Visual evoked electrical and magnetic response to half-field stimulation using pattern reversal stimulation G. F. A. Harding, R. A. Armstrong and B. Janday
Department of Vision Sciences, Aston University, Birmingham, UK (Received 30 Octoher 1991) The visual evoked magnetic response lo half-field stimulation using pattern reversal was studied using a d.c. SQUID coupled to a second ordergrudiometer. The main component of the magnetic response consisted of a positive wave at around 100 ms (PIOOM). At the time this component was present the response to hiilf-field stimulation consisted of an outgoing magnetic field contralateral and extending to the midline. When the left half field was stimulated the outgoing field was over the posterior right visual cortex and when the right half field was stimulated it was over the left anterior visual cortex. These findings would correctly identify a source located in the contralateral visual cortex. The orientation of the dipoles was not that previously assumed to explain the paradoxical lateralization of the visual evoked potential. The results are discussed in terms of both electrical and magnetic models of the calcarine fissure.
Correspondence to: Professor G. F. A. Harding, Department of Vision Sciences, Aston University, Aston Triangle, Birmingham B4 7ET, LfK.
locations on either side of the head that the two contralateral negative components are maximal. Both of the components appear to originate from the same or a closely related source as both components are still present in patients who have received a hemisphereetomy. It was suggested that although the major positivity was therefore generated by the visual cortex contralateral to the field stimulated, the response appears maximal at the midline and ipsilateral electrodes because the orientation of these electrodes in relation to a mid-frontai reference positions them maximally for a radial sourec, whereas the electrodes over the source are seeing a tangential dipole. Since electrical potentials maximize radial signals and not tangential ones, these latter electrodes would not receive a clear signal. It appears that this situation is only true with large fields of stimulation, since the responses are represented for a large distance along the calcarine fissure. If the size of the visual field is reduced below a 0-4 degree radius target, however, the lateralization no longer holds, and the maximal signal may appear contralateral to the half Held stimulated". In other studies it has been shown that when the size of the stimulated visual half field is reduced below 5 degrees the response becomes clearly contralateral in normal subjects^-\ When the size of the checks is also reduced this may increase the contralateral spread of the major positive component^ although this is still the subject of contention {Figure I). Using dipole modelling. Lehmann et al. showed that with large targets the dipole is clearly situated in the contralateral hemisphere, with the positive end pointing towards the electrodes on the ipsilateral scalp^. When the target size was reduced, the orientation of the dipole changed being less ipsilateral and being closer to the midline. With intracerebral recording they identified
© 1992 Butterworth-Heinemann for British College of Optometrists 0275-5408/92/020171-04
Ophthal. Physiol. Opt., 1992, Vol. 12, April
Topographic distribution of the visual evoked electrical potential (VEP) to pattern reversal stimulation has been the subject of many studies. Halliday's group, restricting themselves to a simple stimulus consisting of 50 min black and white checks in a field of 0 16 degrees radius, carried out studies of both full-field and half-field responses. On full-field stimulation, they found that the N75-P100-N145 complex was always maximal at a position ^ 5 cm above the occiput, but that the amplitude gradient rapidly reduced in more temporal locations'. With half-field stimulation, however, the PIOO component of the VEP appeared to be maximal both at the midline and ipsilateral to the half field stimulated. The response showed a more lateral distribution than under full-field stimulation and was clearly present at the most lateral occipital electrodes. However, on the contralateral side of the head the response was of lower amplitude and a separate response of opposite polarity consisting of a positive at 75, a negative at 105 and a positive at 135 ms was obtained. This component was clearly maximal at the most lateral occipital electrode position and reduced in amplitude around the midline^. Blumhardt and Halliday showed that the full-field response is in fact the algebraic summation of the responses to each half iield\ The PIOO complex is of highest amplitude at the mid-occipital electrode simply because contralateral negativity is of lowest amplitude at this point whichever half field is stimulated. When both the half fields are stimulated the positivity is clearly maximal at the occiput since it is not cancelled by both contralateral negativities, whereas at the most occipital electrodes, the fall-off of Ihe PIOO component is marked because it is in these
171
Visual evoked electrical anil magnetic response: G. F. A. Harding et al. ipsilateral positivity and contralateral negativity for large field stimulation suggesting an origin in extra striate visual cortex. Using source derivation techniques Flanagan and Harding showed that with large fields the source was always maximal at the midline occipital electrode whichever half field was stimulated, but found that the sink was always contralateral to the half field being stimulated**. These difficulties inherent in electrophysiology provide the rationale for the more complex techniques of magnetoencephalography. There have been to date relatively few studies on the visual evoked magnetic response (VEMR) to pattern reversal stimulation". As the VEMR reflects current flows in the brain rather than potential differences across the scalp, the response to pattern reversal may well complement VEP studies of the visual cortex, particularly since magnetic responses probably reflect activity in fissures rather than gyri'"-^\ We have therefore carried out a study of the topographic distribution of the PIOOM component of the V E M R using three different check sizes in both full- and half-field stimulation.
