- Email: [email protected]
Presaccadic spike potential to horizontal eye movements
Electroencephalography and clinical Neurophysiology, 1988, 7 0 : 5 5 9 - 5 6 2
559
Elsevier Scientific Publishers Ireland, Ltd. EEG 02275
Short communication
Presaccadic spike potential to horizontal eye movements C.
Boylan and H.R. Doig
Department of Vision Sciences, University of Aston, Aston Triangle, Birmingham B4 7ET (U.K.) (Accepted for publication: 22 August 1988)
Summary Presaccadic spike potentials prior to 20 ° horizontal eye movements were recorded from 4 electrode sites around the eye with Pz as reference. The technique of back-averaging from the onset of the saccade was used and separate wave forms were obtained for abducting and adducting movements. The spike onset began between 12.0 and 4.4 msec before the beginning of the saccade with a significantly earlier onset with adduction (right eye P < 0.001, left eye P < 0.025). The peak occurred between the onset of eye movement and 7.8 msec after the beginning of the movement; the peak also occurred significantly earlier with adduction (right eye P < 0.01, left eye P < 0.01). The amplitude of the onset to peak measured from 32.2 to 47.0/LV with greater amplitudes on adduction compared to abduction although the differences were not statistically significant. Several of the traces also showed a later smaller component, after the main onset-to-peak complex, that has not been previously reported. Key words: Presaccadic spike; Horizontal eye movements; Amplitude; Latency; Abduction; Adduction
A large amplitude presaccadic spike potential (SP), occurring 10-40 msec prior to the onset of horizontal saccades, has been recorded from electrodes around the eye with m a x i m u m amplitude when using Pz as the reference site. The SP is believed to be of extraocular muscle origin representing the summated activity of motor units within these muscles (Thickbroom and Mastaglia 1985, 1986). Although it does not seem possible to determine the relative contribution of any individual muscle to the SP, the technique may provide an alternative to electromyography in clinical investigations and it m a y also give a further method of examining abnormal saccadic eye movements in addition to the measuring of the usual parameters of saccades. The amplitude of the SP is reported to be greater with abducting than adducting saccades, although no statistical analysis is available (Thickbroom and Mastaglia 1985, 1986). There have been no reports of variations in the latency of the SP with abduction and adduction. If the recording of the SP is to have clinical use, its normal parameters must first be established and the purpose of the present study was to further investigate the SP latencies and amplitudes in a normal population.
Correspondence to: Dr. C. B o y l a n , Department of Vision Sciences, University of Aston, Aston Triangle, Birmingham B4 7ET (U.K.).
Method
Five males and 5 females, ranging in age from 18 to 29 years, were examined (mean age 24.6 years males and 23.4 years females). They had no known ophthalmological or neurological defects. Recordings were made from both eyes and from 4 electrode sites (inner canthus, outer canthus, above the eye and below the eye). Pz was used as the reference in accord with previous investigations (Thickbroom and Mastaglia 1985, 1986). Visually triggered saccades were initiated by 2 alternately flashing red LEDs horizontally separated to subtend 20 o at the subject. The responses to 40 saccades (20 abducting and 20 adducting) were recorded with eye position monitored by electro-oculography (EOG). The individual saccadic traces and the SPs were recorded on separate channels by a Nicolet Pathfinder lI with a time sweep of 500 msec and a bandpass filter with a low cut-off of 0.50 Hz and a high cut-off of 100 Hz. Each saccade and SP were stored on hard disc and, following recordings, the individual saccades were recalled and the beginning of the eye movement identified manually using the averager's internal cursor. Visual examination of the eye position data allowed accurate identification of the beginning of the E O G deflection and hence the beginning of the saccade. If this point could not be identified easily, the data were not included in the analysis. Following this, the computer back-averaged the SP from the beginning of the eye movements. Thus, the latency of the spikes relative to the onset of the saccades could be determined. For display purposes the beginning of the saccades was aligned at a pre-determined and fixed latency. The results
0013-4649/88/$03.50 © 1988 Elsevier Scientific Publishers Ireland, Ltd.
560
C. BOYLAN, H.R. D O I G
Right eye Electrode site
Inner
Left eye
21 12 \
Outer
,
ove
50 rnsec
2
,<
X
X
Fig. 1. The group average spike potential wave forms for the right and left eyes during abduction and adduction at each electrode site. 1 = abduction; 2 = adduction; X = onset of saccade and ' b ' marks the second component seen following the main spike on many traces.
X
Fig. 2. The SP recorded from the electrode below the left eye during adduction in all 10 subjects. In many of the traces a second, smaller component can be seen occurring after the main spike. X = onset of sa¢cade.
