Excerpt
Table of Contents
Table of Figures
1 Introduction
1.1 Social Neuroscience
1.2 EEG as a Neuroimaging Tool
1.3 EEG Hyperscanning
1.4 EEG Neurofeedback
2 Inter-Brain Synchronization in Social Interaction
2.1 The Classification of Brain Waves
2.2 Recent Studies on Social Interaction
2.2.1 Intra- and Inter-Brain Synchronization during Guitar Playing
2.2.2 Inter-Brain Synchronization during Social Interaction Tasks
2.3 Summary
3 Neurofeedback
3.1 Recent Studies on Neurofeedback
3.1.1. Neurofeedback Research with FMRI and FNIRS
3.1.2. Neurofeedback Research with Multiple Participants
3.2 Summary
4 Study Proposal
4.1 Methods
4.1.1 Participants
4.1.2 Task Instructions
4.1.3 Electrophysiological Recording
4.1.4 Neurofeedback Calculation
4.1.5 Experimental Design
4.1.6 Questionnaires
4.2 Results
4.3 Discussion
5 Conclusion
References
Table of Figures
Figure 1. Tentative forward model for action coordination
Figure 2. Within brain synchronization
Figure 3. Networks of within-brain modules and hyper-brain modules in the delta band after the onset of coordinated guitar play
Figure 1. Experimental setup of the study by Dumas and colleagues
Figure 2. Inter-brain synchronization in the alpha-mu band, in the beta band, and in the gamma band
Figure 6. The tasks of the study by Yun and colleagues
Figure 7. Experimental setup of the study by Yun and colleagues
Figure 8. Phase synchrony between brains in the theta band and in the beta band
Figure 9. Within-brain networks in the line on top and between brain networks in the line below of the pairs during the kissing experiment
Figure 10. Setup of the experiment by Duan and colleagues
Figure 11. The presentation of feedback by Duan and colleagues
Figure 12. Setup of the experiment by Kovacevic and colleagues
Figure 13. Presentation of the Feedback in the four experimental conditions by Kovacevic and colleagues
Figure 14. Experimental setup of the participants
Figure 15. Presentation of the two versions of the neurofeedback visualization on the screen...
Figure 16. English version of the post-questionnaire
1 Introduction
During their whole lifespan, human individuals interact with other humans every day and in many different situations. Humans are social creatures who have social needs and constructs, and who live in groups. Living in a group requires coordinated social interaction between individuals, which leads to social constructs, and social bonds. Coordinated social interaction can be observed in many daily situations, such as carrying home food from the supermarket together with a friend, cleaning up at home with a relative or helping a roommate getting dressed (Allport, 1924; Argyle & Cook, 1976; Emery, 2000; Sänger). There are several factors which constitute and modify the social interaction context, such as the time point of the social interaction, the mood of the individuals, and the social frame of the situation (Liu & Pelowski, 2014). For instance, a person normally reacts to other individuals differently during daytime compared to nighttime, as the visual input is completely different and therefore information about the identity and the mood of a person who passes by is calculated slower than at daytime. Also, if a person has received bad news recently and therefore is in a bad mood, the person’s reaction will probably differ to input such as personal updates thereafter. Also, time pressure and resulting stress play a role in human social behaviour, among many other factors (McGrath, Kelly, & Machatka, 1984).
Therefore, a healthy human individual is able to modulate and adapt its own behaviour in social interaction and thereby enables successful interaction with others. From an evolutionary perspective, this ability is a fundamental and pivotal requirement for survival (Liu & Pelowski, 2014), because without flexible adaptation to other persons’ behaviour, social interaction is not possible. During evolution, primates formed social bonds, which required adaptation to the given situation and which developed over time. Forming and keeping social bonds is the most important feature that humans have gained during evolution, because it is the foundation of prosocial emotions and therefore forms moral behaviour (Sterelny, 2012). In addition, research has found that social bonds are important for the physical health of human beings (Eisenberger & Cole, 2012). This means that interpersonal interaction is not only a tool, which has grown within the evolution of humans, but that it is an important factor for every single individual.
Further, in present society, it is widely known that working together can lead to better results than working alone. This means that the performance outcome of a team is not simply the sole summation of the capacities of the single individuals, but indeed higher than this, because humans who work in teams are healthier and more motivated (Busch, 2009). In addition, it has been revealed that in companies, teams generate more innovative product solutions compared to a single person (Gemünden & Högl, 1998). Nevertheless, the present essay will not focus on performance measures of group work, but rather on biological correlates of social interaction. Thus, it is interesting to investigate social interaction in a scientific manner to gain a better understanding of both the behaviour of interacting individuals and the ongoing processes in the body, which contribute to social interaction. In the following parts of this chapter, the area of social neuroscience and the method of neurofeedback will briefly be introduced.
1.1 Social Neuroscience
In recent years, the field of social neuroscience has emerged. Research in this field combines neuroimaging techniques and research into social cognitive behaviour. Thus, multiple research areas are connected in order to open up a completely new field which addresses questions from psychology and neuroscience. This field of research emerged as questions about social cognition started to emerge in both the field of psychology and biology about thirty years ago. Because of the increasingly wider application of functional magnetic resonance imaging (fMRI) in neuroscientific research, the field of social neuroscience has also been growing in the past decade (Stanley & Adolphs, 2013). Since then, research recruiting neuroscientific methods into social cognitive processes has successfully contributed not only to the field of psychology, but also to the field of economics (Adolphs, 2010).
In general, neuroscience provides a link between biological and psychological approaches. These approaches can be linked to social behaviour as it has been found that neural processes are not isolated from social context but rather that social context can influence neural processes. There are two main mechanisms in cognition which are interesting for social neuroscience: firstly, neural regulations are reflected by innate mechanisms, which are automatic and cognitively impenetrable; and secondly, volitional mechanisms of self-regulation, which are acquired processes (Adolphs, 2003). Importantly, the first category is shared with other species, whereas the second category has only been found elaborately in human beings (Dunbar, 2003). The architecture of neural processes in social cognition is complex, because first it consists of diverse pathways, which process information, and second, because it is recruited in a variety of different situations and conditions. In addition, the diverse pathways recruited in social cognitive processes can form neural networks which can, depending on the situation, interact with each other. Therefore, complicated processes are recruited to organize interaction in the neural pathways and networks so that it is challenging to find a neuroscientific approach which is able to comprehensively explain processes of social cognition (Adolphs, 2003). It has also not become clear so far whether processing of social information is only done in one domain of the brain or if there are several recruited domains in social interactional processes. Further, it is still an open question whether there are universal principles of social information processing or if factors such as culture or inter-individual difference of cognitive abilities contribute to this (Adolphs, 2010). Therefore, the main aim of social neuroscience is to identify neural processes in the brain and map them to cognitive processes in the mind within the framework of social interaction.
Applications of social neuroscience research are, for instance, the exploration of neural processes of fundamental socio-cognitive tools such as empathy (Jean Decety & Lamm, 2006), the investigation of disorders which are related to an impairment of social cognition such as autism spectrum disorder (Klin, Jones, Schultz, Volkmar & Cohen, 2002), the characterization and treatment of a disruption of social cognition in psychiatric diseases, for instance depression (Davidson, Pizzagalli, Nitschke, & Putnam, 2002; Stanley & Adolphs, 2013), the investigation of what the ‘social brain’ is and what structures it contains (Adolphs, 2010; de Vignemont & Singer, 2006), the investigation of what the nature of social interaction is and what bounds it has in humans in normal social interaction, and also in social transactions in the internet (Stanley & Adolphs, 2013), the perception of social relevant stimuli such as faces (Aylward et al., 2005), and the neuroscience of decision making (Sanfey, 2007) in a social context. Nevertheless, there are tensions about the priority of the different applications of social neuroscience. At the moment, important questions are being addressed concerning the behaviour of both humans and animals, although it is hard to achieve a focus in the field and coherent research questions among the scientific groups (Stanley & Adolphs, 2013).
Research in the field of social neuroscience has been conducted using brain imaging techniques such as electroencephalography (EEG) (Uddin, Iacoboni, Lange, & Keenan, 2007), (functional) near-infrared spectroscopy (NIRS/ fNIRS) (Chatel-Goldman, Schwartz, Jutten, & Congedo, 2013), fMRI (Baron-Cohen et al., 1999; T. Singer & Lamm, 2009), diffusion tensor imaging (Barnea-Goraly, Kwon, Menon, Eliez, Lotspeich, & Reiss, 2004), magnetoencephalography (Halgren, Raji, Marinkovic, Jousmäki, & Hari, 2000), positron emission tomography (Ruby & Decety, 2003), eye-tracking (Boraston & Blakemore, 2007; Riby & Hancock, 2009), and measurement of behaviour (Fiske, Cuddy, & Glick, 2007). Although research in social neuroscience with single individuals has contributed greatly to the field, it is of great interest to also conduct studies with more than one participant as the term ‘social’ per definition implies a certain dynamic, inter-individual interaction. If dynamics of social real life interaction are taken into account, the nature of the mechanisms for social emotions and perception can be understood even better than in studies with single individuals (Liu & Pelowski, 2014). Neural markers of social interaction can be investigated in studies of people who are interacting, even though experiments which involve the measurement of two brains instead of only one are methodologically more advanced and less feasible than studies with only one participant (Konvalinka & Roepstorff, 2012). Given this, the number of neuroimaging studies with two participants has been rather small so far.
There are rhythms within the body, which correlate with or reflect perceptual and cognitive processes. Therefore, it is of general interest to measure actions and reactions of the human body, because the body has specific rhythms which are omnipresent in living individuals (Glass, 2001). These rhythms are physiological and central to life. The most prominent of these rhythms is the rhythm of the heartbeat. Another example is the rhythmic motion of the limbs in movements of the body, for instance in walking. In addition, the sleep-wake cycle is a bodily rhythm, as well as is the menstruation cycle in female individuals. Even hormonal regulation of growth and metabolism, as well as the process of digestion, work in a certain rhythm. The rhythms within the body do not only interact with each other, but also interact with the environment. As the environment outside the body is noisy and quickly-changing, a high number of feedback systems is crucial to make bodily functions work in an orderly manner. As well-working bodily rhythms serve as an indicator of health, disease is linked to a disruption of normal rhythmic processes of the body and to abnormal bodily rhythms (Glass, 2001).
Because of the complexity of these rhythms, it is crucial to investigate how the bodily rhythms interact with each other and how each of them works. Therefore, it is of great interest to research how bodily rhythms influence each other in human social interaction (Müller & Lindenberger, 2011), because here not only one body reacts to its environment, but the other body is also able to react back so that there is a mutual process of feedback and interaction between the different rhythms of the body. Also, the human brain has rhythmic patterns which can be measured already in prenatal development. The electrical activity of the human brain starts at week 17- 23 of embryonal development, and the human brain is assumed to be already fully developed when the child is born, which means that all parts of the brain are existent as in the adult brain(Teplan, 2002). A fully developed human cortex contains about 1011 neurons with a density of approximately 104 neurons per mm3 (Nunez & Cutillo, 1994). During the lifespan, the number of neurons decreases, whereas the number of synaptic fibers between the neurons increases, so that connections between the neurons become stronger. Nevertheless, as the number of neurons decreases over time, the total number of synapses decreases, too. Compared to other large brain structures, i.e., the cerebellum and brain stem, the cerebrum is the most important structure for the investigation of neuronal processes which are involved in social interaction, as it is in charge of the expression of emotions, complex analysis of the situation and behaviour of other individuals, among other functions (Teplan, 2002).
FMRI research in the field of social neuroscience has revealed that there are specific regions in the cortex which are active in social cognition tasks (Babiloni & Astolfi, 2014). Particularly, the temporo-parietal junction was found to be active in tasks involving the estimation of others’ intentions, goals and desires (Schurz, Radua, Aichhorn, Richlan, & Perner, 2014) and it forms a network with the medial prefrontal cortex. This activation was found regardless of whether the judgment of such intentions and goals of others was positive or negative (Babiloni & Astolfi, 2014; Van Overwalle & Baetens, 2009). More specifically, the right supra-marginal gyrus, which is a small area within the tempo-parietal junction, is assumed to play a crucial role in egocentricity (Silani, Lamm, Ruff, & Singer, 2013). Accordingly, this brain area is assumed to be the key area in the avoidance of egocentrically biased judgements and is an addition to the other activated brain networks in social interaction. It is suggested that the right supra-marginal gyrus regulates early processes in perception to disentangle information about the self and others.
In addition to investigating where social processes take place in the brain, EEG studies have been conducted to investigate whether social interaction has an effect on the brain waves in the involved individuals and how bodily rhythms adapt to each other in social interaction.
1.2 EEG as a Neuroimaging Tool
In this essay, EEG is the brain imaging technique in focus, although fMRI studies and fNIRS studies will also be introduced. EEG is a non-invasive imaging technique which is widely used in scientific studies and which displays activity generated by structures inside the brain by recording activity from the scalp’s surface via electrodes filled with a conductive substance (Teplan, 2002). It measures the electric current in the brain which is produced by the activation of neurons in the cerebral cortex and is then conducted during excitation of the dendritic synapses (branches of the neuron) in the pyramidal neurons. The summation of postsynaptic potentials from pyramidal cells causes differences in the electrical potentials. The potentials are graded, because during the flow of current there are electrical dipoles between the body of the neuron (soma) and the dendrites. The electrical current flow in the brain is caused by ions (mostly Na+, K+, Ca++, and Cl-) which are transmitted through neuronal channels in the cell membrane. If the population of the involved neurons is large enough, the electric activity can be recorded on the surface of the head (Teplan, 2002).
EEG is a widely-applied tool in neurology and research as it is a very fast brain imaging technique that can record complex patterns of brain activity within under a second after presentation of a stimulus. However, it has a worse spatial resolution than other brain imaging techniques, such as fMRI and positron-emission tomography. EEG is used in a wide range of clinical applications, which includes locating tumors or damaged areas in the brain, monitoring cognitive engagement, investigating epilepsy and the origin of seizures, monitoring brain development in humans and animals, investigating sleep disorders, and producing biofeedback, among others (Bickford, 1987). Patterns which can be observed in EEG waves are modified by various factors, which are caused by biochemical, neuroelectric, metabolic, circulatory, and hormonal processes, as well as behavioural factors (Bronzino, 1995).
1.3 EEG Hyperscanning
In recent years, neuroimaging techniques have extended from recording one person to the simultaneous recording of multiple subjects although the neural mechanisms underlying social interaction have not been well understood so far. Behavioural studies using the approach of coordination dynamics have shown that interpersonal coupling occurs in action coordination. Research has shown that there are similar phenomena when two people rhythmically synchronize the movement of their limbs and when a single person performs interlimb coordination (Schmidt & Richardson, 2008). Thus, also the brain activity of multiple persons is interesting to be investigated in social interaction. As behavioural dynamic coupling has been detected in interpersonal interaction, it is clearly also of interest whether there is also coupling to be found on a neural basis in social interaction. The technique of simultaneously scanning more than one participant is called hyperscanning and can be applied with brain imaging techniques which measure hemodynamic activity or with techniques which measure electrophysiological activity in the brain, for instance fMRI, EEG, fNIRS (Duan et al., 2013; Dumas, Lachat, Martinerie, Nadel, & George, 2010). A number of reviews have emphasized the potential of hyperscanning studies to address questions about the social brain (Babiloni & Astolfi, 2014; Dumas et al., 2010; Hasson, Ghazanfar, Galantucci, Garrod, & Keysers, 2012; Sänger et al., 2012). Hyperscanning’s potential lies in the fact that only a small technical step is needed to record brain activity of two or more subjects at the same time to achieve a great next step in the methodology. Therefore, the number of hyperscanning studies has been growing more and more (Dumas et al., 2010). More precisely, hyperscanning unveils interpersonal mechanisms of the brain and neural substrates which underlie social interaction,which cannot be measured in single subject recordings (Duan et al., 2013; Scholkmann, Holper, Wolf, & Wolf, 2013). In the last decade, neural coupling dynamics have been repeatedly observed in interpersonal joint action (Konvalinka & Roepstorff, 2012).
1.4 EEG Neurofeedback
Neurofeedback is the ability to voluntarily use learned control of one’s own electrophysiological activity in the brain (Kamiya, 1969). There are parameters of EEG which can be brought under control when the ongoing changes are displayed in real-time in a training process via an EEG biofeedback loop (Egner & Gruzelier, 2004). It is widely agreed that neurofeedback is a conditioning process in the brain based on reinforcement of desired brain states which leads to self-regulatory processes of those brain states (Barth, Strehl, Fallgatter, & Ehlis, 2016). However, not only the principle of trial-and-errorlearning is a key criterion for the acquisition of voluntary control of the brain processes. Other aspects, such as individual motivation and the ability to identify strategies, also play an important role in neurofeedback (Strehl, 2014).
