Post-movement beta synchronisation after complex prosaccade task

2.1. Subjects 

Eight healthy, right-handed subjects were investigated, but two were excluded because they were not able to perform the task without excessive blinking. The average age of the 6 subjects (3 men) was 62±7.1years (range: 51.8 and 73.8years). None of the subjects had history of neurological, ophthalmologic or vestibular disease. Subjects were without medication. No alcoholic beverages or sleeping pills were allowed for at least 2days before the investigation. Brain imaging (CT or MRI) performed prior to our study showed no intracranial abnormality. On the mini mental status scale examination (MMSE) each subject had the maximal 30 points. None of the subjects had impairment of visual attention based on Toulouse–Pieron neuropsychological test. Smooth pursuit eye movements (SPEM), saccades, fixation and optokinetic nystagmus (OKN) were examined horizontally and vertically using electro-oculography (EOG) (PowerLab, Chart5 v5.5© 1994–2006 ADInstrument Ltd.) according to standard protocols (Marmor and Zrenner, 1993; Marmor, 1998). Each subject had normal conjugated eye movements, normal fixation, saccades, SPEM and OKN compared to age-matched data published previously (Zackon and Sharpe, 1987; Peterka et al., 1990; Munoz et al., 1998; Yang et al., 2002). Each subject had normal EEG activity at rest with their eyes open. The study protocol was approved by the Local Ethics Committee, subjects signed informed consent.

2.2. Test procedure 

Subjects were seated in a comfortable chair with neckrest in a darkened, soundproof room at a distance of 1.3m from a 1.5-m-wide black screen, which was seen from 60°. The complex saccade task was stimulated by a white dot (4°) projected onto the screen using a standard unpredictable task protocol (Gaymard et al., 1998; Ettinger et al., 2005). Subjects were instructed to look at the white dot without turning the head. At the beginning of a task the dot appeared at the centre of the screen then sprang to the right or left side (30°) randomly. After 3.1±0.1s the white dot sprang back to the centre of the screen and stayed there for 12s (rest period). After rest a new complex saccade was initiated. Direction of the complex task was named according to the direction of the second saccade. The intersaccadic interval was defined as the period between the end of the first saccade and the return of the white dot to the centre of the screen. The flowchart of the complex saccade task is shown in Fig. 1. One test block contained 60 complex saccade tasks, 30 directed to the right and 30 to the left in a randomly selected order. Tests were repeated until 15 tasks without blinking or any other artifacts could be collected (5 subjects had to repeat the test two or three times, while 1 subject five times because of artifacts). Subjects were provided rest in dark for 10min between each test block.

View Large Image Download to PowerPoint

Fig. 1. Flowchart of the complex prosaccade task (left-directed). The test consisted of two saccades evoked by a moving white dot projected onto a black screen (top row). The subject focused on the dot (A) which after about 12s suddenly appeared on the right side of the screen (B) evoking the first saccade. Approximately 3s later the dot turned back to the centre (C) evoking the second saccade. Saccade latency (grey stripes) is calculated as the delay of eye movement after the appearance of the white dot in a new location (EOG, electro-oculography; R, right; L, left).

2.3. EEG and EOG recording 

Horizontal and vertical movements of the right eye were recorded using EOG. Four Ag/AgCl electrodes were placed around the right eye. DC amplification and 30Hz upper cut off frequency were used (Brain Star EEG-Recording 3.55© Hienert-Brandl Software, Germany). The EEG was recorded simultaneously with the EOG. Twenty-five EEG electrodes were placed on the scalp according to the modified international 10–20 system. C1, Cz, C2, FC1, FCz and FC2 electrodes are referring from the vicinity of the premotor and supplementary motor areas, P1, P2 and P3, P4 electrodes are positioned above the posterior parietal cortex including the intraparietal sulcus and parietal eye field (PEF) (Homan et al., 1987; Steinmetz et al., 1989; Herwig et al., 2003). Electrode impedance was kept below 5kΩ, the time constant was 1s, the upper cut off frequency was 70Hz. Linked earlobes were used as reference similarly to previously published studies (Alegre et al., 2002, Alegre et al., 2004). Three different markers signed online the appearance of the white dot on the right or the left edge or in the centre of the screen. Analogue EEG signals were digitized at a sampling rate of 1024Hz. Data were stored on disk for offline analysis.