50 msec
P R E S A C C A D I C SPIKES A N D H O R I Z O N T A L EYE M O V E M E N T S for abducting and adducting saccades were averaged separately. At each electrode site a pair of SPs were recorded (one abducting and one adducting), and the following parameters were measured for each of the pair using the averager's internal cursor: the latencies of the onset and peak of the SP relative to the beginning of the saccade; a negative value indicating a latency prior to the onset of the saccade and a positive value a latency after the onset of the saccade and the amplitude from the onset to the peak. For the group of subjects the means and standard deviation (S.D.) of each measurement were calculated and group average wave forms at each electrode site computed by the averager.
561
Two-way analysis of variance was performed on the latency and amplitude raw data using the eye movement direction (abduction, adduction) as one treatment and the electrode site (inner canthus, outer canthus, above the eye and below the eye) as the second treatment. The spike onset occurred statistically significantly earlier on adduction than abduction; right eye F1, 9 = 44.32 ( P < 0.001), left eye FI, 9 = 7.39 ( P < 0.025). The SP also peaked statistically significantly earlier on adduction than abduction; right eye F1,9=14.84 ( P < 0 . 0 1 ) , left eye Fa, 9 =11.59 ( P < 0.01). The amplitude differences failed to reach statistical significance; right eye k~.27 = 2.17 (NS), left eye F3, 27 = 1.67 (NS). A further feature of the SP can be seen labelled by the letter 'b" in the traces in Fig. 1. A second, smaller component occurring after the first, major spike is clearly visible in several of the wave forms. This wave is not simply a feature of the group average as it can be seen in many of the individual wave forms. Fig. 2 shows the left eye, adducting wave forms recorded from the electrode below the eye in all 10 subjects and the second, smaller c o m p o n e n t can be seen in the traces.
Results
A clear SP was recorded from each subject at all four electrode sites during abducting and adducting eye movements. The potentials had a similar wave form at each electrode site as shown in the group average responses in Fig. 1 and all could be characterised by a negative spike. The onset of the SP began between 12.0 and 4.4 msec prior to the start of the eye movement. The peak occurred between saccadic onset and 7.8 msec after the onset with an onset-peak amplitude similar at all electrode sites of between 32.0 and 47.7/tV. Table I shows the means and standard deviations for these 3 parameters during abduction and adduction movements at each electrode site. At all electrode sites the onset occurred earlier prior to adducting than abducting saccades and the spike also peaked earlier after adducting saccades; in one case the peak occurred at the time of the beginning of the saccadic movement (recording from inner canthus with right eye adducting). The amplitudes were larger for adducting than abducting movements at all electrode sites.
Discussion
We used the technique of back-averaging from the start of the saccades rather than the re-averaging of data about a peak in the SP as used by Thickbroom and Mastaglia (1986). As these authors did, in some cases, line up the spike potentials rather than the saccadic onsets, they could not possibly determine the latency of the spike onset accurately. With examination of the individual saccadic traces there is no difficulty in determining saccadic onset precisely and clear SPs are averaged. Wave forms similar to those described by these authors were obtained although analysis of the wave form parameters reveals features not previously reported.
TABLE I The mean and standard deviation (S.D.) spike potential latencies (Lat) and amplitudes (Amp) for 10 subjects at 4 electrode sites. Electrode site Inner
Outer
Lat (msec)
A m p (/~V)
Onset
Onset-peak Onset
Peak
Above
Lat (msec) Peak
A m p (/W)
Below
Lat (msec)
Onset-peak Onset
Peak
A m p (ttV)
Lat (msec)
Onset-peak Onset
Peak
A m p (~tV) Onset-peak
Right eye Mean abducting S.D.
- 7 . 9 +6.2 41.3 2.6 3.3 15.5
- 7 . 6 +4.9 34.5 2.0 4.5 17.2
- 5 . 0 +7.8 35.0 5.0 4.6 13.2
-7.2 2.1
Left eye Mean abducting S.D.
- 8 . 0 +6.2 39.8 2.9 3.5 10.3
- 7 . 6 +6.8 32.2 2.0 3.3 12.8
-4.4 3.1
+ 7 . 4 34.7 2.8 8.4
- 6 . 9 +6.0 38.9 2.9 2.9 9.0
Right eye Mean adducting S.D.
- 9.9 1.5
Left eye Mean adducting S.D.
