EFL group – Relaxation task EEGlab)

Fig. 10.   Power spectrum plot for Relaxation task in EFL group using NBT. We can clearly witness the peak at almost 10 Hz. (EFL group – Relaxation task NBT)

Fig. 11.   Example of an individual channel from the right hemisphere of the brain in EFL group using NBT. The colors on the top represent the activity of the specific brain area where the electrode was placed. The more intense the red color, the more intense the activity. The graph below witnesses the power spectrum and the dots in the right show the position of each electrode on the scalp. (EFL group – Relaxation task NBT)

The three last figures present a complete image of the relaxation task, using both EEGlab and NBT. It is clearly indicated that the frequency has been maintained at 10 Hz, as it was in the previous task of familiarization.

After the relaxation task, the participants watched the solution of the maze in a faster pace, almost at double speed of the virtual camera fly-through compared to the previous group. The power spectrum peak, in this case, had slightly been increased in about 11 – 12 Hz. During the last task, the one of the maze, the spectrum maintain the peak at the same level of frequency, varying a bit according to each participant. We observed a range of frequencies from 8 to 12 Hz.

III.  DISCUSSION AND FUTURE WORK The main purpose of brainwave entrainment is to stimulate the brain in a desired frequency via auditory tones,

flashing lights, or a combination of both. There are five categories of these brainwaves, ranging from the least activity to the most activity: delta (0 - 4 Hz), theta (4-8 Hz), alpha (8-12 Hz), beta (12-25 Hz) and gamma (25- 100Hz). Brainwave Entrainment refers to the brain's electrical response to rhythmic sensory stimulation, such as pulses of sound or light. When the brain is given a stimulus, through the ears, eyes or other senses, it emits an electrical charge in response, called a Cortical Evoked Response. These electrical responses travel throughout the brain to become what you see and hear. This activity can be measured using sensitive electrodes attached to the scalp. The need of such a technique was created by the fact that a major amount of the useful brain frequencies are below the hearing range of the human ear. There are several studies though, which support the claim that the use of sound beats, is the most efficient for achieving brain entrainment. [26] However, the efficiency of such a brain entrainment depends on the kind of the beat. The binaural one is considered to be one of the less effective due to its modulation depth. Although listening to binaural beats produces the illusion that sounds are located somewhere within the head [30], it is supported that, by themselves, are unlikely to produce any kind of brain entrainment, but they do have a hypnotic and relaxing effect. Hence, we verified in our experiments that isochronic tones constitute the most effective entrainment method as they cause a strong auditory evoked response and moreover, most people find them enjoyable [29]. Motivated though, by the historical correlation between the relaxation state or the state of hypnogogia and creative insights, we tried to study through our research the benefits of an alpha wave brain entrainment in augmented cognition and improvement of task performance in a unique, VR-induced entrainment, open s/w framework [27]. We were also influenced by the use of virtual reality environment in education and training which brings many innovative advantages to people of all ages. Moreover,

virtual reality has already been used for therapeutically reasons [28] The combination of the 3D environment and the use of a specific kind of beat, isochronic beat, allowed us to have some really promising results for further research in the fields of education, learning and even therapy.

We clearly demonstrated that throughout this VR experiments the participants managed to maintain alpha brainwave state even from the very first task, without the need of an extra brain entrainment, only via the auditory isochronic stimulus. There is a study from 1970, which indicates that simply by watching TV, Beta waves are replaced by alpha waves in a very short time range of less than 1 minute [41]. This phenomenon works adversely as it leads to diminished concentration. Given the needs of our research though, we didn’t notice to have a negative result of this immediate transition to alpha state. Contrariwise, the possibility to achieve a brain entrainment simply with the use of a 3D environment allows us to take advantage of VR opportunities.

Given that we had already succeed to induce via VR brainwave entrainment to our subjects, the use of an isochronic beat led the power spectrum of each signal to a lower frequency peak, which illustrates that we achieved a deeper alpha state entrainment, with enhanced augmented cognition for the EFL group. Hence, subjects in this group were managed to learn faster the demonstrated solution to the maze exit by achieving faster exit times. We can thus conclude that VR-induced brainwave entrainment via modern, wearable h/w and an open s/w architecture is a viable path towards human augmented cognition.

ACKNOWLEDGMENTS The research leading to these results has received funding from the EU FET Proactive grant TIMESTORM - Mind and Time: Investigation of the Temporal Traits of Human-Machine Convergence. Special thanks to Stavros Kateros and Stylianos Georgiou for the VR application development.

REFERENCES [1]   NationalAcademyofEngineering, “Grand Challenges for Engineering,” www.engineeringchallenges.org, pp. 1–56, Sep.

