Open Access
Article  |   December 2024
Voluntary blinks and eye-widenings, but not spontaneous blinks, facilitate perceptual alternation during continuous flash suppression
Author Affiliations
  • Ryoya Sato
    Graduate School of Science and Engineering, Chiba University, Chiba, Japan
    [email protected]
  • Eiji Kimura
    Department of Psychology, Graduate School of Humanities, Chiba University, Chiba, Japan
    [email protected]
Journal of Vision December 2024, Vol.24, 11. doi:https://doi.org/10.1167/jov.24.13.11
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ryoya Sato, Eiji Kimura; Voluntary blinks and eye-widenings, but not spontaneous blinks, facilitate perceptual alternation during continuous flash suppression. Journal of Vision 2024;24(13):11. https://doi.org/10.1167/jov.24.13.11.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

The fact that blinks occur more often than necessary for ocular lubrication has led to the proposal that blinks are involved in altering some aspects of visual cognition. Previous studies have suggested that blinking can modulate the alternation of different visual interpretations of the same stimulus, that is, perceptual alternation in multistable perception. This study investigated whether and how different types of blinks, spontaneous and voluntary, interact with perceptual alternation in a multistable perception paradigm called continuous flash suppression. The results showed that voluntary blinking facilitated perceptual alternation, whereas spontaneous blinking did not. Moreover, voluntary eye-widening, as well as eyelid closing, facilitated perceptual alternation. Physical blackouts, which had timing and duration comparable to those of voluntary blinks, did not produce facilitatory effects. These findings suggest that the effects of voluntary eyelid movements are mediated by extraretinal processes and are consistent with previous findings that different types of blinks are at least partially mediated by different neurophysiological processes. Furthermore, perceptual alternation was also found to facilitate spontaneous blinking. These results indicate that eyelid movements and perceptual alternation interact reciprocally with each other.

Introduction
Humans spontaneously blink approximately 20 times per minute, a rate more than five times higher than what is required for ocular lubrication (Doane, 1980). These “surplus” blinks have been assumed to be associated with cognitive processing of visual information. Research has linked blink frequency during cognitive tasks to internal cognitive processes, such as attention (Ang & Maus, 2020; Irwin, 2011; Nakano, 2015; Nakano, Kato, Morito, Itoi, & Kitazawa, 2013; Oh, Jeong, & Jeong, 2012), prediction (Bonneh, Fried, & Adini, 2015; Brych & Händel, 2020), and cognitive load (Fukuda & Matsunaga, 1983; Fukuda, Stern, Brown, & Russo, 2005; Siegle, Ichikawa, & Steinhauer, 2008). For example, during video viewing, blinks synchronize both among viewers and within individual viewers at implicit breakpoints in the story, such as the conclusion of a main character's action (Nakano et al., 2013; Nakano & Kitazawa, 2010; Nakano, Yamamoto, Kitajo, Takahashi, & Kitazawa, 2009). 
However, external sensory inputs also strongly influence blink frequency (Cardona, Gómez, Quevedo, & Gispets, 2014; Cong, Sharikadze, Staude, Deubel, & Wolf, 2010; Doughty, 2013). Recent studies indicate that blink inhibition or disinhibition depends on lower-level stimulus features such as contrast and spatial frequency (Bonneh, Adini, & Polat, 2016). This complicates the investigation of blink modulation by internal cognitive processes. 
Multistable perception offers a potential solution to this complex situation. In this phenomenon, different visual interpretations of the same stimulus compete for dominance, leading to a continuous alternation of the dominant percept over time. Examples include binocular rivalry and ambiguous figures. Crucially, in multistable perception, perception varies while the stimulus remains unchanged, allowing exploration of interactions between blinking and cognitive processing associated with perceptual, rather than physical, changes. Previous studies examining changes in blink frequency before and after perceptual alternation have suggested a reciprocal relationship between blinking and perceptual alternation. However, results have not always been consistent across studies (Brych, Murali, & Händel, 2021; Ito et al., 2003; Kalisvaart & Goossens, 2013; Nakatani, Orlandi, & van Leeuwen, 2011; van Dam & van Ee, 2005), as will be detailed later. 
The present study aims to elucidate the functional relationship between blinking and cognitive processing by investigating their mutual effects in multistable perception tasks. We employ several approaches, with a key focus on comparing the effects of different types of blinks—namely, spontaneous and voluntary blinks—on perceptual alternation. Blinks are categorized into reflex, voluntary, and spontaneous types (Stern, Walrath, & Goldstein, 1984). Reflex blinks are reflexive eyelid-closing responses to external stimuli, while voluntary blinks are intentionally initiated. Spontaneous blinks occur without reflexive triggers or intentional efforts. Despite similar eyelid movements and retinal changes, spontaneous and voluntary blinks involve different neural mechanisms (Kato & Miyauchi, 2003; Nakano et al., 2013; Nakano & Kuriyama, 2017). For example, Nakano and Kuriyama (2017) found that spontaneous but not voluntary blinks induce transient activation of the sympathetic nervous system, indicated by increased heart rate and skin conductance levels. 
Previous studies have not consistently distinguished between blink types, resulting in varied findings. Voluntary blinks have been shown to facilitate perceptual alternation in ambiguous figures (Flamm & Bergum, 1977) and binocular rivalry (Rainwater & Cogan, 1975). In contrast, spontaneous blinks only facilitate perceptual alternation under some limited conditions with ambiguous figures (Nakatani et al., 2011) and binocular rivalry (Kalisvaart & Goossens, 2013). Other research found no facilitatory effects of spontaneous blinks on perceptual alternation in ambiguous figures (Brych et al., 2021; Ito et al., 2003) and binocular rivalry (van Dam & van Ee, 2005), while some studies report that spontaneous blinks suppress perceptual alternation in ambiguous figures (Murali & Händel, 2024) and in a variant of binocular rivalry (continuous flash suppression) (Van Opstal, De Loof, Verguts, & Cleeremans, 2016). These discrepancies, due to different blink types, tasks, and experimental conditions, hinder a coherent understanding of the relationship between blinking and perceptual alternation. 
To address these inconsistencies, we examine whether spontaneous and voluntary blinks facilitate or suppress perceptual alternation within the same multistable perception task. This approach may also elucidate the role of retinal transients in perceptual alternation. Given that both blink types produce similar retinal changes, any differential effects could suggest that these changes are not pivotal in modulating alternation. Additionally, we directly examine the effects of retinal changes by introducing a physical blackout, designed to mimic the timing and duration of natural blinks. 
Our second approach utilizes continuous flash suppression (CFS) to investigate how blinking influences multistable perception. CFS involves presenting a high-contrast dynamic pattern (the suppressor) to one eye, causing robust interocular suppression that can render even a salient stimulus presented to the other eye (the target) invisible for an extended duration (Tsuchiya & Koch, 2005) (Figure 1). We employed a variant of this method, known as breaking CFS (b-CFS) (Gayet, Van Der Stigchel, & Paffen, 2014; Jiang, Costello, & He, 2007; Stein, 2019), which offers better prediction and control of perceptual alternation. In b-CFS, the suppressor is first presented alone to establish exclusive dominance. The target is then introduced with gradually increasing its contrast from zero to maximum. This sequence ensures that the suppressor is always perceived initially, with the target breaking through suppression as its contrast is sufficiently increased. This consistent perceptual change across trials makes b-CFS more suitable for our study than binocular rivalry and ambiguous figures, where perceptual sequences are unpredictable. 
Figure 1.
 
Stimulus sequence (left side) and typical percept (right side) during the b-CFS paradigm used in the present study. The suppressor was a dynamic Mondrian pattern that was refreshed at 10 Hz and presented to the right eye. The target was a Gabor patch (oriented clockwise or counterclockwise) and presented to the left eye. The observer's task was to indicate the target orientation as soon as possible.
Figure 1.
 
