by Klaus Gramann
Biological Psychology and Neuroergonomics, Berlin Institute of Technology (Germany)
Imaging the human brain during active behavior is essential to understand how the brain supports natural cognitive processes that are based on and make use of our physical structure. A new imaging modality, mobile brain/body imaging (MoBI), uses electroencephalography (EEG) synchronized to motion capture and other data streams to investigate brain activity while participants actively move in and interact with their environment.
Decades of brain imaging studies have described the functional architecture of the human brain to a hitherto unknown level of granularity. However, in all these studies, human participants are not allowed to move, sometimes, mostly in EEG experiments, not even to move their eyes to avoid movement artifact to compromise the signal of interest. Human cognition, however, is embodied in the sense that cognitive processes are based on and make use of our physical structure while being situated in a specific environment (Wilson, 2002). From the viewpoint of embodied cognition a primary function of human cognition and perception is to assist motor control (Churchland et al., 1994) and, as a consequence, the brain dynamics accompanying cognitive processes are tightly coupled to motor behavior in our natural surroundings. This is reflected, inter alia, in the fact that brain areas and activities that originally evolved to organize motor behavior of animals in their three-dimensional environments also support human cognition (Rizzolatti et al., 2002).
Motor behavior influences cognitive processing on several levels, including the integration of movement-related feedback from our senses with the objective of reaching a cognitive goal. A simple example is spatial navigation where the human brain efficiently integrates among other senses, vestibular, proprioceptive, and visual information that accompany every translational or rotational movement of the navigator. Even without vision we are perfectly able to indicate the position of our office chair after a few steps into different directions in the room (yes, you can try this out now). This ability is called path integration (Mittelstaedt and Mittelstaedt, 1980), a continuous integration of movement cues that allows efficient updating of position and orientation even in the absence of vision. Like path integration, many other natural cognitive processes make use of and are based on movement-related sensory information. The brain dynamics supporting this kind of Natural Cognition, however are not systematically investigated. On the contrary, traditional brain imaging methods force participants to sit still or lie prone to avoid any movements. Why is there a lack of studies investigating the brain dynamics underlying cognitive processing during active behavior?
This is primarily due to technical constraints of traditional brain imaging methods like functional magnetic resonance imaging (fMRI) or magnetoencephalography (MEG) that require participants to remain motionless and that are highly sensitive to movement artifacts. This imposes a fundamental mismatch of the recorded brain dynamics and participants’ actions that are usually restricted to a button press after each trial (Makeig et al., 2009). To overcome this mismatch a brain imaging modality has to be used that allows for movement of the participant as part of natural cognitive processes. Only EEG involves sensors light enough to allow movement of the head and body. Furthermore, EEG provides sufficient time resolution to record brain activity on the time scale of natural motor behavior, making this method the clear choice for brain imaging of humans performing tasks involving natural action (Gramann et al., 2010, Gwin et al., 2010, 2011). So, what would a mobile brain/body imaging approach look like?
The mobile brain/body imaging (MoBI) approach
The MoBI approach can be described by three contributing main areas. 1) Hardware that is necessary to record brain activity accompanying natural cognition in mobile humans, 2) the software that allows synchronous recording and analyses of different data streams from different devices, and last but not least 3) the application of MoBI to specific research areas.
> Hardware. There are two essential hardware components to MoBI which can be extended to include several additional modules. The first essential component is the EEG (functional near infrared spectroscopy (fNIRS) is also an option but is not further discussed in this article). Using EEG allows for recordings of brain dynamics in actively moving participants with high temporal resolution without too strong limitations regarding the movement range. To allow fully mobile experimental setups the EEG system should be wireless or small and lightweight enough so that the amplifier can be put in a backpack and carried around by the participant. In addition it will be an advantage to use actively amplified electrodes and/or shielded cables to avoid electromagnetic interference and other artifacts in non-standard laboratory conditions from contaminating the signal.