2008. [2]   M. V. Sanchez-Vives and M. Slater, “From presence to consciousness through virtual reality,” Nature Reviews

Neuroscience, vol. 6, no. 4, pp. 332–339, Apr. 2005. [3]   S. Kateros, S. Georgiou, M. Papaefthymiou, G. Papagiannakis, and M. Tsioumas, “A comparison of gamified,

immersive VR curation methods for enhanced presence and human- computer interaction in digital humanities,” International Journal of Heritage in the Digital Era, pp. 1–14, Aug. 2015.

[4]   M. Slater, “Grand Challenges in Virtual Environments,” Front. Robot. AI, vol. 1, May 2014. [5]   Ponder, M., Papagiannakis, G., Molet, T., Magnenat-Thalmann, N., & Thalmann, D. (2003, Jul 1). VHD++ development

framework: Towards extendible, component based VR/AR Simulation Engine Featuring Advanced Virtual Character Technologies. Proc. of Computer Graphics International03, pp. 96-104, IEEE Computer Society Press, Tokyo.

[6]   G. Papagiannakis, P. Papanikolaou, E. Greassidou, and P. E. Trahanias, “glGA: an OpenGL Geometric Application Framework for a Modern, Shader-based Computer Graphics Curriculum,” Eurographics 2014 Education Papers, 2014, pp. 9–16.

[7]   D. Schmorrow, “Augmented Cognition: Achieving Human Computer Symbiosis with 21st century Human Factors Tools,” presented at the HCI International, 2014, pp. 1–32.

[8]   M. F. S. John, D. A. Kobus, J. G. Morrison, and D. Schmorrow, “Overview of the DARPA Augmented Cognition Technical Integration Experiment.,” IJHCI, vol. 17, no. 2, pp. 131–149, 2004.

[9]   W. Bainbridge and M. Roco, “Converging Technologies for Improving Human Performance.”,” NSF-WTEC Report, Kluwer Academic Publishers, ISBN 1-4020-1254-3, vol. 24, 2003.

[10]  J.H.Gruzelier, T.Thompson, E.Redding, R.Brandt, T.Steffert, “Application of Alpha/Theta neurofeedback and heart rate variability training to young contemporary dancers(State anxiety and creativity),” International Journal of Psychophysiology, Elsevier, May 2013

[11]  Engelbart, D.C. 1962. Augmenting Human Intellect: A Conceptual Framework, October 1962. Air Force Office of Scientific Research.

[12]  Egges, A., Papagiannakis, G., and Magnenat-Thalmann, N. 2007. Presence and Interaction in Mixed Reality Environments. The Visual Computer, Springer-Verlag 23, 5, 317–333.

[13]  Steptoe, W., Normand, J.M., Oyekoya, O., et al. 2012. Acting rehearsal in collaborative multimodal mixed reality environments. Presence (Camb.).

[14]  Zhuang, T., Zhao, H., and Tang, Z. 2009. A study of brainwave entrainment based on EEG brain dynamics. Computer and Information Science.

[15]  Gruzelier, J. 2008. A theory of alpha/theta neurofeedback, creative performance enhancement, long distance functional connectivity and psychological integration. Cognitive Processing 10, S1, 101–109.

[16]  Gruzelier, J.H., Thompson, T., Redding, E., Brandt, R., and Steffert, T. 2013. Application of alpha/theta neurofeedback and heart rate variability training to young contemporary dancers: State anxiety and creativity. International Journal of Psychophysiology, 1–7.

[17]  Kounios, J. and Beeman, M. 2009. The Aha!Moment: The Cognitive Neuroscience of Insight. Current Directions in Psychological Science 18, 4, 210–216.

[18]  Papagiannakis, G., Kim, H., and Magnenat-Thalmann, N. 2005. Believability and Presence in Mobile Mixed Reality Environments. Proc. of IEEE VR2005 Workshop on Virtuality Structures, Bonn, 1–4.

[19]  Papagiannakis, G., & Magnenat-Thalmann, N. (2007). Mobile Augmented Heritage: Enabling Human Life in ancient Pompeii. The International Journal of Architectural Computing, Multi-Science Publishing , 5 (2), pp.395-415.

[20]  R. Azuma, Y. Baillot, R. Behringer, S. Feiner, S. Julier, and B. MacIntyre, “Recent advances in augmented reality,” IEEE Computer Graphics and Applications, pp. 34–47, 2001.

[21]  G. Papagiannakis, G. Singh, and N. Magnenat-Thalmann, “A survey of mobile and wireless technologies for augmented reality systems,” Comp. Anim. Virtual Worlds, vol. 19, no. 1, pp. 3–22, 2008.