The brain imaging techniques which are used for neurofeedback range from EEG, transcranial Doppler sonography (Duschek, Schuepbach, Doll, Werner, & Del Paso, 2011), fMRI (Ruiz, Buyukturkoglu, Rana, Birbaumer, & Sitaram, 2014), to NIRS (Cheng, Li, & Hu, 2015; Kober, Wood, Kurmann, Friedrich et al., 2014). Neurofeedback has been used as a scientific method since the 1980s, receiving reasonable scientific validation in the last decade and experiencing a revival as a scientific method since the Society of Applied Neuroscience has encouraged the more extensive use of it (Gruzelier, 2014c). Since then, it has been applied among a wide range of areas in neuroscientific research.
In the past decade, there has been increasing evidence that specific brain rhythms can be effectively trained by the use of an electroencephalogram (Weber, Köberl, Frank, & Doppelmayr, 2011). In addition, it is an inexpensive method compared to other brain imaging techniques and it is a non-pharmaceutical treatment to support individuals who suffer from neurological diseases, i.e., epilepsy. Therefore, neurofeedback is especially applied in clinical intervention and rehabilitation, for instance in the treatment of attention deficit hyperactivity disorder (ADHD) (Ali, Mahmud, & Samaneh, 2015; Leins, Goth, Hinterberger, Klinger, Rumpf, & Strehl, 2007), treatment of epileptic seizures (Sterman & Egner, 2006; Tan, Thornby, Hammond, Strehl, Canady, Arnemann, & Kaiser, 2009), intervention in autism spectrum disorder (Coben, Linden, & Myers, 2010), the therapy of addiction disorders (Cox, Subramanian, Linden, Lührs, McNamara, Playle, Hood, Watson, Whittakar, Sakhuja, Issen, 2016), the treatment of tinnitus (Milner, Lewandowska, Ganc, Ciesla, Niedzialek, &Skarryzski, 2016), learning disorders (Fernandez, Herrer, Harmony, Diaz-Comas, Santiago, Sanchez, Bosch, Fernandez-Bouzas, et al., 2003), sleeping disorders, such as insomnia (Arns & Kenemans, 2014), emotional disturbances (Joshua Raymond, Sajid, Parkinson, & Gruzelier, 2005), and post-stroke rehabilitation (Sacchet & Gotlib, 2016). These interventional functions of neurofeedback are grounded in the enhancement or suppression of EEG features, such as the power of a certain frequency band (Egner & Gruzelier, 2003). For example, epileptic motor seizures can be controlled with learned enhancement of the sensorimotor rhythm which has a bandwidth of 12-15 Hz as well as through controlling and modulating slow cortical potentials (Rockstroh, Elbert, Birbaumer, Wolf, Düchting-Röth, Reker, Daum, Luzetnberger, Dichgans, 1993; Sterman & Macdonald, 1978).
The enhancement of the sensorimotor rhythm is also used as treatment for ADHD in the cortical frequency components between 15-18 Hz (Fuchs, Birbaumer, Lutzenberger, Gruzelier, & Kaiser, 2003; Lubar, Swartwood, Swartwood, & O’Donnell, 1995). In addition, it is applied to brain-computer communication (Birbaumer, Ghanayim, Hinterberger, Iversen, Kotchoubey, Kübler, Perelmouter, Taub, & Flor, 1999). Furthermore, neurofeedback can be used in relaxation training as an additional therapeutic instrument in the treatment of post-traumatic stress disorder (Peniston & Kulkosky, 1991) and alcohol use disorder (Peniston & Kulkosky, 1989). This can be done by training to increase activity in the frequency band between 5-8 Hz (theta) and suppressing the frequencies between 8-11 Hz (alpha) with closed eyes, whichenables the patients to relax (Egner & Gruzelier, 2003).
Neurofeedback can be also applied in healthy individuals; however, there are only a few studies on the effects of neurofeedback on the cognitive performance of participants not suffering from any neurological disorders (Weber et al., 2011).
In the clinical applications of neurofeedback, a high amount of training sessions is needed to ensure that participants can voluntarily control their brain waves, i.e., the electroencephalographic rhythms, and to follow the given instructions. In some cases, the participants are never able to acquire control over their brain rhythms. Weber et al. (2011) tested whether it is possible to determine the ability to perform neurofeedback in an experiment, in which they compared performance in an early training session to performance in later sessions. They found that in later sessions (from the eleventh training session onward), the performance level can be predicted by earlier sessions with more than 90% accuracy. From this, one can draw the following two conclusions. Firstly, there seems to be a threshold which determines whether the participants are able to learn to control their brain waves or not. Secondly, eleven sessions seem to be sufficient to make clinical patients learn how to use neurofeedback.
Similarly, there is evidence that neurofeedback protocols lead to successful learning of voluntary control over one’s own brain waves in healthy participants (Gruzelier, 2014b, 2014d). In this essay, research on inter-brain synchronization in EEG hyperscanning studies will be reported as well as studies on EEG neurofeedback. In the next two sections, a summary will be given on how EEG hyperscanning and EEG neurofeedback physically work and an overview about recent findings in EEG hyperscanning and EEG neurofeedback will be given. In the subsequent sections, a study proposal will be elaborated on involving a combination of EEG hyperscanning and EEG neurofeedback in social interaction.
2 Inter-Brain Synchronization in Social Interaction
In scientific research, sometimes more than one participant is recorded at the same time in the frame of a testing, which is called hyperscanning. This can be done with several brain imaging techniques. EEG is advantageous as it is portable, has a very good temporal resolution, is completely non-invasive, and has low costs compared to other brain imaging techniques such as fMRI (Liu & Pelowski, 2014; Michel & Murray, 2012). Thus, EEG hyperscanning offers the possibility to measure the brain waves of multiple subjects online in real social interaction (Konvalinka & Roepstorff, 2012; Liu & Pelowski, 2014). In dyadic EEG hyperscanning, two EEG devices are used to record the brain signals from the participants simultaneously in the same laboratory (Liu & Pelowski, 2014). The two EEG devices can be adjusted and calibrated by trigger signals with fixed amplitude to improve synchronization and sensitivity during the recording (Babiloni & Astolfi, 2014). Therefore, EEG hyperscanning studies have been conducted to study inter-brain connectivity in interacting participants (Astolfi et al., 2010; Dumas et al., 2010; Liu & Pelowski, 2014).
Besides the advantages of EEG hyperscanning, there is also one disadvantage, which is the poor spatial and anatomical resolution of the EEG signal. Thus, it is not possible to locate where exactly the signal is generated in the brain (Corbetta, 2012; Cui, Bryant, & Reiss, 2012; Lieberman, 2010; Liu & Pelowski, 2014). This is the case because the EEG signal is measured on the surface of the scalp and only records alternating electrical potentials, but does not match them to anatomical structures in a 3D-manner as fMRI scans do. When it comes to the coupling of two brains in specific brain structures and systems, this limitation of EEG recordings can be an obstacle (Liu & Pelowski, 2014). Nevertheless, as EEG hyperscanning has the advantages named above, it is still a very common method in social neuroscience studies with multiple subjects. Moreover, source localization analyses such as dipole modeling, low resolution electrical tomography (LORETA), local autoregressive average (LAURA), multiple-signal classification (MUSIC) algorithm, artificial neural network (ANN) analyses as well as different beamforming approaches can be useful to improve localization accuracy of EEG activity and underlying neural sources (s. Grech et al., 2008 for review). Furthermore, NIRS is a useful tool in research with multiple brains. Similar to the fMRI technique, NIRS measures changes in the metabolism of blood in the brain. There are advantages of NIRS over fMRI, firstly that it is less expensive, secondly that it is less bound to a place and better portable and thirdly that an NIRS recording is less sensitive to movement than fMRI recordings are. In addition, participants are not exposed to as much noise during a NIRS scan as they are in an fMRI scanner and the preparation is faster than in fMRI and EEG setups. Therefore, the environment of NIRS recordings is assumed to be more natural than in fMRI (Obrig, 2014; Piper et al., 2014).
In the following sections of this chapter, the underlying principle of EEG hyper-brain networks will be explained by introducing the dominant brain waves in the human brain and their frequency bands. Next, a model for joint action will be introduced. Finally, studies on hyper- brain networks in social interaction will be summarized.
2.1 The Classification of Brain Waves
Electrical brain activity recorded with EEG normally forms sinusoidal waves, which are measured from peak to peak. The waves have an amplitude between 0.5 and 100 gV. Compared to an electrocardiogram, the amplitude is 100 times lower (Teplan, 2002). In the analysis of the EEG signal, the spectrum of the sinus-waves is transformed so that dominant frequencies become visible. The spectrum of an EEG signal ranges from 0 to 1/2 of the sampling frequency. Certain frequencies of the spectrum become more dominant depending on the individual brain state and therefore give an insight into ongoing processes in the brain (Teplan, 2002). Classical views have suggested that anatomical correlates of mind activity can be matched to anatomical structures so that activity in a region and functional connectivity between brain areas reflect a certain cognitive process in a one-to-one relationship. More recently, it has been suggested that functional connectivity and anatomical correlates of behaviour can be matched to cognitive processes; however, this is only the geometric dimension of cognitive processes (Klimesch, 2012). There is another component of cognitive processes crucial to the understanding of brain activity, namely time (Buzsaki & Draguhn, 2004; Klimesch, 2012). Accordingly, brain oscillations generate and organize a rhythmic temporal structure in the brain, which contributes to the understanding of brain processes (Arnal & Giraud, 2012; Chakravarthi & VanRullen, 2012). The most dominant brain oscillations have been categorized into four to five groups, which are the delta band (1.5-4 Hz), the theta band (4-10 Hz), the alpha band (10-13 Hz), the beta band (13-30 Hz), and the gamma band (> 30 Hz) (Buzsaki & Draguhn, 2004; Teplan, 2002). In the normal healthy adult human brain, alpha is the dominant frequency in the scalp EEG (Klimesch, 1999). Unlike humans, lower mammals display the theta band as their most dominant frequency in the scalp EEG (Lopes da Silva, 1992). Therefore, the theta band is more widely found in animals than in humans and has larger power, so that changes in frequency and in power can be observed more easily. In contrast, in the human brain, advanced methods are needed to detect changes in the theta frequency (Klimesch, 1999). This fact shows how difficult it is to define strict categories of brain waves, especially if their functionality and interplay is not taken into account (Steriade, 2006).
Klimesch (2012) has explored how to categorize brain oscillations according to their frequency by using harmonic mathematical calculation with the alpha band as the center. Accordingly, frequency bands in the brain can be calculated around the alpha band if the following assumptions hold. Firstly, in the awake, conscious brain, the alpha band is the most dominant and resonant frequency, it is the most dominant frequency in the conjunction of the brain waves and has its center frequency at approximately 10 Hz. Secondly, harmonic frequencies relative to 10 Hz, which is the frequency of the alpha band, provide for ideal communication between other brain oscillations and the alpha band. Thus, the center frequencies of the other brain oscillations are harmonic to the alpha band. From this, the center frequencies belonging to the other frequency bands can be calculated from the center frequency of the alpha band (fa, fa = 10 Hz) (Klimesch, 2012).
fa / 4 = 2.5 Hz for the delta band (range from 2-3 Hz),
fa / 2 = 5 Hz for the theta band, (range from 4-6 Hz),
(fa = 10 Hz), (range from 8-12 Hz),
fa * 2 = 20 HZ for the beta band, (range from 16-25 Hz)
fa * 4 = 40 Hz for the gamma band, (range from 32 Hz).
Thirdly, frequency separation between frequency domains and the width of each frequency band is calculated with the golden mean (g = 1.618) frequencies relative to alpha (Pletzer, Kerschbaum, & Klimesch, 2010). The golden mean denotes a certain ratio of two quantities and has been assumed to underlie the human biological system as it determines the architecture of amino acids (Rakocevic, 1998). Our sense of evolution and time is also assumed to be based on the golden mean (Datta, 2003; Weiss & Weiss, 2003). In human brain oscillations, it is assumed that in each domain, the border frequency is able to interact with the center frequency, meaning that the borders are a part of the specific domain and do not overlap with the adjacent domains. Accordingly, it is assumed that within a frequency band, the frequencies at the border communicate with the center frequency and that they do not overlap with the frequency band of the adjacent domains (Klimesch, 2012). To calculate the frequency bands harmonic to the alpha band, the center frequency of the upper frequency band is divided by the golden mean and the center frequency of the lower frequency band is multiplied by the golden mean. For instance, the borders between the theta band and the alpha band are calculated in the following way:
In the first step, the center frequency of the alpha band is divided by the golden mean: fa / g = 10 Hz / 1.618 = 6.2 Hz. In the second step, the center frequency of the theta band is multiplied by the golden mean: fth * g = 5 Hz * 1.618 = 8.1 Hz.
From this follows that the upper frequency border for theta is 6.2 Hz, which is maximally separated from the alpha frequency but still within the theta band, and that the lower border for the alpha band is 8.1 Hz (see Klimesch, 2012, 613).
Importantly, the center frequency may shift within each frequency band to either couple or decouple with another frequency domain. For example, if the alpha band shifts from 10 Hz to 8 Hz, it separates from theta as the relation is no longer harmonic. Thus, the factor with which alpha (8 Hz) has to be divided to obtain theta (5 Hz) is no longer 2 but is 1.618 (because 8 Hz / 1.618 = 5 Hz). Nevertheless, if the alpha frequency remains at 10 Hz, ideal coupling with the theta frequency can be provided. In summary, this approach gives an explanation of how the frequency domains are distributed in a global structure (Klimesch, 2012).
On these grounds, it is easy to imagine how frequencies which belong to one global system can interact with each other, as brain oscillations are assumed not to consist of non-interacting categorical frequency domains. Research has shown that interaction of different frequencies of brain oscillations represents different brain processes. Thus, it is assumed that cognitive processes are reflected by different frequencies of brain oscillations (Klimesch, 2012). This means that, for instance, oscillations in the theta band frequency are associated with information processing (Klimesch, 1999), whereas alpha oscillations appear to be related to accessing information about the environment (Klimesch, 1997, 2011). Moreover, oscillations at 20 Hz in the beta band are associated with cognitive control and motor activity, and gamma band activity at 40 Hz and higher reflects perceptual and cognitive processes in cortical networks (Klimesch, 2012). The size of neural networks determines the specific frequencies within them. So, the smaller a neural network is, the faster are the frequencies of network oscillations (Buzsaki & Draguhn, 2004; Lakatos, Shah, Knuth, Ulbert, Karmos, & Schroeder, 2005; von Stein & Sarnthein, 2000).
As frequencies within different frequency bands are able to interact with each other, they are able to couple and to build networks across frequencies to enable brain cells to integrate information and to communicate with each other. This phenomenon is called cross-frequency coupling (Varela, Lachaux, Rodriguez, & Martinerie, 2001). Cross-frequency coupling reflects the exact timing between different rhythmic brain oscillations (Jensen & Colgin, 2007; Jirsa & Müller, 2013; Müller et al., 2016). In addition, it permits selective control of distributed functional cell assemblies (Canolty, 2012). Moreover, cross-frequency coupling increases different dimensions of information integration within the brain (Buzsâki, 2006; Buzsaki & Draguhn, 2004; Varela et al., 2001). Research has shown that cross-frequency coupling seems to play an important role in elementary brain functions and processes: in learning, working memory (Fell & Axmacher, 2011), encoding of nouns (Schack & Weiss, 2005), auditory information processing (Isler, Grieve, Czernochowski, Stark, & Friedman, 2008), accurate timing of brain oscillatory rhythms across frequencies, communication, and neuronal computation (Canolty, 2012; Müller et al., 2016). Cross-frequency coupling occurs between several brain frequency bands depending on the task. For instance, Schack & Weiss, (2005) found cross-frequency coupling between the theta and gamma band in encoding and consolidation of nouns. In addition, Jirsa & Müller, (2013) found that timing of oscillatory brain rhythms are reflected in cross-frequency coupling of the delta and alpha frequency band, and Palva & Palva, (2007) investigated the changes of phase-coupling between alpha band (10 Hz) and beta band (20 Hz) oscillations depending on the task. Yet, it is possible that frequency bands of different domains interfere in a spurious way meaning that the cross-frequency synchronization happens erroneously (Klimesch, 2012). This is because the excitatory phases of two frequencies, f1 and f2, can either have a harmonic relationship and meet regularly and frequently or can meet in a non-harmonic relationship and interact irregularly and infrequently. This depends on their numerical ratio. Only harmonic coupling offers an ideal basis for functional exchange of two oscillatory bands (Nikulin, Nolte, & Curio, 2012).