Stimulus sequence (left side) and typical percept (right side) during the b-CFS paradigm used in the present study. The suppressor was a dynamic Mondrian pattern that was refreshed at 10 Hz and presented to the right eye. The target was a Gabor patch (oriented clockwise or counterclockwise) and presented to the left eye. The observer's task was to indicate the target orientation as soon as possible.
Using b-CFS, we investigate whether spontaneous and voluntary blinking accelerates or decelerates perceptual alternation from the suppressor to the target. It is important to note that this represents a unidirectional form of perceptual alternation, distinct from the bidirectional alternation in previous studies (e.g., Flamm & Bergum, 1977; Rainwater & Cogan, 1975). While b-CFS involves contrast manipulation, unlike the constant stimuli in binocular rivalry and ambiguous figures, the consistent manipulation of target contrast across stimulus conditions enables us to study the effects of blinking. Moreover, the gradual increase in target contrast provides observers with ample opportunity for spontaneous blinking and gives experimenters the opportunity to manipulate the observer's voluntary blinks. 
Our third approach is concerned with the effects of perceptual alternation on blinking. We compare blink frequency following perceptual alternation in two types of trials: those with a spontaneous blink before alternation (spontaneous-blink trials) and those without preceding blinks (no-blink trials). This method is critical for addressing confounding factors. During visual tasks, blinks are often suppressed to prevent information loss, leading to a posttask increase in spontaneous blinking due to disinhibition (Fogarty & Stern, 1989). Moreover, suppressing blinks can cause eye dryness, potentially triggering more blinks. These factors could suggest a stronger link between blinking and perceptual alternation than actually exists. We hypothesize that these confounding effects would be less pronounced in spontaneous-blink trials. Spontaneous blinking lubricates the eyes, and its occurrence indicates weaker blink suppression. Therefore, if confounding factors primarily cause increased blinking, we expect a lower blink frequency after perceptual alternation in spontaneous-blink trials. Conversely, if perceptual alternation itself induces blinking, a high blink frequency following perceptual alternation should be observed in both no-blink and spontaneous-blink trials. We tested this prediction to clarify the relationship between perceptual alternation and blinking. 
In summary, by combining these three approaches—comparing the effects of spontaneous and voluntary blinks, utilizing b-CFS, and examining the effects of perceptual alternation on blinking—our study aims to provide a comprehensive understanding of the intricate relationship between blinking and cognitive processing in multistable perception tasks. 
Experiment 1
Methods
Observers
Eight observers participated in Experiment 1 (five females; age, mean = 21.8, SD = 2.0, range = 19–25), but the results from six observers were analyzed. Two observers were excluded from the analysis because they reported intentionally blinking during stimulus observation to prevent their eyes from getting dry or to facilitate target discrimination. All observers in this and the following experiments had normal or corrected-to-normal visual acuity and normal color vision, as assessed with Ishihara pseudoisochromatic plates. Before the experiment, written consent was obtained from the observers after the procedures were explained. The experiments were conducted following the Declaration of Helsinki and were approved by the university's Human Research Ethics Committee. 
Apparatus
A haploscope with three CRT displays located in the front, left, and right of the observer was used to present the stimulus (see Figure 1 of Abe, Kimura, & Goryo, 2011). The stimuli for the left and right eyes were presented on the left and right displays, respectively (Flex Scan T566, Eizo Corp., Ishikawa, Japan; pixel resolution, 1,024 × 768; refresh rate, 100 Hz). They were generated using a VSG2/5 graphics card with 15-bit color resolution (Cambridge Research System Ltd., Cambridge, UK) controlled with MATLAB (The MathWorks Inc., Natick, MA, USA). The observer's response was collected with a CT3 response box (Cambridge Research System Ltd.). 
The equipment was adjusted for each observer before every session. The observer was first asked to look at the center of the front display (Press View 17SR, Radius Ltd., Tokyo, Japan; a pixel resolution of 1,024 × 768 and a refresh rate of 75 Hz), which was used to obtain the observer's natural accommodation and convergence. Then, the stimuli on the left and right displays were superimposed over the stimulus on the front display at two beam splitters in front of the observer's eyes. After the adjustment, the front display was turned off to establish a physical blackout in Session 1 (see Procedure). 
The observer's head was fixed with a chin and forehead rest, with a viewing distance of 57 cm. An infrared eye tracker (MCU400, Arrington Research Inc., Scottsdale, AZ, USA) was used to record pupil diameter and eye position from the observer's left eye with a sampling rate of 90 Hz. 
Stimuli
The suppressor for CFS was a dynamic Mondrian pattern subtending 7° × 7° and refreshed at 10 Hz (Figure 1). Each pattern consisted of 500 rectangles, superimposed upon one another, with their short and long sides independently varying between 0.2° and 1.2°. The center coordinates of each rectangle were randomly assigned within the 7° × 7° frame. The color of each rectangle was one of eight colors, each defined by maximum RGB values: red, green, blue, cyan, magenta, yellow, black, and white. The target stimulus was a Gabor patch (σ = 1°, 1.06 cpd, clockwise or counterclockwise orientation). The suppressor and target were always presented to the right and left eyes, respectively, in the center of an achromatic background field (x = 0.313, y = 0.329, 6 cd/m²), subtending 12° × 12°. A fixation pattern composed of four white crosses (x = 0.313, y = 0.329, 15 cd/m²) was presented to support binocular fusion. Each cross subtended 1.5° × 1.5° and was placed 4.5° away from the center of the background field in the left, right, upper, and lower directions. The observers were instructed to look at the center of the pattern during the experiment. 
The eye receiving the suppressor and target was fixed across experiments, which allowed us to monitor the target-receiving (left) eye with an eye camera. While presenting the suppressor to the dominant eye could enhance suppression, previous research (Shimizu & Kimura, 2020) has shown that strong suppression occurs even when the suppressor is presented to the nondominant eye at lower swapping frequencies. 
Procedure
To investigate the effects of voluntary blinking, observers were instructed to intentionally blink while observing stimuli. However, this instruction could influence their natural, spontaneous blinking behavior. To minimize potential confounding effects, the experiment was divided into two separate sessions conducted on different days. In Session 1, observers were informed that an eye camera would be set up to monitor their gaze position. Then, in Session 2, they were asked to blink voluntarily in specific conditions. After completing Session 2, they were notified that their blinking had been measured during the experiment. They were then asked to confirm the use of their blink data for the analysis. 
Session 1 was designed to investigate the effects of spontaneous blinking and blackout. The main stimulus condition in this session was the b-CFS condition. Trials for this and other stimulus conditions were initiated by the observer's button press. The Mondrian suppressor was presented first to establish steady suppression, and the target was presented 0.5 s later. The target contrast was ramped up from zero to a maximum of 0.8 over 3.0 s, maintained at that level for 4.0 s, and then ramped down over another 3.0 s (for a total duration of 10.0 s). The suppressor was turned off 0.5 s after the target offset, followed by dynamic white noise (7.10° × 7.10°, 10 Hz) for 1.5 s to minimize carryover effects such as afterimages. Observers were asked to indicate the target orientation (clockwise or counterclockwise) as soon as possible using a CT3 response box, where pressing the right button corresponded to clockwise responses and the left button to counterclockwise responses. There was no time limit for the observers’ responses. The time required for orientation discrimination (discrimination response time, DRT), used as an index of the time required for a perceptual alternation, was measured from target onset to the observer's response. Auditory feedback was provided at the end of each trial. The trials in the b-CFS condition were classified into no-blink and spontaneous-blink trials according to whether a blink was observed at a specific time (see Trial classification). 
In Session 1, blackout and no-suppression conditions were included, which were variations of the b-CFS condition. In the blackout condition, a brief physical blackout was introduced by temporarily darkening the stimulus displays. This blackout began 1.0 s after the target onset and lasted 0.3 s, a duration chosen to match the typical length of a spontaneous blink (Stern et al., 1984). During this period, both the target and the Mondrian suppressor were not visible, effectively interrupting the visual input. The no-suppression condition was included to measure DRT in the absence of any visual suppression. In this condition, only the target was presented without the Mondrian suppressor, allowing the observer to view the target directly from the moment it appeared. This condition allowed us to assess how quickly observers could discriminate the target's orientation when no visual suppression was present. 
Session 2 was designed to investigate the effect of voluntary blinking. The stimulus conditions consisted of the voluntary-blink condition and three control conditions, which were variations of the b-CFS condition. In the voluntary-blink condition, observers were instructed to blink immediately after a cue. The cue involved a color change in the fixation pattern from white to red (x = 0.642, y = 0.337, 12 cd/m²) and was presented 1.0 s after the target onset. The three control conditions were included to examine the effects of the mere presence of the cue and conducting dual tasks (target discrimination and intentional blinking). In these control conditions, both the instructions to observers—whether to respond to or disregard the cue—and the presentation of the cue were systematically manipulated. Specifically, in one control condition, observers were instructed to blink in response to the cue, but the cue was not presented. In another condition, observers were instructed to disregard the cue, and the cue was presented. In the final condition, observers were instructed to disregard the cue, and the cue was not presented. Prior to each trial, observers were informed whether they were supposed to respond or disregard the cue. 
