## Introduction

Duing gaze fixation, our eyes are not still but often perform small and involuntary movements. There are three distinct types of fixational eye movements: drift, tremor and microsaccades. Among them, microsaccades represent the fastest component, with amplitudes of generally less than 1°1,2,3. Microsaccades and saccades have similar properties and may share the common mechanisms that are involved in orienting attention in space and time4,5. One intriguing line of research has established a tight link between covert spatial attention and some microsaccade activities. The microsaccade rate sharply dropped after the onset of an attentional cue, followed by a temporary enhancement6,7, and their direction was shown to be modulated by a shift in attention focus induced either endogenously8,9,10 or exogenously11,12. On the other hand, a more recent study revealed that the direction of spontaneous microsaccades inherently reflected the shift of spatial attention13. All these findings suggest that microsaccades are closely correlated with selective spatial attention and could reveal the deployment of attention focus. Therefore, microsaccades could be used as a psychophysiological measurement of cognition8,14.

Selective attention facilitates the perception of stimuli while ignoring or suppressing task-irrelevant distractors. The magnitude of this attention-mediated enhancement depends on the attentional load, which could be defined as either the number of perceived items or the complexity of the perceptual identification15,16. Mechanisms of the attentional load have been proposed as perceptual load, depending on the perceptual demands of the stimulus, and/or cognitive load, traditionally depending on the higher demands of working memory17. The load theory17,18,19 suggests that increasing the perceptual load decreases distractor interference, whereas increasing the working memory load enhances distractor competition. These suggestions have been verified by neuroimaging studies20,21,22.

## Materials and methods

### Animal preparation and surgical procedures

We collected data from two (monkey G and monkey N) adult male rhesus monkeys (Macaca mulatta) that were 8 years of age and weighed 7–9.5 kg. All the animal procedures were approved by the Institutional Animal Care and Use Committee of Shanghai Jiao Tong University and followed the recommendations of the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Similar to our previous surgical procedures30, each macaque was implanted with a titanium head-post for head stabilization and a single sclera eye coil to monitor eye movements. During the behavioral training, the animals had controlled water access and obtained most of their fluid intake (water or juice) by performing the behavioral tasks.

### Apparatus

The stimulus patterns were generated by a computer running Visionworks (Vision Research Graphics, Durham, USA) and were presented on a gamma-corrected CRT monitor (GDM-F520 Monitor, Sony Electronics, Tokyo, Japan, 640 × 480 pixels, 150 Hz refresh rate) positioned 57 cm from the monkey’s eyes. The mean luminance of the screen was 22 cd/m2. Another computer running CORTEX (NIMH Laboratory of Neuropsychology) was used to control the different events of the behavioral task (reward time and amount, maintenance of visual fixation, bar touch, bar release and time of stimulus presentation). A computer running OmniPlex (Plexon Corp., Dallas, USA) was used for data acquisition. The reward system was controlled by a peristaltic pump (BT100M, Chuangrui, Baoding, China). Eye movements were measured by the eye tracking system (ScleraTrak, Crist Instrument, Maryland, USA) using the magnetic induction technique31 and sampling at 1 kHz.

### Data analyses

For each trial, we analyzed microsaccades occurring during three particular time windows (TWs) of 650 ms (see Fig. 1C). For TW 1 and TW 2, we set the time windows as 150 ms before the onsets of the cue and stimulus to 500 ms after them. Both monkeys’ bar release RTs were more than 150 ms; therefore, TW 3 was defined as the interval between 500 ms before and 150 ms after the target color change. The effects of attentional load on microsaccade parameters, including rate, amplitude and directional congruency, were quantified as follows.

### Estimation of the microsaccade amplitude fluctuation

The fluctuation of the microsaccade amplitude may reflect the modulation of the perceptual load. The Fano factor provides a method to describe the variability in the collection of amplitudes and to relate it to the variability associated with a Poisson process. Let Ni for i = 1, …, n be a collection of independent identically distributed samples from a Poisson distribution. The Fano factor for these data is given by the ratio of the sample variance to the sample mean:

$${\rm{F}}=\,\frac{\frac{1}{{\rm{n}}-1}{\sum }_{{\rm{i}}=1}^{{\rm{n}}}{({{\rm{N}}}_{{\rm{i}}}-\bar{{\rm{N}}})}^{2}}{\bar{{\rm{N}}}},\,{\rm{w}}{\rm{h}}{\rm{e}}{\rm{r}}{\rm{e}}\,\bar{{\rm{N}}}=\frac{1}{{\rm{n}}}\sum _{{\rm{i}}=1}^{{\rm{n}}}{{\rm{N}}}_{{\rm{i}}}$$
(1)

We computed Fano factors of the microsaccade amplitude in 50 time bins of Periods 1, 2 and 3 for both task difficulties. A paired t test with a significance level of 0.05 was used to test the statistical significance of the differences in Fano factors under the easy and hard conditions during each period.

