SELF-MOVEMENT, OBSERVATION, AND IMAGINATION EFFECTS ON Mu RHYTHM AND READINESS POTENTIALS (RPs): TOWARDS A BRAIN-COMPUTER INTERFACE (BCI)

 

 

 

 

J. A. Pineda, B.Z. Allison, and A. Vankov

 

 

 

Department of Cognitive Science

University of California, San Diego

La Jolla, CA 92093

 

 

 

Please send correspondence to:

Jaime A. Pineda, Ph.D.

Department of Cognitive Science 0515

University of California, San Diego

La Jolla, CA 92093

(858) 534-7087 (office)

pineda@cogsci.ucsd.edu

 

ABSTRACT

            Current movement-based BCIs utilize spontaneous EEG rhythms associated with movement, such as the mu rhythm, or responses time-locked to movements that are averaged across multiple trials, such as the Readiness Potential (RP) as control signals.  In one study, we report that the mu rhythm is not only modulated by the expression of self-generated movement but also by the observation and imagination of movement.  In another study, we show that simultaneous self-generated multiple limb movements exhibit  properties distinct from those of single limb movements.  Identification and classification of these signals with pattern recognition techniques provides the basis for the development of a practical BCI.

INTRODUCTION

            The concept of a direct interface between the human brain and a sophisticated artificial system, such as a computer, is not a new one. In recent years, there have been advances in a number of fields that make the design and development of a practical brain computer interface (BCI) possible.  Such a BCI would be capable of quickly and reliably extracting meaningful information from the human electroencephalogram (EEG) or other recordable electrical  potentials.  Over the past decade, several working BCI systems have been described in the literature [2, 3, 6, 7, 8].  These systems use a variety of collection mechanisms, pattern recognition approaches, and interfaces, and require different types of cognitive activity on the part of the user.   

One type of BCI that has been examined extensively derives information from a user’s movements or the imagination of movement.  Many of these movement-based BCIs can recognize changes in the human mu rhythm, which is an EEG oscillation recorded in the 8-13 Hz range from the central region of the scalp overlying the sensorimotor cortices [4].  This rhythm is large when a subject is at rest, and is known to be blocked or attenuated by self-generated movement.  Indeed, the mu wave is hypothesized to represent an “idling” rhythm of motor cortex that is interrupted when movement occurs. The free running EEG shows characteristic changes in mu activity, which are unique for the movement of different limbs. These findings have and will continue to be useful in the construction of BCI systems.

The imagination or performance of a movement is also generally accompanied by a readiness potential (RP; also called Bereitshaftspotential or BP) which is most prevalent over cortical motor areas.  The RP is a time-locked response to the movement event, or event related potential (ERP), that is extracted from the ongoing EEG using signal averaging techniques across a number of trials.

The primary goal of the two studies we report was to characterize mu and RP signals in simple, straightforward tasks.  The recognition and discrimination of these signals could then provide a basis for the development of a practical BCI, one that would be useful to both normal and disabled individuals.

STUDY 1

            In this study, we show that the mu rhythm is attenuated not only by self-generated movement but also when a subject observes the movement or imagines making the same, self-generated movement.  According to Rizzolatti and colleagues, the responsiveness of the mu wave to visual input may be the human electrophysiologic analog of a population of neurons in area F5 of the monkey premotor cortex [1,5].  These cells respond both when the monkey performs an action and when the monkey observes a similar action made by another monkey or by an experimenter.  Other, older studies have reported that  mu-like waves are blocked by thinking about moving.  For example, individuals with amputated limbs can block this rhythm by thinking about moving the amputated limb.  The blocking of the mu rhythm by visual and imagery input may have implications for understanding movement-related responses and for the rehabilitation of movement-related neurological conditions. 

METHODS

            Subjects in this study were 17 healthy volunteers (10 men, 7 women, ranging in age between 19-58, with a mean of 27.7 years).  Most subjects were students or employees at the University of California, San Diego (UCSD) and naive to the purposes of the experiment.  Only 10 subjects were used for statistical analysis because of problems with noise.  All subjects signed a consent form that was approved by the UCSD Human Subjects IRB committee.

