PARAMETERS, AND METHODS IN THIS
3.3 METHODS AND DATA ANALYSES
3.3.1 SUBJECTS AND APPARATUS
Neurologically normal volunteers participated in the experiments of this study.
Informed consent was obtained from each participant. Subjects were comfortably seated in an armchair. A pile of two rectangular stainless-steel plates (20 cm long, 10 cm wide and 0.3 cm thick) was fixed on one side (the side of subjects' preferred arm) of the armrest.
Between the two plates was a short piece of wood at one end. A strain gauge was attached
on the upper plate to detect deformation of the plate. The subjects put their preferred
forearm on the upper plate. Subjects contracted the triceps and/or the biceps muscles of
the upper arm, and the force was detected by the strain gauge. This was seen on the monitor window of an oscilloscope (VC-6723, Hitachi Co. Ltd., Japan). The oscilloscope was placed one meter in front of the subjects. The subjects were instructed to keep their eyes open and maintain a stable arousal level of consciousness during the experimental trials.
3.3.2 STIMULATION
In the experiments of this study, somatosensory electrical stimulation was applied to the fingers. This type of stimulation affects all sensory receptors in the skin (pressure, touch, temperature, and pain), and afferent inputs derived from the stimulation are transferred to the CNS by sensory neuron innervating the skin. Pathways for somatosensory information from the sensory receptors to the CNS are known in details, and better understood than those of auditory and visual information. In a number of studies on ERPs, somatosensory stimulation has often been used to stimulate mixed nerves such as the median nerve. Such a stimulation can activate motor nerves as well as sensory nerves. If the stimulation is relatively strong, involuntary movements could appear as a reflex. As this study aimed to investigate the information processing during voluntary movements, somatosensory stimulation was applied to the fingers but not the mixed nerves.
Pain-related potential (Kakigi et al., 1989) is also likely to be mixed with ERPs.
The intensity of stimulus was selected to range from two to three times the subjective sensory threshold measured at each finger, resulting in an intensity strong enough to be perceived accurately without pain.
Electrical square stimuli of 0.2 msec duration were generated by the electrical
stimulators (3F46, NEC Medical Systems Co. Ltd., Japan). These stimuli were delivered
to the fingers of preferred hand through ring electrodes attached at the middle of the first
(cathode) and second phalanx (anode) of each finger. In Experiments 3 and 4, an auditory
click (50-60 dB, 1 msec duration), delivered to the subjects via headphones, was used as a
warning signal (WS). The auditory signal was generated using auditory stimulators (ST-5,
Medelec, UK, or SMP-4100, Nihon-Koden Co. Ltd., Japan). Both the electrical and
auditory stimuli were recorded as the trigger pulses used for subsequent off-line averaging
of both the ERPs and SSR data.
Chapter 3 Issues to be Investigated, Parapeters, and Methods
3.3.3 RECORDING ANALOGUE DATA FOR EACH PARAMETER 3.3.3.1 Electroencephalogram (EEG)
EEG was recorded using Ag/AgC1 disk electrodes placed on the scalp, according to the International 10-20 system with the reference of linked earlobes. The EEG analogue output was amplified through a bandpass filter of 0.53-120 Hz.
3.3. 3.2 Electro-oculogram (EOG)
To monitor likely artifacts due to eye movements, EOG was recorded using a pair of small electrodes placed above and below the eye of one side.
3.3.3.3 Electromyogram (EMG)
EMG was recorded using pair(s) of surface electrodes on the triceps and/or biceps muscles of the preferred arm, mostly the right arm, and was amplified through a bandpass filter of 5.3-1500 Hz.
3.3.3.4 Sympathetic skin response (SSR)
SSR was recorded using Ag/AgC1 disk electrodes placed on both the palmar and dorsal sites of non-preferred hand, mostly the left hand, and was amplified through a bandpass filter of 0.53-1500 Hz.
3.3.3.5 R-R intervals (only Experiment 3)
To examine R-R intervals as an autonomic response of the heart, ECG was recorded using three electrodes, one of which was placed on the sternum and the other two
on each of the fifth ribs.
3.3.4 DATA STORAGE
All analog signals including the electrical signals converted from both electrical stimulation and auditory signals (WS) were recorded both on recording paper of an EEG recorder (1A97, NEC San-ei Co. Ltd., Japan, or EE 1121 A, NEC Medical Systems Co.
Ltd., Japan) and on magnetic tape of a data recorder (XR-710, TEAC Co., Japan, or PC216Ax, Sony Precision Technology Inc., Japan).
3.3.5 DATA ANALYSES
Analog data of EEG, EOG, EMG and SSR stored on magnetic tape were
converted into digital data through an A/D converter installed on a personal computer (PC-9821 Xa7, NEC Co. Ltd., Japan), and analyzed with software (EPLYZER, Kissei Comtec Co. Ltd., Japan). Trials with either excessive muscle activity or eye blinks (detecting from EMG and EOG) were excluded from subsequent analyses in order to eliminate likely artifacts on the averaged waveforms of both ERPs and SSR.
3.3.5.1 EEG, N140 and P300
EEG analogue data were converted into digital data at a sampling rate of 200 or 500 Hz for 800 msec (sampling time ranging from 200 msec before the stimulus onset to 600 msec after the stimulus onset) and then averaged for each stimulus condition. The N140 and P300 components of ERPs were defined at peak amplitudes appearing in two different post-stimulus windows ranging from 120 to 160 and from 245 to 450 msec, respectively. The amplitudes of N140 and P300 were measured as relative potentials from a 200-msec prestimulus baseline. The latencies of N140 and P300 were measured as the time elapsing from the stimulus onset to the peak amplitudes.
3.3.5.2 EMG
EMG data were converted into digital data at a sampling rate of 1000 Hz and used to determine the EMG-RT.
3.3.5.3 SSR
SSR data were converted into digital data at a sampling rate of 200 or 500 Hz with a sampling time of 3700 (Experiment 1), 4500 (Experiments 2 and 4), and 4700 msec (Experiment 3), and then averaged. The amplitude of SSR was measured as a peak-to-peak difference of the averaged waves. The latency of SSR was measured as the duration from the stimulus onset to the response onset determined at an initial rise from the 200-msec prestimulus baseline.
3.3.6 STATISTICS
Repeated measures analyses of variance (ANOVA) were performed on both the amplitudes and latencies of ERPs and SSR in each experiment. To decrease the experiment-wise error rate due to the repeated-measures design, a Greenhouse-Geisser adjustment to the degree of freedom was performed. The student's t-test was also used in part. A level of P<0.05 was accepted as indicating statistical significance.
Chapter 4 Experiments