- Email: [email protected]
The scalp topography of potentials in auditory and visual discrimination tasks
528
Electroencephalography and Clinical Neurophysiology, 1977, 4 2 : 5 2 8 - - 5 3 5 © Elsevier/North-Holland Scientific Publishers, Ltd.
THE SCALP T O P O G R A P H Y OF POTENTIALS IN A U D I T O R Y AND VISUAL DISCRIMINATION TASKS * RICHARD SIMSON, HERBERT G. VAUGHAN JR. and WALTER RITTER Departments o f Neuroscience and Neurology, and the Rose F. Kennedy Center for Research in Mental Retardation and Human Development, Albert Einstein College o f Medicine, 1300 Morris Park Avenue, Bronx, N.Y. 10461 (U.S.A.) (Accepted for publication: July 12, 1976)
This is the second of a series of studies intended to provide information on the distribution of event-related cortical potentials (ERPs) which arise when demands are made for cognitive stimulus evaluation. These potentials include a late positive c o m p o n e n t (LPC or P3) and an earlier negativity that is ordinarily obscured by the concurrent sensory evoked potential (EP). In the first report (Simson et al. 1976), we examined the scalp topography of ERPs emitted in association with the unpredictable deletion of an auditory or visual stimulus from a regular sequence. The 'missing stimulus paradigm' permitted the mapping of these potentials in the absence of stimulus evoked activity. The late positivity (PMSP) possessed a topography centering on the midparietal region which was independent of stimulus modality whereas the preceding negative wave (NMSP) displayed a modality specific topography compatible with an origin in auditory and visual association areas. The topography of task-related activity preceding the LPC cannot be directly analyzed when a sensory EP is present. Although some studies have found that a distinct negative wave (N2) precedes P3 with unpredictable stimulus changes (Courchesne et al. 1975; Squires et al. 1975), others have not (Sutton et al. 1965; Ritter et al. 1968, 1972; Picton et * This research was supported by Grants MH-06723 and HD-01799 from the U.S. Public Health Service.
al. 1974). Presumably, the relative magnitude and timing of the P2 E P c o m p o n e n t and the late negative wave determine the degree to which the negativity appears as a distinct c o m p o n e n t (N2) or merely produces an apparent attenuation in P2 amplitude. This study examines the scalp topography of potentials recorded in a task that requires the detection of infrequent changes in stimulus parameters (Ritter and Vaughan 1969). To analyze the distribution of components differentially elicited by the infrequent (signal) stimuli, potentials evoked by the frequent (non-signal) stimuli were subtracted from those associated with the signals, obtaining a difference waveform whose deflections were mapped and compared with the previously reported topography of the missing stimulus potentials (MSPs).
Method The subjects were one female and seven males, 21--45 years old, all but one of whom had served as a subject in a previous topographic study (Simson et al. 1976). A detailed description of general procedures can be found in that report. One hundred and fifty auditory or visual stimuli were presented in each run at a rate of 1/2 sec (non-signals). Randomly, averaging one in ten presentations, a non-signal was replaced by an easily discriminable different
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS
stimulus (signal) to which the subject was instructed to respond after the next non-signal. Three runs with approximately 15 signals were presented in each modality. Auditory stimuli were 50 msec tone bursts with rise and decay times of 5 msec, delivered binaurally through Koss Pro 4A headphones at 50--60 dB above subjective threshold. The nonsignal tone was 2000 c/sec and the signal was 1000 c/sec. Visual stimuli were black bars within a 5 ° white circle flashed tachistoscopically for 10 msec onto the center of a dimly lit 11 ° X 11 ° field which the subject viewed with the right eye. The black bar was
A
U
D
I
NS
529
oriented vertically as the non-signal, and horizontally as the signal. Thirteen felt-padded chlorided silver disc electrodes were affixed to the scalp over the midline and left hemicranium as illustrated in Fig. 1 arid were referred to an electrode on the tip of the nose. A supraorbital electrode monitored vertical eye movements. The EEG was amplified by Tektronix 2A61 low level differential amplifiers set for a gain of 20,000 with bandpass down 3 dB at 0.06 and 60 c/sec, and recorded on FM magnetic tape. Averages were computed off line by a Nicolet Med-80 computer and written out on a
T
0
R
Y
S
A N2A
NI
P2
P3/,,
io . - , " ~ , ~
L
I
I
I
I
I
I l l l ] l
t ] I lOOms
L I
] ]5,uv
Fig. 1. Mean for eight subjects of the non-signal (NS), signal (S) and difference (A) waveforms at each electrode site in the auditory condition. Isopotential topographic distributions are expressed as percentages of maximum response amplitude for the N1 and P2 components of the non-signal response (left) and the negative (N2 A) and positive (P3A) components of the A waveform (right). Supraorbital (0) and vertex (electrode 3)traces from the 3 runs are superimposed.
530 Houston Instrument Co. X--Y plotter. Averages for each subject were obtained for 45 signals and 300 non-signals in each modality. Grand Mean waveforms for the eight subjects were also computed for 360 signals and 2,400 non-signals. Difference waveforms were obtained by digital subtraction of the average non-signal waveforms from the average signal wave forms. Peaks were visually identified for the Grand Mean and each subject's signal, non-signal and difference waveforms at each electrode site and their amplitudes were measured with reference to the average voltage over the 50 msec period following stimulus onset. These measurements were converted to percentages of the maximum amplitude and isopotential maps were constructed for each deflection as described in Simson et al. (1976). Peak latency measurements to the nearest 5 msec were determined from the recordings of the electrode with the largest amplitude for that component. The similarity of the distributions of comparable components was assessed by the Pearson correlation coefficient (r), calculated from pairs of Grand Mean amplitude measures at the 13 scalp locations. Comparisons between the results of the vigilance and missing stimulus paradigms were restricted to data obtained from the seven subjects common to both studies.