### The amount of task difficulty modulation index

The amount of the spatial attention and perceptual load effects on microsaccades was determined as a difficulty modulation index (DMI) according to the following equation:

$${\rm{D}}{\rm{M}}{\rm{I}}=\frac{{{\rm{M}}{\rm{S}}}_{{\rm{H}}}-{{\rm{M}}{\rm{S}}}_{{\rm{E}}}}{{{\rm{M}}{\rm{S}}}_{{\rm{H}}}+{{\rm{M}}{\rm{S}}}_{{\rm{E}}}}$$
(2)

where MSH are the microsaccade activities, such as the rate, amplitude or directional congruency during the hard task, and MSE are those during the easy task.

We computed the DMIs of the microsaccade activities (rate, amplitude, and directional congruency) in Periods 1, 2 and 3 and then used one-way ANOVA to compare them during different periods and a Bonferroni test for post hoc comparisons of each other. A positive index indicates that the microsaccade activities were enhanced during the hard task, whereas a negative index indicates that the microsaccade activities were suppressed.

### Relationship between microsaccades and behavioral performance

We measured both the RT and the correct rate as a function of time when a microsaccade was triggered relative to the onset of color change in both the easy and hard tasks. To acquire time courses of these effects, we aligned all the trials in which microsaccades occurred during the period between 200 ms before and 150 ms after the target color change and then used a running average with temporal bins (50 ms width) successively moved in 10 ms steps. We measured baselines of the RT and correct rate on all trials during which no microsaccades occurred within 350 ms around the time of color change. To detect the effect of microsaccades on behavioral performance, we compared with- and without-microsaccade situations within 350 ms around the target color change in easy and hard tasks from both monkeys (ANOVA with a criterion of p < 0.05).

All error bars and the ranges of error specified after the values of the mean were generated from SEM, i.e., mean ± SEM.

## Results

We analyzed behavioral data from 1458 easy task trials and 1379 hard task trials (collected from 23 easy and 23 hard blocks, respectively) performed by monkey G and 1435 easy task trials and 1339 hard task trials (collected from 24 easy blocks and 24 hard blocks) performed by monkey N.

Moreover, we cued a position that was different from the location where the grating changed color (10% of invalid cued trials) in several blocks. In these experiments, the RTs were significantly longer in the trials that were cued incorrectly (331.4 ± 1.8 ms vs. 292.7 ± 0.8 ms, F 1,255 = 16.8, p < 0.001, one-way ANOVA), indicating that the animals were utilizing the cue to perform the task.

We first examined the microsaccade rate changes over time in three 650-ms time windows. Figure 2 shows the evolution of the microsaccade rate as a function of time across all trials in the easy and hard tasks for each monkey. The microsaccades had comparable baseline rates during the easy and hard tasks for monkey G (1.85 ± 0.04 Hz vs. 1.73 ± 0.04 Hz, F 1,28 = 3.53, p = 0.06, one way ANOVA) and for monkey N (2.77 ± 0.08 Hz vs. 2.64 ± 0.12 Hz, F 1,28 = 0.81, p = 0.38, one way ANOVA).