2.2 Recent Studies on Social Interaction
Recently, several EEG hyperscanning studies focused on neural synchrony using action coordination paradigms. In these studies, it is hypothesized that interpersonally coordinated actions, as well as synchronous interpersonal brain oscillations, support social interaction. Accordingly, in interpersonal action coordination, the brain oscillations synchronize between subjects by means of feedback of the sensory and motor system (Sänger, Lindenberger, & Müller, 2011). Brain oscillations have certain properties because of which one can assume that they are able to adapt interpersonally as they are fast, spatially distributed, functionally related, support perception, and support motor function. These properties play a crucial functional role in mutual adaptation of brain oscillations in interpersonal coordinated social interaction because of the following three reasons: Firstly, the information which is exchanged in interpersonal action coordination is conducted rapidly so that the interaction happens within milliseconds to allow for coordination with the partner without any time lag (Roelfsema, Engel, König, & Singer, 1997; Sänger et al., 2011). Secondly, it is important that information, which is functionally related, is bound even though the origin of this information is spatially distributed so that it can be conducted rapidly (Sänger et al., 2011; Varela et al., 2001). Thirdly, the feedback within the brain which plays a functional role in joint action coordination, arises within the areas of perception and motor function so that brain oscillations are connected to these areas in order to conduct the feedback rapidly (Kilner, Baker, Salenius, Hari, & Lemon, 2000; Makeig & Jung, 1996; Sanes & Donoghue, 1993; Sänger et al., 2011). From this follows that when individuals interact behaviourally and coordinate their motor activity, inter-brain activity should be observably coherent (Ben-Ari, 2001, 2002; Singer, 1995). Sänger et al., (2011) assume that the representation of action coordination between persons can be explained in a three-layered model.
Abbildung in dieser Leseprobe nicht enthalten
Figure 3. Tentative forward model for action coordination. Source: own illustration based on Sänger et al. (2011: 656).
Figure 1 shows the model for action coordination which consists of three interacting layers. In this model, it is assumed that in interpersonal action coordination each individual’s brain processes can be assumed to be threefold. The first layer on the bottom of the figure is the “Individual Forward Model”, which shows the intended action of one person, which is in turn a result of the joint goal. The joint goal depicted at the left of the figure has an impact on all three layers of the model. The intended action of a person leads to an action intention in the sensory motor system,an efference copy and a motor command which causes an action effect. The action effect is compared to the efference copy via a loop through the sensory system and the result of this comparison is either sensory congruency or sensory discrepancy. The following actions are influenced by the result of the feedback to reach the joint goal. The second layer in the middle of the figure is the “Representation of Other’s Forward Model”. This layer describes the action intention of the other agent(s) in joint action and is assumed to have a corresponding structure to the individual layer. So, the sensory system of one person compares the intended action effects of the other agents to the actual action effects. The third layer on top of the figure is called “Representation of Joint Forward Model”. This layer is assumed particularly in individuals, who are highly skilled in interactive tasks. In this layer, the joint intention and the joint action are represented and the effects of the joint action are predicted in a more abstract manner, separated from the individual contribution. Further, this layer contains a feedback loop, as in the other layers, which measures the outcome of the action and the approximation of the expected joint action effect. The different colours in the model reflect the different representations of the self and other in joint action coordination within one person. Yellow elements reflect elements which are assigned to the actor herself, green elements reflect the other person or other people depending on how many individuals participate in the interaction, blue elements reflect the “suprapersonal representation of the joint action”, and orange elements reflect observable effects of joint action.
In summary, each of the layers has its own sensorimotor feedback and the layers represent the actors in interpersonal action coordination. Importantly, there is not only one layer per actor but also an additional layer for the common goal, which shows that the computation of action coordination requires a high cognitive capacity. Therefore, as mentioned above, it is important that brain oscillations are fast, receive feedback and are functionally bound.
In the following passages, some recent EEG hyperscanning studies in coordinated joint social interaction will be summarized. Previous studies have found that social interaction has neural correlates in the frontal areas of the cortex. For instance, the left inferior frontal cortex (IFC) is activated in face-to-face communication, but not in non-face-to-face communication (Jiang, Li, Wang, D. Zhu, Liu, & Lu, 2012). In addition, research has shown that the right superior frontal cortex is activated in cooperative video games (Cui et al., 2012), and that both the IFC and the right superior frontal cortex are active in joint melody singing (Kimura, 1964). The studies presented in the following sections focus on brain networks within and across frequencies in social interactions, both inside one brain and with two brains involved in the network. Even though the fronto-parietal region are found to predominantly involved in hyper-brain networks, which is in line with the findings of Cui et al., (2012); Jiang et al., (2012); and Kimura, (1964), the main focus of the studies below is not to find out which brain areas exactly build the networks. Rather, the studies attempt to investigate how brains communicate with each other and how neural processes reflect social interaction.
2.2.1 Intra- and Inter-Brain Synchronization during Guitar Playing
In a study on cortical phase synchronization in music playing, EEG data from guitarists were recorded while they played a melody on the guitar together (Lindenberger, Li, Gruber, & Müller, 2009). Participants (n=16) were aligned into eight different dyads. Two analyses of brain synchronization were conducted: intra-brain synchronization, which is the synchronization of brain oscillations within the brain of one of the participants measured by Phase Locking Index (PLI), and inter-brain synchronization between the brains of the dyads measured by Inter-brain Phase Coherence (IPC). The PLI reflects the invariance of phasesacross trials at single electrodes inthe time-frequency domain and is calculated on the data of one brain. The IPC represents the stability of phase differencesacross trials between two electrodes measured from two brains on a certain point of measurement also in the time- frequency domain(Lindenberger et al., 2009; Sänger et al., 2011). In this study, both within- and between-brain phase synchronization arose in preparation for playing, which consisted of a metronome setting in the beginning of each trial, and at the onset of the guitar play. From this, the authors concluded that oscillatory coupling within the brain of one person and between brains of the dyad arise directly before joint interaction and accompanying it.
Abbildung in dieser Leseprobe nicht enthalten
Figure 4. Within brain synchronization. Panel A: synchronization within brains. Panel B: synchronization between brains. Source: own illustration based on Lindenberger et al. (2009: 6).
The figures above show the inter-brain synchronization of two subjects during the guitar play. Panel A shows the topography of the synchronization with the areas marked in the red as the areas of highest synchronization within the brains. In this picture, one can see that inter-brain synchronization takes place predominantly in the fronto-central brain areas. Panel B shows interbrain networks and their topography on the scalp. The phase alignment between brains reflects the synchrony of the play onset of the guitarists showing that behavioural synchrony and neural synchrony seem to be connected to each other. As shown in Figure 2, Panel A, the inter-brain synchronization patterns were observed in frontal and central electrodes and had the ranges of the delta and theta frequency bands. Overall, the results of this study show that both intra-brain and inter-brain effects of the dyad indicate synchronization between brains in joint action. A limitation to this study is that the observed similarities in the brain oscillations of the two participants may be caused by similarities of the input that the participants perceive and the output the participants produce.
In another study, Sänger, Müller, & Lindenberger, (2012) tested interpersonal action coordination in dyads of guitar players. Participants (n=24) were aligned into 12 pairs and were asked to play a two-voiced modified Rondo by C.G. Scheidler while an EEG of the dyad was recorded. As in the study of Lindenberger et al. (2009), the PLI was taken as a measure of invariance across trials at a single electrode within the time-frequency domain, and the intra- brain and inter-brain phase coherence was measured to reflect the uniformity of phase difference between two electrodes across trials within and between two brains, respectively. They analysed within- brain phase coherence and inter-brain phase coherence in the delta frequency band (1-4 Hz) and in the theta frequency band (4-8 Hz). They found that synchronization both within and between the brainswas particularly enhanced when there was higher demand on musical coordination. Importantly, key findings of the study by Lindenberger et al. (2009) were replicated here despite the shift from unison plying in the previous study to playing in two voices: within- and between- brain synchronization in delta and theta frequencies was enhanced at frontal and central regions during preparatory metronome tempo setting and during coordinated play onset. In addition, by conducting a graph-theoretical analysis, they found that the two brains built modules together, which they called “hyper-brain modules”. The hyper-brain modules were enhanced during periods in which the participants coordinated their playing. In the present study, the participant pairs were assigned the roles of leader and follower and phase locking was modulated accordingly.
Unlike the study of (Lindenberger et al., 2009), guitarists played in two voices to prevent the likelihood that brain synchronization only reflects similarities in the produced output. This was done because, in joint motor interaction, similarities in both the perceived input as well as the produced output can lead to a higher synchronization of brain oscillations (Lindenberger et al., 2009).
Abbildung in dieser Leseprobe nicht enthalten
Figure 5. Networks of within-brain modules (Panel A) and hyper-brain modules (Panel B) in the delta band after the onset of coordinated guitar play. Source: own illustration based on Sänger et al. (2012: 11).
Figure 3 shows an example of a modular community structure of a dyad in the guitar duet study. The upper picture shows within brain modularity structures which consist of three modules in the leader (left) and two modules in the follower (right). The upper picture shows the modularity structure of the hyper-brain networks of both the leader and the follower. Here both the within- brain and the between-brain connections are taken into account. The structure in this case consists of five modules in the brain of the leader (left) and three modules in the brain of the follower (right). In this case, the green and yellow modules in the brain of the leader do not belong to the hyperbrain network as they predominantly reflect within-brain modules, while all three modules in the brain of the follower belong to the hyper-brain network.
Importantly, the study replicated the findings of Lindenberger et al., (2009), so also in this study phase locking within and between brains was enhanced both during the preparatory setting of the tempo and during the periods of music coordination in the frontal and central electrodes. This indicates also that when the action and perception of the two partners differ in social interaction coordination, hyper-brain networks are built. This indicates that hyper-brain networks of the two participants’ brains do not only reflect similarities in perceived input and in produced output.
In addition to the phase locking in the fronto-central areas, the authors detected within-brain and hyper-brain networks in the temporal and parietal regions of the brain. These areas at the boundary between the parietal and the temporal cortical have found to play a functional role in the processing of melodies, i.e. the mapping of auditory representations and motor representations of perceived melodies (Hickok, Buchsbaum, Humphries, & Muftuler, 2003). Therefore, the authors assumed that this process might also play an important role when music is jointly produced by more than one person. Moreover, also parietal regions of the cortex have been found to be active in social cognitive processes (Jean Decety, Jackson, Sommerville, Chaminde, & Meltzoff, 2004). Therefore, hyper-brain modules in the fronto-parietal area may connect brain areas which are associated with music production and social cognition. In addition, the authors found that phaselocking was modulated by the roles of leader and follower, which were assigned to the experimental dyads. In summary, the authors found that networks which are built by synchronous brain oscillations between and within brains support musical performance. In addition, the revealed brain mechanisms support action coordination in social interaction through phase-coherence, phaselocking and structural properties of the brain.
In another study on EEG hyperscanning in guitar improvisation, hyper-brain networks were discovered (Müller, Sänger, & Lindenberger, 2013). The findings again replicated previous findings that phase locking within and between brains was enhanced during joint guitar playing (Lindenberger et al., 2009; Sänger et al., 2012). In this study, eight dyads (n=16) were asked to do music improvisation on the guitar. Also in this study, a complex interaction of brain oscillations within and between brains has been found using a graph-theoretical analysis. The results showed that in intra-brain connections higher frequencies were predominantly involved, for instance the beta frequency band, than in inter-brain connections, which predominantly involved the delta and theta frequency band. Moreover, the topographies of the networks between brains were dependent on the frequency band they recruited. In addition, the analysis showed that some of the properties of the hyper-brain networks were related to the task of musical improvisation. This finding extended the previous studies on EEG hyperscanning in joint guitar playing.
In another study on intra- and inter-brain synchronization in guitar playing, it was investigated whether different roles during social interaction are reflects by hyper-brain networks. Twelve dyads (n=24) were tested in this study and functionally connected hyper-brain networks were analyzed in the alpha band and in the beta band. The results showed that the roles of the leader and the follower during music playing are associated with between brain coupling and that the coupling of brain waves reflects the direction of leading and following. The brain coupling was mostly predominant in the onset of the play as in the study of Lindenberger et al., (2009). The finding of this study adds an important component to previous studies on interbrain synchronization in guitar playing as it shows that in musical dyads, directionality is reflected different in leaders and in followers.
In summary, the studies illustrated in this section showed that in joint guitar playing phase locking within and between brains was enhanced. Nevertheless, also in non music-related tasks, hyperbrain networks have been detected recently. Therefore, studies on EEG hyperscanning with gestural related social interaction tasks will be introduced, which are range from gestural imitation (Dumas et al., 2010), motor imaging (Yun, Watanabe, & Shimojo, 2012), to romantic kissing (Müller & Lindenberger, 2014).
2.2.2 Inter-Brain Synchronization during Social Interaction Tasks
In their study, Dumas et al., (2010) researched phase synchronization between brains of participant pairs (n=22) in social interaction with EEG hyperscanning using a gestural imitation task. Figure 4 shows the experimental setup of the study. Participants were connected to each other, but they did not interact with each other directly as they were separated by a wall between their chairs.
Abbildung in dieser Leseprobe nicht enthalten
Figure 6. Experimental setup of the study by Dumas and colleagues (Dumas et al., 2010: 3).
In the study, the authors used a gestural imitation task and computed the Phase Locking Value which is a coupling or synchronization measure like IPC representing the stability of phase difference between two signals across time instead of trials (Lachaux, Rodriguez, Martinerie, & Varela, 1999). Dumas et al. (2010) found a network of neural synchronization when the gestures, which were required in the gesture imitation task, both started and ended at the same time. More specifically, they found that the brain of the model and the imitator synchronized in the centro- parietal regions in the alpha-mu band, in the central and in theright parieto-occipital regions in the beta band, and in the parieto-occipital regions and in the centro-parietal regions in the gamma band.
Abbildung in dieser Leseprobe nicht enthalten
Figure 7. Inter-brain synchronization in the alpha-mu band (Panel A), in the beta band (Panel B), and in the gamma band (Panel C). Source: own illustration based on Dumas et al. (2010: 6).
The picture above illustrates the findings of the authors on inter-brain synchronization in the three investigated frequency bands (alpha-mu, beta, gamma) and their topography. According to the authors, the findings reflect underlying cortical processes in flexibly distributed areas of the cortex, which communicate efficiently with each other. Importantly, they found that synchrony in brain oscillations does not necessarily exclusively reflect the synchronous movement of the limbs as there were no significant differences between trials with similar movement patterns of the other person and without similar movement. In neither of the studies mentioned above (Dumas et al., 2010; Lindenberger et al., 2009),directed effects between the two partners of the dyad were assessed.
In another study, Yun, Watanabe, & Shimojo, (2012) investigated unconscious social motor coordination, i.e., body movement in social interaction, with EEG hyperscanning. In addition, they investigated the functional connectivity and neural correlates within and among brain regions. The experiment consisted of eight session, which were divided into pre-training (session 1 and 2), training (session 3 to 6), and post-training (session 7 and 8). In the first two pre-training sessions, participants (n=20) were assigned partners and were asked to hold their index fingers towards each other while looking at the finger of the other participant. In the next four training sessions (3 to 6), one leader was selected for each pair, who was asked to randomly move his finger, and the assigned follower, who was the other person in each pair, was asked to follow the movement of the leader’s finger with his own finger. In the seventh and eight post- training sessions, the task of the first and second session was repeated.
Abbildung in dieser Leseprobe nicht enthalten
Figure 8. The tasks of the study by Yun and colleagues (Yun et al., 2012: 2).
Figure 6 shows the three different kinds of tasks, i.e., the pre-training session, the training session, and the post-training sessions in the experimental setup. In the training sessions, the movement is indicated with round arrows. In the EEG recording, brain oscillations of the two participants in each pair were assessed at the same time. The EEG was recorded from both scalps with two EEG recording systems and the recordings were then synchronized.
Abbildung in dieser Leseprobe nicht enthalten
Figure 9. Experimental setup of the study by Yun and colleagues (Yun et al., 2012: 2).