Materials and methods Four normal subjects aged 23 40 years participated in the study. They had no neurological or ophthaimological problems and had corrected visual acuities of 6/6 or better. The University Ethical Committee approved the study. The magnetic responses were recorded using a BTi second-order d.c. single channel gradiometer (Model 601. Biomagnetic Technologies, CA. USA). The system had a white noise level of 16.6 fT Hz"^ at 5 Hz. All traces were recorded on a Bio-logic Traveler, Bio-logic, Oxon, UK and 50 responses were averaged at each location. To carry out topographic mapping the magnetometer was moved into a series of positions on a 20-point grid consisting of 4 rows of 5 locations per row, located over the visual cortex with the inion as the mid-position on the bottom row of positions. The distance between the points on the grid was defined as 10% of the half circumference of the head. The stimulus consisted ofthree different check sizes (22, 34 and 70 min) back projected on to a screen to produce a 14 degree circular field. The checkerboard reversed twice per second
and the mean luminance of the screen was 1050 cd m"^ while the contrast was 76%. We have previously shown'- that the most consistent VEMR component to pattern reversal stimulation is a major positive deflection (outgoing magnetic field) observed between 90 and 120 ms. This component we have entitled the PIOOM and the amplitude of this signal peak to baseline at each location on the grid was used as the mapping input. Topographic mapping was performed on a Nicolct Pathfinder II (Warwick. UK) using a rectangular grid and linear interpolation. This data was also used as the basis for source localization using a dipole in a sphere model'-*.
Results Half-field stimulation produced the strongest magnetic responses over the contralateral hemisphere. Pattern reversal response to right half-field stimulation using either 34 or 70 min checks showed an outgoing magnetie field over the anterior left visual cortex and an ingoing field over the posterior left visual cortex. When the left-half field was stimulated the response consisted of an outgoing field over the posterior right visual cortex and an ingoing field over the right anterior visual cortex {Figure J). The maps were often complex and quite frequently showed an ingoing field around the midline or even extending slightly into the ipsilateral hemisphere in some individuals (F/t/nrc 3). Applying Fleming's right hand rule this would suggest that the source is in the contralaterai hemisphere to the field stimulated with current flowing towards the midline. The results of all four subjects were submitted to source localization techniques and these in general confirmed the source assumed from the topographic maps [Figure 4). In each case the dipole was located to the contralateral side of the midline to the field stimulated but surprisingly the orientation derived from the magnetomctry data that was obtained" did not coincide with the orientation of presumed dipoles derived from the VEP but had more ipsilateral and anterior orientation. Discussion The results of half-field stimulation confirm that the source of both the visual evoked magnetic response, and
Figure I Brain maps of the main positive PIOO componcnl of the vi.sual evoked electrical polentiai elicited by stimulation of Ihc lefl half field. The check sizes of ihe paltern reversing stimulus are shown above ihc head maps and consisl of 56. 27 and 11 min checks. The actual lime of occurrence of the PIOO is shown below each map in milliseconds. The colour bar at the bottom of the figure shows the amplitude in microvolts. It can be seen that the PIOO component demonstrates the well known paradoxical lateralization. With whatever check size the Icfl half field is stimulated, the maximal respon.se appears lo be over the left visual cortex Figure 2 The lopographic maps obtained from one of the subjects lo Icfl half lield ( LHF), full field (FF)and right half field IRHF) stimulation. The red end of the spectrum represents positive, thai is outgoing magnetic fields, and the green end of ihe spectrum represents negalivc. thai is ingoing fields. These responses were produced by paltern reversal stimulation using 34 min checks. The circles indicate ihe positions recorded using Ihc magnetometer with ihe occiput being represented as ihc central posilion of the second row from the bottom. It can be seen thai on Icfl half ticld stimulation the response consists of an outgoing field in the right posterior quadrant and in ingoing field in the right upper quadrant, suggesting a dipole anterior lo lh
Department of Vision Sciences, Aston University, Birmingham, UK (Received 30 Octoher 1991) The visual evoked magnetic response lo half-field stimulation using pattern reversal was studied using a d.c. SQUID coupled to a second ordergrudiometer. The main component of the magnetic response consisted of a positive wave at around 100 ms (PIOOM). At the time this component was present the response to hiilf-field stimulation consisted of an outgoing magnetic field contralateral and extending to the midline. When the left half field was stimulated the outgoing field was over the posterior right visual cortex and when the right half field was stimulated it was over the left anterior visual cortex. These findings would correctly identify a source located in the contralateral visual cortex. The orientation of the dipoles was not that previously assumed to explain the paradoxical lateralization of the visual evoked potential. The results are discussed in terms of both electrical and magnetic models of the calcarine fissure.