Secondly, if we are interested in the brain dynamic patterns that reflect the interplay of movement and cognition we have to synchronously record EEG together with participants’ movements. Thus, the second essential hardware component to MoBI is motion capture. Optical motion capture systems allow precise recordings of participants’ motor behavior inside small or larger laboratories including information on absolute position. If participants move outside the lab, inertial motion capture systems can be used.
A possible setup is used here at the Berlin Mobile Brain/Body Imaging Lab (BeMoBIL) providing a 140 m2 laboratory environment and a smaller 50m2 MoBI lab equipped with the Brain Products MOVE system (Brain Products, Gilching, Germany) with 160 channels active amplified electrodes, wirelessly transmitted via the MOVE system, and synchronized to a 30 cameras active infrared motion capture system (PhaseSpace, San Leandro, CA, US). Besides EEG and motion capture additional hardware components can be added to the system, including but not limited to other physiological recordings like electromyography, electrocardiography and virtual reality. The BeMoBIL will integrate the Oculus Rift DK2 (Oculus VR, Irvine, CA, US), a head mounted VR system, that allows for creating virtual worlds up to the size if the BeMoBIL with 140 m2 physical space.
> Software. Besides hardware challenges, there are significant requirements for software solutions to recording MoBI data. First, the synchronization of different physiological data streams with different sample rates from motion capture, virtual reality goggles, and other devices like eye tracking and force platforms is a non-trivial problem. Most data streams come with different sample rates and all need event information with exact and comparable timing. A powerful solution to this recording problem is provided by the Laboratory Streaming Layer (LSL) framework (code.google.com/p/labstreaminglayer) developed by Christian Kothe at the Swartz Center for Computational Neuroscience. LSL drivers for different hardware receive data transported across a local area network (LAN) using a UDP protocol. LSL saves the data streams with time markers allowing later analysis of multi-stream data with respect to synchronous activity in more than one stream. The LSL framework further allows for near real-time computation, and/or visualization of the data for better experiment control and supervision.
A second non-trivial problem is the analyses of multimodal data. Parallel processing of different data streams in synchrony is necessary to allow e.g. artifact detection and preprocessing with respect to the interplay of all modalities. The MoBILAB software framework (sccn.ucsd.edu/wiki/MoBILAB; Ojeda et al., 2014) developed by Alejandro Ojeda at the SCCN is a powerful approach to visualize and process multi-data streams recorded through LSL. MoBILAB is an open source toolbox supporting the analysis and visualization of mixtures of synchronously recorded data streams. MoBILAB can serve as a pre-processing environment for editing event markers for further processing in all data streams, and as a platform to expand analysis of multi stream data recorded simultaneously. Special emphasis in MoBI research has to be placed on EEG-data analyses because surface recordings might be strongly contaminated with non-brain activity like muscle and eye movements as well as also mechanical and electromagnetic artifacts. Data-driven analysis approaches must allow separation of different brain and non-brain activities. While Independent component analysis (ICA) and related blind source separation methods have proven effective for separating brain from non-brain activities from electrophysiological data (Makeig et al., 2004, Gramann et al., 2014a), new analyses approaches will evolve and have to be tested for different experimental protocols and levels of data quality.
> Application. After our seminal paper on MoBI in 2009 (Makeig et al., 2009) a number of labs have adopted the MoBI approach and conducted innovative and exciting experiments that provided completely new insights into the brain dynamics of actively behaving humans. First papers describing the potential and limits of using MoBI in walking and running participants (Gramann et al., 2010, Gwin et al., 2010, 2011) were followed by publications in different areas including gait rehabilitation (Wagner et al., 2012), general kinematics (Presacco et al., 2012), dual motor tasks (De Sanctis et al., 2014), spatial cognition (Ehinger et al., 2014) and many more. An overview on this field is given in a recently published Research Topic on MoBI in Frontiers in Human Neuroscience (Gramann et al., 2014b).