+6.5 38.5 3.8 16.0
0.0 44.9 1.7 19.8
- 1 2 . 0 +3.0 41.4 3.2 4.3 12.2
- 1 0 . 1 +1.3 37.3 3.9 3.8 12.0
- 1 0 . 0 +1.0 45.7 3.2 4.6 16.4
- 1 0 . 0 +1.6 46.4 2.3 4.4 17.1
- 9 . 7 +2.3 45.3 3.2 4.5 15.6
- 7 . 9 +1.7 40.1 5.0 6.6 13.5
10.0 +1.2 47.0 5.2 4.8 16.5
562 Analysis of our data reveals differences between the SPs recorded with abducting and adducting saccades. Unlike Thickbroom and Mastaglia (1986), we found greater amplitudes on adduction than on abduction although the differences did not reach statistical significance. This shows that care must be taken when making assumptions based on raw data, particularly that regarding amplitude. This feature of apparent amplitude differences that are not supported statistically is c o m m o n to other electrophysiological tests such as the recording of visually evoked cortical potentials (Boylan et al. 1984) and is due to the inherent variability of amplitude data. Therefore, any SP amplitude differences found between abduction and adduction m u s t be regarded with caution until a consistent relationship can be demonstrated. Previously unreported latency differences between the position of the SP relative to the saccade were also found for the two different types of movement; this m a y reflect the contribution to the spike by the different extraocular muscles responsible for moving the eyes in the two directions. The finding of a second smaller component suggests that the SP is not simply a single spike but has a more complex wave form with two or more components. This is in agreement with Thickbroom and
C. BOYLAN, H.R. D O I G Mastaglia (1987) who described secondary, intrasaccadic spikes. Although the SP is an intriguing finding, further work must be able to relate the findings to the actions of the extraocular muscles before its recording will gain any cfinical importance.
References
Boylan, C., Clement, R.A. and Harding, G.F.A. Lateralization of the flash visual evoked cortical potential in h u m a n albinos. Invest. Ophthalmol. Vis. Sci., 1984, 25: 1448-1450. Tbickbroom, G.W. and Mastaglia, F.L. Pre-saccadic 'spike' potential: investigation of topography and source. Brain Res., 1985, 339: 271-280. Thickbroom, G.W. and Mastaglia, F.L. Presaccadic spike potential. Relation to eye movement direction. Electroenceph. clin. Neurophysiol., 1986, 64: 211-214. Thickbroom, G.W. and Mastaglia, F.L. Presaccadic spike potential: a computer model based upon motor unit recruitment patterns in the extraocular muscles. Brain Res., 1987, 422: 377-380.
559
Elsevier Scientific Publishers Ireland, Ltd. EEG 02275
Short communication
Presaccadic spike potential to horizontal eye movements C.
Boylan and H.R. Doig
Department of Vision Sciences, University of Aston, Aston Triangle, Birmingham B4 7ET (U.K.) (Accepted for publication: 22 August 1988)
Summary Presaccadic spike potentials prior to 20 ° horizontal eye movements were recorded from 4 electrode sites around the eye with Pz as reference. The technique of back-averaging from the onset of the saccade was used and separate wave forms were obtained for abducting and adducting movements. The spike onset began between 12.0 and 4.4 msec before the beginning of the saccade with a significantly earlier onset with adduction (right eye P < 0.001, left eye P < 0.025). The peak occurred between the onset of eye movement and 7.8 msec after the beginning of the movement; the peak also occurred significantly earlier with adduction (right eye P < 0.01, left eye P < 0.01). The amplitude of the onset to peak measured from 32.2 to 47.0/LV with greater amplitudes on adduction compared to abduction although the differences were not statistically significant. Several of the traces also showed a later smaller component, after the main onset-to-peak complex, that has not been previously reported. Key words: Presaccadic spike; Horizontal eye movements; Amplitude; Latency; Abduction; Adduction
A large amplitude presaccadic spike potential (SP), occurring 10-40 msec prior to the onset of horizontal saccades, has been recorded from electrodes around the eye with m a x i m u m amplitude when using Pz as the reference site. The SP is believed to be of extraocular muscle origin representing the summated activity of motor units within these muscles (Thickbroom and Mastaglia 1985, 1986). Although it does not seem possible to determine the relative contribution of any individual muscle to the SP, the technique may provide an alternative to electromyography in clinical investigations and it m a y also give a further method of examining abnormal saccadic eye movements in addition to the measuring of the usual parameters of saccades. The amplitude of the SP is reported to be greater with abducting than adducting saccades, although no statistical analysis is available (Thickbroom and Mastaglia 1985, 1986). There have been no reports of variations in the latency of the SP with abduction and adduction. If the recording of the SP is to have clinical use, its normal parameters must first be established and the purpose of the present study was to further investigate the SP latencies and amplitudes in a normal population.