Experiments were conducted in a dark room. Before the experiment, the observers were dark-adapted for 5 min and then light-adapted to the background field for 2 min. All stimulus conditions in both Sessions 1 and 2 were tested 36 times. 
Blink measurement
The experimenter monitored the observer's pupil diameter and eye position online, using a video image of the eye from the infrared camera. The occurrence of blinks during stimulus observation was analyzed offline using custom-made software in MATLAB, based on the horizontal pupil diameter. The onset of a blink causes a rapid decrease in the pupil diameter. Therefore, a criterion was set for the rate of change in velocity (0.065/ms), and the blink onset was defined as the point at which the change in pupil diameter exceeded this criterion. The blink offset was defined as the point at which the pupil diameter had returned to its pre-onset value. As the final check, all data were inspected visually by the experimenter, and both the onset and offset timing of blinks were adjusted manually, if necessary. 
Blinks were classified as spontaneous if they occurred at a time other than that instructed by the experimenter. 
Trial classification
The trials in some stimulus conditions were classified according to whether spontaneous blinks occurred between target onset and the target discrimination response. First, incorrect response trials were excluded from classification in both sessions (3.2% in total in Session 1 and 1.9% in Session 2, respectively). In the b-CFS condition in Session 1, trials in which a spontaneous blink occurred before target discrimination were classified as spontaneous-blink trials, while the others were classified as no-blink trials. Trials in which two or more spontaneous blinks occurred before target discrimination were excluded from classification (2.5% of the b-CFS condition) to equate the number of retinal image changes across the classified trials. 
The same classification procedure was applied to the three control conditions in Session 2. Again, trials in which two or more spontaneous blinks occurred before target discrimination were excluded (3.3% in total). Despite our precautions, analysis of the DRT in these control conditions showed no significant difference, F(2, 17) = 1.085, p = 0.375, \(\eta _p^2\) = 0.03, suggesting that target discrimination did not vary depending on the experimental procedure used to produce voluntary blinks. Therefore, trials from these conditions were pooled in the subsequent analysis. 
The blackout and voluntary-blink conditions were simply classified as blackout and voluntary-blink trials, respectively. In the blackout condition, trials in which one or more spontaneous blinks occurred before target discrimination were excluded (16.7%). In the voluntary-blink condition, trials in which a spontaneous blink occurred before the cue and where two or more blinks occurred before target discrimination were excluded (5.9%). 
Data analysis
(a) DRT for classified trials: The DRT was analyzed using both observer-based frequentist statistics and trial-based Bayesian statistics. In the frequentist approach, the DRT was analyzed with a one-way repeated-measures analysis of variance (ANOVA), with trial type serving as the within-participant variable. Trial types included no-suppression, no-blink, spontaneous-blink, and blackout trials for Session 1 and no-blink, spontaneous-blink, and voluntary-blink trials for Session 2. If the main effect of trial type was statistically significant, multiple comparisons using Bonferroni correction were conducted. ANOVA was performed using the anovakun function (Iseki, 2023) in R (R Foundation for Statistical Computing, Vienna, Austria). 
In the Bayesian approach, the DRT was analyzed using generalized linear mixed-effects models (GLMMs) (Baayen, Davidson, & Bates, 2008). The frequency of blinks and their effect on task performance are likely to vary within as well as across individual observers. Therefore, mixed-effects models that include all trials for each observer provide a more accurate and reliable estimate of the effects. Moreover, GLMMs allow us to model the relationships between independent variables (trial type) and the dependent variable (DRT) while appropriately accounting for the distribution of the DRT. By assuming that the DRT follows either a Gamma or Inverse Gaussian distribution to account for the positive skew in the DRT distribution, we can analyze the relationship between trial type and DRT without nonlinearly transforming the DRT to meet the Gaussian assumptions of normality and homoscedasticity (Lo & Andrews, 2015). In the analysis, the link function of the GLMMs was assumed to be identity (Lo & Andrews, 2015). The Inverse Gaussian was chosen for the DRT distribution because it provided a better fit to the observed DRT than the Gamma function as determined by Akaike information criterion (AIC) and Bayesian information criterion (BIC) statistics. 
The GLMMs were computed with a random effect of observer and a fixed effect of trial type. The composition of trial types in Sessions 1 and 2 was the same as that used in the ANOVA. Initially, we examined the full model, which included the random intercept for observer and the by-observer random slope for trial type, in addition to the fixed effect. However, as the full model did not always converge, we adopted a simplified model without the random slope. This simplified model and the reduced version were then compared using Bayes factors (BFs) calculated following the method proposed by Wagenmakers (2007). Multiple comparisons were performed using the emmeans package (Lenth, 2023). 
We report the BFs in this and the following experiments as follows (Hesselmann, Darcy, Rothkirch, & Sterzer, 2018; Shimizu & Kimura, 2023). First, we identified the model with the highest BF compared to an intercept-only model and considered this the best model, setting its BF to 1. We then recalibrated the BFs of all other models relative to this best model. With this procedure, the BF of each model indicates the extent to which the data are more consistent with the best model than with the model being considered. 
(b) Discrimination response frequency after blinking: If blinking facilitates or suppresses perceptual alternation (target discrimination), we would expect the frequency of target discrimination responses to increase or decrease shortly after a blink. To test the significance of any change in the observed discrimination response frequency (ODF), we compared the ODF with the random discrimination response frequency (RDF), which assumes that target discrimination occurs independently of blinking. 
Deriving the ODF: 
  • 1. Calculate relative times: For each blink trial, we calculated the time of each target discrimination response relative to the blink offset.
  • 2. Pool data: These relative times were pooled across all trials for each type of blink (spontaneous or voluntary).
  • 3. Determine frequency: We divided the number of target discrimination responses within each 0.1-s time bin by the total number of discrimination responses to obtain the ODF over time.
Deriving the RDF: 
  • 1. Random pairing: For each blink in the spontaneous- or voluntary-blink trials, we randomly paired it with a target discrimination response from a no-blink trial within the same session.
  • 2. Calculate relative times: We calculated the time difference between the blink offset and the randomly paired target discrimination response, just as we did for the observed data.
  • 3. Repeat for all blinks: Steps 1 and 2 were repeated for all blinks in the spontaneous- or voluntary-blink trials.
  • 4. Iterate simulations: We repeated Steps 1–3 a total of 1,000 times to generate a distribution under the assumption that blinking and target discrimination are independent.
  • 5. Calculate 95% intervals: For each time point after the blink offset, we calculated the 95% interval (and the upper and lower limits, RDF limits) of the discrimination response frequency from these simulations.
  • 6. Compare ODF and RDF: We compared the ODF with the RDF limits to assess the significance of any changes in discrimination response frequency following a blink.
(c) Blink frequency after target discrimination: We investigated the effects of perceptual alternation (target discrimination) on blinking in a manner similar to our previous analysis of how blinking affects perceptual alternation. Specifically, we calculated observed blink frequency (OBF) as a function of time relative to each target discrimination response. This allowed us to see how blink frequency changed following perceptual alternation events. 
To determine if any changes in blink frequency were significant, we compared the OBF with the random blink frequency (RBF), which assumes that blinks occur independently of target discrimination. The RBF was generated by randomly pairing each target discrimination response from one trial with a blink from another trial within the same pool of classified trials. By comparing the OBF and RBF in a similar manner to the comparison between the ODF and RDF, we could assess the significance of any changes in blink frequency following target discrimination. 
Results and discussion
DRT for classified trials
Figure 2a shows the mean DRT for different classified trials in Session 1: no-suppression, no-blink, spontaneous-blink, and blackout trials. A one-way repeated-measures ANOVA showed a significant main effect of the trial type, F(3, 23) = 33.801, p < 0.001, \(\eta _p^2\) = 0.87. Multiple comparisons using Bonferroni correction showed that the DRT in the no-suppression condition was shorter than that in the other three conditions, indicating the effects of the suppressor (ts > 4.956, ps < 0.004). Moreover, the DRT in spontaneous-blink trials was longer than that in no-blink trials, t(5) = 4.022, p = 0.028, d = 1.84, and blackout trials, t(5) = 4.092, p = 0.028, d = 1.69. However, there was no significant difference in DRT between blackout and no-blink trials, t(5) = 0.144, p = 0.891, d = 0.20. 
Figure 2.
 