On the other hand, a profound perceptual load modulation on the microsaccade rate during the rebound phase was found in TW 2 (Fig. 2B,F) but not TW 1. This significant microsaccade rate suppression by the perceptual load was also found immediately before the color change for both monkeys (Fig. 2C,G). We further analyzed the impact of the perceptual load on the microsaccade rate that occurred during the Control period (150 ms before the cue onset), Period 1 (500 ms after the cue onset), Period 2 (500 ms after the stimulus onset) and Period 3 (500 ms before the color change). As shown in the upper panels of Fig. 2D,H, high perceptual load significantly suppressed the microsaccade rate for both monkeys during Period 3 (1.07 ± 0.02 Hz vs. 0.88 ± 0.02 Hz, F 1,98 = 45.27, p < 0.0001 for monkey G and 2.41 ± 0.03 Hz vs. 2.02 ± 0.03 Hz, F 1,98 = 107.44, p < 0.0001 for monkey N, one way ANOVA). There were no significant difference during Period 1 (F 1,98 = 1.17, p = 0.28 for monkey G and F 1,98 = 0.05, p = 0.83 for monkey N, one way ANOVA). In addition, there were no consistent difference during Period 2 (1.05 ± 0.09 Hz vs. 0.92 ± 0.08 Hz, F 1,98 = 1.25, p = 0.27 for monkey G and 2.22 ± 0.14 Hz vs. 1.90 ± 0.10 Hz, F 1,98 = 4.49, p = 0.05 for monkey N, one way ANOVA).

To quantify the perceptual load modulation on the microsaccade rate during the four periods, we calculated the DMI_rate ratios (see methods). The DMI_rate was close to zero during the Control period, showing no modulation of the perceptual load, since there was no peripheral stimulus present. Values of the DMI_rate were less than zero during Periods 2 and 3 for both monkeys, indicating an inhibition of the microsaccade rate by the perceptual load. The analysis revealed a significant effect of time period on the DMI_rate (F 3,161 = 3.53, p = 0.02 for monkey G and F 3,161 = 16.72, p < 0.0001 for monkey N, one way ANOVA). The Bonferroni post hoc comparisons revealed significantly greater DMI_rate during the Period 3 compared to the Period 1 (−0.10 ± 0.01 vs. −0.06 ± 0.01, p = 0.01 for monkey G and −0.09 ± 0.01 vs. −0.0007 ± 0.01, p < 0.0001 for monkey N) and the Period 2 (−0.10 ± 0.01 vs. −0.07 ± 0.02, p = 0.05 for monkey G and −0.09 ± 0.01 vs. −0.07 ± 0.01, p = 0.05 for monkey N), indicating that the strongest modulation effect of the perceptual load on the microsaccade rate appeared immediately before the occurrence of the response event.

In summary, even though the microsaccade rate was suppressed during the rebound phase after the stimulus onset, the sustained and stronger modulation of the perceptual load on them appeared before the target color change.

Generally, both monkeys made small microsaccades even before the cue onset while performing the demanding detection task. For monkey G, the mean amplitudes were 0.23° ± 0.015° and 0.22° ± 0.010°(F 1,28 = 0.16, p = 0.69, one way ANOVA) in the Control period during the easy task and hard task, respectively. For monkey N, the mean amplitudes were 0.21° ± 0.006°in the easy task and 0.21° ± 0.006° in the hard task (F 1, 28 = 0.53, p = 0.47, one way ANOVA).

To explore a possible interaction between the perceptual load and the microsaccade amplitude, we performed an analysis similar to the one we performed on the microsaccade rate. In Fig. 3, the microsaccade amplitude demonstrated a transient increase of 100–110 ms when a stimulus event occurred, exactly at the time of the minimum microsaccade rate. When the microsaccade amplitude increased after the cue onset, an inconsistent effect of the perceptual load on the microsaccade amplitude has been found for the two monkeys (see Fig. 3AE).

To evaluate the effect of the perceptual load modulation on the microsaccade amplitude, we calculated mean amplitudes across all trials and corresponding DMI ratios during each time period under both task difficulties. Here, we observed a suppressing effect of the perceptual load on the microsaccade amplitude for both monkeys. The suppressing effect was statistically significant during all periods after the cue onset, as follows: Period 1, 0.19° ± 0.007° vs. 0.15° ± 0.007°, F 1,98 = 13.51, p < 0.0001 for monkey G and 0.26° ± 0.006° vs. 0.23° ± 0.004°, F 1, 98 = 10.48, p = 0.002 for monkey N, Period 2, 0.13° ± 0.004° vs. 0.09° ± 0.004°, F 1,98 = 37.20, p < 0.0001 for monkey G and 0.23° ± 0.005° vs. 0.20° ± 0.005°, F 1,98 = 19.27, p < 0.0001 for monkey N, and Period 3, 0.12 ± 0.003° vs. 0.11 ± 0.002°, F 1,98 = 4.66, p = 0.03 for monkey G and 0.20 ± 0.002° vs. 0.19 ± 0.002°, F 1,98 = 10.39, p = 0.002 for monkey N (easy vs. hard, one way ANOVA). There was also a significant effect of time period on the DMI_amplitude (F 3,161 = 24.88, p < 0.0001 for monkey G and F 3,161 = 10.63, p < 0.0001 for monkey N). As shown at the bottom panels of Fig. 3D,H, the absolute values of the DMI_amplitude were the greatest in Period 2 for both monkeys, suggesting that the strongest modulation by the perceptual load on the microsaccade amplitude occurred after the stimulus onset.

