The Journal of Neurophysiology Vol. 80 No. 6 December 1998, pp. 2900-2910
Copyright ©1998 by the American Physiological Society

Lateralized Human Cortical Activity for Shifting Visuospatial Attention and Initiating Saccades

Bernd Wauschkuhn, Rolf Verleger, Edmund Wascher, Wolfgang Klostermann, Marcel Burk, Wolfgang Heide, and Detlef Kömpf

Medical University of Lübeck, Department of Neurology, 23538 Lübeck, Germany

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Wauschkuhn, Bernd, Rolf Verleger, Edmund Wascher, Wolfgang Klostermann, Marcel Burk, Wolfgang Heide, and Detlef Kömpf. Lateralized human cortical activity for shifting visuospatial attention and initiating saccades. J. Neurophysiol. 80: 2900-2910, 1998. The relation between shifts of visual attention and saccade preparation was investigated by studying their electrophysiological correlates in human scalp-recorded electroencephalogram (EEG). Participants had to make saccades either to a saliently colored or to a gray circle, simultaneously presented in opposite visual hemifields, under different task instructions. EEG was measured within the short interval between stimulus onset and saccade, focusing on lateralized activity, contralateral either to the side of the relevant stimulus or to the direction of the saccade. Three components of lateralization were found: 1) activity contralateral to the relevant stimulus irrespective of saccade direction, peaking 250 ms after stimulus onset, largest above lateral parietal sites, 2) activity contralateral to the relevant stimulus if the stimulus was also the target of the saccade, largest 330-480 ms after stimulus onset, widespread over the scalp but with a focus again above lateral parietal sites, and 3) activity contralateral to saccade direction, beginning about 100 ms before the saccade, largest above mesial parietal sites, with some task-dependent fronto-central contribution. Because of their sensitivity to task variables, component 1 is interpreted as the shifting of attention to the relevant stimulus, component 2 is interpreted as reflecting the enhancement of the attentional shift if the relevant stimulus is also the saccade target, and component 3 is interpreted as the triggering signal for saccade execution. Thus human neurophysiological data provided evidence both for independent and interdependent processes of saccade preparation and shifts of visual attention.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Shifts of visual attention are accomplished in everyday life by movements of the whole body or of the head or the eyes only. The more minute movements are also the faster ones, but attention may be shifted even more rapidly in covert form, without any overtly visible movement. The relation of these "covert" shifts of attention (Posner 1978) to saccadic eye movements, as the fastest means of "overt" shifts, was the subject of a long and ongoing debate (e.g., Fischer and Weber 1993; Posner and Petersen 1990; Rizzolatti et al. 1987). Recent studies of reaction time and discrimination performance in healthy human subjects provided evidence for a close coupling between saccade target selection and the focusing of attention (Chelazzi et al. 1995; Deubel and Schneider 1996; Hoffman and Subramaniam 1995), and this coupling was also found in electrophysiological recordings from monkeys' inferior temporal cortex (Chelazzi et al. 1993) and superior colliculus (Kustov and Robinson 1996). However, there is also evidence for separation of both processes, both in human behavioral experiments (Stelmach et al. 1997; Zelinsky and Sheinberg 1997) and in recordings from monkeys' lateral intraparietal area (Colby et al. 1996). Thus it would be of considerable interest to study if and how human neurophysiological activity while shifting attention is moderated by performing saccades.

For this purpose, three questions are relevant. First, which cerebral areas are responsible for saccades and for shifts of attention? Second, what is the time course of their activation? Third, how do these activations interact? With regard to the first question, although saccades are finally generated in the brain stem reticular formation [not accessible to scalp electroencephalogram (EEG) recording], this is done under cortical control. Evidence from patients with focal brain lesions (Heide et al. 1995; Pierrot-Deseilligny et al. 1995), experiments in monkeys (Andersen 1989; Goldberg and Segraves 1989), regional cerebral blood flow studies (e.g., Goebel et al. 1998; Paus 1996; Sweeney et al. 1996), and cortical stimulation studies in humans (e.g., Godoy et al. 1990; Lim et al. 1994) suggest that three areas in each cortical hemisphere interact in triggering saccades, mainly toward the contralateral side: the frontal (FEF), the supplementary (SEF), and the parietal eye field (PEF). Of these saccade areas, the FEF and the PEF are situated near or within areas that are also responsible for covert shifts of visual attention. Two centers were described to serve this latter purpose within either hemisphere (Posner and Dehaene 1994), mainly being responsible for shifts toward the contralateral side: a frontal and a posterior-parietal center. Relevant evidence was derived from patients with focal brain lesions (Bisiach 1993; Posner et al. 1984; Rafal et al. 1996), experiments in monkeys (Colby et al. 1996), and regional cerebral blood flow studies (Corbetta et al. 1993; Nobre et al. 1997).

Evidence from patients and from blood flow studies cannot, however, provide much evidence with regard to the second and third questions, the time course and interplay of activation of these areas, and thus of the processes of saccade preparation and attention allocation. In principle, these questions can be studied by electrophysiological measurement of human scalp-recorded cortical activity because of its excellent temporal resolution (Rugg and Coles 1995). Indeed, progress was made with regard to attention allocation. Most relevant to the present purpose, task-relevant lateral stimuli evoke a contralateral negativity, maximum at the occipito-temporal junction, peaking ~250 ms after stimulus onset ("L-250", L for lateralization), perhaps generated in area V4 (Luck et al. 1997) and interpreted as related to the attentional shift induced by the eliciting stimulus (Eimer 1996; Girelli and Luck 1997; Luck and Hillyard 1994a,b; Van der Lubbe and Woestenburg 1997; Wascher and Wauschkuhn 1996) (L-250 was called "N2pc" in several of those studies). Somewhat in contrast, studies of saccade preparation have so far not yielded unequivocally converging evidence for contralaterally enhanced activity of PEF, FEF, and SEF. Although some contralateral enhancement was found (Everling et al. 1998; Evdokimidis et al. 1992; Klostermann et al. 1994; Moster and Goldberg 1990; Thickbroom and Mastaglia 1985a), its topography and timing widely differed between studies and was not found at all in other studies (Evdokimidis et al. 1996; Everling et al. 1997). We reasoned that some part of this variability between studies might be due to the approach of studying saccades in isolation, without requiring the subjects to distinguish between relevant and irrelevant stimulation. Thereby, saccades might have been deprived of their important function of assisting to shift visual attention.