Mean discrimination response time (DRT) for different classified trials in Experiment 1: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 2.
 
Mean discrimination response time (DRT) for different classified trials in Experiment 1: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
The DRTs were also analyzed with the Bayesian approach using GLMMs. The model that predicted the data best (best model) was the one with a fixed effect of trial type and a random effect of observer. The other (intercept-only) model was much less consistent with the data than the best model (BF > 100). The results of multiple comparisons were consistent with those obtained by the ANOVA; that is, spontaneous-blink trials revealed a significant increase in DRT compared to no-blink trials (z = 7.786, p < 0.001) and blackout trials (z = 7.830, p < 0.001). 
Figure 2b shows the mean DRT for different classified trials in Session 2: no-blink, spontaneous-blink, and voluntary-blink trials. A one-way repeated-measures ANOVA showed a significant main effect of the trial type, F(2, 17) = 19.622, p = 0.001, \(\eta _p^2\)= 0.80. Multiple comparisons using Bonferroni correction showed that the DRT in voluntary-blink trials was shorter than that in no-blink trials, t(5) = 3.826, p = 0.012, d = 1.58, and spontaneous-blink trials, t(5) = 5.153, p = 0.011, d = 2.62. The DRT in spontaneous-blink trials was longer than that in no-blink trials, t(5) = 3.309, p = 0.021, d = 0.89. 
Consistent with these results, the analysis using Bayesian GLMMs showed that the best model included a fixed effect of trial type and a random effect of observer. The other (intercept-only) model was much less consistent with the data than the best model (BF > 100). Multiple comparisons showed that voluntary-blink trials revealed a significant decrease in DRT compared to no-blink trials (z = 13.721, p < 0.001) and spontaneous-blink trials (z = 14.261, p < 0.001). Moreover, the DRT in spontaneous-blink trials was longer than that in no-blink trials (z = 5.165, p < 0.001). In the GLMM analysis, we accounted for the skewness in the DRT distribution by assuming that the DRT follows an Inverse Gaussian distribution. In contrast, in the ANOVA, we did not apply a nonlinear transformation to satisfy the assumption of normality (Lo & Andrews, 2015). The finding that both approaches yielded consistent results suggests that the conclusions drawn from the ANOVA were not substantially affected by violations of normality (see also Knief & Forstmeier, 2021)
Collectively, the results suggest that blinking modulates perceptual alternation, and the effects are functionally distinct for different types of blinks; the DRT was longer in spontaneous-blink trials and shorter in voluntary-blink trials. In contrast, a physical blackout of the stimulus displays did not produce observable changes in DRT. These findings provide converging evidence that the effects of blinking are not mediated by retinal transients induced by blinking. 
Discrimination response frequency after blinking
To further investigate the effects of blinking on perceptual alternation, the frequency of the target discrimination response following a blink was examined (Figure 3). If the shorter mean DRT in voluntary-blink trials (Figure 2b) reflects a facilitation of perceptual alternation by blinking, the target would be successfully discriminated shortly after blinking. The results support this prediction (Figure 3b). In voluntary-blink trials, the target was discriminated within 1.0 s after blinking in 97.2% of the trials. 
Figure 3.
 
Frequency of the target discrimination response after blinking in Experiment 1. The histogram shows the observed discrimination response frequency (ODF) as a function of time after the blink in (a) spontaneous-blink trials and (b) voluntary-blink trials. The shaded area in each panel represents the 95% interval of the random discrimination response frequency (RDF), and the upper and lower limits are shown by dotted lines (RDF limits). Uniformly colored bars illustrate instances where the ODF exceeds or falls short of the limits, indicating a significant increase or decrease in the frequency of target discrimination at those times.
Figure 3.
 
Frequency of the target discrimination response after blinking in Experiment 1. The histogram shows the observed discrimination response frequency (ODF) as a function of time after the blink in (a) spontaneous-blink trials and (b) voluntary-blink trials. The shaded area in each panel represents the 95% interval of the random discrimination response frequency (RDF), and the upper and lower limits are shown by dotted lines (RDF limits). Uniformly colored bars illustrate instances where the ODF exceeds or falls short of the limits, indicating a significant increase or decrease in the frequency of target discrimination at those times.
Moreover, comparison with the upper limit of the 95% interval of the random discrimination response frequency (RDF limit, represented by a dotted line in Figure 3), which was calculated under the assumption that target discrimination was independent of blinking, revealed a significant increase in ODF within the time range of 0.2 to 0.8 s after voluntary blinks (Figure 3b). It is crucial to understand in this analysis that the RDF, especially that derived from voluntary blinks, might exhibit a peak at some time after blinking, even if the target discrimination occurs independently of the blinks. This phenomenon arises because both voluntary blinking and target discrimination are temporally dependent on the time elapsed after the target onset. The blink cue was presented in a time-locked fashion to the target onset, while target discrimination was likely to occur within a specific interval after the target onset. Therefore, for the increase in ODF to be considered evidence of significant facilitation, it must exceed the upper RDF limit. 
If the longer mean DRT in the spontaneous-blink trials (Figures 2a, 2b) reflects the suppression of perceptual alternation, the target would be successfully discriminated less frequently immediately after the blinks (suppression) and perhaps more frequently after the suppression weakened (disinhibition). Inconsistent with this prediction, visual inspection of the results indicated no systematic changes in ODF in spontaneous-blink trials (Figure 3a). The frequency was generally low and broadly distributed over several seconds after the blinking. Comparison of the ODF with the upper RDF limit (dotted line) in each time bin indicated that the ODF sporadically exceeded the limit, which was also consistent with the absence of systematic changes in discrimination response frequency. The ODFs in Sessions 1 and 2 were also analyzed separately, but the results were essentially the same. These results indicate that spontaneous blinking itself does not exert a strong suppressive effect on perceptual alternation. The apparent discrepancy between the results of the DRT and ODF will also be confirmed in Experiment 2, and its reconciliation will be discussed in the General Discussion section. 
Blink frequency after target discrimination
To investigate whether perceptual alternation affects spontaneous blinking, blink frequency after target discrimination was compared to the 95% limit of the random blink frequency (RBF limit), which was calculated under the assumption that spontaneous blinks occurred independently of perceptual alternation (Figure 4). The results showed that OBF significantly increased in the time range of 0.3 to 1.0 s after target discrimination in no-blink trials (Figure 4a). In contrast, the OBF in spontaneous-blink trials within the same time range was close to the upper RBF limit (dotted line), and the comparison in each time bin indicated that the OBF rarely exceeded the limit (Figure 4b). 
Figure 4.
 
Blink frequency after target discrimination in Experiment 1. The histogram shows observed blink frequency (OBF) as a function of time after target discrimination in (a) no-blink trials and (b) spontaneous-blink trials. The shaded area in each panel represents the 95% interval of the random blink frequency (RBF), and the upper and lower limits (RBF limits) are shown by dotted lines (the lower limit for the spontaneous-blink trials was very close to zero). Uniformly colored bars illustrate instances where the OBF exceeds the upper RBF limit, indicating a significant increase in blink frequency at those times.
Figure 4.
 