Results
Auditory condition The mean non-signal, signal and difference (A) waveforms computed across subjects for the auditory modality are depicted for each electrode site in Fig. 1. The non-signal auditory evoked potential (AEP) at the vertex (electrode 3), was triphasic with peaks P1 (~ = 50 msec; S.D. = 12 msec), N1 (100 + 11 msec) and P2 (190 -+ 11 msec). In three subjects a small late positive component (350 + 13 msec) was present as well.
R. SIMSON ET AL. The early portion of the signal AEP was similar to that of the non-signal in all subjects, with peaks P1 (50 +- 13 msec) and N1 (100 + 11 msec). The second positive peak, P2 (165 + 18 msec), was earlier in peak latency than the comparable non-signal P2 and of smaller amplitude. Two additional components characteristic of the signal AEP were N2 (220 -+ 22 msec) and P3 (350 -+ 29 msec). Topographic maps of the non-signal N1 and P2 are shown at the left in Fig. 1. The distributions of the non-signal and signal N1 were virtually identicaJ (r = 0.98), whereas the P2 distribution for the signals differed somewhat from that of the non-signals (r = 0.87) evidently due to distortion introduced by a concurrent negative deflection in the signal AEP. This negativity (N2A) is clearly disclosed in the biphasic difference waveforms obtained by subtracting the non-signal AEP from the signal AEP. At the site of maximum amplitude (electrode 3), N2A had a mean onset at 130 + 25 msec and peaked at 200 -+ 17 msec. The distribution of N2A differed from that of the signal N2 due to the interaction between the N2 and P2 components in the signal AEP. In contrast, the topography of P3 (350 _+ 29 msec) for the A waveform was the same as that of the signal P3 (r = 0.95). The topography of N2A depicted in Fig. 1 shows a maximum just anterior to the vertex and a distribution similar to that of the non-signal AEP components N1 (r = 0.71) and P2 (r = 0.76), except for the extension toward the posterior extremity of the lateral fissure. The P3 distribution had a midparietal maximum and extended laterally into the parietotemporal and anteriorly into the midfrontal regions.
Visual condition The mean non-signal, signal and difference waveforms computed across subjects for the visual modality are depicted for each electrode site in Fig. 2. The non-signal visual evoked potential (VEP) at the midoccipital electrode (6), seen in all subjects was tri-
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS
V
I
S
U
NS
[
~
~
531
A
L
S
I
d
J
J
&
I
E -
I J lOOms
I
I
I
I
~v
Fig. 2. Mean waveforms and topographic distributions for the visual condition.
N2A
NMSP
P3A
PMSP
VISUAL
AUDITORY
Fig. 3. Uni-hemipheric distributions of the negative (N2A, NMSP) and positive (P3A, PMSP) components of the different waveforms (A) and the missing stimulus potentials (MSPs) in the auditory and visual modalities. These distributions are derived as explained in the text from the data shown in Fig. 1 and 2 of this paper and in Fig. 1 and 2 of Simson et al. (1976).
532 phasic with peaks P1 (105 -+ 18 msec), N1 (165 + 9 msec) and P2 (245 + 31 msec). In seven subjects' records a low amplitude P3 (375 + 54 msec) could also be seen. P1 was small and could be detected only over the posterior portion of the scalp. The alterations of the VEP to the signals (see Fig. 2) were much like those described for the AEP. P1 (100 -+ 28 msec) and N1 (165 + 7 msec) were indistinguishable from the comparable non-signal components. P2 (230 + 16 msec) was earlier and of smaller amplitude than the non-signal P2. N2 (295 + 43 msec) and a large P3 (465 + 45 msec) completed the signal VEP. The scalp distributions of the principal non-signal VEP components N1 and P2 are depicted in Fig. 2. As in the case of the AEP, the N1 distributions for the non-signal and signal stimuli were virtually identical (r = 0.96), whereas the P2 distributions differed from one another (r = 0.20). The A waveforms were biphasic comprising a negative component (N2A) with onset at 170 +- 30 msec and peak at 310 + 47 msec, and P3 at 470 + 43 msec with essentially the same distribution as the signal P3 (r = 0.98). N2A was restricted to the parieto-occipital region and its distribution was very similar to that of the N1 VEP component (r = 0.93). P3 was widely distributed with a parietal maximum, resembling the topography of the auditory P3 (r = 0.86).