Therefore, in this study, the time course of attention- and saccade-related electrophysiological activity was measured in the short interval between stimulus onset and saccade onset, in fast sequences of trials. This enabled us to study the interplay between attention- and saccade-related processes. On the other hand, besides some technical difficulties associated with the EEG artifact caused by saccades (see RECORDING AND DATA PROCESSING), this approach posed the problem of how to disentangle these two activities and to distinguish them from pure perceptual effects. Several strategies were used to tackle this problem. First, two symmetrical stimuli were presented in each trial, one gray, the other colored. Depending on the task, saccades had to be made to one of these two stimuli. This way, lateralizations caused by stimulus relevance could be separated from lateralizations caused by purely perceptual reasons. Second, although the relevant stimulus was also the saccade target in the easy tasks, this was not necessarily the case in the difficult task. Thus, by comparing between tasks, allocation of attention could be separated from saccade preparation. Third, analysis was restricted to the difference in activation between recording sites located contralaterally and ipsilaterally to saccade direction and/or to the relevant stimuli. Because cortical areas are mainly responsible for attentional and saccadic shifts toward their contralateral side, the contralateral surplus activation (averaged across left and right shifts) is a specific expression of activity toward the contralateral side. The large unspecific activity is filtered out by this contralateral-ipsilateral subtraction (see Fig. 3), as are constant hemispheric asymmetries (by averaging across left and right shifts). We used the term "event-related lateralization" (ERL) for these difference potentials (Wascher and Wauschkuhn 1996; Wauschkuhn et al. 1997). Fourth, EEG was averaged and measured both time locked to stimulus onset to obtain primarily attention-related activity and time locked to saccade onset to obtain primarily saccade-related activity.

View larger version (39K): [in this window] [in a new window]   FIG. 3. Example for event-related potential (ERPs) and event-related lateralizations (ERLs). These are grand averages (i.e., averages across all subjects) of the recordings made in the easy, salient-target task. Stimulus onset is at 0 ms. Left column: ERPs recorded from symmetrical electrode pairs, with potentials from left and right sites separately averaged according to whether they were contralateral or ipsilateral to the direction of the saccade. For example, for FC3/FC4, FC3 is contralateral for saccades to the right and FC4 is contralateral for saccades to the left; both are averaged separately and then averaged together to form the contralateral FC3/4 potential. Likewise, FC3 is ipsilateral for saccades to the left, and FC4 is ipsilateral for saccades to the right. Small differences between contralateral and ipsilateral curves are visible. Right column: these differences are plotted in the right column, as ERLs. Negativity of this contralateral-ipsilateral difference is plotted upward. The peak latencies of L-250 and of L-400 are marked by vertical lines. There is no corresponding peak in the ERP graphs. Note different scale of the y-axis. Horizontal EOG (hEOG) is treated analogously to the EEG. In the ERP column, hEOG is plotted with a small scale to illustrate the mean time course of the whole saccade. In the ERL column, hEOG is plotted with the same scale as the EEG to illustrate its fine-grained time course.

	METHODS

Abstract Introduction Methods Results Discussion References

Subjects

Twelve subjects, four men and eight women, mean age 26 yr (range 23-29 yr), one left-handed, participated in the experiment. They had normal or corrected-to-normal vision and no history of neurological disorders.

Stimuli and procedure

Subjects were seated in a comfortable armchair in a soundproof, electrically shielded chamber and viewed the 14-in. Multisync monitor from a distance of ~120 cm.

A white fixation cross (0.45° wide and 0.35° high) was displayed continuously in the center of the screen. In each trial, two filled circles (diameter 0.65°) were presented for 1,600 ms at symmetrical locations (5.5° left and right from fixation). One circle was gray; the other was red, green, or blue, in random order across trials. The presentation sides of the colored and the gray circles varied randomly; 900 ms after the offset of the two circles, the next trial started. (Three consecutive trials are schematically represented in Fig. 1, left). Stimuli were presented in this bilateral, symmetrical way to avoid early exogenous effects on the ERLs (Valle-Inclán 1996) and to have the same physical stimuli in each task. Luminances of the gray, blue, green, and red circles were 2.2, 3.3, 5.2, and 6.0 cd/m², and the background had 0.2 cd/m².

View larger version (24K): [in this window] [in a new window]   FIG. 1. Left: schematic example for stimuli presented in 3 consecutive trials. The colors used are symbolized by different dash patterns. In each trial, presentation time was 1,600 ms; intertrial interval was 900 ms. Right: table of correct responses to the 3 stimulus arrays in the different tasks.

There were two types of choice-response tasks, differing by the feature relevant for responding (Fig. 1, right): 1) constant target; saccades had to be made to the location of the relevant circle. Relevant were in different blocks either the saliently colored circle (red, blue, and green) or the gray circle. 2) color-defined ("Simon task") (Simon 1990); irrespective of their location, a red circle required a saccade to the left, a blue circle a saccade to the right, and a green circle required continuing fixation (no-go). Thus saccade direction was either compatible or incompatible with the location of the salient task-relevant stimulus: compatible, for example, when a red circle was on the left; incompatible, for example, when a red circle was on the right. Tasks 1a and 1b consisted of 192 trials each; task 2 consisted of 576 trials (with a break in-between), including 192 compatible, 192 incompatible, and 192 no-go trials. Twenty practice trials were presented before each task. Tasks 1a, 1b, and 2 were arranged in pseudo-random order within one session, different for each subject.

Recording and data processing

EEG was recorded from Fz, FC3, FCz, FC4, T7, C3, Cz, C4, T8, CPz, P7, P3, Pz, P4, P8, PO7, PO8, O1, and O2 (Fig. 2) (Pivik et al. 1993) with Ag/AgCl electrodes (Picker-Schwarzer) with electrodes affixed at the mastoids as reference (linked by a 5-k resistor). A ground electrode was affixed at the forehead. Vertical electrooculogram (EOG) was recorded from above versus below the left eye; horizontal EOG was recorded from the outer canthi of both eyes. Resistance was <5 k for all electrodes.

View larger version (18K): [in this window] [in a new window]   FIG. 2. Recording sites used in this study. The focus will be on analyzing the difference between symmetrical electrode pairs, contralateral minus ipsilateral to the direction of the saccade. For example, for the symmetrical pair FC3/FC4, the difference FC3  FC4 is formed for rightward saccades and FC4-FC3 for leftward saccades; the average of these two differences is the mean contralateral-minus-ipsilateral lateralization for FC3/4.