Figure 7 shows the EEG setup of the study. In the analysis, source and phase synchrony were calculated and averaged across and within subjects. In addition, the cross correlation of fingertip movements was measured in each of the conditions. This was done by attaching infrared reflection markers on the fingertips of both participants to measure the motion data of their fingers. Then, the cross-correlation of the finger movement of the two participants was calculated across time to assess whether the assigned follower in fact followed the assigned leader in the experimental session. Functional connectivity was calculated from the data by using postsynaptic potentials to locate highly synchronized voxels within the brain with a LORETA detection of postsynaptic potentials across the scalp. Then, cross spectra were calculated for each participant at each timepoint and frequency band. Time points were six randomly selected intervals of five seconds each and frequency bands were delta (2Hz - 3.5Hz), theta (4Hz - 7.5Hz), alpha (8Hz - 12Hz), and beta (12Hz - 30Hz). Further, phase synchrony between the brain regions of the two participants was calculated for each time point and frequency bin using a Fourier transform. Phase and amplitude was computed for each frequency. The phase corresponds to the position of the oscillation cycle. The authors used this information about the phase to compute a Phase Locking Value, which varies over time and which indicates the neural synchrony at a certain frequency and time window. This value was computed for phase differences of all electrode pairs. In addition, power changes in frequency indicate local neural synchronization and served as an additional marker for the formation of networks, which are built on brain oscillations between cortical regions.
Abbildung in dieser Leseprobe nicht enthalten
Figure 10. Phase synchrony between brains in the theta band (Panel A) and in the beta band (Panel B).
Different lines indicate the phase-locking value (PLV). Source: own illustration based on Yun et al. (2012: 5).
Figure 8 shows the topography of the networks build by phase synchronization between brains. In the picture on the top, the pre-training sessions and post-training sessions were contrasted. The brain on the left depicts the leader , whilethe brain on the right depicts the follower. The inter-brain connections in the theta band (4 Hz- 7.5 Hz)are shown.In the bottom picture, connections in the beta band (12 Hz to 30 Hz) are shown. The authors revealed that inter-brain connections in the particular frequency bands are found predominantly in the anterior cingulate gyrus, in the inferior frontal gyrus, and in the postcentral gyrus. These networks are assumed to be crucial for the understanding of brain mechanisms in interpersonal social interaction. Therefore, phase synchronization both within one brain and across brains was quantified to measure functional connectivity across the brains. During the experiment, participants were asked to try different strategies to achieve synchrony with the partner across brains, including imagining the performance of the motor task while not actually doing it.
The behavioural results of this experiment showed a higher correlation of finger movement between the participant pairs in the post-training trials compared to pre-training trials. Interestingly, correlation was highest at zero time-lag indicating that the correlation was not caused by intentional following but rather unintentional. This is in line with previous findings about unconscious mimicry as the correlation was highest in pairs who sat closely in front of each other and facing each other. The mechanism of unconscious mimicry causes a passive and unintentional adapting of one’s own behaviour to match the behaviour of others (Kendon, 1970). This is done by a link between perception and behaviour, also called “the chameleon effect” (Chartrand & Bargh, 1999). The localization analysis revealed that after training, the theta frequency band was more active in the precuneus , whilethe beta frequency was more active in the middle temporal gyrus.
This is in line with studies that revealed that the right posterior temporal cortex and inferior parietal cortex are functionally involved in social cognition (Decety & Lamm, 2007). Specifically, these areas are involved in empathy and in attentional processes of reorientation to salient stimuli of lower level. On these grounds, the authors concluded that the observed increase in the right posterior medial temporal gyrus reflects a social process in cognition. Thus, they confirmed the hypothesis that social interpersonal interaction has its basis in bodily movement. In addition, fingertip synchrony positively correlated with activity in the theta frequency band in the ventromedial prefrontal cortex. This is in line with studies which have revealed that the ventromedial prefrontal cortex is a circuit of self-representation and representation of others, i.e., introspection versus theory of mind (Keysers & Gazzola, 2007). Moreover, the number of phase synchrony in inter-brain networks was higher after training. In this sense, the training increased both synchronization of fingertips and synchronization between cortical regions of the two participants, specifically in the inferior frontal gyrus, the parahippocampal gyrus, the anterior cingulate gyrus, and in the postcentral gyrus. The authors take this finding as possible evidence for inter-brain synchronization as a neural correlate for interpersonal interaction. Within the inter-brain synchronization network, the brain of the follower influenced the network more, such that the network was not symmetric. This could be explained by the fact that the follower was explicitly instructed to follow the leader and therefore socio-motor regions were activated in the follower person of the dyad in addition to social brain regions, i.e. the medial frontal gyrus, the postcentral gyrus, the precentral gyrus, and the lingual gyrus. Therefore, according to the authors, increased synchrony was found because social interaction was triggered, and not only because sensory inputs gave commands to move synchronously. Therefore, the authors concluded that synchrony in body movement is a measurable correlate of implicit social interaction and that both behavioural and neural correlates of implicit social interaction was identified with the motor imagination task. In addition, participants have learned to control their cross-brain correlation of the motor area. Therefore, the authors were able to assess the change of synchrony in movement in both the experimental and the control group.
In another study by Müller & Lindenberger, (2014), the interaction of brain oscillations in social interaction was studied not only in an interactional task but also when participants had physical contact, specifically when kissing each other. Participants (n=30) consisted of heterosexual couples who were asked to kiss each other in three different conditions while an EEG was recorded. In addition, an electromyogram (EMG) of the lip activity was recorded during the testing. In the first condition, they were asked to kiss each other in a normal romantic manner, in the second condition, participants were asked to kiss each other while being mentally distracted with a silent arithmetic task, and in the third condition asked to kiss their own hand in the presence of the partner. The authors analyzed the data using within-frequency coupling and cross-frequency coupling, and introduced a new method of network construction based on these two coupling types. These networks were designated later as hyper-frequency networks. These complex hyper-brain hyper-frequency networks of kissing pairs were analyzed using a graph-theoretical approach. Modularity analyses allow the authors to identify theta-alpha sub- networks within the hyper-brain networks, which bind the two brainsof kissing partners together. In addition, they found that alpha served as a pacemaker frequency cleaving or binding all frequencies together within the common network.
Abbildung in dieser Leseprobe nicht enthalten
Figure 11. Within-brain networks in the line on top and between brain networks in the line below of the pairs during the kissing experiment. The figure shows results for three experimental conditions of romantic kissing
(rk; Panel A), romantic kissing while performing silent Arithmetiv (k-sa; Panel B) and hand kissing (hk; Panel C). Source: own illustration based on Müller & Lindenberger (2014: 15).
The first line of the figure shows the intra-brain connections in the dyads in the three experimental conditions, which are romantic kissing (rk), kissing while performing silent arithmetic (k-sa), and hand kissing (hk), from left to right. In each condition, the picture on the left shows the brain of the female partner and the picture on the right shows the brain of the male partner, respectively. The second line of the picture shows the inter-brain connection networks of the dyads in the three experimental conditions.
Graph-theoretical analyses showed that intra- and inter-brain strengths were higher during rk and k-sa conditions as compared to hk condition. Moreover, characteristic path length (a measure of network integration) was shorter (stronger network integration) during rk and k-sa than during hk. Importantly, the authors found significant correlations between greater strength and shorter path length within the networks, and partner-oriented kissing satisfaction in both conditions in which the partners kissed each other. Notably, these correlations were strongest for inter-brain connections. Partner-oriented kissing satisfaction was assessed after the entire EEG session. Furthermore, kissing quality, which was assessed immediatelyafter kissing sessions, correlated with the strength of intra-brain alpha band oscillations (10 Hz) in both conditions in which the partners kissed each other, but not with the recorded lip activation or alpha band activity in terms of spectral power. This indicates that neither activity of the lip muscles nor the activity of the alpha band alone are responsible for kissing quality and satisfaction. From their data, the authors concluded that hyper-brain networks which are based on the interaction and coupling across frequencies do not only support social interaction between persons but also support bonding behaviour.
The studies presented in this section show that phase synchrony and hyper-brain networks of brain oscillations support voluntary social interaction in coordinated interaction such as gestural imitation (Dumas et al., 2010) and in motor imaging (Yun et al., 2012). Moreover, hyper-brain networks have also been detected when participants stay in physical contact during a kissing task, and that coupling of brain oscillations across frequencies fosters social bonding (Müller & Lindenberger, 2014). This shows that oscillations of the brain reflect social behaviour both in task performance and in intimate situations.
2.3 Summary
In summary, recent EEG studies have revealed that in social interaction, brain oscillations of participants synchronize with each other and build hyper-brain networks (Dumas et al., 2010; Lindenberger et al., 2009; Müller & Lindenberger, 2014; Sänger et al., 2012; Yun et al., 2012). Moreover, it has been shown that this phenomenon is not only caused by similarities in input information and produced output but reflect, at least in part, neural processes of social interaction (Sänger et al., 2012). In the next section, the method of EEG neurofeedback will be introduced and research on neurofeedback will be presented in both single person EEG recordings and in scanning of multiple persons.
3 Neurofeedback
As mentioned in the introduction, EEG neurofeedback has a wide range of applications in the clinical domain, for instance in the treatment of ADHD and epileptic seizures. Research has revealed that also in healthy subjects, neurofeedback can be applied for certain purposes. The more positive outcomes that research with healthy individuals have gained, the more clinical neurofeedback protocols have become enriched, especially in research with older people. Clinical research in neurofeedback has led to scientific application of neurofeedback and in turn, scientific findings inform clinical questions. In particular, applications of neurofeedback included training to increase power in one frequency band while inhibiting an increase in another frequency band (Gruzelier, 2014b), which for instance led to an improvement in memory performance when power of the alpha band was increased whereas an increase in the theta band power was inhibited (Lecomte & Juhel, 2011). In addition, down-regulation of individualized maxima, neurofeedback was used as a treatment against cognitive disorder in older patients (Becerra et al., 2012), in upregulation of frontal midline theta band power while theta and gamma band were inhibited, attention and working memory improved in the elderly (Wang & Hsieh, 2013). Moreover, the upregulation of frontal gamma oscillations positively influenced intelligence and memory in older participants (Staufenbiel, S.W. Brouwer, Keizer, & Van Wouwe, 2014).
In EEG sensorimotor rhythm ratio studies, which are the most applied protocol of EEG neurofeedback, studies with healthy participants have revealed several findings, even though sensorimotor rhythm ratio studies have their origin in a clinical framework. Research has shown that sensorimotor rhythm ratio treatment led to a gain in attention in participants (Gruzelier, 2014c; Vernon et al., 2003). Moreover, it has been helpful in memory skills in adults and children (Barnea, Rassis, & Zaidel, 2005; Hoedlmoser et al., 2008), in psychomotor skills and fastness of reaction time (Tomas Ros et al., 2009). In addition, sensorimotor rhythm ratio studies have helped to reduce anxiety and to support calmness (Faridnia, Shojaei, & Rahimi, 2012; Tomas Ros et al., 2009) and in spatial rotation abilities (Doppelmayr & Weber, 2011; Gruzelier, 2014b). Especially because of its potential to increase attention, sensorimotor rhythm training is also suggested to be a good treatment tool for ADHD.
Studies of EEG neurofeedback with healthy participants have not only been conducted with the sensory motor rhythm, but also in the beta and gamma band, . In studies with ADHD patients, healthy participants have widely served as the control group on whom beta/theta ratio training has been applied. Nevertheless, there are only few studies of beta and theta training concerned exclusively with healthy participants. For instance, abilities of fluid intelligence were improved through training of neurofeedback control in the beta band (Keizer, Verschoor, Verment, & Hommel, 2010; Staufenbiel, Brouwer et al., 2014).
EEG neurofeedback studies with theta protocols have revealed the following promising outcomes. In theta down-regulation training, general verbal intelligence was improved, and executive functions and attention were enhanced in the elderly (Becerra et al., 2012). In addition, up-regulation of theta in frontal parts caused an improvement of executive attention, recognition memory, and the orientation sense in older participants, as well as executive attention in university students (Wang & Hsieh, 2013). Further, motor procedural learning was improved and gains in sleep could be achieved with only one up-regulation training session of the theta band in the centro- parietal area (the electrode Pz) (Reiner, Rozengurt, & Barnea, 2013). Also, impulsive errors could be reduced in children (Gruzelier, 2014a), and improvements of mood and general well-being (Joshua Raymond, Varney, Parkinson, & Gruzelier, 2005) could be achieved with up-regulation of the theta/alpha ratio. Thus, alpha/theta ratio training does not only reduce anxiety, but generally increases the energy level of the participants and fosters a sense of well-being (Joshua Raymond, Varney, et al., 2005).
The training of alpha power with EEG neurofeedback has also given promising results in the facilitation of working memory (Escolano, Aguilar, & Minguez, 2011), with increasing power of the upper alpha spectrum. Also, procedural memory abilities were improved with alpha training, i.e., with down-regulation of the alpha band originating in the motor cortex, which reduced reaction times reflecting procedural memory processes (Ros, Munneke, Parkinson, & Gruzelier, 2014).
Not only EEG is a useful tool in neurofeedback research. Functional magnetic resonance imaging (fMRI) has also been used in several neurofeedback studies and has some advantages compared to EEG, as it has a higher spatial resolution, scans the anatomic structures of the whole brain and does not only reflect processes going on on the surface of the scalp. In fMRI neurofeedback experiments, the feedback is normally presented with a scale on a screen, but in some studies it can be implemented into virtual reality, for instance with virtual fire (DeCharms, 2008). Also auditory neurofeedback is an option, although visual feedback is the normal way to present neurofeedback to the participants (Gruzelier, 2014b). In fMRI neurofeedback experiments, the neurofeedback mechanisms can even be combined with stimulation. In particular, brain activation can be controlled to respond to pain sensation induced through painful stimuli (deCharms et al., 2005). Research on biofeedback has also been conducted with other methods besides brain imaging, i.e., galvanic skin response, also known as the electrodermal response (Gruber & Moore, 1997), which is a tool that measures the electrical properties of the skin by capturing nerve responses to measure the sweat gland function (bin Syed Noh, 2015). It is a relatively simple, non-invasive, transportable method. Galvanic skin response correlates with affective states and cognitive processes. The signal rises when participants are exposed to stress (bin Syed Noh, 2015), and when the cognitive load of a task increases (Shi et al., 2007).
Crucially, achievements in certain EEG neurofeedback protocols can already be observed after one single training session (Gruzelier, 2014d), particularly in up-regulation of alpha band oscillations, posterior theta, in alpha desynchronization, in the slowing of left temporal theta band, and in the sensorimotor rhythm (Gruzelier, 2014b). Additional to learning outcomes after only one session, also short-term gains in cognition and affectiveness have been observed, for instance memory consolidation overnight, mental rotation, calmness, higher musical performance, and decrease of mind wandering, and instrumental learning (Reiner et al., 2013). In the clinical application of neurofeedback, normally 10 to 12 training sessions are needed to measure significant improvement of behaviour regardless of the particular brain imaging method, i.e., EEG, fMRI, NIRS (Sulzer et al., 2013; Weiskopf, 2012). To guarantee a better comparability of results, it has been suggested to adapt the amount of training sessions according to individual learning outcomes (Scharnowski, Hutton, Josephs, Weiskopf, & Rees, 2012; Sulzer et al., 2013).
Besides variation in the speed of learning, one study has shown that some individuals do not learn to voluntarily control their brain waves at all (Logemann, Lansbergen, Van Os, Böcker, & Kenemans, 2010). This finding raised the question whether there is a generic capacity for learning neurofeedback or if neurofeedback is not learned because the task is not specific enough, which is more likely according to the authors. Nevertheless, whether neurofeedback is a generic capacity is an interesting question and could raise a new onset for research. In the research of Hanslmayr, Sauseng, Doppelmayr, Schabus, & Klimesch, (2005), participants were trained in a neurofeedback alpha protocol or a theta protocol, and results have shown that the ability to learn to control one’s own brainwaves depends on the neurofeedback protocol. Therefore, it is unlikely that there is a generic learning facility of neurofeedback (Gruzelier, 2014d). The existence of such a tool has not yet been demonstrated, but interestingly, the fact that the learning process depends on the neurofeedback protocol is a crucial finding not only for research in the field but also for clinical applications of neurofeedback (Gruzelier, 2014d).
3.1 Recent Studies on Neurofeedback
Neurofeedback studies have been conducted recently not only with a single participant, but also with more than one subject. In the next two sections, recent studies of neurofeedback will be introduced to show the possibilities of neurofeedback research beyond a single individual. The studies will contain the research methods of fMRI, EEG, and fNIRS.