Discussion
Subtracting EPs to non-signal stimuli from those to signal stimuli in a vigilance task yields a biphasic negative--positive waveform similar in configuration to the MSPs described in the first paper of this series (Simson et al. 1976). The subtraction procedure discloses a negative deflection (N2A) of the signal EP which is otherwise partially obscured by the concurrent P2 component. The earlier component, N1, being identical in signal and nonsignal responses, is absent from the A wave-
R. SIMSON ET AL. form and P3A is unchanged in both latency and topography from the signal P3. The subtraction technique used in the present study has previously been employed to differentiate sensory and motor components of cortical event-related potentials in reaction time tasks (Vaughan et al. 1965; Vaughan and Costa 1968). Hatter and Salmon (1972) applied this m e t h o d in a visual discrimination task to define the difference in cortical response to signal and non-signal stimuli, and reported biphasic A waveforms like those described here for a similar paradigm. N2A might be affected by alterations in the amplitude of P2 to signals and non-signals. Such a contribution would represent a decrease in P2 amplitude for the signal EP. AIthough it is not possible to rule this out, the timing of P2 and N2A peaks is substantially different. Furthermore, there is no deflection or inflection in the falling phase of N2A which might indicate a contribution of P2 changes to the difference waveform. Thus, any influence of EP changes to the N2A waveform must be small and is unlikely to affect its peak amplitude distribution. In the auditory modality N2A did not present any definite differences in waveshape over the scalp. Its topography was very similar to that of the auditory NMSP, although decrementing more rapidly with distance from the site of maximum amplitude. Individual subjects' distributions were quite comparable across experiments. Auditory N2A and NMSP distributions were not only similar to one another, but resembled in part the N1 and P2 auditory EP distributions. The main difference was in the extension of N2A and NMSP over the posterior sylvian region where polarity inversion was seen for the AEP components. Cortical and intracerebral mapping of the AEP in the rhesus monkey (Arezzo et al. 1975) has shown that sources in the supratemporal plane produce volume-conducted surface potential distributions like those found in human scalp recordings of AEP, NMSP, and N2A. Considering this, the present auditory NMSP and N2A dis-
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS tributions appear to be best explained by a source within the supratemporal plane projecting to the cortical surface in the central region, which sums with activity generated by a secondary source on the surface of the posterior portion of the superior temporal gyrus. It is possible that some of the central activity could be generated within m o t o r cortex, as is the case for some components of the monkey AEP (Goldring et al. 1970; Arezzo et al. 1975), but Goldring failed to find either gross potentials (Goldring et al. 1970) or unit activity (Goldring and Ratcheson 1972) evoked by auditory stimuli in human motor cortex. Although N2A appears to be a distinct c o m p o n e n t of the signal EP, it may not be a single waveform. Examination of the visual N 2 ~ traces obtained at frontal compared to posterior placements discloses a difference in the waveshape, with the negative wave beginning and reaching its peak considerably later in the midfrontal region than elsewhere on the scalp. Frontal activity did not contribute appreciably to the visual N2A distribution, but the visual NMSP did extend into the posterior frontal region. Individual maps for the seven subjects who participated in both studies confirmed the differential distribution of the visual negative c o m p o n e n t between the two paradigms in five cases (for one subject the distribution could not be accurately mapped in one condition, and for another subject there was no difference). VEP P2 distributions similarly showed a less prominent frontal extension in the vigilance than in the missing stimulus condition. The topography of the visual N2A is indicative of a principal source within the prestriate cortex. Thus, both NMSP and N2A activity appears to be generated principally within secondary auditory and visual cortical areas, overlapping the sources of the N1 and P2 components of AEP and VEP (Vaughan 1969; Vaughan and Ritter 1970; Simson et al. 1 9 7 6 ) . As in the MSP study, the P3 topography in the vigilance task was quite similar across modalities. However, distributions differed
533 between the two studies, with the vigilance P3 extending more anteriorly in the midline than the PMSP. Examination of the vigilance Grand Mean waveforms discloses that in the visual modality the timing of the predominant positive peak is approximately 50 msec earlier in frontal than parieto-occipital areas. The posterior waveforms also return to baseline more slowly than those recorded over the rest of the scalp. Auditory LPCs are prolonged in the parieto-occipital region but do n o t show distinct differences in peak latency. Squires et al. (1975) have reported a similar prolonged return to baseline (the slow wave or SW), of greatest magnitude parietally, when unpredictable intensity shifts are target stimuli (signals). Subtle but distinct regional variations in LPC waveshape were seen in the data of the preceding study as well and are suggestive of more than one intracranial generator. Assuming approximate symmetry of ERP distributions (viz., Vaughan and Ritter 1970; Hillyard et al. in press), it is possible to obtain the distributions due to sources within one hemisphere by halving the observed midline values and adjusting parasagittal amplitudes when necessary to equalize the medial and lateral potential gradients. Fig. 3 depicts these distributions for the negative components, NMSP and N2A, and for the LPC, PMSP and P3A. Substantial similarities between the distributions of the negative components for each modality are evident between the two experimental conditions, although the auditory and visual distributions are different. Topographic differences between PMSP and P3 are less prominent in the single hemisphere maps than in the raw distributions. Both LPCs are seen maximally overlying the inferior parietal lobule, implicating this modality unspecific association area as a main source of late positive activity in both experimental conditions. Furthermore, distributions in both paradigms show some extension into the frontal region, differing mainly in the strength of this activity. These topographies are compatible with either a single
534 source extending into the central region from the inferior parietal region, or with two spatially distinct parietal and frontal sources. The postulate of distixlct frontal and parietal sources is consistent with recent reports of more than one LPC that differ in scalp distribution (Courchesne et al. 1975; Squires et al. 1975). The data of neither the present nor the preceding study, however, disclose the distinctly earlier P3a reported by Squires et al. The fact that N2A and P3 possess different topographic distributions, and thus, derive from different intracranial sources suggests differences in the functional significance of these two potentials. N2A occurs early enough to represent a necessary part of the information processing sequence leading to a motor response, which is consistent with our previous suggestion (Ritter et al. 1972) that the cortical processes associated with sensory discrimination precede the LPC. Several studies have reported variations in N2 which further support this (Bostock and Jarvis 1970; Karlin et al. 1970; Ford et al. 1973). The modality specific topography of the N2A potentials described here suggests that they reflect the operation of the discriminative mechanism for stimuli within a single sensory modality. Since an immediate motor response was not required it is likely that the potentials we recorded were uncontaminated by activity related to the organization of a m o t o r response, thus representing 'pure' indices of the discriminative process. Inasmuch as N2A is concurrent with much of the P2 c o m p o n e n t of the sensory EP and derives from similar brain regions, it seems likely that the discriminative mechanism it reflects involves neural systems related to the later portions of the obligatory cortical response to sensory input.
Summary Averaged event-related cortical potentials (ERPs) were obtained from an array of scalp electrodes overlying the left hemicranium in
R. SIMSON ET AL. response to regularly presented visual or auditory stimuli (non-signals) and to infrequent random replacements by different stimuli (signals) in the same modality. A delayed m o t o r response was required to the signals. Non-signal ERPs were subtracted from signal ERPs and the topographic distributions of the negative (N2A) and positive (P3A) components were plotted as isopotential maps. N2A distributions differed for the auditory and visual modalities, whereas P3A was modality unspecific. These topographic data were compared to those from the previous study of missing stimulus potentials (Simson et al. 1976) using maps representing the contributions from unilateral cerebral sources. The N2A and negative missing stimulus potential distributions ascribed to cortical activity within the secondary auditory and visual regions, whereas the late positive c o m p o n e n t (positive missing stimulus potential or P3A) were considered to derive principally from inferior parietal association cortex.