            EEG signals were recorded from 6 sites on an electrode cap placed over frontal (F3, F4), central (C3, C4), and occipital (O1, O2) areas according to the standard 10-20 International Electrode Placement System.  Blinks and eye movements were monitored with an electrode in the bony orbit dorsolateral to the right eye.  EEG was amplified by a Grass model 7D polygraph using 7P5B pre-amplifiers with bandpass of 1-35 Hz.  For computerized data collection and analysis, the ADAPT ( ã A. Vankov, 1997) scientific software was used.   EEG was digitized on-line for two minutes at a sampling rate of 256 Hz.  All electrode sites showed impedance of less than 5 kOhms.

            Subjects participated in four conditions: 1) rest: in which subjects sat in a comfortable chair inside and acoustic chamber, but no particular task was required; 2) self-generated movement: subjects were asked to move their opposing thumb to middle fingers of the right hand (making a “duck” movement); 3) observation: subjects watched a confederate of the experimenter perform the “duck” movement; and 4) imagination: subjects were instructed to imagine performing the self-generated “duck” movement without actually doing it.  The confederate faced the subject who was seated approximately four feet away throughout all conditions of the experiment.   The power spectrum was calculated for each second of the EEG, and mean power within the mu range (8-13 Hz) was calculated for each condition over two minutes.

RESULTS

            During the rest condition, subjects exhibited significant power in the 8-13 Hz frequency range.  This rhythm showed statistically significant changes during the various conditions (F(3,27)=4.98, P<0.01).  Pairwise comparisons showed that the main differences were a reduction in power during the self-generated movement and the observation conditions (see Fig. 1).   Post-hoc analysis of the data showed that during the imagination condition, mu power decreased at frontal sites but was less affected at central and occipital sites (site x condition, (F(15,135)=2.22, P< 0.01).

STUDY 2    

             Numerous studies have explored the RPs and mu changes associated with single movements of the finger and hand. However, the electrophysiology of left and right foot movement, or those preceding the voluntary simultaneous movement of multiple limbs, has not been thoroughly explored. This information is necessary to better understand how the brain’s activity gives rise to different movements, and also expands the range of input signals that could be used in a BCI.

            This study recorded EEGs from human subjects performing voluntary movements of either one or two limbs at self-paced intervals. Results confirmed that each type of movement is associated with unique EEG characteristics that were classified using Thoughtform Interpretation Studio (c) software. 

METHODS

            A total of 18 subjects (mean age 23.7 +/- 2.8 years) were run in this experiment. Seven subjects were female, with 3 of the female and two of the male subjects being left handed. Most were undergraduate students at UCSD who were compensated with either class credit or monetary payment. All subjects were native English speakers, with no sensory or motor deficits and no history of psychological disorders.

            EEG activity was recorded using an electrode cap with monopolar Ag/AgCl electrodes overlying nine cortical sites: F3, Fz, F4, C3, Cz, C4, P3, Pz, and P4 (according to the standard 10-20 International Electrode Placement System) referenced to linked mastoids. Signals were amplified 10,000 times and recorded with a bandpass between .1-100 Hz. EOG activity was recorded through an electrode placed over the right orbital bone. Eye activity was amplified 5,000 times and with a bandpass of  .3-100 Hz. All electrode sites had an impedance of less than 5 kOhms. Subjects’ hand movements were detected through two joysticks, while a foot pedal device recorded foot movements. All data, including subjects’ movements and EEG, were sampled at 256 Hz and recorded on a computer for later analysis.

            In single movement trials, subjects made voluntary movements of either left or right hand or left or right foot during a 10 minute-long trial block. The movements could be of any limb and there was at least a five-second delay between each movement.  Subjects were instructed not to worry about randomizing which limb was moved or ensuring a fair distribution of different limb movements. Instructions were identical for multiple movement trials except that they were asked to move any two limbs simultaneously.