Furthermore, a lower fluctuation of amplitudes during the hard task after the cue onset was found. Table 1 presents a comparison of the Fano factors of the microsaccade amplitudes during the easy and hard tasks. The Fano factors always exhibited a significant difference during Periods 1, 2 and 3 in both monkeys, indicating lower fluctuations of microsaccade amplitude under the high perceptual load.

Taken together, these observations revealed that the microsaccade amplitude and its fluctuation were suppressed by the perceptual load immediately after the cue onset, and the strongest effect emerged after the stimulus onset.

### Effect of microsaccades on behavioral performance

We also analyzed the influence of microsaccades that occurred around the onset of the color change on behavioral performance. For comparison to microsaccade-free situations, we measured the baseline RT and correct percentage on all trials during which no microsaccades occurred within 350 ms of the target color change onset. For monkey G, the no-microsaccade baselines of the RT were 282.5 ± 0.74 ms in the easy task and 343.4 ± 1.5 ms in the hard task, and they were 276.5 ± 1.2 ms and 332.9 ± 2.1 ms for monkey N. The no-microsaccade baselines of the correct percentage were 89.8 ± 2.3% for the easy task and 75.7 ± 2.2% for the hard task for monkey G and 84.3 ± 2.4% and 72.4 ± 3.4% for monkey N.

The temporal patterns of the RT and correct rate associated with the microsaccades occurring relative to the target color change are shown in Fig. 5. As depicted in Fig. 5A,E, both monkeys exhibited longer RT when microsaccades occurred around the target color change. We then compared these reaction times in the easy and hard tasks with those observed on the microsaccade-free trials. As shown in Fig. 5C,G, the reaction times of trials with microsaccades were significantly longer than those without microsaccades during the hard task (monkey G, 370.6 ± 12.0 ms vs. 345.0 ± 1.5 ms, F 1,1212 = 4.57, p = 0.03, monkey N, 337.3 ± 3.8 ms vs. 326.2 ± 2.1 ms, F 1,1207 = 13.54, p < 0.001, one way ANOVA).

Microsaccades occurring near the response event were also associated with changes in the monkeys’ accurate percentages. We performed an analysis on the accurate rate (see Fig. 5B,F) and found a manifest tendency toward lower accuracy in the cases with microsaccades than those without microsaccades, especially during the hard tasks (see Fig. 5D,H). For monkey G, the mean accurate percentages were 83.8 ± 3.3% vs. 89.7 ± 2.6%, F 1,1417 = 13.58, p < 0.001 and 72.6 ± 2.3% vs. 75.6 ± 2.3%, F 1,1212 = 4.35, p = 0.04 in the easy and hard tasks, respectively. For monkey N, they were 81.6 ± 4.1% vs. 84.0 ± 3.7%, F 1,136 = 3.23, p = 0.07 in the easy task and 69.3 ± 3.3% vs. 71.9 ± 4.8%, F 1,1207 = 3.88, p = 0.04 in the hard task (with vs. without microsaccades, one way ANOVA).

Generally speaking, microsaccades impaired the monkeys’ behavioral performance, especially in the hard tasks with high perceptual load. The increase in the RT and the decrease in the accuracy are likely a manifestation of microsaccade suppression.

## Discussion

### The influence of microsaccades at the time of color change on behavioral performance

Even though many studies concerning microsaccades focus on their roles report that they triggered excitatory neural responses in most brain areas4,41,42, and agree that they generally benefit vision41,43,44,45,46, microsaccades have also been shown to suppress neural activity47,48 and inhibit behavioral performance10 during stimulus presentation in monkeys. Our results revealed a detrimental effect of microsaccades on the task performance as well. Microsaccades appeared before the target color change seemed to decrease the speed and efficiency of the behavioral response, especially in the hard tasks with high perceptual load, suggesting that the inhibition of microsaccade generation might aid the detection of target change.