R6sum6 Topographie des potentiels sur le scalp lors de taches de discrimination auditives et visuelles
Sur un faisceau d'~lectrodes de scalp recouvrant l'h~micr~ne gauche les ERP moyens sont obtenus en r~ponse ~ des stimuli visuels ou auditifs pr~sent6s r~guli~rement (non signaux) et ~ diff~rents stimuli (signaux) de m~me modalit~ qui les remplacent de faqon peu fr~quente et al~atoire. I1 est demand6 au sujet de r~pondre aux signaux par une action motrice retard~e. Les-ERP aux stimuli nonsignaux sont soustraits des ERP aux signaux et les distributions topographiques des composantes n~gatives (N2A) et positives (P3A) sont relev~es sous forme de nappes isopotentielles. Les distributions de N2A different pour les modalit~s auditive et visuelle, tandis que celles de P3A sont non sp6cifique de la modalit~ de stimulation. Ces donn6es topographiques sont compar6es ~ celles obtenues,
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS l o r s d ' 6 t u d e s a n t 6 r i e u r e s , s u r les p o t e n t i e l s a u x s t i m u l i s m a n q u a n t s ( M S P , S i m s o n e t al. 1 9 7 6 ) ~ l ' a i d e d e n a p p e s r e p r 6 s e n t a n t les c o n tributions de sources c6r6brales unilat6rales. L e s d i s t r i b u t i o n s d e N 2 A e t N M S P se r 6 f ~ r e n t une activit6 corticale ~ l'int6rieur des r6gions v i s u e l l e s e t a u d i t i v e s s e c o n d a i r e s t a n d i s q u e les LPC (PMSP ou P3A) paraissent d6river princ i p a l e m e n t d u c o r t e x d ' a s s o c i a t i o n p a r i 6 t a l inf6rieur.
References Arezzo, J., Pickoff, A. and Vaughan, H.G., Jr. The sources and intracerebral distribution of auditory evoked potentials in the alert rhesus monkey. Brain Res., 1975, 90: 57--73. Bostock, H. and Jarvis, M.J. Changes in the form of the cerebral evoked response related to the speed of simple reaction time. Electroenceph. clin. Neurophysiol., 1970, 29: 137--145. Courchesne, E., Hillyard, S.A. and Galambos, R. Stimulus novelty, task relevance and the visual evoked potential in man. Electroenceph. clin. Neurophysiol., 1975, 39: 131--143. Ford, J.M., Roth, W.T., Dirks, S.J. and Kopell, B.S. Evoked potential correlates of signal recognition between and within modalities. Science, 1973, 181: 465--466. Goldring, S. and Ratcheson, R. Human motor cortex: sensory input data from single neuron recordings. Science, 1972, 175: 1493--1495. Goldring, S., Aras, E. and Weber, P.C. Comparative study of sensory input to motor cortex in animals and man. Electroenceph. clin. Neurophysiol., 1970, 29: 537--550. Hatter, M.R. and Salmon, L.E. Intra-modality selective attention and evoked cortical potentials to randomly presented patterns. Electroenceph. clin. Neurophysiol., 1972, 32: 605--613. Hillyard, S.A., Courchesne, E., Krausz, H.I. and Picton, T.W. Scalp topography of the 'P3' wave in different auditory decision tasks. In W.C. McCallum and J.R. Knott (Eds.), Event related slow potentials of the brain. John Wright and Sons, Bristol, in press. Karlin, L., Martz, M.J. and Mordkoff, A.M. Motor
535 performance and sensory evoked potentials. Electroenceph, clin. Neurophysiology, 1970, 28: 307--313. Picton, T.W., Hillyard, S.A. and Galambos, R. [Cortical evoked responses to omitted stimuli. In M.N. Livanov (Ed.), Major problems of brain electrophysiology.] (in Russian). Academy of Sciences, U.S.S.R., 1974: 302--311. Ritter, W. and Vaughan, H.G., Jr. Averaged evoked responses in vigilance and discrimination: A reassessment. Science, 1969, 164: 326--328. Ritter, W., Simson, R. and Vaughan, H.G., Jr. Association cortex potentials and reaction time ih auditory discrimination. Electroenceph. clin. Neurophysiol., 1972, 33: 547--555. Ritter, W., Vaughan, H.G., Jr. and Costa, L.D. Orienting and habituation to auditory stimuli: A study of short term changes in average evoked responses. Electroenceph. clin. Neurophysiol., 1968, 25: 550--556. Simson, R., Vaughan, H.G., Jr. and Ritter, W. The scalp topography of potentials associated with missing visual or auditory stimuli. Electroenceph. clin. Neurophysiol., 1976, 40: 33--42. Squires, N.K., Squires, K.C. and Hillyard, S.A. Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroenceph, clin. Neurophysiol., 1975, 38: 387--401. Sutton, S., Braren, M., Zubin, J. and John, E.R. Evoked-potential correlates of stimulus uncertainty. Science, 1965, 150: 1187--1188. Vaughan, H.G., Jr. The relationship of brain activity to scalp recordings of event-related potentials. In E. Donchin and D.B. Lindsley (Eds.), Averaged evoked potentials: methods, results, evaluations. National Aeronautics and Space Administration (NASA SP-191), Washington, D.C., 1969: 45--94. Vaughan, H.G., Jr. and Costa, L.D. Analysis of electroencephalographic correlates of human sensorimotor processes. Electroenceph. clin. Neurophysiol., 1968, 24: 288. Vaughan, H.G., Jr. and Ritter, W. The sources of auditory evoked responses recorded from the human scalp. Eleetroenceph. clin. Neurophysiol., 1970, 28: 360--367. Vaughan, H.G., Jr. Costa, L.D., Gilden, L. and Schimmel, H. Identification of sensory and motor components of cerebral activity in simple reaction-time tasks. Proc. 73 Conf. Amer. Psychol. Ass., 1965, 1: 179--180.