Correspondence to: Professor G. F. A. Harding, Department of Vision Sciences, Aston University, Aston Triangle, Birmingham B4 7ET, LfK.
locations on either side of the head that the two contralateral negative components are maximal. Both of the components appear to originate from the same or a closely related source as both components are still present in patients who have received a hemisphereetomy. It was suggested that although the major positivity was therefore generated by the visual cortex contralateral to the field stimulated, the response appears maximal at the midline and ipsilateral electrodes because the orientation of these electrodes in relation to a mid-frontai reference positions them maximally for a radial sourec, whereas the electrodes over the source are seeing a tangential dipole. Since electrical potentials maximize radial signals and not tangential ones, these latter electrodes would not receive a clear signal. It appears that this situation is only true with large fields of stimulation, since the responses are represented for a large distance along the calcarine fissure. If the size of the visual field is reduced below a 0-4 degree radius target, however, the lateralization no longer holds, and the maximal signal may appear contralateral to the half Held stimulated". In other studies it has been shown that when the size of the stimulated visual half field is reduced below 5 degrees the response becomes clearly contralateral in normal subjects^-\ When the size of the checks is also reduced this may increase the contralateral spread of the major positive component^ although this is still the subject of contention {Figure I). Using dipole modelling. Lehmann et al. showed that with large targets the dipole is clearly situated in the contralateral hemisphere, with the positive end pointing towards the electrodes on the ipsilateral scalp^. When the target size was reduced, the orientation of the dipole changed being less ipsilateral and being closer to the midline. With intracerebral recording they identified
© 1992 Butterworth-Heinemann for British College of Optometrists 0275-5408/92/020171-04
Ophthal. Physiol. Opt., 1992, Vol. 12, April
Topographic distribution of the visual evoked electrical potential (VEP) to pattern reversal stimulation has been the subject of many studies. Halliday's group, restricting themselves to a simple stimulus consisting of 50 min black and white checks in a field of 0 16 degrees radius, carried out studies of both full-field and half-field responses. On full-field stimulation, they found that the N75-P100-N145 complex was always maximal at a position ^ 5 cm above the occiput, but that the amplitude gradient rapidly reduced in more temporal locations'. With half-field stimulation, however, the PIOO component of the VEP appeared to be maximal both at the midline and ipsilateral to the half field stimulated. The response showed a more lateral distribution than under full-field stimulation and was clearly present at the most lateral occipital electrodes. However, on the contralateral side of the head the response was of lower amplitude and a separate response of opposite polarity consisting of a positive at 75, a negative at 105 and a positive at 135 ms was obtained. This component was clearly maximal at the most lateral occipital electrode position and reduced in amplitude around the midline^. Blumhardt and Halliday showed that the full-field response is in fact the algebraic summation of the responses to each half iield\ The PIOO complex is of highest amplitude at the mid-occipital electrode simply because contralateral negativity is of lowest amplitude at this point whichever half field is stimulated. When both the half fields are stimulated the positivity is clearly maximal at the occiput since it is not cancelled by both contralateral negativities, whereas at the most occipital electrodes, the fall-off of Ihe PIOO component is marked because it is in these
171
Visual evoked electrical anil magnetic response: G. F. A. Harding et al. ipsilateral positivity and contralateral negativity for large field stimulation suggesting an origin in extra striate visual cortex. Using source derivation techniques Flanagan and Harding showed that with large fields the source was always maximal at the midline occipital electrode whichever half field was stimulated, but found that the sink was always contralateral to the half field being stimulated**. These difficulties inherent in electrophysiology provide the rationale for the more complex techniques of magnetoencephalography. There have been to date relatively few studies on the visual evoked magnetic response (VEMR) to pattern reversal stimulation". As the VEMR reflects current flows in the brain rather than potential differences across the scalp, the response to pattern reversal may well complement VEP studies of the visual cortex, particularly since magnetic responses probably reflect activity in fissures rather than gyri'"-^\ We have therefore carried out a study of the topographic distribution of the PIOOM component of the V E M R using three different check sizes in both full- and half-field stimulation.