Blink frequency after target discrimination in Experiment 1. The histogram shows observed blink frequency (OBF) as a function of time after target discrimination in (a) no-blink trials and (b) spontaneous-blink trials. The shaded area in each panel represents the 95% interval of the random blink frequency (RBF), and the upper and lower limits (RBF limits) are shown by dotted lines (the lower limit for the spontaneous-blink trials was very close to zero). Uniformly colored bars illustrate instances where the OBF exceeds the upper RBF limit, indicating a significant increase in blink frequency at those times.
The results were inconclusive and might have been influenced by observers’ awareness of the manipulation involving voluntary blinks in Session 2, as most of the trials analyzed were sampled in this session. Therefore, with a modified procedure aimed at increasing the sample size of no- and spontaneous-blink trials in Session 1, we further explored the effects of perceptual alternation on blink frequency in Experiment 2
Experiment 2
The aim of Experiment 2 was to explore the mechanism underlying the facilitatory effect of voluntary blinking on perceptual alternation. The differential effects of spontaneous and voluntary blinking found in Experiment 1 suggest that voluntary eyelid movements, rather than blinking itself, could be a critical factor in the facilitatory effect. Thus, we tested the possibility that voluntary eye-widening, another type of voluntary eyelid movement, facilitates perceptual alternation. 
Another aim of Experiment 2 was to examine the replicability of some of the important findings from Experiment 1 using modified procedures (see Methods). 
Methods
The same methods as in Experiment 1 were used with a few exceptions. In Experiment 2, the sample size, which included the number of observers and trials, was determined based on subsampling simulations (Baker et al., 2021) using data from Session 2 of Experiment 1. In this session, the discrimination response time in no-blink trials was compared to that in spontaneous-blink and voluntary-blink trials. The effect size of the main effect of trial types was estimated to be \(\eta _p^2\) = 0.80. We then performed an ANOVA on the randomly subsampled data set 10,000 times and counted the number of times the main effect was statistically significant. The results indicated that a statistical power of over 0.80 would require four observers and 20 trials for each condition. Based on these results, we set the number of trials for the b-CFS condition for each observer to 88 trials. In Experiment 1, approximately 25% of trials from the b-CFS condition were classified as spontaneous-blink trials, and thus we expected that 88 trials would produce a sufficient number of spontaneous trials (i.e., 22 trials). The number of trials for the other conditions was set to 44 trials in order to balance the total number of trials between sessions. We recruited five new naive observers for Experiment 2 (two females; age, mean = 19.8, SD = 0.8, range = 19–21). 
Experiment 2 was also divided into two sessions, similar to Experiment 1. However, both Sessions 1 and 2 were repeated on different days to ensure that the stimulus conditions were repeated enough times to obtain the required number of trials. In Session 1, only the b-CFS and blackout conditions were tested. The b-CFS condition was conducted similarly to Experiment 1, but the number of repetitions was increased to 44 times per day (88 times in total). In the blackout condition, the starting time and duration of the physical blackout were adjusted to 1.5 s after the target onset and 0.4 s, respectively. This adjustment was made to resemble the voluntary blinks observed in Experiment 1. The blackout condition was repeated 22 times per day (44 times in total). 
In Session 2, the effects of voluntary blinking and eye-widening were investigated. This session included the voluntary-blink, eye-widening, and cue-disregarding conditions. Observers were asked to respond to a cue presented 1.0 s after target onset in three distinct ways. In the voluntary-blink condition, observers blinked immediately following the cue. In the eye-widening condition, observers widened their eyes immediately after the cue. In the cue-disregarding condition, observers disregarded the cue. The experimenter monitored the observer's response to the cue in each trial online using the video image of the eye from the infrared camera. The same trial was repeated on rare occasions when the response deviated from the instructions. Each of the three conditions was conducted 22 times per day, totaling 44 times. 
Trial classification
The trials were classified similarly as in Experiment 1. First, incorrect response trials were excluded from classification in both sessions (2.0% in total in Session 1 and 1.5% in Session 2, respectively). The trials in the b-CFS condition in Session 1 and the cue- disregarding condition in Session 2 were classified into no-blink and spontaneous-blink trials. However, the spontaneous-blink trials in Session 2 were very few (31 trials from all observers) and thus excluded from the analysis. Trials in which two or more spontaneous blinks occurred before target discrimination were excluded from classification (1.4% of the b-CFS condition and 0.9% of the cue-disregarding condition, respectively). The effects of perceptual alternation on the occurrence of spontaneous blinks were analyzed using data only from Session 1. 
The trials in the blackout, voluntary-blink, and eye-widening conditions were classified as blackout, voluntary-blink, and eye-widening trials, respectively. In the blackout and eye-widening conditions, trials in which one or more spontaneous blinks occurred before target discrimination were excluded (15.3% and 3.5%, respectively). In the voluntary-blink condition, the same criterion for data exclusion as in Experiment 1 was applied, but none of the trials were excluded. 
Data analysis
The DRT in Experiment 2 was also analyzed using both ANOVA and GLMMs. A one-way repeated-measures ANOVA was used, with trial type serving as the within-participant variable. Trial types included no-blink, spontaneous-blink, blackout trials for Session 1 and no-blink, voluntary-blink, and eye-widening trials for Session 2. 
In the Bayesian approach, Inverse Gaussian GLMMs with the identity link function were used to analyze the DRT as in Experiment 1. The models were computed with random effects of observer and daily session (day) and a fixed effect of trial type. The composition of trial types in Sessions 1 and 2 was the same as that used in ANOVA. We first examined the full model, including all random effects with observer and day, in addition to the fixed effect. However, as the model did not always converge, we adopted a simplified model without the random slopes. This and reduced versions of the model were compared based on BFs. We report the BFs of all tested models in Tables 1 and 2 (see the Results section). 
Table 1.
 
Results of Bayes factor analysis of the DRT in Session 1 using GLMMs in Experiment 2.
Table 1.
 
Results of Bayes factor analysis of the DRT in Session 1 using GLMMs in Experiment 2.
Table 2.
 
Results of Bayes factor analysis of the DRT in Session 2 using GLMMs in Experiment 2.
Table 2.
 
Results of Bayes factor analysis of the DRT in Session 2 using GLMMs in Experiment 2.
Both the ODF after blinking and the OBF after target discrimination were analyzed similarly as in Experiment 1. However, it was difficult to determine the timing of eye-widening offline using the pupil diameter. Any changes in pupil diameter and eye position caused by eye-widening were minimal, if observable at all. Consequently, determining the onset of eye-widening was not possible in most trials, even though eye-widening was confirmed to occur in real time during the measurement. Therefore, an analysis of the ODF after eye-widening was not conducted. 
Results and discussion
DRT for classified trials
Figure 5a shows the DRT for different classified trials in Session 1. A one-way repeated-measures ANOVA showed a significant main effect of the trial type, F(2, 14) = 17.848, p = 0.001, \(\eta _{p\ }^2\)= 0.82. Multiple comparisons using Bonferroni correction showed that the DRT in spontaneous-blink trials was longer than that in no-blink trials, t(4) = 9.863, p = 0.002, d = 2.26, and blackout trials, t(4) = 3.267, p = 0.031, d = 1.62. The difference in DRT between blackout and no-blink trials was not statistically significant, t(4) = 1.487, p = 0.211, d = 0.20. 
Figure 5.
 
Mean DRT for different classified trials in Experiment 2: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 5.
 
Mean DRT for different classified trials in Experiment 2: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
In the analysis of the DRT using Bayesian GLMMs, the best model was the one with a fixed effect of trial type and two random effects of observer and day (daily session) (Table 1). The other models were less consistent with the data than the best model (BFs > 9.9). Multiple comparisons revealed that the DRT in spontaneous-blink trials was significantly longer than in no-blink trials (z = 10.513, p < 0.001) and blackout trials (z = 7.432, p < 0.001). 
Figure 5b shows the mean DRT for different classified trials in Session 2. A one-way repeated-measures ANOVA showed a significant main effect of the trial type, F(2, 14) = 13.287, p = 0.003, \(\eta _{p\ }^2\)= 0.77. Multiple comparisons using Bonferroni correction showed that the DRT in voluntary-blink trials was shorter than in no-blink trials, t(4) = 4.701, p = 0.028, d = 3.38. Similarly, the DRT in eye-widening trials was shorter than in no-blink trials, t(4) = 2.912, p = 0.044, d = 2.42. In addition, the difference in DRT between voluntary-blink and eye-widening trials was not statistically significant, t(4) = 2.593, p = 0.061, d = 0.63. 
Consistent with these results, the analysis using Bayesian GLMMs showed that the best model included a fixed effect of trial type and two random effects of observer and day (Table 2). The other models were less consistent with the data than the best model (BFs > 22.9). Multiple comparisons revealed a significant decrease in DRT in voluntary-blink trials compared to no-blink trials (z = 22.177, p < 0.001). Similarly, the DRT in eye-widening trials was shorter than in no-blink trials (z = 17.232, p < 0.001). 
These results reaffirmed the findings of Experiment 1 (i.e., the longer DRT in spontaneous-blink trials, the shorter DRT in voluntary-blink trials, and no detectable change in DRT in blackout trials). One new finding was that voluntary eye-widening led to a shorter DRT by an amount similar to that of voluntary blinking, suggesting a facilitatory effect on perceptual alternation. Eye-widening could cause minute eye movements because the eyelid muscle is connected to the extraocular muscles that control the movement of the eyeball (Collewijn, Steen, & Steinman, 1985; Evinger, Shaw, Peck, Manning, & Baker, 1984). Such eye movements can produce transient changes in the retinal images. However, if retinal transients associated with eyelid movements were critical in modulating perceptual alternation, spontaneous blinking would have exhibited the same modulating effect as that of voluntary eyelid movements (i.e., voluntary blinking and eye-widening). Thus, these findings provide evidence against the interpretation that retinal transients account for the effects of voluntary eyelid movements. 
Discrimination response frequency after blinking
To investigate whether blinking modulates perceptual alternation, the temporal relationship between blinking and target discrimination was examined (Figure 6). The results of the voluntary-blink trials (Figure 6b) confirmed the findings of Experiment 1. The target was successfully discriminated immediately after blinking. The comparison with the upper RDF limit (dotted line) revealed that the ODF significantly increased in the time range of 0.2 to 0.6 s after voluntary blinks. These findings further support the idea that voluntary blinking facilitates perceptual alternation. 
Figure 6.
 