Although our results clearly demonstrate the influence of microsaccades on perceptual performance, a potential contribution from attentional allocation to this effect must be carefully evaluated. For example, it could be argued that more microsaccades could result from inattentiveness, which in turn could have lowered perceptual performance. There are several reasons why attentional allocation is unlikely to affect our measurements. First, the color change was very subtle even in the easy task, therefore the monkeys would perform the tasks with great effort. Second, we observed that the monkeys executed saccades to the target location after response in most trials, indicating they were paying attention to the perceptual target. Finally, since the monkeys were primarily driven by the magnitude of reward (the faster the response, the larger the reward), it is unlikely that they would decrease their effort during the easy trials.

The superior colliculus (SC) is traditionally divided into two areas, the rostral fixation zone and the caudal saccade zone, with neurons that are activated respectively during microsaccades and saccades49. Previous electrophysiological research50 has found that fluctuations in SC activity at the rostral pole play a causal role in microsaccade generation during fixation. Furthermore, the rostral SC receives excitatory inputs from the frontal and parietal areas and inhibitory inputs from the basal ganglia. Given that these cortical areas are involved in visual attention51, varying the perceptual load should modulate rostral SC activity and, thus, the microsaccade rate and amplitude.

In conclusion, we assessed the effectiveness of two perceptual load levels to modulate microsaccades in monkeys. We demonstrated that high perceptual load suppressed the rate and amplitude of microsaccades, suggesting that these two parameters could be indicators of the perceptual load.

## References

1. 1.

Martinez-Conde, S., Macknik, S. L. & Hubel, D. H. Microsaccadic eye movements and firing of single cells in the striate cortex of macaque monkeys. Nat. Neurosci. 3, 251–258 (2000).

2. 2.

Rolfs, M. Microsaccades: small steps on a long way. Vision Res. 49, 2415–2441 (2009).

3. 3.

Martinez-Conde, S., Macknik, S. L., Troncoso, X. G. & Hubel, D. H. Microsaccades: a neurophysiological analysis. Trends Neurosci. 32, 463–475 (2009).

4. 4.

Martinez-Conde, S., Otero-Millan, J. & Macknik, S. L. The impact of microsaccades on vision: towards a unified theory of saccadic function. Nat. Rev. Neurosci. 14, 83–96 (2013).

5. 5.

Otero-Millan, J., Macknik, S. L., Langston, R. E. & Martinez-Conde, S. An oculomotor continuum from exploration to fixation. Proc. Natl. Acad. Sci. USA 110, 6175–6180 (2013).

6. 6.

Engbert, R. & Kliegl, R. Microsaccades uncover the orientation of covert attention. Vision Res. 43, 1035–1045 (2003).

7. 7.

Laubrock, J., Engbert, R. & Kliegl, R. Microsaccade dynamics during covert attention. Vision Res. 45, 721–730 (2005).

8. 8.

Hafed, Z. M. & Clark, J. J. Microsaccades as an overt measure of covert attention shifts. Vision Res. 42, 2533–2545 (2002).

9. 9.

Pastukhov, A. & Braun, J. Rare but precious: microsaccades are highly informative about attentional allocation. Vision Res. 50, 1173–1184 (2010).

10. 10.

Hafed, Z. M., Lovejoy, L. P. & Krauzlis, R. J. Modulation of microsaccades in monkey during a covert visual attention task. J. Neurosci. 31, 15219–15230 (2011).

11. 11.

Galfano, G., Betta, E. & Turatto, M. Inhibition of return in microsaccades. Exp. Brain. Res. 159, 400–404 (2004).

12. 12.

Rolfs, M., Engbert, R. & Kliegl, R. Crossmodal coupling of oculomotor control and spatial attention in vision and audition. Exp. Brain. Res. 166, 427–439 (2005).

13. 13.

Yuval-Greenberg, S., Merriam, E. P. & Heeger, D. J. Spontaneous microsaccades reflect shifts in covert attention. J. Neurosci. 34, 13693–13700 (2014).

14. 14.