EEG signals were amplified between 0.032 and 25 Hz, EOG signals between 0.032 and 70 Hz by a Nihon-Kohden 4421. Triggered by the control computer, the EEG amplifier was reset to baseline after each trial to avoid contamination of the EEG of the next trial with residual EOG artifacts by the saccade. This was an important methodological detail (first used in Wauschkuhn et al. 1997) allowing to present the trials in rapid succession, adequate to the fast, flexible nature of the saccadic response. (Long intertrial intervals of 10 s were used in previous research for this purely technical reason) (e.g., Klostermann et al. 1994).

Triggered by the control computer, the data (EEG and EOG) were stored on another PC and were digitized with 200 Hz from 100 ms before to 1,800 ms after stimulus onset. Off-line, trials were excluded when there were zero lines, out-of-scale values, slow drifts >80 µV, or fast shifts >100 µV/500 ms. The transmission of vertical and horizontal EOG into the EEG, as ocular artifacts, was estimated separately in areas of maximum EOG variance and was subtracted out from the EEG data. Reliability and validity of such a procedure to remove the ocular artifact from the EEG were repeatedly demonstrated (e.g., Anderer et al. 1992; Kenemans et al. 1991; Verleger et al. 1982). Further, of the several measurements taken, only one component, L-400 in the easy task, was measured on the average after saccade onset and was therefore at risk of being affected by incompletely removed saccade artifacts.

Data analysis

RESPONSE PARAMETERS. Response time was defined in each trial as the moment when the amplitude of horizontal EOG exceeded one-half of the mean peak amplitude of all correctly performed saccades (~2.75°). Mean latencies of the correct responses and frequencies of error trials were evaluated statistically by analysis of variance (ANOVA) for repeated measurements with the factors task (easy vs. difficult, i.e., task 1 vs. task 2) and target saliency (colored circle vs. gray circle, i.e., task 1a and compatible trials of task 2 vs. task 1b and incompatible trials of task 2). Because error frequencies could additionally be measured in the no-go trials of the color-defined task (task 2), further ANOVAs compared these error frequencies with the compatible and incompatible trials of this task.

EEG PARAMETERS. Trials with incorrect responses (wrong direction or saccades in no-go trials) were rejected from further analyses. Only difference potentials between hemispheres will be reported. These ERLs were calculated as the difference potential contralateral-ipsilateral with respect to saccade direction (in no-go trials: with respect to position of the salient circle) separately for each symmetrical electrode pair (FC3/FC4, T7/T8, C3/C4, P7/P8, P3/P4, PO7/PO8, and O1/O2) and separately for each condition: forming the difference left minus right recording in the average over right-response trials, right minus left recording in the average over left-response trials, and averaging these two differences to yield the general difference contralateral-minus-ipsilateral.

Averaging over trials was done in two ways: stimulus-locked, i.e., all trials were temporally aligned to target onset, and response-locked, i.e., all trials were temporally aligned to saccade onset. Response locking was, of course, not possible for no-go trials. For both the stimulus-locked and the response-locked ERLs, the 100 ms before stimulus presentation was taken as baseline.

ERL components were determined from inspection of the grand means and were then measured in each subject's ERL curves. Two components were measured in the stimulus-locked averages: L-250 and L-400, with L standing for lateralization and 250 and 400 denoting the approximate latency in milliseconds relative to stimulus onset. L-250 appeared as a distinct peak; therefore the amplitude and latency of this peak was measured in the time interval 180-310 ms after stimulus presentation. L-400 did not have a clear peak in each subject's ERL and was therefore measured as mean amplitude 340-480 ms after stimulus presentation because this interval covered activity in all conditions (Fig. 4). One component was measured in the response-locked averages as mean amplitude in the time interval 100-50 ms before the saccade and was labeled SORL (saccade-onset-related lateralization).

View larger version (39K): [in this window] [in a new window]   FIG. 4. ERLs in the easy, constant-target task (left column, for salient targets same as right column of Fig. 3) and in the difficult, color-defined task (middle and right column). Grand averages, across all subjects. Negativity of the contralateral-ipsilateral difference is plotted upward. Stimulus onset is at 0 ms. The time window for measurement of the L-250 and of the L-400 is marked by the dashed lines (L-250 between left and middle dashed line, L-400 between middle and right dashed line).

Depending on saccadic response times, the measurement epoch of SORL inevitably overlapped with L-250 or with L-400. SORL activity would probably not, at least not only, be due to overlap from these stimulus-locked components if it had larger amplitudes in the response-locked than in the stimulus-locked averages and if its topography would remain constant across tasks in spite of different overlapping components. To detail, mean saccadic response times were 334 ms in the easy and 467 ms in the difficult task (Table 1). Thus, on the average, SORL covered the epoch of 234-284 ms after stimulus onset in the easy task and of 367-417 ms in the difficult task. Therefore SORL overlapped with L-250 in the easy task and with L-400 in the difficult task. If the topography of SORL would be constantly different from these two components across tasks, it would probably not or not only be caused by these two components. The same considerations apply for L-250 and L-400. If their amplitudes would be larger in stimulus-locked than in response-locked averages and if their topography would be constant across tasks, they would not only be caused by the overlapping saccade-locked SORL.

The same ANOVA designs as for response times and errors were applied to the ERL components, measured at that electrode pair where they were maximum: L-250 at P7/P8, L-400 at PO7/PO8, and SORL at P3/P4. Further, the topographic distribution of amplitudes was tested in an ANOVA on vector-sum normalized amplitudes (Naumann et al. 1992), measured from all seven electrode pairs. If not otherwise noted (effects of topography), degrees of freedom were 1/11. Because topography was a repeated-measurement factor with more than two levels, its degrees of freedom were corrected by the Huynh-Feldt  coefficient.

	RESULTS

Abstract Introduction Methods Results Discussion References

Behavioral results

Response times and percentages of correct trials are compiled in Table 1. As could be expected, latencies of saccades were shorter in the easy constant-target than in the difficult color-defined task (F = 190.72, P < 0.001) and shorter with saccades to the salient than to the gray circle (F = 41.34, P < 0.001). This latter effect was more marked in the easy task (task × target saliency: F = 5.19, P = 0.04) but was nevertheless still reliable in the difficult task (F = 7.54, P < 0.05).

More errors were made in the difficult than in the easy task (F = 21.82, P < 0.001) and with saccades to the gray than to the salient circle (F = 29.21, P < 0.001) equally for both tasks (task × target saliency: F = 0.39, n.s.). The error rate for no-go trials in the color-defined task was larger than both for compatible (F = 27.59, P < 0.001) and incompatible trials (F = 13.53, P = 0.004).