Biofeedback has been shown to be combinable with the extension of a virtual environment (Bersak et al., 2001). A computer can be connected to the biofeedback loop as an additional component. In the above study, the authors attempted to show that participants are not only able to learn to voluntarily control of their body states in a situation decoupled from the environment, but that it is also possible to control one’s own biofeedback when another component is added. This is called “Affective Feedback”. The term is a combination of “biofeedback” and “affective computing”. Affective computing means that brain processes that are computed are influenced by emotional states (Picard, 1995). Affective feedback assumes that the biofeedback influences affective states of the participant as there is a third component in the feedback loop with which participants interact. Studies with a human computer interface therefore have been the step between single-person neurofeedback and multi-person biofeedback experiments, because another component in added to the loop, but this component could be controlled by the experimenters and was not a second individual.
The method of using neural synchronization between two participants as the target of regulation in social interaction is called cross-brain neurofeedback (Duan et al., 2013). It is a new concept and has emerged to study whether participants are able to extend the regulation of their brain activity to an external domain, which is cross-brain synchronization. Because of the components of voluntary control of interacting body rhythms, cross-brain neurofeedback is a promising tool in social neuroscience research. To address the question whether participants can adapt their brain oscillations to each other, two steps have to be undertaken in order to calculate the feedback. First, the feedback must be calculated for each participant separately and must then be compared to the brain oscillations of the other participant. Then, the feedback is given to both participants. Using this feedback, participants are asked to regulate their neural synchronization within the cross-brain domain. As research in neurofeedback training in single persons has shown that both the functional brain organization and behavioural patterns change with training, it is interesting to investigate the implications of cross-brain neurofeedback over time. Thus, in addition to short-term neurofeedback studies, which aim to investigate if and how participants are able to synchronize their brains, long-term studies on cross-brain neurofeedback give an insight into whether the brain reorganizes and changes social behaviour patterns when it is trained to adapt to the oscillations of another participants over long time (Weiskopf, 2012). The first attempt to combine neurofeedback with brain imaging with multiple persons was carried out by Goebel, Sorger, Kaiser, Birbaumer, & Weiskopf (2004) in a study called “BOLD brain pong”. In this fMRI study, participants learned to control their brain activity in a social interaction situation.
3.1.1. Neurofeedback Research with FMRI and FNIRS
In a study in 2009, feedback regulation was measured in prefrontal activity in healthy adults in a brain-computer interface task (Ayaz, Shewokis, Bunce, Schultheis, & Onaral, 2009) with fNIRS. The aim of the study was to determine whether direct communication of biomarkers generated by cognitive activity, and measured by neural activity, are able to directly communicate with a computer. Participants (n=5) were asked to perform cognitive tasks with different cognitive activity levels over two experimental days. Feedback was provided through a visual bar on a screen and showed how much cognitive load was needed in the task. In addition, self-assessment from the participants indicated their judgement of the level of cognitive load of the task. There were resting conditions between the trials. Functional NIRS can detect an increase in oxygenation within the brain, thus giving an online physiological measure of cognitive load (Ayaz, Izzetoglu, Bunce, Heiman-Patterson, & Onaral, 2007). Overall, the training of rating the cognitive load with neurofeedback led to a higher performance level across all subjects than individual classification without neurofeedback only. This provides evidence that an adaption process took place over the two days of training. In addition, the measure of blood oxygenation showed that there was a significantly higher cognitive demand in task conditions compared to in rest conditions. This shows that cognitive load can be measured with a fNIRS based brain-computer interface and that information about cognitive load can lead to a higher correctness in participants’ judgements thus speaking in favor of a learning effect induced by the neurofeedback loop.
In another study, Duan et al., (2013) investigated the neural synchronization of dyads of participants in a cross- brain neurofeedback paradigm with fNIRS with 44 channels on each participants’ scalp. As such, the relationship between social behaviour and the neural synchronization of the participants was investigated to achieve new findings about the neural processes underlying social interaction. In the present experiment, the authors used an experimental platform for the cross-brain neurofeedback, which consisted of three modules: firstly, the data acquisition, secondly the immediate online processing of the data, and thirdly an interface. Participants (n=2) were asked to play a neurofeedback game on the computer screen sitting next to each other while being connected to a multi-person brain computer interface. The game was called “tug-of-war”, in which participants tried to regulate their brain activity with motor imagery. The game setup displayed a ribbon on the screen, which was put on a rope.
Abbildung in dieser Leseprobe nicht enthalten
Figure 12. Setup of the experiment by Duan and colleagues (Duan et al., 2013 :2).
Figure 10 shows the experimental setup. On the left side, connections between the recording devices and the computer are shown; on the right side the experimental setup with the participantsisshown. Theamplitudeoftheparticipants’brainactivityintheleftsensorymotor area was measured and compared to that of the other participant. Depending on which of the participants had the higher motor activity, the ribbon was shifted toward the side of that participant. Regions of interest for the neurofeedback calculation were the left central channels on the scalp, which are on top of the left motor area. In the experiment, participants performed “fighting rounds” of 80s (40s baseline and 40s “fighting”) with a break in the middle. Before the “tug-of-war” game, participants did a separate single neurofeedback training of kinesthetic motor imagery. During the mental “fighting”, participants imagined kinesthetic motor activity to “fight” against each other. In addition, they were asked to adjust their strategies according to the situation in during the trials. Because of the pre-training, participants were able to use the strategies they had learned in the single neurofeedback session. The neurofeedback was calculated with the hemoglobin oxygen signal of the participants. In each round of the game, the difference between the averaged signal amplitudes in the regions of interest was used to calculate the feedback for the two participants and the feedback in turn was represented by the shift of the ribbon on the screen. In addition, the signal amplitude of the two participants was drawn on the screen with two bars, one in red and one in blue.
Abbildung in dieser Leseprobe nicht enthalten
Figure 13. The presentation of feedback by Duan and colleagues. Source: own illustration based on Duan et al. (2013: 3).
Figure 11 shows the visual feedback the participants received in the study. After the experiment, an additional off-line analysis was conducted. The analysis showed that the results of the game reflected interaction states between the participants, i.e. a cross-brain relationship of the two participants. This revealed that cross-brain correlation was higher in each participants’ brain when neither of the participants were detected as being the winner of the game compared to rounds in which there was a clear winner.
Overall, the study showed that the test subjects were able to voluntarily regulate their brain activity in the social interaction situation. The finding about neural synchronization in the different fighting rounds were, according to the authors, of preliminary nature, such that further research is needed. The most crucial finding of the study is that it is possible to implement a platform which calculates integrated neurofeedback of the brains of two participants in cross- brain interaction. This new approach can give new insights into neural processes human interaction not only in observing the results but with an active control of the neural processes of the participants themselves as the two methods of hyperscanning with functional NIRS and neurofeedback were combined.
The studies in this section aimed to train neurofeedback with fMRI and fNIRS in a brain computer interface task and with social interaction. In the next section, an EEG study on neurofeedback training with multiple participants will be introduced.
3.1.2. Neurofeedback Research with Multiple Participants
In a recent study, Kovacevic, Ritter, Tays, Moreno, & McIntosh, (2015) aimed to create a more natural environment for the test subjects than a testing session in the laboratory. It is assumed that human brains compute variable and complex tasks in daily situations, which are significantly reduced in laboratory conditions (Kovacevic et al., 2015). To create an environment which matches the complexity of daily real-world situations better than experiments in the laboratory, the authors conducted an EEG experiment within a multi-media installation combining science and art. Participants (n=523) were connected to a brain computer interface and played a collective computer game. Test subjects were asked to manipulate two of their mental states, which were relaxation and concentration. In addition, they received visual feedback based on their spectral power in the alpha and beta band. The electroencephalogram was recorded with wireless EEG headsets, which were connected to a computer each via Bluetooth using an interface software. The recorded areas for the neurofeedback were the bilateral frontal poles of each participant. Participants were assigned to groups of twenty and then subdivided into four groups of five from whom the EEG data was collected simultaneously and visualized as collective feedback on a screen. During the recording, participants sat in a semicircle in front of an audience and were instructed by an experimental supervisor. The collective neurofeedback visualization was displayed on a screen in front of each participant. Additional screens faced towards the auditory and the ceiling of the room was illustrated with changing colours.
Abbildung in dieser Leseprobe nicht enthalten
Figure 14. Setup of the experiment by Kovacevic and colleagues (Kovacevic et al., 2015: 3).
Figure 12 shows the experimental setup during the testing in the artistic environment. As aimed in this study, the environment provided more auditory and visual input than in a laboratory environment, since there were many people in the room and the ceiling was illuminated with pictures. EEG neurofeedback was calculated from relative spectral power (RSP) in the alpha band (8Hz - 12Hz) and in the beta band (18Hz - 30Hz). The experiment started with a tutorial part to measure the individual bounds of the alpha and beta band for each participant. The individual feedback of each participant was compared to the collective neurofeedback within both the alpha and the beta band.
The game was divided into six steps and the screen was divided into five vertical strips, one for each participant. The first two steps consisted of explaining how to identify one’s own brain waves in the visualization of the collective feedback, a welcome message, and a baseline measure of brain activity. In the third step, participants did a tutorial session. In this session, they were first asked to relax for 20s and the feedback was calculated by measuring the increase of alpha band power. Feedback was visualized with colourful particles on the screen, which increased the more they relaxed. Crucially, only positive feedback was provided, such that the particles on the screen never declined even if alpha power decreased. In the next step, participants were asked to concentrate for 20s. In this condition, the feedback was calculated by measuring the increase of beta band power. Here, the particles on the screen brightened and intensified as more beta power was gained. At the end of the tutorial, the particles on the screen each participant gained were launched into a ‘firework’. To prevent participants from being too competitive and to create the most relaxing atmosphere possible, no scores were presented. The tutorial session also served to set the individual default upper and lower thresholds for each participant in the two frequency bands. In the fourth step, the tutorial session was repeated with the difference that the individual thresholds were used in the calculation of the neurofeedback and that the concentration period lasted 30s. In the fifth step, a group-guided game was performed in which each group of five people received collective feedback from their brain in addition to the individual feedback on the screen, and again with 20s of relaxation and then 30s of concentration.
The collective feedback was displayed in the middle of the screen and the feedback information of the individuals was formed in a semi-circular shape around it. In addition to the individual feedback information, the visual feedback of the group increased in size whenever three or more participants reached the target state, providing the participants with a common goal. In the last step, another group game was performed. In this step, no specific instructions of what to do were given, so that the participants could either relax or concentrate as they wanted for 90s. The only instruction was that the participants synchronize with other group members without talking about the strategy, i.e. to either relax or concentrate and draw from the neurofeedback information whether they synchronized with the other participants. Again, the size of the collective feedback visualization increased when three or more of the participants in the group reached the same brain state.
Abbildung in dieser Leseprobe nicht enthalten
Figure 15. Presentation of the Feedback in the four experimental conditions by Kovacevic and colleagues (Kovacevic et al., 2015: 5). Panel A shows the welcome message, after which the training condition followed. Panel B shows the visual feedback seen in a relaxation trial, Panel C shows that seen in a concentration trial, and Panel D shows that from a trial in which participants were asked to decide themselves whether to relax or to concentrate.
Figure 13 shows the four different conditions of the experiment. Due to the large sample size, the results of this study showed differences in learning outcomes of neurofeedback. More specifically, the speed of learning to change the power spectrum of one’s own brain oscillations varied within a range of approximately one minute. Further, the baseline activity of the participants, which was measured at the first stage of the testing, predicted neurofeedback learning in the beta band. From this fact, the authors concluded that neurofeedback learning is state dependent.
In summary, the present study presented a first approach to creating a natural environment of neurofeedback learning and combined a scientific study with an artistic environment. In the setup, the authors used both the brain computer interface as a connection between each participant, and the computer as well as brain synchronization between two or more participants at the same time. For future research, it would be interesting to give more quantified feedback to the participants, especially in the condition in which they synchronize with each other and receive feedback about how many participants they are connected with. In the present setup, there was no information for the participants about who connected with whom and in one condition they did not even know which strategy made them receive positive collective feedback at a certain point in time.
3.2 Summary
In summary, studies on general social interaction and on neurofeedback in social interaction have been conducted with fNIRS, fMRI, and EEG (Hasan Ayaz et al., 2009; Duan et al., 2013; Kovacevic et al., 2015). There are still very few studies in which neuroscientific research on social interaction has been combined with neurofeedback. In the next section, a study combining EEG hyperscanning and EEG neurofeedback in social interaction will be introduced.
4 Study Proposal
Research has shown that in social interaction, the brain waves of participants synchronize and build hyper-brain networks (Jirsa & Müller, 2013; Lindenberger et al., 2009; Müller et al., 2016, 2013; Müller & Lindenberger, 2014). Moreover, it has been shown that voluntary control of brain oscillations is possible (Gruzelier, 2014b; Kovacevic et al., 2015; Raymond, Varney, & Gruzelier, 2005; Vernon et al., 2003). Furthermore, studies have revealed that voluntary adaption of brain waves to those of another person is possible as well as adaption with a brain- computer interface and playing a game with another person with the help of a neurofeedback protocol (Ayaz et al., 2007; Kovacevic et al., 2015; Sulzer et al., 2013).
On these grounds, the present study aims to combine the methods of EEG neurofeedback and EEG hyperscanning. It assesses whether brains synchronize in the absence of a task, which includes any motor movement or motor movement, which is contrary to previous studies (Lindenberger et al., 2009; Müller & Lindenberger, 2014; Müller et al., 2013). Instead of a specific task, participants are presented with visualized neurofeedback on a computer screen (Duan et al., 2013; Kovacevic et al., 2015; Yun et al., 2012). Crucially, participants do not communicate with each other or coordinate their movement with each other, but rather sit together in the EEG cabin back to back and interact with each other only by knowing that they are performing the neurofeedback task together. As previous studies have shown, participants who are able to see each other during the experiment unconsciously mirror the movement of the other, as in the motor imaging study of Yun et al., (2012). In addition, movement of the tongue can influence coordination of brain activity (Ouyang et al., 2016). Therefore, to control for these important factors, participants were not allowed to look at each other during the experiment. The study took place at Max-Planck- Institute for Human Development, Berlin.
4.1 Methods
4.1.1 Participants
Fifty-two (25 female, 27 male) healthy right-handed young adults participate in the study (age range 18-35 years). Ethical approval was granted by the Max-Planck Ethics Committee and all of the procedures were in accordance with the principles of the Declaration of Helsinki (2013). All participants were informed with a detailed explanation of the procedures during the experiment and provided written consent. Participants did not have any history of developmental or neurological disorders such as stroke or seizures. Three participants (1 female, 2 male) were excluded from the analysis because of technical issues during the testing. To control for gender effects, a balanced number of dyads with either two female subjects, two male subjects, or one female and one male subject took part in the experiment.
4.1.2 Task Instructions
The experiment consisted of twenty-six sessions. The first session was a pre-rest condition to record a baseline of the brain oscillations of each participant. In this session, participants were asked to sit calmly and relaxed in front of the computer screen without moving their bodies, specifically their limbs, tongue, chin, gums, eyes and head. The rest condition lasted for four minutes. Participants were asked to sit two minutes with opened eyes and another two minutes with eyes closed. In the second to twenty-fifth trials, participants performed the neurofeedback task. Each trial lasted 210 seconds. The last trial was a repetition of the rest condition in the beginning of the experiment and consisted of the same procedure.
Abbildung in dieser Leseprobe nicht enthalten
Figure 16. Experimental setup of the participants.
Figure 14 shows how the participants were seated in the EEG cabin. They were not able to see each other so that they were not able to communicate via mimicry. In addition, they were not allowed to talk to each other during the experiment or to communicate through any other sounds or movement.
In the present experiment, participants were asked to look at a computer screen with a neurofeedback visualization, which consisted of either two pendulums with different colours on the screen (red and blue) or two balls with different colours (red and blue, respectively). The instruction was to look at the presented feedback and adapt their brain waves to each other. In the pendulum condition, each of the two pendulums represented the brain oscillations of one of the participants. They were asked to make them swing at the same speed, in phase, so that pendulum movements were parallel. In the condition with the balls, each of the balls was assigned to the neurofeedback of one of the participants, while the balls moved towards or away from the middle of the screen. The task in this condition was to move the balls as close as possible to each other and to make them visually overlap.