CLASSIFICATION PROCEDURE

            Thoughtform Interpretation Studio (TIS 2.0) is a commercially available product that makes EEG pattern recognition and single trial event detection possible.  During the learning mode, input signals are decomposed into features on a multidimensional phase space using a variety of techniques, including time-frequency expansion, feature coherence analysis, and principal component analysis (PCA).  This is followed by a state discriminant analysis to find feature clusters that are most reliably different between two epoch types.   The resulting Interpretation Maps are then used in the interpretation mode to classify new single trial data.  We used the single-limb RP averaged data collected in this study as input to TIS to determine whether trial data for left/right-hand and left/right-foot movements could be easily discriminated. 

RESULTS

            The ERP data obtained show that the RPs preceding 2 types of combined movement (left foot/right hand movement and right foot/left hand movement) have a significantly larger peak amplitude than any other single or combined movement (see Fig. 2). In addition, each of the four single movement types shows unique RP and mu rhythm characteristics.  TIS classification of averaged data resulted in reliable discrimination of movement categories with 60-100% confidence.  We are currently examining the results of single trial analyses.

GENERAL DISCUSSION

                Our initial attempts to elicit reliable mu and RP signals by self-generated movement, observation of movement, and by the imagination of movement have been successful.  Furthermore, attempts to discriminate such signals at the single trial level have also been encouraging.  Future work will involve detecting RPs to imagined single and multiple movements and comparing those to RPs evoked by overt movements.  We also intend to evaluate other preprocessing approaches, such as Independent Component Analysis (ICA), as well as neural networks for the classification of signals.   

LITERATURE CITED

1.   L. Fadiga, L. Fogassi, G. Pavesi, G. Rizzolatti.    Motor facilitation during action observation: A magnetic stimulation study. Journal of Neurophysiology, 73 (6): 2608-2611, 1995.

2.   L. A. Farwell and E. Donchin.   Talking off the top of your head:  Toward a mental prosthesis utilizing event-related brain potentials.  Electroenceph. Clin. Neurophysiol., 59: 236-245, 1988.

3.   D. J. McFarland, T. Lefkowicz, and J. R. Wolpaw.   Design and operation of an EEG-based brain-computer interface with digital signal processing technology.  Beh. Res. Meth. Inst & Comp., 29(3): 337-345, 1997.

4.   G. Pfurtscheller.  Functional topography during sensorimotor activation studied with event-related desynchronization mapping.  J. Clin. Neurophysiol., 6: 75-84, 1989.

5.   G. Rizzolatti and L. Fadiga.  Grasping objects and grasping action meanings:  dual role of monkey rostroventral premotor cortex (area F5).  Novartis Foundation Symposium, 218: 81-95, 1998.

6.   T. M. Vaughan, J. R. Wolpaw, and E. Donchin.  EEG-based communication:  prospects and problems.  IEEE Trans Rehab Eng., 4: 425-430, 1996.

7.   J. R. Wolpaw and D. J. McFarland. Multichannel EEG-based brain-computer communication.  Electroenceph. Clin. Neurophysiol., 90: 444-449, 1994.

8.      J. R. Wolpaw, D. J. McFarland, G. W. Neat, and C. A. Forneris.   An EEG-based brain-computer interface for cursor control.  Electroenceph. Clin. Neurophysiol., 78: 252-259, 1991.

FIGURE LEGENDS

Fig. 1.  Left panel represents twenty seconds of EEG data for a single subject during the REST, SELF-GENERATED MOVEMENT (opposing thumb to middle two fingers or “duck” movement), OBSERVATION, and IMAGINATION conditions.  Center panels represent the integrated power for the entire 120 seconds, expressed in mV2/sec.  Panels on the right depict both time and frequency information.  For each of the 120 seconds, the integrated power in the 8-13 Hz range (|FFT[VC3(t) – VC4(t)](w)|2dw) is shown in mV2/sec so as to be comparable to the scale in center panels.

Fig. 2.   Grand average ERPs recorded at the vertex site (Cz) during the second preceding voluntary movement combinations of two limbs.  The Readiness Potential (RP) is the increased negativity that develops hundreds of milliseconds prior to the movement at time zero.  Note that the movements of diagonally opposing limbs, shown in the center, produce larger RPs than other movement combinations.