Delayed saccade to perceptually demanding locations in Parkinson's disease: analysis from the perspective of the speed-accuracy trade-off.

Parkinson's disease (PD) patients reportedly have shortened, normal, or prolonged latency of visually guided saccades (VGSs). This inconsistency seems to be partly derived from differences in experimental conditions, such as target eccentricity and direction. Another etiology may be a physiological saccade property, the speed-accuracy trade-off. VGS latency tends to increase along with its gain in certain conditions; however, this relationship has not been addressed in PD saccade studies. In this study, we measured VGS latency and gain in 47 PD patients and 48 normal controls (NCs). VGS was evoked by a target, which was presented at the central position initially and pseudo-randomly jumped to the horizontal (10° or 20° eccentricity) or vertical (10° or 15°) meridian. For each target location, the logarithm of the latency (log-latency) was modeled with subject type (PD or NC), age, and gain in the linear-mixed regression analysis. Subsequently, for target locations where PD patients showed an abnormality, the log-latency was similarly modeled with additional clinical variables measured by the mini-mental state examination (MMSE) and unified Parkinson's disease rating scale Part III. PD saccade latency was prolonged and influenced by the MMSE score when targets were presented at the 20° horizontal and upper vertical meridians. Furthermore, gain was a consistently significant variable in all models. The target locations of the delayed saccade corresponded to perceptually demanding locations, indicating that PD subclinical visual dysfunction prolonged the latency. The influence of the MMSE score supports this reasoning. Moreover, the speed-accuracy trade-off appeared to contribute to the accurate saccade analysis.