Frequency of the target discrimination response after blinking in Experiment 2. Other aspects are the same as those in Figure 3.
Figure 6.
 
Frequency of the target discrimination response after blinking in Experiment 2. Other aspects are the same as those in Figure 3.
As described in the Methods section, the same analysis could not be conducted for the results of eye-widening trials. However, the facilitatory effects of eye-widening are similar to those of voluntary blinking. Within 2.0 s after the offset of the blink cue, the target was successfully discriminated in 93.5% of the voluntary-blink trials and 92.4% of the eye-widening trials. 
The results of the spontaneous-blink trials differed from those of Experiment 1. As in Experiment 1, the ODF was broadly distributed over several seconds after blinking (Figure 6a). However, the ODF was extremely low up to 1.3 s after blinking and exceeded the upper RDF limit in the time range of 2.5 to 3.2 s after blinks. This temporal relationship might be considered consistent with the suppression and subsequent disinhibition of target discrimination caused by spontaneous blinking. However, the time scale for the suppressive effect was very long, and the temporal relationship could not be replicated in different experiments (see also Figures 3a, 3b). This issue is discussed in the General Discussion section. 
Blink frequency after target discrimination
The blink frequency after target discrimination showed a similar distribution in both no-blink (Figure 7a) and spontaneous-blink trials (Figure 7b). Across different classified trials, the OBF consistently exceeded the upper RBF limit, resulting in a significant increase around the time range of 0.5 to 0.9 s after target discrimination. Time-locked increases in blink frequency suggest that perceptual alternation induces spontaneous blinking. 
Figure 7.
 
Blink frequency after target discrimination in Experiment 2. Other aspects are the same as those in Figure 4.
Figure 7.
 