[22]  T.L.Huang, C.Charyton, “A comprehensive review of the psychological effects of brainwave entrainement.”, 3rd ed., vol. 14, No5, Alterntive therapies, September 2008, pp.38-49, in press

[23]  A.Delorme, S.Makeig, “EEGLAB: an open sourxe toolbox for analysis of single trial EEGdynamics including independent component analysis,”, Journal of Neuroscience Methods, Elsevier, vol. 134, Issue 1, March 2004, pp. 9-21, in press

[24]  F.C.Crow, “The aliasing problem in computer – generated shaded images”, Magazine Communication of the ACM, Vol.20, pp.799-805, November 1987, in press

[25]  I.Bilinskis, “Digital Alias – free Signal Processing”, pp.14-17, 71-76, ISBN 978-0-470-02738-7, 2007 [26]  D.Vernon, G.Peryer, J.Louch, M.Shaw, “Tracking EEG changes in response to alpha and beta binaural beats”,

International Journal of Psychophysiology, Elsevier, vol.93, pp.134-139, 2014, in press [27]  T.Thompson, T.Steffetr, T.Ros, J.Leach, J.Gruzelier, “EEG application for sport and performance”, Methods, Elsevier,

vol.45, pp.279-288, 2008, in press [28]  H.G.Hoffman, “Virtual Reality Therapy”, Scientific American, July 2004, in press [29]  D.Siever, “Entraining tones and binaural beats”, Mind Alive, unpublished [30]  G.Oster, “Auditory Beats In the Brain”, Scientific American, vol.4,pp.94-102, 1973 [31]  F.Casciaro, V.Laterza, S.Conte, M.Pieralice, A.Federici, O.Todarello, F.Orsucci, E.Conte, “ Alpha-rhythm stimulation

using brain entrainment enhances heart rate variability in subjects with reduced HRV”, World Journal of Neuroscience, Vol.3, No.4 , 2013

[32]  D.Siever, “Audio-Visual Entrainment: History anf Physiological Mechanisms”, Biofeedback Magazine, Vol.31, No2, 2003

[33]  G.Goldman, J.Segal, M.Segalis, “L’action d’une excitation inermittente sur le rythme de Berger” ( The effects of intermittent excitation on the Berger rhythms), C.R Societe de Biologie Paris, Vol.127, pp.1217-1220, 1938, in press

[34]  S.Bartley, “Relation of intensity and duration of brief retinal stimulation by light to the electrical response of the optiv cortex of the rabbit”, American Journal of Physiology, Vol.108, pp.397-408, 1934, in press

[35]  R.Jung ,” Das Elektroencephalogram und seine klinische Anwendung” (The electroencephalogram ans its clinical application, Nervenarzt,Vol.12,pp.569-591, 1973, in press

[36]  E.Adrian, B.Matthews, “The Berger rhythm: Potential changes from the occipital lobes in man”, Brain, Vo.57, pp.355-384, 1934, in press

[37]  E.Dempsey, R.Morison, “The interaction of certain spotaneous and induced cortical potentials”, American Journal of Physiology, Vol.135, pp.301-307, 1942, in press

[38]  G.Chaitran, M.Petersen, J.Lazarte, “Response to clicks from the human brain:Some depth electrographic observations”, Electroencephalography and Clinical Neurophysiology, Vol.12, pp.479-489, 1959, in press

[39]  G.W.Walter, “Colour illusions and abberations during stimulation by flickering light”, Nature, Vol.177, pp.710, 1956 [40]  J.Toman, “Flicker potentials and the alpha rhythm in man”, Journal of Neurophysiology, Vol.4, pp.51-61, 1941 [41]  H.E.Krugman, E.L.Hartley, “Passive Learning from television,” Oxford Journals,Vol.34,pp.184-190,1970, [42]  Wagner, D., Billinghurst, M., & Schmalstieg, D. (2006, Jan 1). How real should virtual characters be? Proc. of the 2006

ACM SIGCHI international conference on Advances in computer entertainment technology . [43]  Dondlinger, M J. 2007. Educational video game design: A review of the literature . Journal of Applied Educational

Technology 4 (1): 21-31. [44]  Mayo, M.J. 2007. Games for science and engineering education . Communications of the ACM 50: 30,35. [45]  National Research Council. 2000. How People Learn: Brain, Mind, Experience, and School: Expanded Edition.

Washington, USA: The National Academies Press. [46]  C. Centeno, “The Potential of Digital Games for Empowerment and Social Inclusion,” JRC Scientific and Technical

Report, pp. 1–172, Jul. 2013. [47]  Kapp, K.M., Blair, L., And Mesch, R. 2013. The Gamification of Learning and Instruction Fieldbook. John Wiley &

Sons.