EEG parameters

To illustrate how lateralizations (ERLs) were computed, Fig. 3 displays the grand averages of the event-related potentials (ERPs) time locked to stimulus onset for salient targets in the easy constant-target task. The left and right recording sites are rearranged to be contralateral and ipsilateral to saccade direction. The difference potentials (contralateral minus ipsilateral) are plotted in the right column. In these ERLs, negativity increased contralaterally to saccade direction at ~250 ms after stimulus onset (L-250), most pronounced at inferior-lateral parietal sites P7/P8 and PO7/PO8. After the L-250 returned to baseline, negativity again increased contralaterally to saccade direction (L-400), again pronounced at inferior-lateral parietal sites (P7/P8 and PO7/PO8) but also well visible at superior-mesial parietal (P3/P4) and frontocentral sites (FC3/FC4). Inspection of the ERPs shows that L-250 was barely visible as a small difference on the descending flank of the N200 peak and that L-400 occurred briefly before the peak of the P3 complex. Evidently, these contralateral-ipsilateral differences can be more simply determined and analyzed when the large common activity is subtracted out. The ERLs of the constant-target task (for salient targets same as in Fig. 3) and of the difficult color-defined task are displayed in Fig. 4. Results are summarized in Table 2.

L-250. There was one obvious difference between tasks. In the easy constant-target task, the L-250 was negative contralateral to the actual direction of the saccade, both with saccades to the salient and to the gray circle. In contrast, in the color-defined task L-250 was always more negative contralateral to the relevant colored circle and thus had an inverted polarity with saccades to the gray circle. Therefore, for statistical comparison of L-250 between conditions (but not in the figures), polarity of L-250 was inverted in this condition.

Peak latencies of L-250 were prolonged in trials with saccades to the less salient, gray circle (F = 15.65, P = 0.002). This effect was more marked in the easy task (task × target saliency, F = 8.08, P = 0.02, see Table 2) but was still reliable in the difficult task (F = 5.25, P < 0.05). On average, these delays were very similar to the delays of response times but did not correlate across subjects with those delays. L-250 was also delayed in the no-go trials of the color-defined task compared with compatible trials (F = 15.13, P < 0.01).

L-250 amplitude measured at P7/P8 was not affected by task or target saliency (except, of course, for the drastic inversion when the gray circle was target in the color-defined task, mentioned previously). However, topography differed between tasks [F(6,66) = 2.22,  = 1.0, P = 0.05] because amplitudes were relatively larger at P3/P4 in the easy than in the difficult task (F = 7.56, P = 0.02). (The apparent difference at FC3/4, cf. Fig. 4, was not significant). This effect probably reflects the overlap of the saccade-onset-related lateralization in the easy task, as discussed in METHODS.

L-400 was a negativity contralateral to the direction of the saccade in the easy task, both with saccades to the saliently colored circle and to the gray circle. Likewise, it occurred in compatible trials of the difficult task but was virtually absent with saccades to the gray circle in this task (task × target saliency: F = 6.02, P = 0.03; target saliency in the difficult task: F = 7.37, P = 0.02). It might be argued that this effect is due to the preceding L-250 running to the opposite direction and thus lowering L400's base level. However, L-400 was also absent for no-go trials, where this L-250 effect did not occur (F = 14.04, P = 0.003, compared with compatible trials) as small as with incompatible trials (F = 1.25, n.s.). In contrast, L-400 did not differ between saccades to the salient and gray circle in the easy task (F = 0.75, n.s.). Although L-400's topography appears to differ in Fig. 4 between the two tasks, this effect did not become significant in the ANOVA on normalized amplitudes [F(6,66) = 1.55, n.s.]. (Only saccades to the salient circle entered this comparison because of the lacking L-400 with gray circles in the difficult task).

SORL. ERLs time locked to saccade onset are displayed in Fig. 5. Negativity increased contralaterally to saccade direction, beginning ~100 ms before saccade onset, most pronounced at the parietal site P3/P4 in the difficult task. Distinct components are also well visible around saccade onset at T7/T8 and at FC3/FC4 but will not be further dealt with, being probably the lateralized part of the myogenic "spike potential" (Thickbroom and Mastaglia 1985b). To avoid confounding by these potentials, the time window for measuring SORL ended 50 ms before saccade initiation. At P3/P4, SORL was neither affected by task (F = 0.01, n.s.) nor by target saliency (F = 0.59, n.s.) nor by their interaction (F = 1.75, n.s.) However, SORL's topography differed according to task and target saliency [F(6,66) = 3.69,  = 0.50, P = 0.02]. This triple interaction reflected two results. First, in the color-defined task topography differed between saccades to the colored and to the gray circle (target saliency × topography in the difficult task [F(6,66) = 7.07,  = 0.53, P <0.001]; SORL was relatively larger before saccades to the gray than to the colored circle at FC3/FC4 (F = 14.90, P = 0.003). (In contrast, the apparently large difference at PO7/PO8 was not significant, neither in normalized nor in raw data). Second, topography tended to differ between the two tasks for saccades to the salient circle (task × topography for salient-circle saccades: F(6,66) = 2.12,  = 1.0, P = 0.06); SORL was relatively larger at T7/T8 in the simple task than in the difficult task (F = 20.30, P = 0.001). This latter effect probably reflects overlap with L-250 in the easy task; L-250 is focused at lateral inferior parietal sites, in particular P7/8, but also extending to T7/8 (Fig. 4); therefore its overlap can be expected to make the topography of SORL more inferior-lateral.

View larger version (42K): [in this window] [in a new window]   FIG. 5. Response-locked ERLs. Grand averages, across all subjects. Different from Fig. 3 and Fig. 4, 0 ms is time of saccade onset. Negativity of the contralateral-ipsilateral difference is plotted upward. The time window for measurement of the SORL is marked by the 2 dashed lines.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Three components of lateralization were found, with different topographical distributions and latencies: 1) the L-250, peaking in different conditions between 220 and 270 ms after stimulus onset (means across subjects), maximum over inferior-lateral parietal scalp sites (PO7/8 and P7/8), 2) the L-400, measured from 340 to 480 ms after stimulus onset, likewise maximum over inferior-lateral parietal sites (PO7/8 and P7/8) but spreading to superior-mesial parietal (P3/4) and even fronto-central sites (FC3/4 and C3/4), and 3) a SORL, measured 100-50 ms before the saccade, maximum at superior-mesial parietal sites (P3/4) but likewise displaying a widespread topography, including inferior-lateral parietal sites (P7/8 and PO7/8) and fronto-central sites (FC3/4 and C3/4).