Correspondence to: Dr. C. B o y l a n , Department of Vision Sciences, University of Aston, Aston Triangle, Birmingham B4 7ET (U.K.).
Method
Five males and 5 females, ranging in age from 18 to 29 years, were examined (mean age 24.6 years males and 23.4 years females). They had no known ophthalmological or neurological defects. Recordings were made from both eyes and from 4 electrode sites (inner canthus, outer canthus, above the eye and below the eye). Pz was used as the reference in accord with previous investigations (Thickbroom and Mastaglia 1985, 1986). Visually triggered saccades were initiated by 2 alternately flashing red LEDs horizontally separated to subtend 20 o at the subject. The responses to 40 saccades (20 abducting and 20 adducting) were recorded with eye position monitored by electro-oculography (EOG). The individual saccadic traces and the SPs were recorded on separate channels by a Nicolet Pathfinder lI with a time sweep of 500 msec and a bandpass filter with a low cut-off of 0.50 Hz and a high cut-off of 100 Hz. Each saccade and SP were stored on hard disc and, following recordings, the individual saccades were recalled and the beginning of the eye movement identified manually using the averager's internal cursor. Visual examination of the eye position data allowed accurate identification of the beginning of the E O G deflection and hence the beginning of the saccade. If this point could not be identified easily, the data were not included in the analysis. Following this, the computer back-averaged the SP from the beginning of the eye movements. Thus, the latency of the spikes relative to the onset of the saccades could be determined. For display purposes the beginning of the saccades was aligned at a pre-determined and fixed latency. The results
0013-4649/88/$03.50 © 1988 Elsevier Scientific Publishers Ireland, Ltd.
560
C. BOYLAN, H.R. D O I G
Right eye Electrode site
Inner
Left eye
21 12 \
Outer
,
ove
50 rnsec
2
,<
X
X
Fig. 1. The group average spike potential wave forms for the right and left eyes during abduction and adduction at each electrode site. 1 = abduction; 2 = adduction; X = onset of saccade and ' b ' marks the second component seen following the main spike on many traces.
X
Fig. 2. The SP recorded from the electrode below the left eye during adduction in all 10 subjects. In many of the traces a second, smaller component can be seen occurring after the main spike. X = onset of sa¢cade.
50 msec
P R E S A C C A D I C SPIKES A N D H O R I Z O N T A L EYE M O V E M E N T S for abducting and adducting saccades were averaged separately. At each electrode site a pair of SPs were recorded (one abducting and one adducting), and the following parameters were measured for each of the pair using the averager's internal cursor: the latencies of the onset and peak of the SP relative to the beginning of the saccade; a negative value indicating a latency prior to the onset of the saccade and a positive value a latency after the onset of the saccade and the amplitude from the onset to the peak. For the group of subjects the means and standard deviation (S.D.) of each measurement were calculated and group average wave forms at each electrode site computed by the averager.
561
Two-way analysis of variance was performed on the latency and amplitude raw data using the eye movement direction (abduction, adduction) as one treatment and the electrode site (inner canthus, outer canthus, above the eye and below the eye) as the second treatment. The spike onset occurred statistically significantly earlier on adduction than abduction; right eye F1, 9 = 44.32 ( P < 0.001), left eye FI, 9 = 7.39 ( P < 0.025). The SP also peaked statistically significantly earlier on adduction than abduction; right eye F1,9=14.84 ( P < 0 . 0 1 ) , left eye Fa, 9 =11.59 ( P < 0.01). The amplitude differences failed to reach statistical significance; right eye k~.27 = 2.17 (NS), left eye F3, 27 = 1.67 (NS). A further feature of the SP can be seen labelled by the letter 'b" in the traces in Fig. 1. A second, smaller component occurring after the first, major spike is clearly visible in several of the wave forms. This wave is not simply a feature of the group average as it can be seen in many of the individual wave forms. Fig. 2 shows the left eye, adducting wave forms recorded from the electrode below the eye in all 10 subjects and the second, smaller c o m p o n e n t can be seen in the traces.
Results
A clear SP was recorded from each subject at all four electrode sites during abducting and adducting eye movements. The potentials had a similar wave form at each electrode site as shown in the group average responses in Fig. 1 and all could be characterised by a negative spike. The onset of the SP began between 12.0 and 4.4 msec prior to the start of the eye movement. The peak occurred between saccadic onset and 7.8 msec after the onset with an onset-peak amplitude similar at all electrode sites of between 32.0 and 47.7/tV. Table I shows the means and standard deviations for these 3 parameters during abduction and adduction movements at each electrode site. At all electrode sites the onset occurred earlier prior to adducting than abducting saccades and the spike also peaked earlier after adducting saccades; in one case the peak occurred at the time of the beginning of the saccadic movement (recording from inner canthus with right eye adducting). The amplitudes were larger for adducting than abducting movements at all electrode sites.