Blink frequency after target discrimination in Experiment 2. Other aspects are the same as those in Figure 4.
General discussion
This study investigated the interaction between blinking and perceptual alternation during CFS to explore the functional roles of blinking in the cognitive processing of visual information. It is important to note that the perceptual alternation in this study is unidirectional (i.e., from the suppressor to the target). One of the main objectives of this study was to investigate whether and how different types of blinks affect perceptual alternation. The results clearly demonstrate the differential effects of voluntary and spontaneous blinks. The results from the voluntary-blink trials (i.e., the shorter mean DRT [Figures 2b and 5b] and the more frequent target discrimination shortly after blinking [Figures 3b and 6b]) suggest that voluntary blinking facilitates perceptual alternation. In contrast, the longer mean DRT in the spontaneous-blink trials (Figures 2 and 5a) suggests a suppressive effect of spontaneous blinking. However, its modulating effect on the ODF was not evident or replicable (Figures 3a and 6a). The present study also shows that a physical blackout, which had timing and duration comparable to blinks, did not significantly affect perceptual alternation (Figures 2a and 5a). Moreover, voluntary eye-widening led to shorter mean DRT (Figure 5b), suggesting a facilitatory effect on perceptual alternation. Collectively, these findings indicate that retinal transients caused by blinking do not play a critical role in modulating perceptual alternation during CFS. Furthermore, the findings suggest that voluntary eyelid movements, rather than blinking itself, are crucial for modulating perceptual alternation. To the best of our knowledge, this is the first study to demonstrate that voluntary eyelid movements (including eye-widening as well as blinking) significantly affect visual cognition. 
Regarding the effect of perceptual alternation on blinking, Experiment 2 showed that with a larger sample size, blinks occurred more frequently shortly after perceptual alternation, even in the spontaneous-blink trials (Figure 7). This finding suggests that perceptual alternation induces blinking. Overall, the present findings indicate that blinking and perceptual alternation mutually affect each other during CFS. 
Before discussing the implications of our findings, we need to consider a possible confounding effect of manual responses on the timing of blinks. The delay between target discrimination and the observed response (likely several hundred milliseconds) could result in some blinks being incorrectly identified as occurring before target discrimination. Consequently, some trials might have been wrongly classified as spontaneous-blink trials. This may explain the relatively higher ODF just after spontaneous blinks in Experiment 1 (Figure 3a). However, in Experiment 2, the ODF remained low for about 1 s after blinks (Figure 6a), suggesting that this confounding effect did not significantly impact our results. 
Effects of blinks on perceptual alternation: Spontaneous blinks
The present results of the spontaneous-blink trials are not straightforward to interpret in terms of the effects on perceptual alternation. The longer mean DRT (Figures 2 and 5a) was not associated with time-dependent modulation of the ODF (Figures 3a and 6a). While these results do not completely rule out the possibility that spontaneous blinking suppresses perceptual alternation, a more plausible explanation could be a confounding effect of bias in the trial classification. Trials with longer DRT were more likely to be classified as spontaneous-blink trials. This is because spontaneous blinks were more likely to occur before perceptual alternation in trials where the alternation required a longer time. Therefore, even if spontaneous blinking does not affect perceptual alternation, DRT could still be longer. This bias may have also influenced a previous finding that spontaneous blinks were associated with longer DRT in b-CFS (Van Opstal et al., 2016), which also used a similar rule for trial classification. 
Another plausible account is that a psychophysiological factor underlies the longer DRT in spontaneous-blink trials.1 Research has shown that the rate of spontaneous blinks decreases when attentional demand is high, but it increases when participants are fatigued or bored (Bentivoglio et al., 1997; Maffei & Angrilli, 2018). Thus, fatigue and/or lapses in attention during observation may induce a spontaneous blink. Moreover, this lower arousal or lower sustained attention could increase DRT (Kim, Lokey, & Ling, 2017; Paffen, Alais, & Verstraten, 2006). Therefore, the same underlying cause might produce both a spontaneous blink and longer DRT, without one directly causing the other. Importantly, this account is not mutually exclusive with the classification bias hypothesis mentioned earlier. In fact, it may offer a psychophysiological basis for the bias in trial classification. 
If spontaneous blinking does not modulate perceptual alternation, it seems inconsistent with the previous findings. Previous studies have suggested that spontaneous blinking actually facilitates perceptual alternation using ambiguous figures (Nakatani et al., 2011) and binocular rivalry (Kalisvaart & Goossens, 2013). The effects of spontaneous blinking are mainly attributed to the contribution of retinal transients. Visual transients produced by a brief flash presented near a bistable stimulus have been shown to induce perceptual alternation in various bistable perception tasks (Kanai, Moradi, Shimojo, & Verstraten, 2005). However, this apparent discrepancy could be reconciled by considering the temporal characteristics of the CFS used in this study. The dynamic Mondrian used as the suppressor produced repetitive visual transients during observation, which may have masked or attenuated the effects of retinal transients caused by blinks. Importantly, this could also explain why physical blackouts did not modulate perceptual alternation in this study. Therefore, CFS may not be a suitable paradigm for retinal transients to be effective. 
Effects of blinks on perceptual alternation: Voluntary blinks
If retinal transients are ineffective during CFS, what mechanisms could explain the facilitatory effects of voluntary blinks and eye-widenings (voluntary eyelid movements)? Based on this elimination process, it seems reasonable to assume the contribution of extraretinal processes. Previous studies have reported cortical modulation associated with blinking (Bristow, Frith, & Rees, 2005; Bristow, Haynes, Sylvester, Frith, & Rees, 2005; Golan et al., 2016). For example, an fMRI study by Bristow, Haynes, et al. (2005) demonstrated that neural activity is modulated during voluntary blinks in various regions of the parietal and prefrontal cortices, in addition to the visual cortices. This modulation is believed to be driven by an extraretinal motor signal in the form of corollary discharge. Importantly, these higher cortical regions have been associated with fluctuations in visual consciousness (Lumer, Friston, & Rees, 1998). Similar extraretinal processes may be involved in the effects of voluntary eyelid movements. However, the cortical modulation associated with blinking has mainly been studied to understand the neural mechanisms underlying blink suppression. These studies generally assumed that blink suppression is essentially the same for different types of blinks. Therefore, although many previous studies have used voluntary blinks (Bristow, Frith, et al., 2005; Bristow, Haynes, et al., 2005; Golan et al., 2016), the specific neural processes related to voluntary blinking have yet to be identified. 
It has been shown that voluntary self-motion signals modulate visual perception in binocular rivalry (Alais, Keys, Verstraten, & Paffen, 2021). Moreover, in other visual tasks, voluntary action has been shown to modulate visual perception (Benazet, Thenault, Whittingstall, & Bernier, 2016; Buaron, Reznik, Gilron, & Mukamel, 2020; Mifsud et al., 2018; Stenner, Bauer, Haggard, Heinze, & Dolan, 2014). However, it is not likely that all voluntary actions facilitate perceptual alternation. Our preliminary observations indicated that a voluntary keypress in response to the same blink cue used in the present study did not modulate perceptual alternation during CFS. Facilitatory effects seem to be specific to voluntary eyelid movements. This issue needs to be further explored in future studies. 
Concerning the neural basis of different types of blinks, Nakano and Kuriyama (2017) suggested that spontaneous and voluntary blinks are associated with distinct changes in the activity of the autonomic nervous system. They showed that both heart rate and skin conductance level increased after spontaneous blinks, indicating an increase in sympathetic nerve activity. In contrast, the heart rate did not increase after voluntary blinks but decreased, although the pattern of heart rate change after voluntary blinks was complicated. They discussed these findings in the context of the neural activity of the hypothalamus and locus coeruleus (LC). These two brain structures are involved in a variety of functions, including the modulation of autonomic activities and regulation of the arousal level of the cerebral cortex, and they reciprocally interact with each other (Aston-Jones & Cohen, 2005; Samuels & Szabadi, 2008). In particular, the LC has recently received growing attention because it is involved in various cognitive processes, including voluntary attention, learning, and memory formation (Poe et al., 2020; Waterhouse & Navarra, 2019). Another line of evidence indicates an association between LC activity and eyelid movements. The eyelid muscle (the levator palpebrae muscle) has been shown to have a neural connection with the LC, and its contraction promotes the secretion of noradrenaline in the LC (Dauvergne et al., 2008; Matsuo, Ban, Hama, & Yuzuriha, 2015). Moreover, a recent study has indicated that noradrenaline modulates perceptual awareness (Gelbard-Sagiv, Magidov, Sharon, Hendler, & Nir, 2018). This pharmacological study showed that detection sensitivity and subjective visibility varied depending on the noradrenaline levels. Taken together, it seems possible to assume that voluntary eyelid movements modulate perceptual alternation, or visual awareness in general, by way of a neural connection to the LC, although the specific neural basis for generating voluntary eyelid movements needs to be further elucidated. 
Recently, Ang and Maus (2020) explored the functional role of voluntary blinking and demonstrated that visual performance was enhanced shortly after voluntary blinks. This enhancement was more prominent than that caused by physical blackouts. They argued that blinking has a resetting function for visual attention (see also Nakano et al., 2013) and that the refresh of attention mediates the enhancement. Moreover, they argued that enhancing effects could be found for all types of blinking. However, when considering the present findings together, the effects of blinking may be much stronger for voluntary blinks. Voluntary blinking, or voluntary eyelid movements in general, may have transient enhancing effects on visual performance. 
Effects of perceptual alternation on blink frequency
