Purchase this article with an account.
Konstantin-Friedrich Willeke, Xiaoguang Tian, Joachim Bellet, Araceli Ramirez-Cardenas, Ziad Hafed; Memory-guided microsaccades. Journal of Vision 2017;17(10):895. doi: 10.1167/17.10.895.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Microsaccades are commonly described as "involuntary". However, under a variety of task conditions, microsaccade directions, amplitudes, and frequencies can be systematically modulated. While these results hint at voluntary control, they are primarily derived from likelihood measures rather than "at will" triggering of individual movements. Here we asked whether individual movements in the microsaccade amplitude range can be triggered: (1) "on demand" based on an arbitrary instruction, (2) without special training, (3) without visual guidance, and (4) in a spatially- and temporally-accurate manner. Two macaque monkeys and 7 human subjects performed a memory-guided saccade task. In this task, an eccentric flash was presented briefly (~50 ms). Subjects maintained flash location in memory for ~300-1100 ms, after which the fixation spot disappeared, providing a "go" command to generate a saccade to the remembered location. After an additional grace period, the remembered stimulus re-appeared allowing subjects to visually correct any remaining errors. The monkeys were only trained on the task with saccades >3 deg in amplitude, and humans were only given verbal instructions and minimal training. We then collected data from trials of random target eccentricities (0.1-16 deg) and directions (0-360 deg). All subjects naturally generated memory-guided movements even less than 0.5 deg in amplitude. These movements were highly directionally accurate, and while their amplitudes may have been quantitatively less accurate than visually-guided corrections, dependence of landing error on eccentricity was not qualitatively different between the memory and visually-corrective movements. Importantly, we ran the human subjects on control versions of the task with manual pointing, and the memory accuracy for foveal targets was comparable to that of "memory-guided microsaccades". We conclude that microsaccades can be generated voluntarily, and that any spatial inaccuracies in memory-guided microsaccades are likely not due to an inability of the occulomotor system to voluntarily trigger a tiny saccade.
Meeting abstract presented at VSS 2017
This PDF is available to Subscribers Only