Saccadic reaction times in patients with frontal and parietal lesions.

The effect of unilateral circumscribed lesions in different areas of the frontal and parietal cortex on the distributions of saccadic reaction times (SRTs) was investigated in 32 patients under four stimulus conditions: (i) gap-random: a target light appeared 200 ms after extinction of a central fixation light randomly at 8 degrees either left or right; (ii) overlap-random: like (i), but the fixation light remained on ('overlap'); (iii) gap-simultaneous: two peripheral lights appeared simultaneously left and right 200 ms after the extinction of the fixation light, one predesignated by instruction as saccadic target; (iv) overlap-simultaneous: like (iii), but the fixation light remained on. Depending on SRT and condition and based on the data of control subjects, we evaluated the percentages of anticipatory (SRT: 0-90 ms), express (91-157 ms) and regular (158-400 ms) saccades, of time (> 400 ms) and direction errors as well as the mean latencies and the standard deviations. These data were compared across control subjects, frontal lobe patients with a lesion in the region of the frontal eye field (FEF), and frontal patients with a lesion outside the FEF, as well as across control subjects, parietal lobe patients with a lesion in the dorsolateral region, and parietal patients with a lesion outside the dorsolateral region. (i) Frontal patients with a lesion in the FEF region showed an increased percentage of express saccades (gap-random condition), especially with saccades directed towards the side of the lesion (ipsilateral). If fixation was not interrupted prior to target appearance (overlap-random), express saccades were largely suppressed, similar to normals. All patients with a frontal lesion, whether in the region of the FEF or not, showed a decreased percentage of contralateral anticipatory saccades (gap-simultaneous), whereas the percentages of direction and time errors were in the normal range, even if more than one stimulus was presented (e.g. overlap-simultaneous condition). (ii) Patients with dorsolateral parietal lesions showed decomposed SRT patterns (high SRT variability, increased percentages of time and direction errors, decreased percentages of express and anticipatory saccades). The decomposition had a contralateral preponderance and increased if more than one stimulus was visible beside the saccadic target. Our results suggest that FEF lesions impair mainly the volitional control over visually triggered express or reflex-like saccades if fixation is disrupted, whereas dorsolateral parietal lesions impair mainly the selection of the visual targets.