2.4. Data analysis 

EEG was analyzed offline using Brain Vision Analyzer (Version 1.05.0003 Brain Products GmbH 1998–2006) after ocular correction performed by the analysis system prior to re-referencing to the common average of scalp electrodes. The amplitude of movement related EEG changes might be reduced when recorded using the common average reference method compared to the reference-free application, but it does not affect the localisation of event-related activity changes (Pfurtscheller, 1992).

We selected 15–15 epochs including saccades directed both to right and left. Saccade latency was determined as the time delay from the appearance of the dot in a new position to the start of the eye movement measured by EOG. In each task termination of the second saccade was marked according to the EOG. EEG spectral analysis from 6s before to 3s after the termination of the second saccade was carried out using Fast Fourier Transformation (FFT) with Hahn window.

We determined the most reactive beta frequency (MRBF) in the 15 to 30Hz range of each individual at each electrode position using average time–frequency–power distribution maps (Lopes da Silva, 1999; Tamás et al., 2003) (Fig. 2).

View Large Image Download to PowerPoint

Fig. 2. Average time–frequency–power map of 15 left-directed saccade-tasks recorded at FC1 (1.row) and FC2 (2.row) electrode positions (Subject Nr6). The flowchart shows the timing of ocular movements (EOG). The 1-s reference interval for EEG data analysis started 6s before the termination of the task. The intersaccadic interval was defined as a period between the termination of first saccade and the return of the white dot to the centre of the screen. The absolute power measured at FC1 and FC2 electrodes plotted against time shows the PMBS (black thick arrows) 1250 and 1375ms after the termination of the complex saccade task. Note that PMBS does not develop after the first saccade (MRBF, most reactive beta frequency; R, right; L, left).

We plotted the power changes of the MRBF against time at each electrode position individually for every single trial. Relative power changes were calculated using the following equation: Power %=(A−R)/R×100, where A is the absolute power at a given time, and R is the mean power of the reference interval (Pfurtscheller and Lopes da Silva, 1999). The reference interval was the first 1s of the 6-s-long period prior to the offset of the second saccade (Fig. 2). The mean power % of the reference period, by definition, was “0%”. The intersaccadic beta peak power % value was identified as the highest power % value between the end of the first saccade and the beginning of the second saccade (Fig. 2). PMBS was defined as the highest power % value after the termination of the ocular movement. Latency of PMBS, i.e. the elapsed time between the termination of ocular movement and the beta power peak following the saccades showed substantial intra- and interindividual differences (Fig. 3A). Therefore we averaged the maximum power % values of the intersaccade and post-saccade period according to the method published previously (Tamás et al., 2003). The average PMBS values at FC1 and FC2 positions of 15 epochs of each subject were used for further analysis.

View Large Image Download to PowerPoint

Fig. 3. PMBS curves. (A) Individual power % curves of 15 saccades directed to the right plotted against time at FC2 in Subject Nr3. The arrows show the highest power % values after termination of eye movement. The thick black line represents the average of the 15 curves. (B) Average power % curves of all subjects at FC1 and FC2 electrodes in the saccade task to the left. The arrows show the highest power % values after termination of eye movement (AVR, average; Nr, subject Nr).

Statistical analysis was performed using repeated measures ANOVA test considering three factors (DIRECTION, ELECTRODE and PERIOD) followed by Newman–Keuls post hoc test (Statistica 6.0, StatSoft Inc., USA). For PERIOD the statistics were given for the mean and the maximum power % values of the reference period, and the maximum power % data of the intersaccadic and postsaccadic interval. If statistically significant results were found correction of p-value was performed according to Bonferroni.

3.1. Latency of saccades 

The latency of the right-directed first saccade was 319±49ms and the right-directed second saccade was 275±21ms (t=2.37; p=0.06). The latency of the first left saccade was 331±59ms and the second left saccade was 293±37ms (t=1.13; p=0.3). Differences between the first right- and first left-directed saccades (t=−1.08; p=0.33) as well as between the second right- and second left-directed saccades (t=−1.45; p=0.2) were not significant.

3.2. MRBF 

MRBF was different in each subject and it also showed intraindividual variability at different electrode positions ranging between 16 and 26Hz. Table 1 shows MRBF during saccades directed to right and left at frontocentral electrode positions. In a given subject the direction of the task did not have an effect on MRBF measured at a given electrode (F=0.48; p=0.77).