L-250

L-250 was not a "mechanical," exogenously produced consequence of asymmetrical stimulus configuration because, in the easy task, L-250 was negative contralateral either to the colored or to the gray circle, whichever was defined as target in that block. Neither was L-250 related to intended saccade direction because, in the difficult task, L-250 was always negative contralateral to the colored circle, even when the saccade had to be performed to the gray circle. Because it was always the colored circle that was relevant in this task, defining saccade direction, L-250 evidently reflected which of the two circles was task-relevant. Therefore, in line with the literature quoted in the INTRODUCTION, L-250 can be interpreted as related to the attentional shift induced by a task-relevant stimulus.

This indicator of attentional shift proved to be largely independent of saccade preparation. First, there was the previously mentioned polarity reversal relative to saccade direction in the color-defined task. Second, its amplitude was neither larger nor smaller in no-go trials than in Go trials. Third, L-250's peak preceded saccade onset by variable intervals, on the average by ~90 ms in the easy task and by ~230 ms in the difficult task. The only argument in favor of a closer relation of L-250 to saccades is that L-250 was delayed by a similar amount as the saccades by decreased (gray) target saliency, both in the easy task (45 ms) and in the difficult task (13 ms). However, these delays did not correlate between subjects. Thus, in spite of this intriguing numerical similarity, we conclude that L-250 is not directly related to programming and execution of the saccade. Thus L-250 appears to be a pure indicator of covert attentional shifts, and the equal delay of L-250 and of saccades indicates that lower saliency of the relevant stimulus affected covert and overt attentional shifts to the same degree.

The delay of L-250 in no-go trials was not expected. Participants perhaps searched for red or blue circles in a first step and, if this search failed, searched for the green circle.

The present findings about L-250 might be related to findings from experiments in monkeys. Neurons in area LIP (the parietal eye field of monkeys) show not only visual and saccade-related activity, but their activity is enhanced in response to a behaviorally relevant stimulus of a certain spatial location, even if the monkey is still waiting for the stimulus to appear (Colby et al. 1996). Further, during the preparation of antisaccades (saccades away from the eliciting stimulus), LIP neurons are tuned more to the direction of the task-relevant target than to the direction of the saccade (Gottlieb and Goldberg 1997); this finding directly corresponds to the present polarity reversal in the color-defined task. Similar to L-250, the LIP activity was interpreted as cueing of the attentional vector to a behaviorally relevant spatial location (Goldberg et al. 1990). On the other hand, being maximum at posterior-lateral sites (P7/8), L-250 seems to be generated in cortical areas that are located more posterior and more lateral to the intraparietal sulcus (the assumed location of the LIP homolog in humans), possibly in V4 (Luck et al. 1997), and therefore a relation to LIP activity might rather be seen in the more superior-mesially located SORL.

L-400

L-250 was followed by a second contralateral negativity, termed here L-400, with a topographical maximum at inferior-lateral parietal sites (PO7/8 and P7/8) similar to L-250 but displaying a more widespread topography, including superior-mesial parietal (P3/4) and even fronto-central sites (FC3/4 and C3/4). L-400 was absent for saccades in incompatible trials in the difficult task as well as for no-go trials, that is, L-400 was evoked by stimuli only if they were both task-relevant and targets of the required saccade, not if they were saccade targets only (gray circle in incompatible trials of the difficult task) or task-relevant only (no-go stimuli).

Obviously, L-400 cannot be interpreted as a direct correlate of the saccade. First, it appears with similar latency in the easy task and in the difficult task while saccade onsets were delayed by ~130 ms in the difficult task. Second, L-400 is less pronounced in the response-locked than in the stimulus-locked data. (Although still being visible, in the easy task, with its mean saccade onset of 334 ms, L-400 should occur briefly after saccade onset, and indeed there is a widespread peak visible at 60-100 ms in Fig. 5. In the difficult task, with its mean saccade onset of 467 ms, L-400 should occur from ~120 ms before saccade onset to about saccade onset, and indeed such a component is visible in Fig. 5, appearing as a continuation of the SORL). This weak relation to saccade onset renders it also unlikely that L-400 is affected by artifacts caused by incorrect removal of saccadic voltages from the EEG by our off-line regression procedure (in the easy task, if at all, because in the difficult task L-400 occurred on the average before saccade onset). Therefore, although affected by the requirement to make a saccade, L-400 is evidently evoked by the stimulus, as is the P3 component of the ERP with which L-400 overlaps (Fig. 3).

The modulation by the requirement to make a saccade is a most interesting finding. If confirmed by further studies, L-400 may be a tool for studying the interaction between ocular and attentional movements; this issue was investigated so far by complex dual-task behavioral studies only (e.g., Stelmach et al. 1997). L-400 might reflect some second step of allocating attention after the step indicated by L-250. Such an additional step may make sense because the relevant stimulus has not only to be attended but is also the target of the new fixation.

Any further interpretation of the function of L-400 must remain speculative at this moment. We know of only one similar finding in the literature. Yamaguchi et al. (1994 1995) reported a second lateralization in their study on shifts of visual attention, beginning ~450 ms after onset of the lateral cue at posterior sites, spreading toward anterior sites. They tentatively related this lateralization to the controlled shift of covert attention, demonstrated by Müller and Rabbitt (1989) as the slower way of shifting attention, compared with the fast reflexive shift evoked by the abrupt onset of lateral stimuli. In our data, however, this interpretation of L-400 would imply that the preceding L-250 indicates the preceding reflexive, strongly automatic stage of attention, and this is not the case (as noted by one referee) because L-250 was not simply evoked by any abrupt onset of lateral stimuli but rather by the task-relevant member of a bilaterally presented pair. Alternatively, the overlap of L-400 with the P3 component of the ERP might be more than pure accidence. Like P3 is known to be enhanced in response to task-relevant targets (Verleger 1988) so is L-400 here evoked by those task-relevant stimuli that are also saccade targets. P3 was recently proposed to be related both to processing of stimulus meaning in the ventral stream and to processing of stimulus-response mapping in the dorsal stream (Verleger 1998). This might be consistent with the widespread topographical distribution of L-400. However, this interpretation of L-400 as asymmetrically enhanced P3, contralateral to the relevant stimulus, meets with the difficulty that P3 is a positive potential and L-400 indicates more contralateral negativity. Therefore all that can be said at this moment is that L-400 probably reflects some second step of allocating attention, after the step indicated by L-250, with this second step occurring only when the relevant stimulus is also target of the saccade.