Laubrock, J., Kliegl, R., Rolfs, M. & Engbert, R. When do microsaccades follow spatial attention? Atten. Percept. Psychophys. 72, 683–694 (2010).

15. 15.

Lavie, N. & Tsal, Y. Perceptual load as a major determinant of the locus of selection in visual Atten. Percept. Psychophys. 56, 183–197 (1994).

16. 16.

Urbach, D. & Spitzer, H. Attentional effort modulated by task difficulty. Vision Res. 35, 2169–2177 (1995).

17. 17.

Lavie, N., Hirst, A., de Fockert, J. W. & Viding, E. Load theory of selective attention and cognitive control. J. Exp. Psychol. Gen. 133, 339–354 (2004).

18. 18.

Lavie, N. Distracted and confused? selective attention under load. Trends Cogn. Sci. 9, 75–82 (2005).

19. 19.

Lavie, N. Attention, distraction, and cognitive control under load. Current Directions in Psychological Science 19, 143–148 (2010).

20. 20.

Schwartz, S. et al. Attentional load and sensory competition in human vision: modulation of fMRI responses by load at fixation during task-irrelevant stimulation in the peripheral visual field. Cereb. Cortex 15, 770–786 (2005).

21. 21.

Rissman, J., Gazzaley, A. & D’Esposito, M. The effect of non-visual working memory load on top-down modulation of visual processing. Neuropsychologia 47, 1637–1646 (2009).

22. 22.

Kelley, T. A. & Lavie, N. Working memory load modulates distractor competition in primary visual cortex. Cereb. Cortex 21, 659–665 (2011).

23. 23.

Benedetto, S., Pedrotti, M. & Bridgeman, B. Microsaccades and exploratory saccades in a naturalistic environment. J. Eye Mov. Res. 4 (2011).

24. 24.

Hicheur, H., Zozor, S., Campagne, A. & Chauvin, A. Microsaccades are modulated by both attentional demands of a visual discrimination task and background noise. J. Vis. 13, 18 (2013).

25. 25.

Valsecchi, M., Betta, E. & Turatto, M. Visual oddballs induce prolonged microsaccadic inhibition. Exp. Brain Res. 177, 196–208 (2007).

26. 26.

Valsecchi, M. & Turatto, M. Microsaccadic responses in a bimodal oddball task. Psychol. Res. 73, 23–33 (2009).

27. 27.

Siegenthaler, E. et al. Task difficulty in mental arithmetic affects microsaccadic rates and magnitudes. Eur. J. Neurosci. 39, 287–294 (2014).

28. 28.

Gao, X., Yan, H. & Sun, H. J. Modulation of microsaccade rate by task difficulty revealed through between- and within-trial comparisons. J. Vis. 15 (2015).

29. 29.

Dalmaso, M., Castelli, L., Scatturin, P. & Galfano, G. Working memory load modulates microsaccadic rate. J. Vis. 17, 6 (2017).

30. 30.

Chen, Y. et al. Task difficulty modulates the activity of specific neuronal populations in primary visual cortex. Nat. Neurosci. 11, 974–982 (2008).

31. 31.

Judge, S. J., Richmond, B. J. & Chu, F. C. Implantation of magnetic search coils for measurement of eye position: an improved method. Vision Res. 20, 535–538 (1980).

32. 32.

Engbert, R. & Mergenthaler, K. Microsaccades are triggered by low retinal image slip. Proc. Natl. Acad. Sci. USA 103, 7192–7197 (2006).

33. 33.

Engbert, R. Microsaccades: a microcosm for research on oculomotor control, attention, and visual perception. Prog. Brain Res. 154, 177–192 (2006).

34. 34.

McCamy, M. B. et al. Microsaccadic efficacy and contribution to foveal and peripheral vision. J. Neurosci. 32, 9194–9204 (2012).

35. 35.

Moller, F., Laursen, M. L., Tygesen, J. & Sjolie, A. K. Binocular quantification and characterization of microsaccades. Graefes. Arch. Clin. Exp. Ophthalmol. 240, 765–770 (2002).

36. 36.

Zuber, B. L., Stark, L. & cook, G. Microsaccades and the velocity–amplitude relationship for saccadic eye movements. Science 150, 1459–1460 (1965).

37. 37.

Martinez-Conde, S., Macknik, S. L. & Hubel, D. H. The role of fixational eye movements in visual perception. Nat. Rev. Neurosci. 5, 229–240 (2004).