MoBI is a new approach to imaging the human brain dynamics supporting natural cognition. It overcomes the restrictions of traditional brain imaging methods and allows to investigate the brain dynamics supporting cognition that unfold in synchrony with active behavior. The first publications already revealed impressive new insights into human brain function and this new research area develops quickly alongside developments in hardware and new software approaches including new algorithms to analyze the complex and often artifact-loaded data.
The first meeting of international MoBI researchers took place at the Hanse-Wissenschaftskolleg Delmenhorst in 2013. Here a small group of enthusiastic researchers developed ideas and approaches for future MoBI research. Following a MoBI workshop at this years’ OHBM meeting in Hamburg the MoBI Society was established which has officially launched its’ websites in October 2014 (mobi-research.org). As the community grows I invite all interested researchers working in this area to join us, to discuss their work, to exchange ideas, data, and code, and to excel this exciting research on mobile brain/body imaging.
 Churchland PS, Ramachandran VS, Sejnowski TJ (1994) A Critique of Pure Vision. In: Large-scale neuronal theories of the brain (Koch, C. and Davis, J. L., eds), pp 23-60 Cambridge, MA: MIT Press.
 De Sanctis P, Butler JS, Malcolm BR, Foxe JJ (2014) Recalibration of inhibitory control systems during walking-related dual-task interference: A Mobile Brain-Body Imaging (MOBI) Study. Neuroimage 94:55-64.
 Ehinger BV, Fischer P, Gert AL, Kaufhold L, Weber F, Pipa G, König P (2014) Kinesthetic and Vestibular Information Modulate Alpha Activity during Spatial Navigation: A Mobile EEG Study. Front Hum Neurosci 8.
 Gramann K, Ferris DP, Gwin J, Makeig S (2014a) Imaging natural cognition in action. Int J Psychophysiol 91:22-29.
 Gramann K, Gwin JT, Bigdely-Shamlo N, Ferris DP, Makeig S (2010) Visual evoked responses during standing and walking. Front Hum Neurosci 4:202.
 Gramann K, Jung T, Ferris DP, Lin C, Makeig S (2014b) Towards a New Cognitive Neuroscience: Modeling Natural Brain Dynamics. Name: Frontiers in Human Neuroscience 8:444.
 Gwin JT, Gramann K, Makeig S, Ferris DP (2010) Removal of Movement Artifact From High-Density EEG Recorded During Walking and Running. J Neurophysiol 103:3526-3534.
 Gwin JT, Gramann K, Makeig S, Ferris DP (2011) Electrocortical activity is coupled to gait cycle phase during treadmill walking. Neuroimage 54:1289-1296.
 Makeig S, Debener S, Onton J, Delorme A (2004) Mining event-related brain dynamics. Trends Cogn Sci 8:204-210.
 Makeig S, Gramann K, Jung TP, Sejnowski TJ, Poizner H (2009) Linking brain, mind and behavior. International journal of psychophysiology : official journal of the International Organization of Psychophysiology 73:95-100.
 Mittelstaedt ML, Mittelstaedt H (1980) Homing by Path Integration in a Mammal. Naturwissenschaften 67:566-567.
 Ojeda A, Bigdely-Shamlo N, Makeig S (2014) MoBILAB: an open source toolbox for analysis and visualization of mobile brain/body imaging data. Front Hum Neurosci 8.
 Presacco A, Forrester LW, Contreras-Vidal JL (2012) Decoding Intra-Limb and Inter-Limb Kinematics During Treadmill Walking From Scalp Electroencephalographic (EEG) Signals. Ieee T Neur Sys Reh 20:212-219.
 Rizzolatti G, Fogassi L, Gallese V (2002) Motor and cognitive functions of the ventral premotor cortex. Curr Opin Neurobiol 12:149-154.
 Wagner J, Solis-Escalante T, Grieshofer P, Neuper C, Muller-Putz G, Scherer R (2012) Level of participation in robotic-assisted treadmill walking modulates midline sensorimotor EEG rhythms in able-bodied subjects. Neuroimage 63:1203-1211.
 Wilson M (2002) Six views on embodied cognition. Psychon Bull Rev 9:625-636.
©Brain Products GmbH 2014