Electroencephalography and Clinical Neurophysiology, 1977, 4 2 : 5 2 8 - - 5 3 5 © Elsevier/North-Holland Scientific Publishers, Ltd.
THE SCALP T O P O G R A P H Y OF POTENTIALS IN A U D I T O R Y AND VISUAL DISCRIMINATION TASKS * RICHARD SIMSON, HERBERT G. VAUGHAN JR. and WALTER RITTER Departments o f Neuroscience and Neurology, and the Rose F. Kennedy Center for Research in Mental Retardation and Human Development, Albert Einstein College o f Medicine, 1300 Morris Park Avenue, Bronx, N.Y. 10461 (U.S.A.) (Accepted for publication: July 12, 1976)
This is the second of a series of studies intended to provide information on the distribution of event-related cortical potentials (ERPs) which arise when demands are made for cognitive stimulus evaluation. These potentials include a late positive c o m p o n e n t (LPC or P3) and an earlier negativity that is ordinarily obscured by the concurrent sensory evoked potential (EP). In the first report (Simson et al. 1976), we examined the scalp topography of ERPs emitted in association with the unpredictable deletion of an auditory or visual stimulus from a regular sequence. The 'missing stimulus paradigm' permitted the mapping of these potentials in the absence of stimulus evoked activity. The late positivity (PMSP) possessed a topography centering on the midparietal region which was independent of stimulus modality whereas the preceding negative wave (NMSP) displayed a modality specific topography compatible with an origin in auditory and visual association areas. The topography of task-related activity preceding the LPC cannot be directly analyzed when a sensory EP is present. Although some studies have found that a distinct negative wave (N2) precedes P3 with unpredictable stimulus changes (Courchesne et al. 1975; Squires et al. 1975), others have not (Sutton et al. 1965; Ritter et al. 1968, 1972; Picton et * This research was supported by Grants MH-06723 and HD-01799 from the U.S. Public Health Service.
al. 1974). Presumably, the relative magnitude and timing of the P2 E P c o m p o n e n t and the late negative wave determine the degree to which the negativity appears as a distinct c o m p o n e n t (N2) or merely produces an apparent attenuation in P2 amplitude. This study examines the scalp topography of potentials recorded in a task that requires the detection of infrequent changes in stimulus parameters (Ritter and Vaughan 1969). To analyze the distribution of components differentially elicited by the infrequent (signal) stimuli, potentials evoked by the frequent (non-signal) stimuli were subtracted from those associated with the signals, obtaining a difference waveform whose deflections were mapped and compared with the previously reported topography of the missing stimulus potentials (MSPs).
Method The subjects were one female and seven males, 21--45 years old, all but one of whom had served as a subject in a previous topographic study (Simson et al. 1976). A detailed description of general procedures can be found in that report. One hundred and fifty auditory or visual stimuli were presented in each run at a rate of 1/2 sec (non-signals). Randomly, averaging one in ten presentations, a non-signal was replaced by an easily discriminable different
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS
stimulus (signal) to which the subject was instructed to respond after the next non-signal. Three runs with approximately 15 signals were presented in each modality. Auditory stimuli were 50 msec tone bursts with rise and decay times of 5 msec, delivered binaurally through Koss Pro 4A headphones at 50--60 dB above subjective threshold. The nonsignal tone was 2000 c/sec and the signal was 1000 c/sec. Visual stimuli were black bars within a 5 ° white circle flashed tachistoscopically for 10 msec onto the center of a dimly lit 11 ° X 11 ° field which the subject viewed with the right eye. The black bar was
A
U
D
I
NS
529
oriented vertically as the non-signal, and horizontally as the signal. Thirteen felt-padded chlorided silver disc electrodes were affixed to the scalp over the midline and left hemicranium as illustrated in Fig. 1 arid were referred to an electrode on the tip of the nose. A supraorbital electrode monitored vertical eye movements. The EEG was amplified by Tektronix 2A61 low level differential amplifiers set for a gain of 20,000 with bandpass down 3 dB at 0.06 and 60 c/sec, and recorded on FM magnetic tape. Averages were computed off line by a Nicolet Med-80 computer and written out on a
T
0
R
Y
S
A N2A
NI
P2
P3/,,
io . - , " ~ , ~
L
I
I
I
I
I
I l l l ] l
t ] I lOOms
L I
] ]5,uv
Fig. 1. Mean for eight subjects of the non-signal (NS), signal (S) and difference (A) waveforms at each electrode site in the auditory condition. Isopotential topographic distributions are expressed as percentages of maximum response amplitude for the N1 and P2 components of the non-signal response (left) and the negative (N2 A) and positive (P3A) components of the A waveform (right). Supraorbital (0) and vertex (electrode 3)traces from the 3 runs are superimposed.
530 Houston Instrument Co. X--Y plotter. Averages for each subject were obtained for 45 signals and 300 non-signals in each modality. Grand Mean waveforms for the eight subjects were also computed for 360 signals and 2,400 non-signals. Difference waveforms were obtained by digital subtraction of the average non-signal waveforms from the average signal wave forms. Peaks were visually identified for the Grand Mean and each subject's signal, non-signal and difference waveforms at each electrode site and their amplitudes were measured with reference to the average voltage over the 50 msec period following stimulus onset. These measurements were converted to percentages of the maximum amplitude and isopotential maps were constructed for each deflection as described in Simson et al. (1976). Peak latency measurements to the nearest 5 msec were determined from the recordings of the electrode with the largest amplitude for that component. The similarity of the distributions of comparable components was assessed by the Pearson correlation coefficient (r), calculated from pairs of Grand Mean amplitude measures at the 13 scalp locations. Comparisons between the results of the vigilance and missing stimulus paradigms were restricted to data obtained from the seven subjects common to both studies.