SORL

SORL was measured 100-50 ms before the saccade when the single trials were aligned in time to saccade onset. It was maximum at superior-mesial parietal sites (P3/4) but displayed a widespread topography, including the inferior-lateral parietal sites (P7/8, PO7/8) and fronto-central sites (FC3/4 and C3/4).

Occurring so close before the saccade, SORL might reflect a parietal triggering signal for the saccade (to be executed by brain stem nuclei), possibly generated by the PEF. The PEF is known from clinical studies and from animal experiments as relevant to visually guided saccades (Pierrot-Deseilligny et al. 1995) and is probably located in humans' intraparietal sulcus (Heide et al. 1997; Müri et al. 1996), approximately underlying recording sites P3/P4. Thus SORL would be the analog to the contralateral activation of the hand-motor area before hand movements (e.g., Coles 1989). Alternatively SORL might reflect a presaccadic signal used to compute the remapping of receptive fields of parietal neurons (Heide and Kömpf 1997). These neurons were found in monkeys' area LIP (Duhamel et al. 1992), and the human intraparietal sulcus appears to have a corresponding function (Heide et al. 1995).

On the other hand, the specificity of the parietal SORL for saccades may be questioned. It may well be argued that the SORL at P3/4 is a more general "action" signal generated by the parietal cortex. Indeed, activity at P3/4, coincident with the early phase of lateralization over the hand-motor areas, occurs frequently in studies that investigate lateralization before hand movements (Wascher and Wauschkuhn 1996) and may be differentiated from the more inferior-laterally located, L-250-type lateralization.

SORL's topography differed between tasks and saccade types in two ways. The task difference for saccades to the salient circle was most probably caused simply by overlap with L-250 in the easy task and will not be further discussed. In addition, however, topography differed between compatible and incompatible trials in the difficult task because SORL was larger before saccades in incompatible trials at FC3/4. Because in incompatible trials, the saccade had to be made away from the relevant stimulus, similar to antisaccades, this difference fits the assumption that the frontal eye fields are involved in voluntary saccades, in line both with positron emission tomography data, which suggest that humans' FEF and SEF are more active during antisaccade than prosaccade tasks (O'Driscoll et al. 1995; Sweeney et al. 1996) and with recordings from monkeys' SEF (Schlag-Rey et al. 1997).

There appear to be three reasons an SORL-type lateralization was not demonstrated previously (except by Evdokimidis et al. 1992). First, as argued in the INTRODUCTION, the overlap of unspecific nonlateralized components was removed in the present data. Second, this study may have investigated saccades in an ecologically more valid manner than many previous studies where subjects' activity consisted in nothing more than making one saccade about every 20 s (e.g., Klostermann et al. 1994). Third, SORL may be obscured by other potentials (L-250 and L-400), which are probably not specific to saccades. For example, in our previous study (Wauschkuhn et al. 1997), a SORL might have been hidden in the "ETPL" (the L-250, possibly with L-400 contribution) after the imperative stimulus. In the current study, this overlap became more transparent, by having tasks that produced different saccadic response times and by forming both stimulus- and response-locked averages.

	CONCLUSIONS

This study could demonstrate for the first time the time course and topography of lateralized cortical activity for the control of visually guided saccades and of associated shifts of visuospatial attention. Three components of lateralized activity could be differentiated: 1) L-250, inferior-lateral parietal activity, reflecting identification of the relevant stimulus; 2) L-400, widespread activity, with a focus again at inferior-lateral parietal sites, possibly reflecting the enhancement of the attentional shift by the requirement to make a saccade; and 3) SORL, superior-mesial parietal activity 100-50 ms before saccade onset, possibly reflecting the triggering signal for the saccade, with the enhanced fronto-central activity before saccades away from the relevant stimulus reflecting increased voluntary control. The specificity of the SORL to saccades still has to be shown in further studies. L-250, as the earliest indicator of attention allocation, was independent of saccade preparation, whereas L-400 reflected some step of processing that integrated the stimulus features of being task-relevant and being the saccade target.

	ACKNOWLEDGEMENTS

  We thank the two referees for numerous constructive suggestions.

  This study was supported by Grant Ve 110/7-1 from the Deutsche Forschungsgemeinschaft to R. Verleger and W. Klostermann. Present address for E.Wascher: Institute of Clinical and Physiological Psychology, University of Tübingen, Tübingen, Germany.

	FOOTNOTES

  Address for reprint requests: R. Verleger, Medical University of Lübeck, Dept. of Neurology, Ratzeburger Allee 160, 23538 Lübeck, Germany.