Abbildung in dieser Leseprobe nicht enthalten
Figure 17. Presentation of the two versions of the neurofeedback visualization on the screen: visualisation with balls (Panel A) and with pendulums (Panel B) (adapted from Müller, Viktor, Max-Planck Institute for Human Development, Berlin).
Figure 15 shows the feedback which was displayed on the computer screen during the testing. A shows the visualization with the balls, which only move horizontally on the screen. B shows the visualization with the pendulums, which swing on the screen. In each visualization, the red ball (or pendulum, respectively) reflects the brain oscillations of one of the subjects, and the blue colour reflects the brain oscillations of the other subject, to prevent them from being confused.
4.1.3 Electrophysiological Recording
The measurement of EEG took place in an electromagnetically and acoustically shielded cabin. EEG was recorded from both participants simultaneously using electrode caps with 64 electrodes for each participant. The EEG caps were placed on the scalp according to the 10-10 system. In this setup, the reference electrode is the right mastoid. The electrooculogram in vertical and horizontal dimension was measured as well to control for eye movement. The sampling rate was 1000Hz. Each participant was recorded with a separate amplifier with separate grounds which were coupled to the same computer. In addition to the measure of EEG, the heartrate frequency, galvanic skin response, and breathing rate were recorded during the testing.
4.1.4 Neurofeedback Calculation
Neurofeedback was calculated online using six fronto-central channels from each participant. These are the F3, F4, Fz, C3, C4, Cz of the EEG caps according to the 10-10 system. The neurofeedback was calculated for two different frequency bands, which were the delta band at 2.5 Hz and the theta band at 5 Hz. The information of the six EEG channels contained the phase in the chosen frequency band, i.e., the theta band or the delta band, and was averaged for each participant and compared to the averaged phase of the other participant. In neurofeedback visualization with the balls, the feedback of the two participants was subtracted from each other so the visualization reflected the difference between the two brain oscillations. As such, the balls always moved synchronously towards the center of the screen or towards the borders of the screen. The balls only moved horizontally. In the pendulum visualization, the movement of each pendulum was calculated from the brain oscillations of each participant independently. Differently to the balls, the pendulums did not move in a parallel manner, and they did not necessarily move with the same speed. The brain oscillations were recorded with BrainVision Recorder software (Brain Products GmbH, Gilching, Germany). This computer sent packages to a second computer which ran a script (MATLAB and Statistics Toolbox Release 2012b, The MathWorks, Inc., Natick, Massachusetts, n.d.) to calculate the neurofeedback.
4.1.5 Experimental Design
In the experiments, there were three different types of trials. Firstly, normal trials, secondly “fake” trials in which the feedback was improved with a factor, and thirdly “inverted” trials in which the feedback was better the more the brain oscillations were in antiphase. The trials with enhanced feedback were implemented to motivate the test subjects by giving them the impression that they were performing well in the experiment, and the inverted trials served as a control condition to control whether performance is as good as in normal trials when the random activity of brain oscillations is taken as the foundation for the feedback. It was expected that performance is better in fake than in normal trials because of enhanced feedback, and that performance in inverted trials is worse than in normal trials, since the feedback does not reflect synchrony of the brain and therefore does not reflect the ongoing processes in social interaction which were previously found in other studies. No roles of leader and follower were assigned, but rather the participants were partners within the experiment. The design of the task was block-wise, meaning that it either started with a block of ball trials or of pendulum trials. There was no change of trials inside one block. Within the blocks, there were two sub-blocks, either the calculation of the theta band or of the delta band. Within these sub-blocks, there were six trials, always in the same order. The block started with two normal trials, followed by a fake trial with enhanced feedback. The fourth and fifth trials were again normal trials followed by a final inverted trial.
4.1.6 Questionnaires
Before the testing session, participants were asked to do a shortened online version of the Saarbrückener Persönlichkeits-Fragebogen (SPF) based on the Interpersonal Reactivity Index (IRI) (Paulus, 2009), an online version of the Neo Five-Factor Inventory (NEO-FFI) (Costa & McCrae, 1992), and a demographic questionnaire. Psychological questionnaires were filled by the participants before the session using the online tool “SoSci Survey” (Leiner & Leiner, n.d.). Immediately before testing, participants were asked to fill in a consumption questionnaire about factors influencing the EEG experiment, for instance consumption of caffeine and alcohol. In addition, educational information was assessed for each participant.
After testing, participants filled in two other questionnaires while still sitting in the EEG cabin. These were a post-questionnaire about the experiment and a likeability questionnaire (Reysen, 2005). The likeability questionnaire assesses the perceived likeability between each of the individualswhich could be an influencing factor in testing. The post-questionnaire assessed the strategies of the two participants. As they were asked not to talk to each other about the strategies before and during testing, it may be interesting to investigate whether brain synchrony was higher when participants used the same strategies. Figure 16 shows the post-questionnaire which was developed for this study.
Abbildung in dieser Leseprobe nicht enthalten
Figure 18. English version of the post-questionnaire.
4.2 Results
Feedback performance was stored during the recording thus providing an insight into how well the participants were able to synchronize their brain oscillations. Brain coupling will be analyzed to investigate whether participants were able to synchronize their brain oscillation. If and how performance changed between trials (normal, fake, inverted) shall be analyzed. Moreover, within- and cross-frequency coupling will be computed within and across brains to detect and identify involved brain networks similar to previous studies (Müller et al., 2016, 2013; Müller & Lindenberger, 2014; Sänger et al., 2012; Yun et al., 2012).
4.3 Discussion
For future studies, it will be interesting to investigate whether there are specific strategies participants can use to improve their synchronization. This could manifest itself for instance in the ability to synchronize faster or to a higher degree, or in a more stable synchronization within trials. One approach could be to ask them to think of motor synchronization as in previous studies on social interaction (Yun et al., 2012). In addition, relaxation and concentration could be given as possible instructions in future research (Konvalinka & Roepstorff, 2012).
One objection to the study is that there was no explicit control condition in the experiment. Nevertheless, if there is a significant correlation between the perceived performance of participant on their performance and the performance measure, it may be ruled out that variation in performance across subjects only reflects normal random distribution and is not caused by the task itself.
Another objection is that it is difficult to practice the neurofeedback response because either a fake trial or an inverted trial appeared every three trials and might have confused the participants. Nevertheless, the likelihood that the improvement of feedback in fake trials motivates participants might have contributed to more successful learning of neurofeedback. In addition, the appearance of the inverted trials in fact may have confused the participants as the relevant feedback did not reflect synchronization but desynchronization of the brain waves and therefore looked different from the visualization in the normal trials. However, this variation could also have been taken as a motivational factor. If the feedback was better in the preceding trials there is the danger that participants wondered why their performance became worse in the current trial, reacting with higher motivation to make the feedback improve again. Furthermore, the inverted trials at the end of each block can be interpreted as a type of control condition. Therefore, it is good to have both the fake trials and the inverted trials in the experimental paradigm as they do not only contribute to the motivation of the participants, but also provide a control condition.
5 Conclusion
Research has shown that in social interaction, certain areas of the human brain are more active than others, the temporo-parietal junction (Babiloni & Astolfi, 2014; Van Overwalle & Baetens, 2009) and the medial prefrontal cortex (Schurz et al., 2014) in particular. In addition, hyperscanning experiments have revealed that in socially coordinated interaction, networks emerge within and between the brains of the participants (Müller & Lindenberger, 2014; Müller et al., 2013; Sänger et al., 2011; Yun et al., 2012). This essay gave a summary of EEG research on social interaction as well as neurofeedback research. In recent years, studies have been conducted in which EEG hyperscanning and EEG neurofeedback or fNIRS neurofeedback were combined (Duan et al., 2013; Kovacevic et al., 2015). Still, more research is needed to find out whether participants are able to control their brain oscillations in a hyperscanning paradigm. It has been shown that motor imagination leads to successful control of one's own brain activity (Duan et al., 2013). Therefore, it might be helpful to assess the strategies which are used by participants in more detail in order to gain insight into which cognitive processes underlie successful synchronization.
EEG is a brain imaging technique with very good temporal resolution, but it has the disadvantage of a bad spatial resolution. Thus, in future studies it might be helpful to combine EEG measure with fMRI to benefit from both the good temporal resolution of EEG and the good spatial resolution of fMRI.
In general, the neural processes of social interaction are central to human life and social bonding, but they are still poorly understood. Therefore, it is a great achievement that hyperscanning is able to measure brain activity of more than one individual at the very moment of social interaction. Furthermore, neurofeedback is a good tool to investigate whether people can learn to voluntarily control their brain oscillations. On these grounds, the combination of these two methods can inform future research on the neural processes in social interaction and can reveal to which extent people can learn to consciously synchronize their brain oscillations with and without specific task instructions.
References
Adolphs, R. (2003). Cognitive neuroscience of human social behaviour. Nat Rev Neurosci, 4 (3), 165-178. http://doi.org/10.1038/nrn1056
Adolphs, R. (2010a). Conceptual Challenges and Directions for Social Neuroscience. Neuron, 65 (6), 752-767. http://doi.org/10.1016Zj.neuron.2010.03.006
Adolphs, R. (2010b). Conceptual Challenges and Directions for Social Neuroscience. Neuron, 65 (6), 752-767. http://doi.org/10.1016/j.neuron.2010.03.006
Ali, Y., Mahmud, N. A., & Samaneh, R. (2015). Current Advances in Neurofeedback Techniques for the Treatment of ADHD. Biomedical & Pharmocology Journal, 8, 165- 177. http://doi.org/10.13005/bpj/573
Allport, F. (1924). Social Psychology. Boston: Houghton Mifflin. Argyle, M., & Cook, M. (1976). Gaze and mutual gaze. Cambridge: Cambridge University
Press. Arnal, L. H., & Giraud, A. L. (2012). Cortical oscillations and sensory predictions. Trends in Cognitive Sciences. http://doi.org/10.1016/j.tics.2012.05.003
Arns, M., & Kenemans, J. L. (2014). Neurofeedback in ADHD and insomnia: Vigilance stabilization through sleep spindles and circadian networks. Neuroscience and BiobehaviouralReviews, 44, 183-194. http://doi.org/10.1016/j.neubiorev.2012.10.006
Astolfi, L., Toppi, J., De Vico Fallani, F., Vecchiato, G., Salinari, S., Mattia, D., ... Babiloni, F. (2010). Neuroelectrical hyperscanning measures simultaneous brain activity in humans. Brain Topography, 23 (3), 243-256. http://doi.org/10.1007/s10548-010-0147-9
Ayaz, H., Izzetoglu, M., Bunce, S., & Heiman-Patterson, T. Onaral, B. (2007). Detecting cognitive activity related hemodynamic signal for brain computer interface using functional near infrared spectroscopy. In 3rd IEEE/EMBS Conference on Neural Engineering (pp. 342345).
Ayaz, H., Shewokis, P. A., Bunce, S., Schultheis, M., & Onaral, B. (2009). Assessment of cognitive neural correlates for a functional near infrared-based brain computer interface system. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 5638 LNAI, 699-708. http://doi.org/10.1007/978-3-642-02812-0_79
Aylward, E. H., Park, J. E., Field, K. M., Parsons, a C., Richards, T. L., Cramer, S. C., & Meltzoff, a N. (2005). Brain activation during face perception: evidence of a developmental change. Journal of Cognitive Neuroscience, 17 (2), 308-319. http://doi.org/10.1162/0898929053124884
Babiloni, F., & Astolfi, L. (2014). Social neuroscience and hyperscanning techniques: Past, present and future. Neuroscience and Biobehavioural Reviews, 44, 76-93. http://doi.org/10.1016Zj. neubiorev. 2012.07.006
Barnea-Goraly, N., Kwon, H., Menon, V., Eliez, S., Lotspeich, L., & Reiss, A. L. (2004). White matter structure in autism: Preliminary evidence from diffusion tensor imaging. Biological Psychiatry, 55 (3), 323-326. http://doi.org/10.1016Zj.biopsych.2003.10.022
Barnea, A., Rassis, A., & Zaidel, E. (2005). Effect of neurofeedback on hemispheric word recognition. Brain and Cognition, 59 (3), 314-321. http://doi.org/10.1016/j.bandc.2004.05.008
Baron-Cohen, S., Ring, H. A., Wheelwright, S., Bullmore, E. T., Brammer, M. J., Simmons, A., & Williams, S. C. R. (1999). Social intelligence in the normal and autistic brain: An fMRI study. European Journal of Neuroscience, 11 (6), 1891-1898. http://doi.org/10.1046/j. 1460-9568.1999.00621.x
Barth, B., Strehl, U., Fallgatter, A. J., & Ehlis, A.-C. (2016). Near-Infrared Spectroscopy based Neurofeedback of Prefrontal Cortex Activity: A Proof-of-Concept Study. Frontiers in Human Neuroscience, 10 (December), 633. http://doi.org/10.3389/fnhum.2016.00633
Becerra, J., Fernandez, T., Roca-Stappung, M., Diaz-Comas, L., Galân, L., Bosch, J., ... Harmony, T. (2012). Neurofeedback in healthy elderly human subjects with electroencephalographic risk for cognitive disorder. Journal of Alzheimer’s Disease, 28 (2), 357-367. http://doi.org/10.3233/JAD-2011-111055
Ben-Ari, Y. (2001). Developing networks play a similar melody. Trends in Neurosciences, 24, 353-360.
Ben-Ari, Y. (2002). Excitatory actions of gaba during development: the nature of the nurture. Nat Rev Neurosci, 3 (9), 728-739. http://doi.org/10.1038/nrn920
Bersak, D., McDarby, G., Augenblick, N., McDarby, P., McDonnell, D., McDonald, B., &
Karkun, R. (2001). Intelligent biofeedback using an immersive competitive environment. Retrieved from http://medialabeurope. org/mindgames/publications/publicationsAtlanta2001rev3. pdf
Bickford, R. D. (1987). Encephalography. In G. Adelman (Ed.), Encyclopedia of Neuroscience (pp. 371-373). Birkhauser, Cambridge (USA).
bin Syed Noh, S. A. (2015). Galvanic Skin Response. Retrieved from http://ir.unikl.edu.my/jspui/handle/123456789/10511
Birbaumer, N., Ghanayim, N., Hinterberger, T., Iversen, I., Kotchoubey, B., Kübler, A., ... Flor, H. (1999). A spelling device for the paralysed. Nature, 398 (6725), 297-298. http://doi.org/10.1038/18581
Boraston, Z., & Blakemore, S.-J. (2007). The application of eye-tracking technology in the study of autism. The Journal of Physiology, 581 (Pt 3), 893-898. http://doi.org/10.1113/jphysiol.2007.133587
Bronzino, J. D. (1995). Principles of Electroencephalography. In J. D. Bronzino (Ed.), The Biomedical Engeneering Handbook (pp. 201-212). CRC Press, Florida.
Busch, C. (2009). Teamarbeit und Gesundheit. In Fehlzeiten-Report 2009: Arbeit und Psyche: Belastungen reduzieren - Wohlbefinden fördern (pp. 137-146). http://doi.org/10.1007/978- 3-642-01078-1_15
Buzsâki, G. (2006). Rhythms of the Brain. Oxfors: University Press. Buzsaki, G., & Draguhn, A. (2004). Neuronal oscillations in cortical networks. Science, 304 (5679), 1926-1929. http://doi.org/10.1126/science. 1099745
Canolty, R T, K. R. (2012). The functional role of cross-frequncy coupling. Trends in Cognitive Sciences, 14 (11), 506-515. http://doi.org/10.1016/j.tics.2010.09.001.The
Chakravarthi, R., & VanRullen, R. (2012). Conscious updating is a rhythmic process. Proceedings of the National Academy of Sciences, 109 (26), 10599-10604. http://doi.org/10.1073/pnas.1121622109
Chartrand, T. L., & Bargh, J. a. (1999). The chameleon effect: the perception-behaviour link and social interaction. Journal of Personality and Social Psychology, 76(6), 893-910. http://doi.org/10.1037/0022-3514.76.6.893
Chatel-Goldman, J., Schwartz, J. L., Jutten, C., & Congedo, M. (2013). Non-local mind from the perspective of social cognition 31. Frontiers in Human Neuroscience, 7. Retrieved from wos:000316987800001
Cheng, X., Li, X., & Hu, Y. (2015). Synchronous brain activity during cooperative exchange depends on gender of partner: A fNIRS-based hyperscanning study. Human Brain Mapping, 36 (6), 2039-2048. http://doi.org/10.1002/hbm.22754
Coben, R., Linden, M., & Myers, T. E. (2010). Neurofeedback for autistic spectrum disorder: A review of the literature. Applied Psychophysiology Biofeedback, 35 (1), 83-105. http://doi.org/10.1007/s10484-009-9117-y
Corbetta, M. (2012). Functional connectivity and neurological recovery. Developmental Psychobiology, 54 (3), 239-253. http://doi.org/10.1002/dev.20507
Costa, P., & McCrae, R. (1992). Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological Assessment, 4 (1), 5-13. http://doi.org/10.1037//1040- 3590.4.1.5
Cox, W. M., Subramanian, L., Linden, D. E. J., Lührs, M., McNamara, R., Playle, R., ... Ihssen, N. (2016). Neurofeedback training for alcohol dependence versus treatment as usual: study protocol for a randomized controlled trial. Trials, 17 (1), 480. http://doi.org/10.1186/s13063-016-1607-7
Cui, X., Bryant, D. M., & Reiss, A. L. (2012). NIRS-based hyperscanning reveals increased interpersonal coherence in superior frontal cortex during cooperation. NeuroImage, 59 (3), 2430-2437. http://doi.org/10.1016/j.neuroimage.2011.09.003
Datta, D. P. (2003). A new class of scale free solutions to linear ordinary differential equations and the universality of the golden mean (5- 1)/2= 0.618033 Chaos, Solitons & Fractals, 17 (4), 621-630.