The present results indicate that a spontaneous blink occurred in a time-locked fashion after perceptual alternation, regardless of whether a blink occurred before perceptual alternation. This finding suggests that confounding factors, such as eye dryness and task-dependent disinhibition of blinking (which modulate blink frequency), are not directly related to perceptual alternation. Moreover, the temporal association of the blink with the perceptual alternation rather than the stimulus change indicates that external sensory factors are less likely to induce spontaneous blinks. 
We believe that a more likely possibility is that internal cognitive factors mediate increased blink frequency, as suggested in previous studies (Brych & Händel, 2020; Fukuda & Matsunaga, 1983; Fukuda et al., 2005; Nakano & Kitazawa, 2010; Nakano et al., 2009; Oh et al., 2012). For example, spontaneous blinks during video viewing are synchronized across and within viewers at semantic breakpoints in video stories rather than physical scene breaks (Nakano & Kitazawa, 2010; Nakano et al., 2009). Moreover, these blinks have been shown to involve massive and dynamic alterations in brain activity (Nakano, 2015; Nakano et al., 2013). When spontaneous blinks occur, the dorsal and ventral attentional networks transiently decrease in activity. Simultaneously, the default mode network, which is implicated in internal processing such as introspection, increases in activity. Based on these findings, it has been proposed that spontaneous blinks are linked to the disengagement of attention from external stimuli (Nakano et al., 2013; Nakano, Kuriyama, Himichi, & Nomura, 2015). According to this proposal, spontaneous blinks are induced at implicit points at which cognitive changes occur. Our findings may also reflect this link. Perceptual alternation involves a change in attentional states because it is the event to which participants are asked to respond. Thus, it is plausible to infer that internal cognitive changes associated with perceptual alternation induce spontaneous blinking. 
Conclusions
The present study revealed that blinking and perceptual alternation interact with each other during b-CFS (i.e., voluntary blinks facilitate perceptual alternation, and perceptual alternation induces spontaneous blinks). Facilitation of perceptual alternation was not observed when blinks were spontaneous. These findings indicate that different types of blinks can differentially affect perceptual alternation and visual cognition in general. These results also emphasize the importance of investigating the effects of different types of blinks within the same experimental paradigm. This approach may be essential for elucidating the functional roles of blinking, or eyelid movements in general, that are not attributable to visual transients. 
Acknowledgments
The authors thank Motomi Shimizu for his advice on data analysis. This work was partly supported by JSPS KAKENHI Grant Numbers 26285162, 18K18686, and 20H01781. 
Commercial relationships: none. 
Corresponding author: Eiji Kimura. 
Address: Department of Psychology, Graduate School of Humanities, Chiba University, Chiba 263-8522, Japan. 
Footnotes
1  We thank Dr. Surya Gayet for suggesting this explanation.
References
Abe, S., Kimura, E., & Goryo, K. (2011). Eye- and feature-based modulation of onset rivalry caused by the preceding stimulus. Journal of Vision, 11(13):6, 1–18, https://doi.org/10.1167/11.13.6. [CrossRef]
Alais, D., Keys, R., Verstraten, F. A. J., & Paffen, C. L. E. (2021). Vestibular and active self-motion signals drive visual perception in binocular rivalry. iScience, 24, 103417, https://doi.org/10.1016/j.isci.2021.103417. [CrossRef] [PubMed]
Ang, J. W. A., & Maus, G. W. (2020). Boosted visual performance after eye blinks. Journal of Vision, 20(10):2, 1–9, https://doi.org/10.1167/jov.20.10.2. [CrossRef]
Aston-Jones, G., & Cohen, J. D. (2005). An integrative theory of locus coeruleus-norepinephrine function: Adaptive gain and optimal performance. Annual Review of Neuroscience, 28, 403–450, https://doi.org/10.1146/annurev.neuro.28.061604.135709. [CrossRef] [PubMed]
Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language, 59, 390–412, https://doi.org/10.1016/j.jml.2007.12.005. [CrossRef]
Baker, D. H., Vilidaite, G., Lygo, F. A., Smith, A. K., Flack, T. R., Gouws, A. D., ... Andrews, T. J. (2021). Power contours: Optimising sample size and precision in experimental psychology and human neuroscience. Psychological Methods, 26, 295–314, https://doi.org/10.1037/met0000337. [CrossRef] [PubMed]
Benazet, M., Thenault, F., Whittingstall, K., & Bernier, P. M. (2016). Attenuation of visual reafferent signals in the parietal cortex during voluntary movement. Journal of Neurophysiology, 116, 1831–1839, https://doi.org/10.1152/jn.00231.2016. [CrossRef] [PubMed]
Bentivoglio, A. R., Bressman, S. B., Cassetta, E., Carretta, D., Tonali, P., & Albanese, A. (1997). Analysis of blink rate patterns in normal subjects. Movement Disorders, 12, 1028–1034, https://doi.org/10.1002/mds.870120629. [CrossRef]
Bonneh, Y., Fried, M., & Adini, Y. (2015). Blinking by surprise: Eye-blink rate and latency uncover stimulus predictability. Journal of Vision, 15(12):779, 1, https://doi.org/10.1167/15.12.779. [CrossRef]
Bonneh, Y. S., Adini, Y., & Polat, U. (2016). Contrast sensitivity revealed by spontaneous eyeblinks: Evidence for a common mechanism of oculomotor inhibition. Journal of Vision, 16(7):1, 1–15, https://doi.org/10.1167/16.7.1. [CrossRef] [PubMed]
Bristow, D., Frith, C., & Rees, G. (2005). Two distinct neural effects of blinking on human visual processing. NeuroImage, 27, 136–145, https://doi.org/10.1016/j.neuroimage.2005.03.037. [CrossRef] [PubMed]
Bristow, D., Haynes, J.-D., Sylvester, R., Frith, C. D., & Rees, G. (2005). Blinking suppresses the neural response to unchanging retinal stimulation. Current Biology, 15, 1296–1300, https://doi.org/10.1016/j.cub.2005.06.025. [CrossRef]
Brych, M., & Händel, B. (2020). Disentangling top-down and bottom-up influences on blinks in the visual and auditory domain. International Journal of Psychophysiology, 158, 400–410, https://doi.org/10.1016/j.ijpsycho.2020.11.002. [CrossRef]
Brych, M., Murali, S., & Händel, B. (2021). The role of blinks, microsaccades and their retinal consequences in bistable motion perception. Frontiers in Psychology, 12, 647256, https://doi.org/10.3389/fpsyg.2021.647256. [CrossRef] [PubMed]
Buaron, B., Reznik, D., Gilron, R., & Mukamel, R. (2020). Voluntary actions modulate perception and neural representation of action-consequences in a hand-dependent manner. Cerebral Cortex, 30, 6097–6107, https://doi.org/10.1093/cercor/bhaa156. [CrossRef]
Cardona, G., Gómez, M., Quevedo, L., & Gispets, J. (2014). Effects of transient blur and VDT screen luminance changes on eyeblink rate. Contact Lens and Anterior Eye, 37, 363–367, https://doi.org/10.1016/j.clae.2014.05.005. [CrossRef]
Collewijn, H., Steen, J. v. d., & Steinman, R. M. (1985). Human eye movements associated with blinks and prolonged eyelid closure. Journal of Neurophysiology, 54, 11–27, https://doi.org/10.1152/jn.1985.54.1.11. [CrossRef] [PubMed]
Cong, D. K., Sharikadze, M., Staude, G., Deubel, H., & Wolf, W. (2010). Spontaneous eye blinks are entrained by finger tapping. Human Movement Science, 29, 1–18, https://doi.org/10.1016/j.humov.2009.08.003. [CrossRef] [PubMed]
Dauvergne, C., Smit, A. E., Valla, J., Diagne, M., Buisseret-Delmas, C., Buisseret, P., ... VanderWerf F. (2008). Are locus coeruleus neurons involved in blinking? Neuroscience Research, 61, 182–191, https://doi.org/10.1016/j.neures.2008.02.005. [CrossRef] [PubMed]
Doane, M. G. (1980). Interaction of eyelids and tears in corneal wetting and the dynamics of the normal human eyeblink. American Journal of Ophthalmology, 89, 507–516, https://doi.org/10.1016/0002-9394(80)90058-6. [CrossRef] [PubMed]
Doughty, M. J. (2013). Effects of background lighting and retinal illuminance on spontaneous eyeblink activity of human subjects in primary eye gaze. Eye & Contact Lens, 39, 138–146, https://doi.org/10.1097/ICL.0b013e31827124b7. [PubMed]
Evinger, C., Shaw, M. D., Peck, C. K., Manning, K. A., & Baker, R. (1984). Blinking and associated eye movements in humans, guinea pigs, and rabbits. Journal of Neurophysiology, 52, 323–339, https://doi.org/10.1152/jn.1984.52.2.323. [PubMed]
Flamm, L. E., & Bergum, B. O. (1977). Reversible perspective figures and eye movements. Perception and Motor Skills, 44, 1015–1019, https://doi.org/10.2466/pms.1977.44.3.1015.
Fogarty, C., & Stern, J. A. (1989). Eye movements and blinks: Their relationship to higher cognitive processes. International Journal of Psychophysiology, 8, 35–42, https://doi.org/10.1016/0167-8760(89)90017-2.
Fukuda, K., & Matsunaga, K. (1983). Changes in blink rate during signal discrimination tasks. Japanese Psychological Research, 25, 140–146, https://doi.org/10.4992/psycholres1954.25.140.
Fukuda, K., Stern, J. A., Brown, T. B., & Russo, M. B. (2005). Cognition, blinks, eye-movements, and pupillary movements during performance of a running memory task. Aviation, Space, and Environmental Medicine, 76, C75–C85, https://doi.org/N.A. [PubMed]
Gayet, S., Van Der Stigchel, S., & Paffen, C. (2014). Breaking continuous flash suppression: Competing for consciousness on the pre-semantic battlefield. Frontiers in Psychology, 5, 460, https://doi.org/10.3389/fpsyg.2014.00460. [PubMed]
Gelbard-Sagiv, H., Magidov, E., Sharon, H., Hendler, T., & Nir, Y. (2018). Noradrenaline modulates visual perception and late visually evoked activity. Current Biology, 28, 2239–2249, https://doi.org/10.1016/j.cub.2018.05.051.
Golan, T., Davidesco, I., Meshulam, M., Groppe, D. M., Mégevand, P., Yeagle, E. M., ... Malach R. (2016). Human intracranial recordings link suppressed transients rather than ‘filling-in’ to perceptual continuity across blinks. eLife, 5, e17243, https://doi.org/10.7554/eLife.17243. [PubMed]
Hesselmann, G., Darcy, N., Rothkirch, M., & Sterzer, P. (2018). Investigating masked priming along the “vision-for-perception” and “vision-for-action” dimensions of unconscious processing. Journal of Experimental Psychology: General, 147, 1641–1659, https://doi.org/10.1037/xge0000420. [PubMed]
Irwin, D. E. (2011). Where does attention go when you blink? Attention, Perception, & Psychophysics, 73, 1374–1384, https://doi.org/10.3758/s13414-011-0111-0. [PubMed]
Iseki, R. (2023). Anovakun (version 4.8.9), https://riseki.cloudfree.jp/?ANOVA%E5%90%9B.
Ito, J., Nikolaev, A. R., Luman, M., Aukes, M. F., Nakatani, C., & van Leeuwen, C. (2003). Perceptual switching, eye movements, and the bus paradox. Perception, 32, 681–698, https://doi.org/10.1068/p5052. [PubMed]
Jiang, Y., Costello, P., & He, S. (2007). Processing of invisible stimuli: Advantage of upright faces and recognizable words in overcoming interocular suppression. Psychological Science, 18, 349–355, https://doi.org/10.1111/j.1467-9280.2007.01902.x. [PubMed]
Kalisvaart, J. P., & Goossens, J. (2013). Influence of retinal image shifts and extra-retinal eye movement signals on binocular rivalry alternations. PLoS ONE, 8(4), e61702, https://doi.org/10.1371/journal.pone.0061702. [PubMed]
Kanai, R., Moradi, F., Shimojo, S., & Verstraten, F. A. (2005). Perceptual alternation induced by visual transients. Perception, 34, 803–822, https://doi.org/10.1068/p5245. [PubMed]
Kato, M., & Miyauchi, S. (2003). Human precentral cortical activation patterns during saccade tasks: An fMRI comparison with activation during intentional eyeblink tasks. NeuroImage, 19, 1260–1272, https://doi.org/10.1016/S1053-8119(03)00223-4. [PubMed]
Kim, D., Lokey, S., & Ling, S. (2017). Elevated arousal levels enhance contrast perception. Journal of Vision, 17(2):14, 1–10, https://doi.org/10.1167/17.2.14.
Knief, U., & Forstmeier, W. (2021). Violating the normality assumption may be the lesser of two evils. