Figures and data
Experimental Paradigm.
A) Observers began each trial by fixating on a central dot surrounded by two squares (5’ x 5’ in size) each offset 18 arcmin to the left/right of the center. A central cue instructed observers to either shift their gaze to the left/right location or maintain fixation at the center (neutral condition). After a variable delay period two fine spatial stimuli, bars tilted ±45deg, were briefly (50 milliseconds) presented at both locations. After 500 milliseconds a response cue appeared above either the left or right location. Observers were instructed to report the orientation of the stimulus previously shown at the location highlighted by this cue. B) The motor cue could be either spatially congruent or incongruent with the response cue location. In the neutral trials a fixation cross instructed subjects to maintain the gaze at the center. C) Examples of gaze position during a typical motor (Left) and neutral (Right) trial. Dashed vertical lines indicate stimuli presentation. D) Probability of stimuli presentation time relative to the motor cue (solid line) and neutral cue (dashed lines) trials. E) Probability of stimuli presentation relative to the microsaccade onset. Dashed line marks microsaccades onset time. Probability distributions in D) and E) were calculated based on a nonparametric kernel-smoothing. Individual observers are color-coded.
Microsaccade characteristics.
Microsaccade amplitude (A) and latency (B) distributions. Dashed lines in (A) show the target distance from the center of the display. C) 68% confidence ellipses of microsaccade landing endpoints. The gray square represents the location of the target drawn to scale. Probability distributions in A) and B) were calculated based on a nonparametric kernel-smoothing. Individual observers are color-coded.
Temporal dynamics of peri-microsaccadic visual discrimination within the foveola
A) 2-D probability distribution of gaze position during the stimuli presentation before, during, or after the microsaccade. B) Average performance measured as d′ smoothed as a function of time (window = ±24 ms, step = 1 ms) relative to the microsaccade onset in the congruent (blue) and incongruent (orange) conditions. Shaded error bars represent the margin of error for a 95% within-subject confidence interval for each time bin. Shaded grey region shows microsaccade approximate duration. C) Average performance at different times with respect to the microsaccade onset and in delay-matched neutral trials. Shaded dots represents single subject data. *p<0.05, ***p<0.001, ****p<0.0001 (one-way repeated-measures ANOVA (factor: condition) per each time frame (before, during and after), Tukey–Kramer corrected t-tests; comparisons relative to neutral trials are shown). (D) Average difference in d′ between the congruent and incongruent trials at different time points before the onset of the microsaccade. *p<0.05, **p<0.01, ***p<0.001 (one-way repeated-measures ANOVA (factor: time) with Tukey-Kramer corrected t-tests). (E) Average performance in congruent condition at different time points after the offset of the microsaccade. *p<0.05, **p<0.01 (one-way repeated-measures ANOVA (factor: time) with Tukey-Kramer corrected t-tests). Vertical error bars in C), D), and E) are 95% within-subject confidence intervals and horizontal error bars in D) and E) represent standard deviations.
Assessing the influence of covert attention
Average performance as a function of time relative to microsaccade (A) and relative to the cue (B) for congruent (blue) and incongruent (orange) trials. Each non-overlapping time bin in A and B covers 40 milliseconds. Shaded gray region indicates 95% within-subject confidence interval of microsaccade onset time. C) Average performance relative to the cue onset in congruent and incongruent trials in which observers failed to perform microsaccades. Each non-overlapping time bin in C covers 50 milliseconds. In A-C, *p<0.05, **p<0.01 (two-way repeated-measures ANOVA (factors: condition x time) with Tukey-Kramer corrected t-tests; comparisons between congruent and incongruent at each time bin are shown). D) Same data shown in B and C averaged over time within each group. *p<0.05, **p<0.01 (paired two-sided t-tests with Bonferroni corrected p-values). Vertical error bars in graphs are 95% within-subject confidence intervals and horizontal error bars represent standard deviations.
Average horizontal gaze position and two-dimensional eye speed relative to microsaccade onset.
Shaded error bars are standard deviation across individual subjects.