Davidson, R. J., Pizzagalli, D., Nitschke, J. B., & Putnam, K. (2002). Depression: perspectives from affective neuroscience. Annual Review of Psychology, 53, 545-574. http://doi.org/10.1146/annurev.psych.53.100901.135148
de Vignemont, F., & Singer, T. (2006). The empathic brain: how, when and why? Trends in Cognitive Sciences, 10 (10), 435-441. http://doi.org/10.1016/j.tics.2006.08.008
Decety, J., Jackson, P. L., Sommerville, J. a., Chaminde, T., & Meltzoff, a. N. (2004). The neural bases of cooperation and competition: an FMRI investigation. NeuroImage, 23 (2), 744751. http://doi.org/10.1016Zj.neuroimage.2004.05.025 Decety, J., & Lamm, C. (2006). Human Empathy Through the Lens of Social Neuroscience. The Scientific World JOURNAL, 6, 1146-1163. http://doi.org/10.1100/tsw.2006.221
Decety, J., & Lamm, C. (2007). The role of the right temporoparietal junction in social interaction: How low-level computational processes contribute to metacognition. Neuroscientist, 13, 580-593.
de Charms, R. C. (2008). Applications of real-time fMRI. Nature Reviews Neuroscience, 9, 720729.
de Charms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M., Soneji, D., ... Mackey, S. C. (2005). Control over brain activation and pain learned by using real-time functional MRI. Proceedings of the National Academy of Sciences of the United States of America, 102 (51), 18626-31. http://doi.org/10.1073/pnas.0505210102
Doppelmayr, M., & Weber, E. (2011). Effects of SMR and theta/beta neurofeedback on reaction time, spatial abilities and creativity. Journal of Neurotherapy. Journal of Neurotherapy, 15, 115-129.
Duan, L., Liu, W. J., Dai, R. N., Li, R., Lu, C. M., Huang, Y. X., & Zhu, C. Z. (2013). Cross- Brain Neurofeedback: Scientific Concept and Experimental Platform. PLoS ONE, 8 (5), 1-5. http://doi.org/10.1371/journal.pone.0064590
Dumas, G., Lachat, F., Martinerie, J., Nadel, J., & George, N. (2010). From social behaviour to brain synchronization: Review and perspectives in hyperscanning. Irbm, 32(1), 48-53. http://doi.org/10.1016/j.irbm.2011.01.002
Dunbar, R. I. M. (2003). The Social Brain: Mind, Language, and Society in Evolutionary Perspective. Annual Review of Anthropology, 32 (2003), 163-181. http://doi.org/10.2307/25064825
Duschek, S., Schuepbach, D., Doll, A., Werner, N. S., & Del Paso, G. A. R. (2011). Self- regulation of cerebral blood flow by means of transcranial doppler sonography biofeedback. Annals of Behavioural Medicine, 41 (2), 235-242. http://doi.org/10.1007/s12160-010-9237-x
Egner, T., & Gruzelier, J. (2003). Ecological validity of neurofeedback: modulation of slow wave EEG enhances musical performance. Neuroreport, 14 (9), 1221-1224. http://doi.org/10.1097/01.wnr.0000081875.45938.d1
Egner, T., & Gruzelier, J. H. (2004). EEG Biofeedback of low beta band components: Frequencyspecific effects on variables of attention and event-related brain potentials. Clinical Neurophysiology, 115 (1), 131-139. http://doi.org/10.1016/S1388- 2457(03)00353-5
Emery, N. (2000). The eye have it: The neuroethology, function and evolution of social gaze. Neuroscience andBiobehaviouralReviews, 24(6), 581-604. http://doi.org/10.1016/S0149- 7634(00)00025-7
Escolano, C., Aguilar, M., & Minguez, J. (2011). EEG-based upper alpha neurofeedback training improves working memory performance. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (pp. 23272330). http://doi.org/10.1109/IEMBS.2011.6090651
Faridnia, M., Shojaei, M., & Rahimi, A. (2012). The effect of neurofeedback on the anxiety of elite female swimmers. Annals of Biological Research, 3, 1020-1028.
Fell, J., & Axmacher, N. (2011). The role of phase synchronization in memory processes. Nature Reviews. Neuroscience, 12 (2), 105-118. http://doi.org/10.1038/nrn2979
Fernandez, T., Herrera, W., Harmony, T., Diaz-Comas, L., Santiago, E., Sanchez, L., ... Valdés, R. (2003). EEG and behavioural changes following neurofeedback treatment in learning disabled children. Clinical EEG (Electroencephalography), 34 (3), 145-152. http://doi.org/10.1177/155005940303400308
Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). Universal dimensions of social cognition: warmth and competence. Trends in Cognitive Sciences. http://doi.org/10.1016/j.tics.2006.11.005
Fuchs, T., Birbaumer, N., Lutzenberger, W., Gruzelier, J. H., & Kaiser, J. (2003). Neurofeedback Treatment for Attention-Deficit/ Hyperactivity Disorder in Children: A Comparison With Methylphenidate. Applied Psychophysiology and Biofeedback, 28(1), 1-12. http://doi.Org/10.1023/A:1022353731579
Gemünden, H. G., & Högl, M. (1998). Teamarbeit in innovativen Projekten: Eine kritische Bestandsaufnahme der empirischen Forschung. Zeitschrift Für Personalforschung, 1970(3), 277-301. http://doi.org/10.1007/978-3-322-92053-9
Glass, L. (2001). Synchronization and rhythmic processes in physiology. Nature, 410 (6825), 277284. http://doi.org/10.1038/35065745
Goebel, R., Sorger, B., Kaiser, J., Birbaumer, N., & Weiskopf, N. (n.d.). BOLD brain pong: Self regulation of local brain activity during synchronously scanned, interacting subjects. In 34th Annual Meeting of the Society for Neuroscience.
Gruber, M., & Moore, P. (1997). Galvanic skin response. The Science Teacher, 64 (9), 52.
Gruzelier, J. H. (2014a). Differential effects on mood of 12-15 (SMR) and 15-18 (beta1) Hz neurofeedback. International Journal of Psychophysiology, 93 (1), 112-115. http://doi.org/10.1016/j.ijpsycho.2012.11.007
Gruzelier, J. H. (2014b). EEG-neurofeedback for optimising performance. I: A review of cognitive and affective outcome in healthy participants. Neuroscience and Biobehavioural Reviews, 44, 124-141. http://doi.org/10.1016/j.neubiorev.2013.09.015
Gruzelier, J. H. (2014c). EEG-neurofeedback for optimising performance. II: Creativity, the performing arts and ecological validity. Neuroscience and Biobehavioural Reviews, 44, 142-158. http ://doi.org/10.1016/j.neubiorev.2013.11.004
Gruzelier, J. H. (2014d). EEG-neurofeedback for optimising performance. III: A review of methodological and theoretical considerations. Neuroscience and Biobehavioural Reviews, 44, 159-182. http://doi.org/10.1016Zj.neubiorev.2014.03.015
Halgren, E., Raji, T., Marinkovic, K., Jousmäki, V., & Hari, R. (2000). Cognitive Response Profile of the Human Fusiform Face Area as Determined by MEG. Cerebral Cortex, 10 (1), 69-81. http://doi.org/10.1093/cercor/10.1.69
Hanslmayr, S., Sauseng, P., Doppelmayr, M., Schabus, M., & Klimesch, W. (2005). Increasing individual upper alpha power by neurofeedback improves cognitive performance in human subjects. Applied Psychophysiology Biofeedback, 30 (1), 1-10. http://doi.org/10.1007/s10484-005-2169-8
Hasson, U., Ghazanfar, A. A., Galantucci, B., Garrod, S., & Keysers, C. (2012). Brain-to- brain coupling: A mechanism for creating and sharing a social world. Trends in Cognitive Sciences, 16 (2), 114-121. http://doi.org/10.1016/j.tics.2011.12.007
Hickok, G., Buchsbaum, B., Humphries, C., & Muftuler, T. (2003). Auditory-Motor Interaction Revealed by fMRI: Speech, Music, and Working Memory in Area Spt. Journal of Cognitive
Neuroscience, 15 (5), 673-682. http://doi.org/10.1162/jocn.2003.15.5.673
Hoedlmoser, K., Pecerstorfer, T., Gruber, G., Anderer, P., Doppelmayr, M., Klimesch, W., & Schabus, M. (2008). Instrumental Conditioning of Human Sensorimotor Rhythm (12-15 Hz) and Its Impact on Sleep as Well as Declarative Learning. SLEEP, 31 (10), 1401- 1408.
Isler, J. R., Grieve, P. G., Czernochowski, D., Stark, R. I., & Friedman, D. (2008). Cross- frequency phase coupling of brain rhythms during the orienting response. Brain Research, 1232, 163172. http://doi.org/10.1016/j.brainres.2008.07.030
Jensen, O., & Colgin, L. L. (2007). Cross-frequency coupling between neuronal oscillations. Trends in Cognitive Sciences. http://doi.org/10.1016/j.tics.2007.05.003
Jiang, J., Li, F., Wang, D. Zhu, C., Liu, L., & Lu, C. (2012). Neural synchronization during face- to-face communication. J Neuroscience, 32, 16064-16069.
Jirsa, V., & Müller, V. (2013). Cross-frequency coupling in real and virtual brain networks. Frontiers in Computational Neuroscience, 7 (July), 78. http://doi.org/10.3389/fncom.2013.00078
Kamiya, J. (1969). Operant control of the EEG alpha rhythm and some of its reported effects on consciousness. In Tart, C.T. (Ed.), Altered States of Consciousness (pp. 519-529).
Keizer, A. W., Verschoor, M., Verm ent, R. S., & Hommel, B. (2010). The effect of gamma enhancing neurofeedback on the control of feature bindings and intelligence measures. International Journal of Psychophysiology, 75, 25-32.
Kendon, A. (1970). Movement coordination in social interaction: Some examples described. Acta Psychologica, 32, 101-125.
Keysers, C., & Gazzola, V. (2007). Integrating simulation and Theory of Mind: from self to social cognition. Trends in Cognitive Sciences, 11, 194-196.
Kilner, J., Baker, S., Salenius, S., Hari, R., & Lemon, R. (2000). Human cortical muscle coherence is directly related to specific motor parameters. Journal of Neuroscience, 20, 8838-8845.
Kimura, D. (1964). Left-right differences in the perception of melodies. Quarterly Journal of Experimental Psychology, 16 (4), 355-358.
Klimesch, W. (1997). EEG-alpha rhythms and memory processes. In International Journal of Psychophysiology (Vol. 26, pp. 319-340). http://doi.org/10.1016/S0167-8760(97)00773- 3
Klimesch, W. (1999). EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Research. Brain Research Reviews, 29 (2-3), 169-95. http://doi.org/10.1016/S0165-0173(98)00056-3
Klimesch, W. (2011). Evoked alpha and early access to the knowledge system: the P1 inhibition timing hypothesis. Brain Research, 1408, 52-71.
Klimesch, W. (2012). Alpha-band oscillations, attention, and controlled access to stored information. Trends in Cognitive Sciences, 16 (12), 606-617. http://doi.org/10.1016/j.tics.2012.10.007
Klin, A., Jones, W., Schultz, R., Volkmar, F., Cohen, D., T, L., ... A, K. (2002). Visual Fixation Patterns During Viewing of Naturalistic Social Situations as Predictors of Social Competence in Individuals With Autism. Archives of General Psychiatry, 59 (9), 809. http://doi.org/10.1001/archpsyc.59.9.809
Kober, S. E., Wood, G., Kurzmann, J., Friedrich, E. V. C., Stangl, M., Wippel, T., ... Neuper, C. (2014). Near-infrared spectroscopy based neurofeedback training increases specific motor imagery related cortical activation compared to sham feedback. Biological Psychology, 95 (1), 21-30. http://doi.org/10.1016/j.biopsycho.2013.05.005
Konvalinka, I., & Roepstorff, A. (2012). The two-brain approach: how can mutually interacting brains teach us something about social interaction? Frontiers in Human Neuroscience, 6, 215. http://doi.org/10.3389/fnhum.2012.00215
Kovacevic, N., Ritter, P., Tays, W., Moreno, S., & McIntosh, A. R. (2015). “My virtual dream”: Collective neurofeedback in an immersive art environment. PLoS ONE, 10(7), 1-18. http ://doi.org/10.1371/j ournal.pone.0130129
Lachaux, J., Rodriguez, E., Martinerie, J., & Varela, F. (1999). Measuring phase synchrony in brain signals. Hum Brain Mapp., S (4), 194-208.
Lakatos, P., Shah, A. S., Knuth, K. H., Ulbert, I., Karmos, G., & Schroeder, C. E. (2005). An oscillatory hierarchy controlling neuronal excitability and stimulus processing in the auditory cortex. Journal of Neurophysiology, 94 (3), 1904-11. http://doi.org/10.1152/jn.00263.2005
Lecomte, G., & Juhel, J. (2011). The Effects of Neurofeedback Training on Memory Performance in Elderly Subjects. Psychology, 2 (8), 846-852. http://doi.org/10.4236/psych.2011.28129
Leiner, D., & Leiner, S. (n.d.). SoSci Survey GmbH. Retrieved from
https://www.soscisurvey.de/index.php?page=home
Leins, U., Goth, G., Hinterberger, T., Klinger, C., Rumpf, N., & Strehl, U. (2007). Neurofeedback for children with ADHD: A comparison of SCP and Theta/Beta protocols. Applied Psychophysiology Biofeedback, 32(2), 73-88. http://doi.org/10.1007/s10484-007-9031-0
Lieberman, M. D. (2010). Social Cognitive Neuroscience. Handbook of Social Psychology, 143193. http://doi.org/10.1002/9780470561119.socpsy001005
Lindenberger, U., Li, S.-C., Gruber, W., & Müller, V. (2009). Brains swinging in concert: cortical phase synchronization while playing guitar. BMC Neuroscience, 10, 22. http://doi.org/10.1186/1471-2202-10-22
Liu, T., & Pelowski, M. (2014). A new research trend in social neuroscience: Towards an interactive-brain neuroscience. PsyCh Journal, 3 (3), 177-188. http://doi.org/10.1002/pchj .56
Logemann, H. N. A., Lansbergen, M. M., Van Os, T. W. D. P., Böcker, K. B. E., & Kenemans, J. L. (2010). The effectiveness of EEG-feedback on attention, impulsivity and EEG: A sham feedback controlled study. Neuroscience Letters, 479 (1), 49-53. http://doi.org/10.1016/j.neulet. 2010.05.026
Lopes da Silva, F. H. (1992). The rhythmic slow activity (theta) of the limbic cortex: an oscillation in search of a function. In E. Basar & T. H. Bullock (Eds.), Induced Rhythms in the Brain (pp. 83-102). Boston: Birkhäuser.
Lubar, J. F., Swartwood, M. O., Swartwood, J. N., & O’Donnell, P. H. (1995). Evaluation of the effectiveness of EEG neurofeedback training for ADHD in a clinical setting as measured by changes in T.O.V.A. scores, behavioural ratings, and WISC-R performance. Biofeedback and Self-Regulation, 20(1), 83-99. http://doi.org/10.1007/BF01712768
Makeig, S., & Jung, T. P. (1996). Tonic, phasic, and transient EEG correlates of auditory awareness in drowsiness. Cognitive Brain Research, 4 (1), 15-25.