Table 1. The most reactive beta frequency (MRBF) of all subjects (S) in right- and left-directed saccade tasks at different electrode positions.

	Electrode	MRBF (Hz) in right saccade task							MRBF (Hz) in left saccade task
		S Nr1	S Nr2	S Nr3	S Nr4	S Nr5	S Nr6	Means±SD	S Nr1	S Nr2	S Nr3	S Nr4	S Nr5	S Nr6	Means±SD
	FC3	23	19	20	20	20	24	21±2	20	22	22	17	21	20	20.33±1.9
	FC1	21	19	20	23	19	18	20±1.8	21	22	21	20	21	18	20.5±1.4
	FCz	22	19	20	19	20	18	19.67±1.4	23	21	21	22	18	18	20.5±2.1
	FC2	20	19	20	26	16	18	19.83±3.4	23	22	19	23	20	18	20.83±2.1
	FC4	21	19	20	18	22	18	19.67±1.6	23	22	22	19	25	17	21.33±2.9
	C1	21	21	22	23	20	18	20.83±1.7	24	22	21	15	20	18	20±3.2
	Cz	21	19	22	20	21	17	20±1.8	20	22	21	20	17	18	19.67±1.9
	C2	23	18	20	23	19	18	20.17±2.3	25	15	22	24	17	18	20.17±4.1
	CPz	20	21	22	23	22	17	20.83±2.1	23	22	20	23	21	24	22.17±1.5

3.3. PMBS 

The most pronounced post-saccadic beta power increase was observed in all subjects at FC1 and FC2 electrode positions (Fig. 4).

View Large Image Download to PowerPoint

Fig. 4. Average PMBS power % of 15 right-directed tests of Subject Nr6 at the most reactive beta frequency plotted against time at different electrode positions. The most pronounced PMBS can be identified at FC electrode positions.

Group average PMBS latency of 15 epochs in right saccade tasks at FC1 was 1159±178ms (ranged between 354 and 1854ms) and at FC2 it was 1050±187ms (ranged between 333 and 1854ms). In the left task the mean PMBS latency was 1040±186ms (ranged between 270 and 1854ms) at FC1 and 1176±148ms (ranged between 416 and 1875ms) at FC2 electrode position. There was no significant difference between PMBS latencies according to direction of the task (F=0.107; p=0.955) or to electrode position (F=0.447; p=0.722) or to either of them (F=1.029; p=0.403).

The maximum beta power % value in the intersaccadic interval was not significantly higher when compared to either the mean power % or the maximum power % of the reference period (p=0.38 and 0.99 at FC1; p=0.34 and 0.98 at FC2 in right task as well as p=0.13 and 0.78 at FC1 and p=0.28 and 0.85 at FC2 in left task) (Fig. 2, Fig. 3, Fig. 4, Fig. 5). However we found significant beta increase after the offset of the second saccade compared to both the mean and maximum power % of the reference period and also to the maximum power % of the intersaccadic interval (PERIOD: F(3,15)=208.14; p=0.000001). There was no significant difference between PMBS amplitude s according to the DIRECTION of the task (F(1,5)=0.176; p=0.9), to the ELECTRODE position (F(1,5)=0.651; p=0.46), or to both of them (DIRECTION×ELECTRODE: F(1,5)=1.056;p=0.35) (Fig. 5, p-values calculated using post hoc statistical analysis).

View Large Image Download to PowerPoint

Fig. 5. Average power % of the reference (ref) period, average intersaccadic (intersacc) peak power % and average peak PMBS values (15–15 tasks of 6 subjects) at FC1 and FC2 electrode positions in tasks directed to the right and left. The post-saccadic beta power increase (PMBS) was significant compared to the reference period in each direction at both the FC1 and FC2 electrodes, while the power increase in the intersaccadic interval was not. There was no significant difference in PMBS according to the direction of the ocular movement and the localisation of the electrodes. Statistical results are given using the mean power % value of the reference interval and the maximum power % values of the intersaccadic and post-saccadic periods (∗comparison to mean power % value of reference period at the same electrode position in the same direction; †comparison to intersaccadic power % at the same position in the same direction; $comparison to PMBS values at the opposite position in the same direction; #comparison to PMBS value at the same position in the opposite direction) (box: means±SE; whiskers: SD).