Results
Auditory condition The mean non-signal, signal and difference (A) waveforms computed across subjects for the auditory modality are depicted for each electrode site in Fig. 1. The non-signal auditory evoked potential (AEP) at the vertex (electrode 3), was triphasic with peaks P1 (~ = 50 msec; S.D. = 12 msec), N1 (100 + 11 msec) and P2 (190 -+ 11 msec). In three subjects a small late positive component (350 + 13 msec) was present as well.
R. SIMSON ET AL. The early portion of the signal AEP was similar to that of the non-signal in all subjects, with peaks P1 (50 +- 13 msec) and N1 (100 + 11 msec). The second positive peak, P2 (165 + 18 msec), was earlier in peak latency than the comparable non-signal P2 and of smaller amplitude. Two additional components characteristic of the signal AEP were N2 (220 -+ 22 msec) and P3 (350 -+ 29 msec). Topographic maps of the non-signal N1 and P2 are shown at the left in Fig. 1. The distributions of the non-signal and signal N1 were virtually identicaJ (r = 0.98), whereas the P2 distribution for the signals differed somewhat from that of the non-signals (r = 0.87) evidently due to distortion introduced by a concurrent negative deflection in the signal AEP. This negativity (N2A) is clearly disclosed in the biphasic difference waveforms obtained by subtracting the non-signal AEP from the signal AEP. At the site of maximum amplitude (electrode 3), N2A had a mean onset at 130 + 25 msec and peaked at 200 -+ 17 msec. The distribution of N2A differed from that of the signal N2 due to the interaction between the N2 and P2 components in the signal AEP. In contrast, the topography of P3 (350 _+ 29 msec) for the A waveform was the same as that of the signal P3 (r = 0.95). The topography of N2A depicted in Fig. 1 shows a maximum just anterior to the vertex and a distribution similar to that of the non-signal AEP components N1 (r = 0.71) and P2 (r = 0.76), except for the extension toward the posterior extremity of the lateral fissure. The P3 distribution had a midparietal maximum and extended laterally into the parietotemporal and anteriorly into the midfrontal regions.
Visual condition The mean non-signal, signal and difference waveforms computed across subjects for the visual modality are depicted for each electrode site in Fig. 2. The non-signal visual evoked potential (VEP) at the midoccipital electrode (6), seen in all subjects was tri-
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS
V
I
S
U
NS
[
~
~
531
A
L
S
I
d
J
J
&
I
E -
I J lOOms
I
I
I
I
~v
Fig. 2. Mean waveforms and topographic distributions for the visual condition.
N2A
NMSP
P3A
PMSP
VISUAL
AUDITORY
Fig. 3. Uni-hemipheric distributions of the negative (N2A, NMSP) and positive (P3A, PMSP) components of the different waveforms (A) and the missing stimulus potentials (MSPs) in the auditory and visual modalities. These distributions are derived as explained in the text from the data shown in Fig. 1 and 2 of this paper and in Fig. 1 and 2 of Simson et al. (1976).
532 phasic with peaks P1 (105 -+ 18 msec), N1 (165 + 9 msec) and P2 (245 + 31 msec). In seven subjects' records a low amplitude P3 (375 + 54 msec) could also be seen. P1 was small and could be detected only over the posterior portion of the scalp. The alterations of the VEP to the signals (see Fig. 2) were much like those described for the AEP. P1 (100 -+ 28 msec) and N1 (165 + 7 msec) were indistinguishable from the comparable non-signal components. P2 (230 + 16 msec) was earlier and of smaller amplitude than the non-signal P2. N2 (295 + 43 msec) and a large P3 (465 + 45 msec) completed the signal VEP. The scalp distributions of the principal non-signal VEP components N1 and P2 are depicted in Fig. 2. As in the case of the AEP, the N1 distributions for the non-signal and signal stimuli were virtually identical (r = 0.96), whereas the P2 distributions differed from one another (r = 0.20). The A waveforms were biphasic comprising a negative component (N2A) with onset at 170 +- 30 msec and peak at 310 + 47 msec, and P3 at 470 + 43 msec with essentially the same distribution as the signal P3 (r = 0.98). N2A was restricted to the parieto-occipital region and its distribution was very similar to that of the N1 VEP component (r = 0.93). P3 was widely distributed with a parietal maximum, resembling the topography of the auditory P3 (r = 0.86).
Discussion
Subtracting EPs to non-signal stimuli from those to signal stimuli in a vigilance task yields a biphasic negative--positive waveform similar in configuration to the MSPs described in the first paper of this series (Simson et al. 1976). The subtraction procedure discloses a negative deflection (N2A) of the signal EP which is otherwise partially obscured by the concurrent P2 component. The earlier component, N1, being identical in signal and nonsignal responses, is absent from the A wave-
R. SIMSON ET AL. form and P3A is unchanged in both latency and topography from the signal P3. The subtraction technique used in the present study has previously been employed to differentiate sensory and motor components of cortical event-related potentials in reaction time tasks (Vaughan et al. 1965; Vaughan and Costa 1968). Hatter and Salmon (1972) applied this m e t h o d in a visual discrimination task to define the difference in cortical response to signal and non-signal stimuli, and reported biphasic A waveforms like those described here for a similar paradigm. N2A might be affected by alterations in the amplitude of P2 to signals and non-signals. Such a contribution would represent a decrease in P2 amplitude for the signal EP. AIthough it is not possible to rule this out, the timing of P2 and N2A peaks is substantially different. Furthermore, there is no deflection or inflection in the falling phase of N2A which might indicate a contribution of P2 changes to the difference waveform. Thus, any influence of EP changes to the N2A waveform must be small and is unlikely to affect its peak amplitude distribution. In the auditory modality N2A did not present any definite differences in waveshape over the scalp. Its topography was very similar to that of the auditory NMSP, although decrementing more rapidly with distance from the site of maximum amplitude. Individual subjects' distributions were quite comparable across experiments. Auditory N2A and NMSP distributions were not only similar to one another, but resembled in part the N1 and P2 auditory EP distributions. The main difference was in the extension of N2A and NMSP over the posterior sylvian region where polarity inversion was seen for the AEP components. Cortical and intracerebral mapping of the AEP in the rhesus monkey (Arezzo et al. 1975) has shown that sources in the supratemporal plane produce volume-conducted surface potential distributions like those found in human scalp recordings of AEP, NMSP, and N2A. Considering this, the present auditory NMSP and N2A dis-