38. 38.

Hafed, Z. M. & Ignashchenkova, A. On the dissociation between microsaccade rate and direction after peripheral cues: microsaccadic inhibition revisited. J. Neurosci. 33, 16220–16235 (2013).

39. 39.

Engbert, R. Computational modeling of collicular integration of perceptual responses and attention in microsaccades. J. Neurosci. 32, 8035–8039 (2012).

40. 40.

Rolfs, M., Kliegl, R. & Engbert, R. Toward a model of microsaccade generation: the case of microsaccadic inhibition. J. Vis. 8, 51–23 (2008).

41. 41.

Martinez-Conde, S., Macknik, S. L. & Hubel, D. H. Microsaccadic eye movements and firing of single cells in the striate cortex of macaque monkeys. Nat. Neurosci. 3, 251–258 (2000).

42. 42.

Troncoso, X. G. et al. V1 neurons respond differently to object motion versus motion from eye movements. Nat. commun. 6, 8114–8123 (2015).

43. 43.

Ko, H. K., Poletti, M. & Rucci, M. Microsaccades precisely relocate gaze in a high visual acuity task. Nat. Neurosci. 13, 1549–1553 (2010).

44. 44.

Chen, C. Y., Ignashchenkova, A., Thier, P. & Hafed, Z. M. Neuronal Response Gain Enhancement prior to Microsaccades. Curr. Biol. 25, 2065–2074 (2015).

45. 45.

Martinez-Conde, S., Macknik, S. L., Troncoso, X. G. & Dyar, T. A. Microsaccades counteract visual fading during fixation. Neuron 49, 297–305 (2006).

46. 46.

McCamy, M. B., Otero-Millan, J., Di Stasi, L. L., Macknik, S. L. & Martinez-Conde, S. Highly Informative Natural Scene Regions Increase Microsaccade Production during Visual Scanning. J. Neurosci. 34, 2956–2966 (2014).

47. 47.

Hafed, Z. M. & Krauzlis, R. J. Microsaccadic suppression of visual bursts in the primate superior colliculus. J. Neurosci. 30, 9542–9547 (2010).

48. 48.

Herrington, T. M. et al. The effect of microsaccades on the correlation between neural activity and behavior in middle temporal, ventral intraparietal, and lateral intraparietal areas. J. Neurosci. 29, 5793–5805 (2009).

49. 49.

Munoz, D. P. & Wurtz, R. H. Fixation cells in monkey superior colliculus. I. Characteristics of cell discharge. J. Neurophysiol. 70, 559–575 (1993).

50. 50.

Hafed, Z. M., Goffart, L. & Krauzlis, R. J. A neural mechanism for microsaccade generation in the primate superior colliculus. Science 323, 940–943 (2009).

51. 51.

Schall, J. D. On the role of frontal eye field in guiding attention and saccades. Vision Res. 44, 1453–1467 (2004).

## Acknowledgements

This research was supported by the National Basic Research Program of China (973 Program, 2011CB707502/3) and the National Natural Science Foundation of China (31471081, 91120304).

## Author information

L.X. and Y.C. devised the project and drafted the manuscript; L.X. and T.W. carried out the data collection and analyses; D.H., Q.H., X.C. and L.L. participated in the design of the study and contributed to writing the manuscript. All authors gave final approval for publication.

Correspondence to Yao Chen.

## Ethics declarations

### Competing Interests

The authors declare that they have no competing interests.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Rights and permissions

Reprints and Permissions

Xue, L., Huang, D., Wang, T. et al. Dynamic modulation of the perceptual load on microsaccades during a selective spatial attention task. Sci Rep 7, 16496 (2017). https://doi.org/10.1038/s41598-017-16629-2

• Accepted:

• Published:

• ### The Attentional Blink is Related to the Microsaccade Rate Signature

• Mark J Roberts
• , Gesa Lange
• , Tracey Van Der Veen
• , Eric Lowet
•  & Peter De Weerd

Cerebral Cortex (2019)

• ### Microsaccadic rate and pupil size dynamics in pro-/anti-saccade preparation: the impact of intermixed vs. blocked trial administration

• Mario Dalmaso
• , Luigi Castelli
•  & Giovanni Galfano

Psychological Research (2019)