  Received 28 January 1998; accepted in final form 9 September 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ANDERER, P., SEMLITSCH, H. V., SALETU, B., BARBANOJ, M. J. Artifact processing in topographic mapping of electroencephalographic activity in neuropsychopharmacology. Psychiatry Res. Neuroimaging 45: 79-93, 1992.
• ANDERSEN, R. A. Visual and eye movement functions of the posterior parietal cortex. Annu. Rev. Neurosci. 12: 377-403, 1989.[Medline]
• BISIACH, E. Mental representation in unilateral neglect and related disorders: the twentieth Bartlett memorial lecture. Q. J. Exp. Psychol. A 46: 435-461, 1993.[Medline]
• CHELAZZI, L., BISCALDI, M., CORBETTA, M., PERU, A., TASSINARI, G., BERLUCCHI, G. Oculomotor activity and visual spatial attention. Behav. Brain Res. 71: 81-88, 1995.[Medline]
• CHELAZZI, L., MILLER, E. K., DUNCAN, J., DESIMONE, R. A neural basis for visual search in inferior temporal cortex. Nature 363: 345-347, 1993.[Medline]
• COLBY, C. L., DUHAMEL, J. R., GOLDBERG, M. E. Visual, presaccadic, and cognitive activation of single neurons in monkey lateral intraparietal area. J. Neurophysiol. 76: 2841-2852, 1996.[Abstract/Free Full Text]
• COLES, M.G.H. Modern mind-brain reading: psychophysiology, physiology, and cognition. Psychophysiology 26: 251-269, 1989.[Medline]
• CORBETTA, M., MIEZIN, F. M., SHULMAN, G. L., PETERSEN, S. E. A PET study of visuospatial attention. J. Neurosci. 40: 1202-1226, 1993.
• DEUBEL, H., SCHNEIDER, W. X. Saccade target selection and object recognition: evidence for a common attentional mechanism. Vision Res. 36: 1827-1837, 1996.[Medline]
• DUHAMEL, J.-R., COLBY, C. L., GOLDBERG, M. E. The updating of the representation of visual space in parietal cortex by intended eye movements. Science 255: 90-92, 1992.[Abstract/Free Full Text]
• EIMER, M. The N2pc component as an indicator of attentional selectivity. Electroencephalogr. Clin. Neurophysiol. 99: 225-234, 1996.[Medline]
• EVDOKIMIDIS, I., LIAKOPOULOS, D., CONSTANTINIDIS, T. S., PAPAGEORGIOU, C. Cortical potentials with antisaccades. Electroencephalogr. Clin. Neurophysiol. 98: 377-384, 1996.[Medline]
• EVDOKIMIDIS, I., MERGNER, T., LÜCKING, C. H. Dependence of presaccadic cortical potentials on the type of saccadic eye movement. Electroencephalogr. Clin. Neurophysiol. 83: 179-191, 1992.[Medline]
• EVERLING, S., KRAPPMANN, P., FLOHR, H. Cortical potentials preceding pro- and antisaccades in man. Electroencephalogr. Clin. Neurophysiol. 102: 356-362, 1997.[Medline]
• EVERLING, S., SPANTEKOW, A., KRAPPMANN, P., FLOHR, H. Event-related potentials associated with correct and incorrect responses in a cued antisaccade task. Exp. Brain Res. 118: 27-34, 1998.[Medline]
• FISCHER, B., WEBER, H. Express saccades and visual attention. Behav. Brain Sci. 16: 553-610, 1993.
• GIRELLI, M., LUCK, S. J. Are the same attentional mechanisms used to detect visual search targets defined by color, orientation, and motion? J. Cognit. Neurosci. 9: 238-253, 1997.[Abstract]
• GODOY, J., LÜDERS, H., DINNER, D. S., MORRIS, H. H., WYLLIE, E. Versive eye movements elicited by cortical stimulation of the human brain. Neurology 40: 296-299, 1990.[Abstract/Free Full Text]
• GOEBEL, R., LINDEN, D.E.J., LANFERMANN, H., ZANELLA, F. E., SINGER, W. Functional imaging of mirror and inverse reading reveals separate coactivated networks for oculomotion and spatial transformations. NeuroReport 9: 713-719, 1998.[Medline]
• GOLDBERG, M. E., COLBY, C. L., DUHAMEL, J.-R. Representation of visuomotor space in the parietal lobe of the monkey. Cold Spring Harbor Symp. 55: 729-739, 1990.[Medline]
• GOLDBERG, M. E., SEGRAVES, M. A. The visual and frontal cortices. In: The Neurobiology of Saccadic Eye Movements, edited by R. H. Wurtz, and M. E. Goldberg Amsterdam: Elsevier, 1989, p. 283-313.
• GOTTLIEB, J., GOLDBERG, M. E. Encoding of stimulus and saccade direction in rhesus monkey lateral intraparietal area (LIP). Soc. Neurosci. Abstr. 23: 17, 1997.
• HEIDE, W., BINKOFSKI, F., POSSE, S., SEITZ, R. J., KÖMPF, D., FREUND, H.-J. Functional anatomy of memory-guided saccade sequences in human cerebral cortex. Soc. Neurosci. Abstr. 23: 2222, 1997.
• HEIDE, W., BLANKENBURG, M., ZIMMERMANN, E., KÖMPF, D. Cortical control of double-step saccades: implications for spatial orientation. Ann. Neurol. 38: 739-748, 1995.[Medline]
• HEIDE, W., KÖMPF, D. Specific parietal lobe contribution to spatial constancy across saccades. In: Parietal Lobe Contributions to Orientation in 3D Space, edited by P. Their, and H.-O. Karnath Heidelberg, Germany: Springer-Verlag, 1997, p. 149-172.
• HOFFMAN, J. E., SUBRAMANIAM, B. The role of visual attention in saccadic eye movements. Percept. Psychophys. 57: 787-795, 1995.[Medline]
• KENEMANS, J. L., MOLENAAR, P.C.M., VERBATEN, M. N., SLANGEN, J. L. Removal of the ocular artifact from the EEG: a comparison of time and frequency domain methods with simulated and real data. Psychophysiology 28: 114-121, 1991.[Medline]
• KLOSTERMANN, W., KÖMPF, D., HEIDE, W., VERLEGER, R., WAUSCHKUHN, B., SEYFERT, T. The presaccadic cortical negativity prior to self-paced saccades with and without visual guidance. Electroencephalogr. Clin. Neurophysiol. 91: 219-228, 1994.[Medline]
• KUSTOV, A. A., ROBINSON, D. L. Shared neural control of attentional shifts and eye movements. Nature 384: 74-77, 1996.[Medline]
• LIM, S. H., DINNER, D. S., PILLAY, P. K., LÜDERS, H., MORRIS, H. H., KLEM, G., WYLLIE, E., AWAD, I. A. Functional anatomy of the human supplementary sensorimotor area: results of extraoperative electrical stimulation. Electroencephalogr. Clin. Neurophysiol. 91: 179-193, 1994.[Medline]
• LUCK, S. J., GIRELLI, M., MCDERMOTT, M. T., FORD, M. A. Bridging the gap between monkey neurophysiology and human perception: An ambiguity resolution theory of visual selective attention. Cognit. Psychol. 33: 64-87, 1997.[Medline]
• LUCK, S. J., HILLYARD, S. A. Electrophysiological correlates of feature analysis during visual search. Psychophysiology 31: 291-308, 1994a.[Medline]