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS tributions appear to be best explained by a source within the supratemporal plane projecting to the cortical surface in the central region, which sums with activity generated by a secondary source on the surface of the posterior portion of the superior temporal gyrus. It is possible that some of the central activity could be generated within m o t o r cortex, as is the case for some components of the monkey AEP (Goldring et al. 1970; Arezzo et al. 1975), but Goldring failed to find either gross potentials (Goldring et al. 1970) or unit activity (Goldring and Ratcheson 1972) evoked by auditory stimuli in human motor cortex. Although N2A appears to be a distinct c o m p o n e n t of the signal EP, it may not be a single waveform. Examination of the visual N 2 ~ traces obtained at frontal compared to posterior placements discloses a difference in the waveshape, with the negative wave beginning and reaching its peak considerably later in the midfrontal region than elsewhere on the scalp. Frontal activity did not contribute appreciably to the visual N2A distribution, but the visual NMSP did extend into the posterior frontal region. Individual maps for the seven subjects who participated in both studies confirmed the differential distribution of the visual negative c o m p o n e n t between the two paradigms in five cases (for one subject the distribution could not be accurately mapped in one condition, and for another subject there was no difference). VEP P2 distributions similarly showed a less prominent frontal extension in the vigilance than in the missing stimulus condition. The topography of the visual N2A is indicative of a principal source within the prestriate cortex. Thus, both NMSP and N2A activity appears to be generated principally within secondary auditory and visual cortical areas, overlapping the sources of the N1 and P2 components of AEP and VEP (Vaughan 1969; Vaughan and Ritter 1970; Simson et al. 1 9 7 6 ) . As in the MSP study, the P3 topography in the vigilance task was quite similar across modalities. However, distributions differed
533 between the two studies, with the vigilance P3 extending more anteriorly in the midline than the PMSP. Examination of the vigilance Grand Mean waveforms discloses that in the visual modality the timing of the predominant positive peak is approximately 50 msec earlier in frontal than parieto-occipital areas. The posterior waveforms also return to baseline more slowly than those recorded over the rest of the scalp. Auditory LPCs are prolonged in the parieto-occipital region but do n o t show distinct differences in peak latency. Squires et al. (1975) have reported a similar prolonged return to baseline (the slow wave or SW), of greatest magnitude parietally, when unpredictable intensity shifts are target stimuli (signals). Subtle but distinct regional variations in LPC waveshape were seen in the data of the preceding study as well and are suggestive of more than one intracranial generator. Assuming approximate symmetry of ERP distributions (viz., Vaughan and Ritter 1970; Hillyard et al. in press), it is possible to obtain the distributions due to sources within one hemisphere by halving the observed midline values and adjusting parasagittal amplitudes when necessary to equalize the medial and lateral potential gradients. Fig. 3 depicts these distributions for the negative components, NMSP and N2A, and for the LPC, PMSP and P3A. Substantial similarities between the distributions of the negative components for each modality are evident between the two experimental conditions, although the auditory and visual distributions are different. Topographic differences between PMSP and P3 are less prominent in the single hemisphere maps than in the raw distributions. Both LPCs are seen maximally overlying the inferior parietal lobule, implicating this modality unspecific association area as a main source of late positive activity in both experimental conditions. Furthermore, distributions in both paradigms show some extension into the frontal region, differing mainly in the strength of this activity. These topographies are compatible with either a single
534 source extending into the central region from the inferior parietal region, or with two spatially distinct parietal and frontal sources. The postulate of distixlct frontal and parietal sources is consistent with recent reports of more than one LPC that differ in scalp distribution (Courchesne et al. 1975; Squires et al. 1975). The data of neither the present nor the preceding study, however, disclose the distinctly earlier P3a reported by Squires et al. The fact that N2A and P3 possess different topographic distributions, and thus, derive from different intracranial sources suggests differences in the functional significance of these two potentials. N2A occurs early enough to represent a necessary part of the information processing sequence leading to a motor response, which is consistent with our previous suggestion (Ritter et al. 1972) that the cortical processes associated with sensory discrimination precede the LPC. Several studies have reported variations in N2 which further support this (Bostock and Jarvis 1970; Karlin et al. 1970; Ford et al. 1973). The modality specific topography of the N2A potentials described here suggests that they reflect the operation of the discriminative mechanism for stimuli within a single sensory modality. Since an immediate motor response was not required it is likely that the potentials we recorded were uncontaminated by activity related to the organization of a m o t o r response, thus representing 'pure' indices of the discriminative process. Inasmuch as N2A is concurrent with much of the P2 c o m p o n e n t of the sensory EP and derives from similar brain regions, it seems likely that the discriminative mechanism it reflects involves neural systems related to the later portions of the obligatory cortical response to sensory input.
Summary Averaged event-related cortical potentials (ERPs) were obtained from an array of scalp electrodes overlying the left hemicranium in
R. SIMSON ET AL. response to regularly presented visual or auditory stimuli (non-signals) and to infrequent random replacements by different stimuli (signals) in the same modality. A delayed m o t o r response was required to the signals. Non-signal ERPs were subtracted from signal ERPs and the topographic distributions of the negative (N2A) and positive (P3A) components were plotted as isopotential maps. N2A distributions differed for the auditory and visual modalities, whereas P3A was modality unspecific. These topographic data were compared to those from the previous study of missing stimulus potentials (Simson et al. 1976) using maps representing the contributions from unilateral cerebral sources. The N2A and negative missing stimulus potential distributions ascribed to cortical activity within the secondary auditory and visual regions, whereas the late positive c o m p o n e n t (positive missing stimulus potential or P3A) were considered to derive principally from inferior parietal association cortex.