MATLAB and Statistics Toolbox Release 2012b, The MathWorks, Inc., Natick, Massachusetts, U. S. (n.d.). MATLAB.
McGrath, J., Kelly, J., & Machatka, D. (1984). The social psychology of time: Entrainment of behaviour in social and organizational settings. Applied Social Psychology Retrieved from http://psycnet.apa.org/psycinfo/1985-24014-001
Michel, C. M., & Murray, M. M. (2012). Towards the utilization of EEG as a brain imaging tool. NeuroImage. http://doi.org/10.1016/j.neuroimage.2011.12.039
Milner, R., Lewandowska, M., Ganc, M., Ciesla, K., Niedzialek, I., & Skarryzski, H. (2016). Slow Cortical Potential Neurofeedback in Chronic Tinnitus Therapy: A Case Report. Applied Psychophysiology Biofeedback, 41 (2), 225-249. http://doi.org/10.1007/s10484- 015-93185
Müller, V., & Lindenberger, U. (2011). Cardiac and respiratory patterns synchronize between persons during choir singing. PLoS ONE, 6 (9), 1-14. http://doi.org/10.1371/journal.pone.0024893
Müller, V., & Lindenberger, U. (2014). Hyper-brain networks support romantic kissing in humans. PLoSONE, 9 (11). http://doi.org/10.1371/journal.pone.0112080
Müller, V., Perdikis, D., Oertzen, T. Von, Sleimen-Malkoun, R., Jirsa, V. K., & Lindenberger, U. (2016). Structure and Topology Dynamics of Hyper-Frequency Networks during Rest and Auditory Oddball Performance. Frontiers in Computational Neuroscience, 10 (October). http://doi.org/10.3389/fncom.2016.00108
Müller, V., Sänger, J., & Lindenberger, U. (2013). Intra- and Inter-Brain Synchronization during Musical Improvisation on the Guitar. PLoS ONE, 8 (9). http://doi.org/10.1371/journal.pone.0073852
Nikulin, V. V, Nolte, G., & Curio, G. (2012). Cross-frequency decomposition: a novel technique for studying interactions between neuronal oscillations with different frequencies. Clin. Neurophysiol, 123, 1353-1360.
Nunez, P. L., & Cutillo, B. A. (1994). Neocortical dynamics and human EEG rhythms. Oxford University Press, USA.
Obrig, H. (2014). NIRS in clinical neurology - a “promising” tool? NeuroImage. http://doi.org/10.1016yj. neuroimage.2013.03.045
Ouyang, G., Sommer, W., Zhou, C., Aristei, S., Pinkpank, T., & Abdel Rahman, R. (2016). Articulation Artifacts During Overt Language Production in Event-Related Brain Potentials: Description and Correction. Brain Topography, 29(6), 1-23. http://doi.org/10.1007/s10548-016-0515-1
Palva, S., & Palva, J. M. (2007). New vistas for alpha-frequency band oscillations. Trends in Neurosciences, 30 (4), 150-8. http://doi.org/10.1016/j.tins.2007.02.001
Paulus, C. (2009). DER SAARBRÜCKER PERSÖNLICHKEITSFRAGEBOGEN SPF(IRI) ZUR MESSUNG VON EMPATHIE: Psychometrische Evaluation der deutschen Version des Interpersonal Reactivity Index. Retrieved October 7, 2016, from http://hdl.handle.net/20.500.11780/3343
Peniston, E. G., & Kulkosky, P. J. (1989). Alpha-theta brainwave training and beta-endorphin levels in alcoholics. Alcoholism, Clinical and Experimental Research, 13 (2), 271-279. http://doi.org/10.1111/j.1530-0277.1989.tb00325.x
Peniston, E. G., & Kulkosky, P. J. (1991). Alpha-theta brainwave neuro-feedback therapy for Vietnam veterans with combat-related post-traumatic stress disorder. Medical Psychotherapy: An International Journal, 4, 47-60.
Picard, R. W. (1995). Affective Computing. MIT Press, (321), 1-16.
http://doi.org/10.1007/BF0123 8028
Piper, S. K., Krueger, A., Koch, S. P., Mehnert, J., Habermehl, C., Steinbrink, J., ... Schmitz, C. H. (2014). A wearable multi-channel fNIRS system for brain imaging in freely moving subjects. NeuroImage, 85, 64-71. http://doi.org/10.1016/jmeuroimage.2013.06.062
Pletzer, B., Kerschbaum, H., & Klimesch, W. (2010). When frequencies never synchronize: The golden mean and the resting EEG. Brain Research, 1335, 91-102.
http ://doi.org/10.1016/j. brainres.2010.03.074
Rakocevic, M. M. (1998). The genetic code as a Golden mean determined system. BioSystems, 46(3), 283-291. http://doi.org/10.1016/S0303-2647(98)00010-0
Raymond, J., Sajid, I., Parkinson, L. A., & Gruzelier, J. H. (2005). Biofeedback and dance performance: A preliminary investigation. Applied Psychophysiology Biofeedback, 30(1), 65-73. http://doi.org/10.1007/s10484-005-2175-x
Raymond, J., Varney, C., & Gruzelier, J. H. (2005). The effects of alpha/theta neurofeedback on personality and mood. Cognitive Brain Research, 23, 287-292.
Raymond, J., Varney, C., Parkinson, L. A., & Gruzelier, J. H. (2005). The effects of alpha/theta neurofeedback on personality and mood. Cognitive Brain Research, 23 (2-3), 287-292. http://doi.org/10.1016Zj. cogbrainres.2004.10.023
Reiner, M., Rozengurt, R., & Barnea, A. (2013). Theta neurofeedback training and enhancement of motor performance. Biological Psychology, (Neurofeedback special issue).
Reysen, S. (2005). Construction of a new scale: The Reysen likability scale. Social Behaviour & Personality: An International Journal, 33 (2), 201-208. http://doi.org/10.2224/sbp.2005.33.2.201
Riby, D. M., & Hancock, P. J. B. (2009). Do faces capture the attention of individuals with Williams syndrome or autism? Evidence from tracking eye movements. Journal of Autism and Developmental Disorders, 39 (3), 421-431. http://doi.org/10.1007/s10803- 008-0641-z
Rockstroh, B., Elbert, T., Birbaumer, N., Wolf, P., Düchting-Röth, A., Reker, M., ... Dichgans, J. (1993). Cortical self-regulation in patients with epilepsies. Epilepsy Research, 14 (1), 6372. http://doi.org/10.1016/0920-1211(93)90075-1
Roelfsema, P. R., Engel, A. K., König, P., & Singer, W. (1997). Visuomotor integration is associated with zero time-lag synchronization among cortical areas. Nature, 385(6612), 157-61. http://doi.org/10.1038/385157a0
Ros, T., Moseley, M. J., Bloom, P. A., Benjamin, L., Parkinson, L. A., & Gruzelier, J. H. (2009). Optimizing microsurgical skills with EEG neurofeedback. BMC Neuroscience, 10 (1), 8797. http://doi.org/10.1186/1471-2202-10-87
Ros, T., Munneke, M. A. M., Parkinson, L. A., & Gruzelier, J. H. (2014). Neurofeedback facilitation of implicit motor learning. Biological Psychology, 95 (1), 54-58. http://doi.org/10.1016/j.biopsycho.2013.04.013
Ruby, P., & Decety, J. (2003). What you believe versus what you think they believe: A neuroimaging study of conceptual perspective-taking. European Journal of Neuroscience, 17 (11), 2475-2480. http://doi.org/10.1046/j.1460-9568.2003.02673.x
Ruiz, S., Buyukturkoglu, K., Rana, M., Birbaumer, N., & Sitaram, R. (2014). Real-time fMRI brain computer interfaces: Self-regulation of single brain regions to networks. Biological Psychology, 95(1), 4-20. http://doi.org/10.1016/j.biopsycho.2013.04.010
Sacchet, M. D., & Gotlib, I. H. (2016). Neurofeedback training for major depressive disorder: recent developments and future directions. Expert Review of Neurotherapeutics, 7175 (June), 14737175.2016.1199959. http://doi.org/10.1080/14737175.2016.1199959
Sanes, J. N., & Donoghue, J. P. (1993). Oscillations in local field potentials of the primate motor cortex during voluntary movement. Proceedings of the National Academy of Sciences of the United States of America, 90 (10), 4470-4474. http://doi.org/10.1073/pnas.90.10.4470
Sanfey, A. G. (2007). Social decision-making: insights from game theory and neuroscience. Science (New York, N.Y.), 318 (5850), 598-602. http://doi.org/10.1126/science.1142996
Sänger, J., Lindenberger, U., & Müller, V. (2011). Interactive brains, social minds. Communicative & Integrative Biology, 4(6) 655-63. http://doi.org/10.4161/cib.17934
Sänger, J., Müller, V., & Lindenberger, U. (2012). Intra- and interbrain synchronization and network properties when playing guitar in duets. Frontiers in Human Neuroscience, d (November), 312. http://doi.org/10.3389/fnhum.2012.00312
Schack, B., & Weiss, S. (2005). Quantification of phase synchronization phenomena and their importance for verbal memory processes. Biol.Cybern., 92, 275-287.
Scharnowski, F., Hutton, C., Josephs, O., Weiskopf, N., & Rees, G. (2012). Improving Visual Perception through Neurofeedback. Journal of Neuroscience, 32 (49), 17830-17841. http://doi.org/10.1523/JNEUROSCI.6334-11.2012
Schmidt, R. C., & Richardson, M. J. (2008). Dynamics of interpersonal coordination. Understanding Complex Systems, 2008, 281-308. http://doi.org/10.1007/978-3-540- 74479-5_14
Scholkmann, F., Holper, L., Wolf, U., & Wolf, M. (2013). A new methodical approach in neuroscience: assessing inter-personal brain coupling using functional near-infrared imaging (fNIRI) hyperscanning. Frontiers in Human Neuroscience, 7 (November), 813. http://doi.org/10.3389/fnhum.2013.00813
Schurz, M., Radua, J., Aichhorn, M., Richlan, F., & Perner, J. (2014). Fractionating theory of mind: meta-analysis of functional brain imaging studies. Neuroscience and Biobehavioural Reviews, 42, 9-34. http://doi.org/10.1016/j.neubiorev.2014.01.009
Shi, Y., Park, T., Ruiz, N., Taib, R., Choi, E. H. C., & Chen, F. (2007). Galvanic Skin Response ( GSR ) as an Index of Cognitive Load. CHI EA ’07 CHI ’07 Extended Abstracts on Human Factors in Computing Systems, 2651-2656. http://doi.org/10.1145/1240866.1241057
Silani, G., Lamm, C., Ruff, C. C., & Singer, T. (2013). Right Supramarginal Gyrus Is Crucial to Overcome Emotional Egocentricity Bias in Social Judgments. Journal of Neuroscience, 33 (39), 15466-15476. http://doi.org/10.1523/JNEUROSCI.1488-13.2013
Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York Academy of Sciences, 1156, 81-96. http://doi.org/10.1111/j.1749-6632.2009.04418.x
Singer, W. (1995). Development and Plasticity of Cortical Processing Architectures. Science, 270, 758-764. http://doi.org/10.1126/science.270.5237.758
Stanley, D. A., & Adolphs, R. (2013). Toward a neural basis for social behaviour. Neuron, 80 (3), 816-826. http ://doi.org/10.1016/j.neuron.2013.10.038
Staufenbiel, S.W. Brouwer, A. M., Keizer, A. W., & Van Wouwe, N. C. (2014). Effect of beta and gamma neurofeedback on memory and intelligence in the elderly. Biological Psychology, 95, 74-85.
Sterelny, K. (2012). Morality’s dark past\rDie dunkle Vergangenheit der Moral. Analyse Und Kritik : Zeitschrift Für Sozialtheorie, 34 (1), 95. Retrieved from http://www.ub.uni- heidelberg.de/cgi- bin/edok?dok=http%3A%2F%2Fwww.gesis.org%2Fsowiport%2Fsearch%2Fid%2Fiz- solis-90580874%5Cnhttp://www.ub.uni-heidelberg.de/cgi- bin/edok?dok=http%3A%2F%2Fsearch.proquest. com%2Fdocview%2F 1029924532
Steriade, M. (2006). Grouping of brain rhythms in corticothalamic systems. Neuroscience, 137 (4), 1087-1106. http:// doi.org/10.1016/j. neuroscience.2005.10.029
Sterman, M. B., & Egner, T. (2006). Foundation and practice of neurofeedback for the treatment of epilepsy. Applied Psychophysiology Biofeedback, 31 (1), 21-35. http://doi.org/10.1007/s10484-006-9002-x
Sterman, M. B., & Macdonald, L. R. (1978). Effects of Central Cortical EEG Feedback Training on Incidence of Poorly Controlled Seizures. Epilepsia, 19 (3), 207-222. http://doi.org/10.1111/j.1528-1157.1978.tb04483.x
Strehl, U. (2014). What learning theories can teach us in designing neurofeedback treatments. Frontiers in Human Neuroscience, 8 (November), 894. http://doi.org/10.3389/fnhum.2014.00894
Sulzer, J., Haller, S., Scharnowski, F., Weiskopf, N., Birbaumer, N., Blefari, M. L., ... Sitaram, R. (2013). Real-time fMRI neurofeedback: Progress and challenges. NeuroImage, 76, 386399. http://doi.org/10.1016/j.neuroimage.2013.03.033
Tan, G., Thornby, J., Hammond, D. C., Strehl, U., Canady, B., Arnemann, K., & Kaiser, D. A. (2009). Meta-Analysis of EEG Biofeedback in Treating Epilepsy, 40 (3), 1-8.
Teplan, M. (2002). Fundamentals of EEG measurement. Measurement Science Review, 2(2), 1-11. http://doi.org/10.1021/pr070350l
Uddin, L. Q., Iacoboni, M., Lange, C., & Keenan, J. P. (2007). The self and social cognition: the role of cortical midline structures and mirror neurons. Trends in Cognitive Sciences, 11(4), 153-157. http://doi.org/10.1016Zj.tics.2007.01.001
Van Overwalle, F., & Baetens, K. (2009). Understanding others’ actions and goals by mirror and mentalizing systems: A meta-analysis. NeuroImage. http://doi.org/10.1016/j. neuroimage.2009.06.009
Varela, F., Lachaux, J. P., Rodriguez, E., & Martinerie, J. (2001). The brainweb: phase synchronization and large-scale integration. Nature Reviews. Neuroscience, 2 (4), 229- 239. http://doi.org/10.103 8/35067550
Vernon, D., Egner, T., Cooper, N., Compton, T., Neilands, C., Sheri, A., & Gruzelier, J. (2003). The effect of training distinct neurofeedback protocols on aspects of cognitive performance. International Journal of Psychophysiology, 47 (1), 75-85. http://doi.org/10.1016/S0167- 8760(02)00091-0
von Stein, A., & Sarnthein, J. (2000). Different frequencies for different scales of cortical integration: from local gamma to long range alpha/theta synchronization. International Journal of Psychophysiology, 38, 301-313.
Wang, J.-R., & Hsieh, S. (2013). Neurofeedback training improves attention and working memory performance. Clinical Neurophysiology, 124 (12), 2406-2420. http://doi.org/10.1016/j. clinph.2013.05.020
Weber, E., Köberl, A., Frank, S., & Doppelmayr, M. (2011). Predicting successful learning of SMR neurofeedback in healthy participants: Methodological considerations. Applied Psychophysiology Biofeedback, 36(1), 37-45. http://doi.org/10.1007/s10484-010-9142-x
Weiskopf, N. (2012). Real-time fMRI and its application to neurofeedback. NeuroImage, 62 (2), 682.
Weiss, H., & Weiss, V. (2003). The golden mean as clock cycle of brain waves. Chaos, Solitons and Fractals, 18 (4), 643-652. http://doi.org/10.1016/S0960-0779(03)00026-2
Yun, K., Watanabe, K., & Shimojo, S. (2012). Interpersonal body and neural synchronization as a marker of implicit social interaction. Scientific Reports, 2, 959. http://doi.org/10.103 8/srep00959.
[...]
- Quote paper
- Melinda Jeglinski (Author), 2017, The Combination of Electroencephalography (EEG) Hyperscanning and EEG Neurofeedback in Social Interaction, Munich, GRIN Verlag, https://www.grin.com/document/1297872
Similar texts
Publish now - it's free
Comments