ELSEVIER Electroencephalography and clinical Neurophysiology 102 (1997) 356-362 o Cortical potentials preceding pro- and antisaccades in man S. Everling*, P. Krappmann, H. Flohr Brain Research Institute, Universi~ of Bremen, POB 33 04 40, 28334 Bremen, Germany Accepted for publication: 10 October 1996 Abstract The antisaccade task has been used widely to assess a frontal lobe deficit. In the present study cortical potentials preceding prosaccades and anti saccades were recorded in 7 healthy subjects with 19 scalp electrodes according to the international 10-20 system. The main results were as follows: (1) In both saccade types, a slow presaccadic negative shift was observed at dorso-medial frontal recording sites. The integral over these potentials was significantly greater for antisaccades at the C3, C4 and Cz location than in prosaccades. (2) The integral over presaccadic positivity was significantly lower for antisaccades compared with prosaccades at the Cz location. It is concluded that these results support an important role of the supplementary eye fields in generating antisaccades. © 1997 Elsevier Science Ireland Ltd. Keywords: Saccades; Antisaccades; Premovement cortical potentials 1. Introduction The 'visual grasp reflex' (Hess et al., 1946) is triggered by the sudden appearance of a target in the peripheral visual field. Subjects, however, can be instructed to inhibit a saccade toward the target and to initiate a saccade to the opposite side. This antisaccade task (Hallett, 1978; Hallett and Adams, 1980) is a simple way to examine voluntary control over a reflexive behavior. Traditionally, an intact prefrontal cortex is considered to be necessary to suppress a reflexive in favor of a deliberate behavior. In fact, Guitton et al. (1985) have shown that patients with lesions in the frontal lobe either could rarely inhibit a (pro)saccade toward the visual target or did not generate the antisaccade. This observation has led to a number of studies with patients having disorders which are known or suspected to involve the frontal cortex. Schi- zophrenics (Clementz et al., 1994; Fukushima et al., 1988; Fukushima et al., 1990; Matsue et al., 1994; Rosse et al., 1993; Sereno and Holzman, 1995), patients with Alzhei- mer's disease (Currie et al., 1991) and patients with * Corresponding author, MRC Group in Sensory-Motor Physiology, Queen's University, Kingston, Canada K7L 3N6. e-maih stefan@ss2.biomed.queensu.ca. Chorea Huntington (Currie et al., 1991) have been shown to be impaired in the antisaccade task. However, it is still controversial which frontal areas are involved in the execution of correct antisaccades. Guitton et al. (1985) discovered that lesions implicating the frontal eye fields and the supplementary motor area impaired the ability to generate antisaccades. On the other hand, Pier- rot-Deseilligny et al. (1991) have reported that only lesions of the dorsolateral prefrontal cortex, but not of the frontal eye fields and supplementary motor areas led to an increased error rate during antisaccades performance. Recently, O'Driscoll et al. (1995) have supported an invol- vement of the frontal eye fields and the supplementary motor area in the generation of antisaccades by PET ima- ging data. However, only measurements of electric poten- tials have the temporal resolution to distinguish between activation preceding and accompanying saccadic eye movements (Kurtzberg and Vaughan, 1982; Moster and Goldberg, 1990). Three distinct scalp potentials can be recorded prior to saccades: 1. Self-paced saccades are preceded by a presaccadic negativity (PSN) which is similar to the Bereitschaft- spotential that precedes voluntary movement of the extremities (Becket et al., 1972). 2. Both, self-paced and visually triggered saccades are 0013-4694/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0921-884X(96)96569-8 EEG 96569