R6sum6 Topographie des potentiels sur le scalp lors de taches de discrimination auditives et visuelles
Sur un faisceau d'~lectrodes de scalp recouvrant l'h~micr~ne gauche les ERP moyens sont obtenus en r~ponse ~ des stimuli visuels ou auditifs pr~sent6s r~guli~rement (non signaux) et ~ diff~rents stimuli (signaux) de m~me modalit~ qui les remplacent de faqon peu fr~quente et al~atoire. I1 est demand6 au sujet de r~pondre aux signaux par une action motrice retard~e. Les-ERP aux stimuli nonsignaux sont soustraits des ERP aux signaux et les distributions topographiques des composantes n~gatives (N2A) et positives (P3A) sont relev~es sous forme de nappes isopotentielles. Les distributions de N2A different pour les modalit~s auditive et visuelle, tandis que celles de P3A sont non sp6cifique de la modalit~ de stimulation. Ces donn6es topographiques sont compar6es ~ celles obtenues,
SCALP TOPOGRAPHY IN DISCRIMINATION TASKS l o r s d ' 6 t u d e s a n t 6 r i e u r e s , s u r les p o t e n t i e l s a u x s t i m u l i s m a n q u a n t s ( M S P , S i m s o n e t al. 1 9 7 6 ) ~ l ' a i d e d e n a p p e s r e p r 6 s e n t a n t les c o n tributions de sources c6r6brales unilat6rales. L e s d i s t r i b u t i o n s d e N 2 A e t N M S P se r 6 f ~ r e n t une activit6 corticale ~ l'int6rieur des r6gions v i s u e l l e s e t a u d i t i v e s s e c o n d a i r e s t a n d i s q u e les LPC (PMSP ou P3A) paraissent d6river princ i p a l e m e n t d u c o r t e x d ' a s s o c i a t i o n p a r i 6 t a l inf6rieur.
References Arezzo, J., Pickoff, A. and Vaughan, H.G., Jr. The sources and intracerebral distribution of auditory evoked potentials in the alert rhesus monkey. Brain Res., 1975, 90: 57--73. Bostock, H. and Jarvis, M.J. Changes in the form of the cerebral evoked response related to the speed of simple reaction time. Electroenceph. clin. Neurophysiol., 1970, 29: 137--145. Courchesne, E., Hillyard, S.A. and Galambos, R. Stimulus novelty, task relevance and the visual evoked potential in man. Electroenceph. clin. Neurophysiol., 1975, 39: 131--143. Ford, J.M., Roth, W.T., Dirks, S.J. and Kopell, B.S. Evoked potential correlates of signal recognition between and within modalities. Science, 1973, 181: 465--466. Goldring, S. and Ratcheson, R. Human motor cortex: sensory input data from single neuron recordings. Science, 1972, 175: 1493--1495. Goldring, S., Aras, E. and Weber, P.C. Comparative study of sensory input to motor cortex in animals and man. Electroenceph. clin. Neurophysiol., 1970, 29: 537--550. Hatter, M.R. and Salmon, L.E. Intra-modality selective attention and evoked cortical potentials to randomly presented patterns. Electroenceph. clin. Neurophysiol., 1972, 32: 605--613. Hillyard, S.A., Courchesne, E., Krausz, H.I. and Picton, T.W. Scalp topography of the 'P3' wave in different auditory decision tasks. In W.C. McCallum and J.R. Knott (Eds.), Event related slow potentials of the brain. John Wright and Sons, Bristol, in press. Karlin, L., Martz, M.J. and Mordkoff, A.M. Motor
535 performance and sensory evoked potentials. Electroenceph, clin. Neurophysiology, 1970, 28: 307--313. Picton, T.W., Hillyard, S.A. and Galambos, R. [Cortical evoked responses to omitted stimuli. In M.N. Livanov (Ed.), Major problems of brain electrophysiology.] (in Russian). Academy of Sciences, U.S.S.R., 1974: 302--311. Ritter, W. and Vaughan, H.G., Jr. Averaged evoked responses in vigilance and discrimination: A reassessment. Science, 1969, 164: 326--328. Ritter, W., Simson, R. and Vaughan, H.G., Jr. Association cortex potentials and reaction time ih auditory discrimination. Electroenceph. clin. Neurophysiol., 1972, 33: 547--555. Ritter, W., Vaughan, H.G., Jr. and Costa, L.D. Orienting and habituation to auditory stimuli: A study of short term changes in average evoked responses. Electroenceph. clin. Neurophysiol., 1968, 25: 550--556. Simson, R., Vaughan, H.G., Jr. and Ritter, W. The scalp topography of potentials associated with missing visual or auditory stimuli. Electroenceph. clin. Neurophysiol., 1976, 40: 33--42. Squires, N.K., Squires, K.C. and Hillyard, S.A. Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroenceph, clin. Neurophysiol., 1975, 38: 387--401. Sutton, S., Braren, M., Zubin, J. and John, E.R. Evoked-potential correlates of stimulus uncertainty. Science, 1965, 150: 1187--1188. Vaughan, H.G., Jr. The relationship of brain activity to scalp recordings of event-related potentials. In E. Donchin and D.B. Lindsley (Eds.), Averaged evoked potentials: methods, results, evaluations. National Aeronautics and Space Administration (NASA SP-191), Washington, D.C., 1969: 45--94. Vaughan, H.G., Jr. and Costa, L.D. Analysis of electroencephalographic correlates of human sensorimotor processes. Electroenceph. clin. Neurophysiol., 1968, 24: 288. Vaughan, H.G., Jr. and Ritter, W. The sources of auditory evoked responses recorded from the human scalp. Eleetroenceph. clin. Neurophysiol., 1970, 28: 360--367. Vaughan, H.G., Jr. Costa, L.D., Gilden, L. and Schimmel, H. Identification of sensory and motor components of cerebral activity in simple reaction-time tasks. Proc. 73 Conf. Amer. Psychol. Ass., 1965, 1: 179--180.