• LUCK, S. J., HILLYARD, S. A. Spatial filtering during visual search: Evidence from human electrophysiology. J. Exp. Psychol. Hum. Percept. Perform. 20: 1000-1014, 1994b.[Medline]
• MOSTER, M. L., GOLDBERG, G. Topography of scalp potentials preceding self-initiated saccades. Neurology 40: 644-648, 1990.[Abstract/Free Full Text]
• MÜLLER, H. J., RABBITT, P. M. A. Reflexive and voluntary orienting of visual attention: time course of activation and resistance to interruption. J. Exp. Psychol. Hum. Percept. Perform. 15: 315-330, 1989.[Medline]
• MÜRI, R. M., IBA-ZIZEN, M. T., DEROSIER, C., CABANIS, E. A., PIERROT-DESEILLIGNY, C. Location of the human posterior eye field with functional magnetic resonance imaging. J. Neurol. Neurosurg. Psychiatry 60: 445-448, 1996.[Abstract]
• NAUMANN, E., HUBER, C., MAIER, S., PLIHAL, W., WUSTMANS, A., DIEDRICH, O., BARTUSSEK, D. The scalp topography of P300 in the visual and auditory modalities: a comparison of three normalization methods and the control of statistical type II error. Electroencephalogr. Clin. Neurophysiol. 83: 254-264, 1992.[Medline]
• NOBRE, A. C., SEBESTYEN, G. N., GITELMAN, D. R., MESULAM, M. M., FRACKOWIAK, R.S.J., FRITH, C. D. Functional localization of the system for visuospatial attention using positron emission tomography. Brain 120: 515-533, 1997.[Abstract/Free Full Text]
• O'DRISCOLL, G. A., ALPERT, N. M., MATTHYSSE, S. W., LEVY, D. L., RAUCH, S. L., HOLZMAN, P. S. Functional neuroanatomy of antisaccade eye movements investigated with positron emission tomography. Proc. Natl. Acad. Sci. USA 92: 925-929, 1995.[Abstract/Free Full Text]
• PAUS, T. Location and function of the human frontal eye-field: a selective review. Neuropsychologia 34: 475-483, 1996.[Medline]
• PIERROT-DESILLIGNY, C., RIVAUD, S., GAYMARD, B., MÜRI, R., VERMERSCH, A. I. Cortical control of saccades. Ann. Neurol. 37: 557-567, 1995.[Medline]
• PIVIK, R. T., BROUGHTON, R. J., COPPOLA, R., DAVIDSON, R. J., FOX, N., NUWER, M. R. Guidelines for the recording and quantitative analysis of electroencephalographic activity in research contexts. Psychophysiology 30: 547-558, 1993.[Medline]
• POSNER, M. I. In: Chronometric Explorations of the Mind. Hillsdale, NJ: Erlbaum, 1978.
• POSNER, M. I., DEHAENE, S. Attentional networks. Trends Neurosci. 17: 75-79, 1994.[Medline]
• POSNER, M. I., PETERSEN, S. E. The attention system of the human brain. Annu. Rev. Neurosci. 13: 25-42, 1990.[Medline]
• POSNER, M. I., WALKER, J. A., FRIEDRICH, F. J., RAFAL, R. D. Effects of parietal injury on covert orienting of attention. Journal Neurosci. 4: 1863-1874, 1984.[Abstract]
• RAFAL, R., GERSHBERG, F., EGLY, R., IVRY, R., KINGSTONE, A., RO, T. Response channel activation and the lateral prefrontal cortex. Neuro-psychologia 34: 1197-1202, 1996.[Medline]
• RIZZOLATTI, G., RIGGIO, L., DASCOLA, I., UMILTÀ, C. Reorienting attention across the horizontal and vertical meridians: evidence in favor of a premotor theory of attention. Neuropsychologia 25: 31-40, 1987.[Medline]
• RUGG, M. D., COLES, M.G.H. In: Electrophysiology of Mind. Event-Related Brain Potentials and Cognition. Oxford, UK: Oxford Univ. Press, 1995.
• SCHLAG-REY, M., AMADOR, N., SANCHEZ, H., SCHLAG, J. Antisaccade performance predicted by neuronal activity in the supplementary eye field. Nature 390: 398-401, 1997.[Medline]
• SIMON, J. R. The effects of an irrelelevant directional cue on human information processing. In: Stimulus-Response Compatibility, edited by R. W. Proctor, and T. G. Reeve Amsterdam: Elsevier, 1990, p. 31-86.
• STELMACH, L. B., CAMPSALL, J. M., HERDMAN, C. M. Attentional and ocular movements. J. Exp. Psychol. Hum. Percept. Perform. 23: 823-844, 1997.
• SWEENEY, J. A., MINTUN, M. A., KWEE, S., WISEMAN, M. B., BROWN, D. L., ROSENBERG, D. R., CARL, J. R. Positron emission tomography study of voluntary saccadic eye movements and spatial working memory. J. Neurophysiol. 75: 454-468, 1996.[Abstract/Free Full Text]
• THICKBROOM, G. W., MASTAGLIA, F. L. Cerebral events preceding self-paced and visually triggered saccades. A study of presaccadic potentials. Electroencephalogr. Clin. Neurophysiol. 62: 277-289, 1985a.[Medline]
• THICKBROOM, G. W., MASTAGLIA, F. L. Presaccadic "spike" potential: investigation of topography and source. Brain Res. 339: 271-280, 1985b.[Medline]
• VALLE-INCLÁN, F. The locus of interference in the Simon effect: an ERP study. Biol. Psychol. 43: 147-162, 1996.[Medline]
• VAN DER LUBBE, R.H.J., WOESTENBURG, J. C. Modulation of early ERP components with peripheral precues: a trend analysis. Biol. Psychol. 45: 143-158, 1997.[Medline]
• VERLEGER, R. Event-related potentials and cognition: a critique of the context updating hypothesis and an alternative interpretation of P3. Behav. Brain Sci. 11: 343-356, 1988.
• VERLEGER, R. Towards an integration of P3 research with cognitive neuroscience. (Author's response on continuing commentary.) Behav. Brain Sci. 21: 150-154, 1998.
• VERLEGER, R., GASSER, T., MÖCKS, J. Correction of EOG artifacts in event related potentials of the EEG: aspects of reliability and validity. Psychophysiology 19: 472-480, 1982.[Medline]
• WASCHER, E., WAUSCHKUHN, B. The interaction of stimulus- and response-related processes measured by event-related lateralisations of the EEG. Electroencephalogr. Clin. Neurophysiol. 99: 149-162, 1996.[Medline]
• WAUSCHKUHN, B., WASCHER, E., VERLEGER, R. Lateralised cortical activity due to preparation of saccades and finger movements: a comparative study. Electroencephalogr. Clin. Neurophysiol. 102: 114-124, 1997.[Medline]
• YAMAGUCHI, S., TSUCHIYA, H., KOBAYASHI, S. Electroencephalographic activity associated with shifts of visuospatial attention. Brain 117: 553-562, 1994.[Abstract/Free Full Text]
• YAMAGUCHI, S., TSUCHIYA, H., KOBAYASHI, S. Electrophysiologic correlates of age effects on visuospatial attention shift. Cognit. Brain. Res. 3: 41-49, 1995.[Medline]
• ZELINSKY, G. J., SHEINBERG, D. L. Eye movements during parallel-serial visual search. J. Exp. Psychol. Hum. Percept. Perform. 23: 244-262, 1997.[Medline]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society