Ch. Klein1, P. Berg2, Ch. Franz1, E. Hafstad1
1Research Group Psychophysiology, University of Freiburg,
Belfortstrasse 20, D-79098 Freiburg, Germany (e-mail:klein@psychologie.uni-freiburg.de);
2Department of Psychology, University of Konstanz,
D-78434 Konstanz, Germany
The preparation of saccadic eye movements is associated with a surface-negative potential shift, the presaccadic negativity (PSN). A series of experiments aimed at investigating the impact of (a) the saccade type (task blocks of visually-guided versus antisaccades), (b) the task instruction (speed versus no speed), and (c) the degree of information about the forthcoming saccade (direction versus no direction information, delivered by a peripheral cue) on the PSN amplitude and topography, as well as (d) the reliability of PSN measurement (4 weeks retest interval).
Visually-guided and antisaccades were elicited within the two-stimulus paradigm, with a 3.5 s SOA between the central fixation point and the saccade eliciting stimulus presented randomly at the left or right side. The EEG was recorded with a DC amplifier (MES, Munich), and AgAgCl electrodes were attached at 25 to 64 recordings sites over both hemispheres, using a sampling rate of 100 Hz (bandpass DC - 30 Hz). Healthy subjects (N1 = 23, N2 = 28, N3 = 16) participated in the experiments. In one of the experiments, the EyeLink system (SMI, Teltow) was additionally used in order to measure eye movements.
The following results were obtained. First, in all experiments the PSN had a topographical maximum at central sites, its amplitude being significantly larger preceding anti- than visually-guided saccades. Second, the instruction to respond "as quickly as possible" reduced this task effect and augmented saccadic reaction times (SRT) for visually-guided and antisaccades, when compared to an instruction that did not stress response speed. Third, direction information delivered by a peripheral cue 3.5 s before the forthcoming saccade, augmented the PSN amplitude both before the anti- and visually-guided saccades when compared to a condition with unreliable directional cueing. Fourth, for a retest interval of 4 weeks, the PSN amplitude at Cz turned out sufficiently reliable (anti: rtt = .80, visually-guided: .57 < rtt < .70), but less reliable than SRT (anti: rtt = .88, visually-guided: rtt = .77).
Altogether, our results indicate that the PSN is a retest-reliable measure of saccade preparation, which is sensitive to the experimental variation of the saccade type, the task instruction, and the amount of information enabling saccade preparation.
Research was supported by the Deutsche Forschungsgemeinschaft (DFG; Kl 985/6-1)