The effects of fixation target size and luminance on microsaccades and square-wave jerks

Abstract

A large amount of classic and contemporary vision studies require subjects to fixate a target. Target fixation serves as a normalizing factor across studies, promoting the field’s ability to compare and contrast experiments. Yet, fixation target parameters, including luminance, contrast, size, shape and color, vary across studies, potentially affecting the interpretation of results. Previous research on the effects of fixation target size and luminance on the control of fixation position rendered conflicting results, and no study has examined the effects of fixation target characteristics on square-wave jerks, the most common type of saccadic intrusion. Here we set out to determine the effects of fixation target size and luminance on the characteristics of microsaccades and square-wave jerks, over a large range of stimulus parameters. Human subjects fixated a circular target with varying luminance and size while we recorded their eye movements with an infrared video tracker (EyeLink 1000, SR Research). We detected microsaccades and SWJs automatically with objective algorithms developed previously. Microsaccade rates decreased linearly and microsaccade magnitudes increased linearly with target size. The percent of microsaccades forming part of SWJs decreased, and the time from the end of the initial SWJ saccade to the beginning of the second SWJ saccade (SWJ inter-saccadic interval; ISI) increased with target size. The microsaccadic preference for horizontal direction also decreased moderately with target size . Target luminance did not affect significantly microsaccades or SWJs, however. In the absence of a fixation target, microsaccades became scarcer and larger, while SWJ prevalence decreased and SWJ ISIs increased. Thus, the choice of fixation target can affect experimental outcomes, especially in human factors and in visual and oculomotor studies. These results have implications for previous and future research conducted under fixation conditions, and should encourage forthcoming studies to report the size of fixation targets to aid the interpretation and replication of their results.

An oculomotor continuum from exploration to fixation

“…The hypothesis that lack of visual content is responsible for the reduction of saccadic rates during the exploration of blank versus natural scenes is consistent with the observation of higher saccadic rates during the attempted fixation of large versus small targets (28,29). Because a large fixation target has no visual content within the local area of fixation/exploration, one should expect lower saccade rates in such a scenario.…”

Changes in quick phases of downbeat nystagmus during visual fixation

“…Nystagmus frequency was, as expected [6], increased at dot fixation In this study, we could show that quick phases of downbeat nystagmus were influenced by the presence of a fixation target: they became smaller and more frequent. Similar changes in fixational eye movements were observed in normal subjects, where microsaccade size is increased and their rate decreased without a visual target [7]. Nevertheless, in DBN, the quick phase was not purely compensatory to the upward drift, but showed a significant, unexpected horizontal component, which is not the case in vestibular nystagmus [8].…”

“…Therefore, in these patients we observed two fixational deficits: a vertical one, caused by the vertical slow drift and the compensatory vertical quick component, that was less influenced by visual fixation, and a horizontal component that depended largely on the presence of a target. The pattern of this horizontal component resembled square wave jerks, that are observed also in normal subjects [7], but are more frequent and larger in cerebellar disease [9][10][11], most likely due to a fastigial nucleus dysfunction [12]. Square wave jerks, both in normal subjects [7] and in cerebellar patients [9] tend to become larger and less frequent without a fixation target, similar to the quick phases of downbeat nystagmus, although in normal subjects there was no influence on their direction [7].…”

Saccadic Eye Movement Metrics Reflect Surgical Residents' Fatigue

“…48,49 Saccades were identified with a modified version of the algorithm developed by Engbert and Kliegl 47,50-53 with λ = 6 (threshold used for saccade detection) and a minimum saccadic duration of 6 ms. To reduce the amount of potential noise, we considered only binocular saccades, that is, saccades with a minimum overlap of one data sample in both eyes. [50][51][52][53][54] Additionally, we imposed a minimum intersaccadic interval of 20 ms so that potential overshoot corrections might not be categorized as new saccades. 55 .…”