Materials and methods Four normal subjects aged 23 40 years participated in the study. They had no neurological or ophthaimological problems and had corrected visual acuities of 6/6 or better. The University Ethical Committee approved the study. The magnetic responses were recorded using a BTi second-order d.c. single channel gradiometer (Model 601. Biomagnetic Technologies, CA. USA). The system had a white noise level of 16.6 fT Hz"^ at 5 Hz. All traces were recorded on a Bio-logic Traveler, Bio-logic, Oxon, UK and 50 responses were averaged at each location. To carry out topographic mapping the magnetometer was moved into a series of positions on a 20-point grid consisting of 4 rows of 5 locations per row, located over the visual cortex with the inion as the mid-position on the bottom row of positions. The distance between the points on the grid was defined as 10% of the half circumference of the head. The stimulus consisted ofthree different check sizes (22, 34 and 70 min) back projected on to a screen to produce a 14 degree circular field. The checkerboard reversed twice per second
and the mean luminance of the screen was 1050 cd m"^ while the contrast was 76%. We have previously shown'- that the most consistent VEMR component to pattern reversal stimulation is a major positive deflection (outgoing magnetic field) observed between 90 and 120 ms. This component we have entitled the PIOOM and the amplitude of this signal peak to baseline at each location on the grid was used as the mapping input. Topographic mapping was performed on a Nicolct Pathfinder II (Warwick. UK) using a rectangular grid and linear interpolation. This data was also used as the basis for source localization using a dipole in a sphere model'-*.
Results Half-field stimulation produced the strongest magnetic responses over the contralateral hemisphere. Pattern reversal response to right half-field stimulation using either 34 or 70 min checks showed an outgoing magnetie field over the anterior left visual cortex and an ingoing field over the posterior left visual cortex. When the left-half field was stimulated the response consisted of an outgoing field over the posterior right visual cortex and an ingoing field over the right anterior visual cortex {Figure J). The maps were often complex and quite frequently showed an ingoing field around the midline or even extending slightly into the ipsilateral hemisphere in some individuals (F/t/nrc 3). Applying Fleming's right hand rule this would suggest that the source is in the contralaterai hemisphere to the field stimulated with current flowing towards the midline. The results of all four subjects were submitted to source localization techniques and these in general confirmed the source assumed from the topographic maps [Figure 4). In each case the dipole was located to the contralateral side of the midline to the field stimulated but surprisingly the orientation derived from the magnetomctry data that was obtained" did not coincide with the orientation of presumed dipoles derived from the VEP but had more ipsilateral and anterior orientation. Discussion The results of half-field stimulation confirm that the source of both the visual evoked magnetic response, and
Figure I Brain maps of the main positive PIOO componcnl of the vi.sual evoked electrical polentiai elicited by stimulation of Ihc lefl half field. The check sizes of ihe paltern reversing stimulus are shown above ihc head maps and consisl of 56. 27 and 11 min checks. The actual lime of occurrence of the PIOO is shown below each map in milliseconds. The colour bar at the bottom of the figure shows the amplitude in microvolts. It can be seen that the PIOO component demonstrates the well known paradoxical lateralization. With whatever check size the Icfl half field is stimulated, the maximal respon.se appears lo be over the left visual cortex Figure 2 The lopographic maps obtained from one of the subjects lo Icfl half lield ( LHF), full field (FF)and right half field IRHF) stimulation. The red end of the spectrum represents positive, thai is outgoing magnetic fields, and the green end of ihe spectrum represents negalivc. thai is ingoing fields. These responses were produced by paltern reversal stimulation using 34 min checks. The circles indicate ihe positions recorded using Ihc magnetometer with ihe occiput being represented as ihc central posilion of the second row from the bottom. It can be seen thai on Icfl half ticld stimulation the response consists of an outgoing field in the right posterior quadrant and in ingoing field in the right upper quadrant, suggesting a dipole anterior lo lh