Discussion
We used the technique of back-averaging from the start of the saccades rather than the re-averaging of data about a peak in the SP as used by Thickbroom and Mastaglia (1986). As these authors did, in some cases, line up the spike potentials rather than the saccadic onsets, they could not possibly determine the latency of the spike onset accurately. With examination of the individual saccadic traces there is no difficulty in determining saccadic onset precisely and clear SPs are averaged. Wave forms similar to those described by these authors were obtained although analysis of the wave form parameters reveals features not previously reported.
TABLE I The mean and standard deviation (S.D.) spike potential latencies (Lat) and amplitudes (Amp) for 10 subjects at 4 electrode sites. Electrode site Inner
Outer
Lat (msec)
A m p (/~V)
Onset
Onset-peak Onset
Peak
Above
Lat (msec) Peak
A m p (/W)
Below
Lat (msec)
Onset-peak Onset
Peak
A m p (ttV)
Lat (msec)
Onset-peak Onset
Peak
A m p (~tV) Onset-peak
Right eye Mean abducting S.D.
- 7 . 9 +6.2 41.3 2.6 3.3 15.5
- 7 . 6 +4.9 34.5 2.0 4.5 17.2
- 5 . 0 +7.8 35.0 5.0 4.6 13.2
-7.2 2.1
Left eye Mean abducting S.D.
- 8 . 0 +6.2 39.8 2.9 3.5 10.3
- 7 . 6 +6.8 32.2 2.0 3.3 12.8
-4.4 3.1
+ 7 . 4 34.7 2.8 8.4
- 6 . 9 +6.0 38.9 2.9 2.9 9.0
Right eye Mean adducting S.D.
- 9.9 1.5
Left eye Mean adducting S.D.
+6.5 38.5 3.8 16.0
0.0 44.9 1.7 19.8
- 1 2 . 0 +3.0 41.4 3.2 4.3 12.2
- 1 0 . 1 +1.3 37.3 3.9 3.8 12.0
- 1 0 . 0 +1.0 45.7 3.2 4.6 16.4
- 1 0 . 0 +1.6 46.4 2.3 4.4 17.1
- 9 . 7 +2.3 45.3 3.2 4.5 15.6
- 7 . 9 +1.7 40.1 5.0 6.6 13.5
10.0 +1.2 47.0 5.2 4.8 16.5
562 Analysis of our data reveals differences between the SPs recorded with abducting and adducting saccades. Unlike Thickbroom and Mastaglia (1986), we found greater amplitudes on adduction than on abduction although the differences did not reach statistical significance. This shows that care must be taken when making assumptions based on raw data, particularly that regarding amplitude. This feature of apparent amplitude differences that are not supported statistically is c o m m o n to other electrophysiological tests such as the recording of visually evoked cortical potentials (Boylan et al. 1984) and is due to the inherent variability of amplitude data. Therefore, any SP amplitude differences found between abduction and adduction m u s t be regarded with caution until a consistent relationship can be demonstrated. Previously unreported latency differences between the position of the SP relative to the saccade were also found for the two different types of movement; this m a y reflect the contribution to the spike by the different extraocular muscles responsible for moving the eyes in the two directions. The finding of a second smaller component suggests that the SP is not simply a single spike but has a more complex wave form with two or more components. This is in agreement with Thickbroom and
C. BOYLAN, H.R. D O I G Mastaglia (1987) who described secondary, intrasaccadic spikes. Although the SP is an intriguing finding, further work must be able to relate the findings to the actions of the extraocular muscles before its recording will gain any cfinical importance.
References
Boylan, C., Clement, R.A. and Harding, G.F.A. Lateralization of the flash visual evoked cortical potential in h u m a n albinos. Invest. Ophthalmol. Vis. Sci., 1984, 25: 1448-1450. Tbickbroom, G.W. and Mastaglia, F.L. Pre-saccadic 'spike' potential: investigation of topography and source. Brain Res., 1985, 339: 271-280. Thickbroom, G.W. and Mastaglia, F.L. Presaccadic spike potential. Relation to eye movement direction. Electroenceph. clin. Neurophysiol., 1986, 64: 211-214. Thickbroom, G.W. and Mastaglia, F.L. Presaccadic spike potential: a computer model based upon motor unit recruitment patterns in the extraocular muscles. Brain Res., 1987, 422: 377-380.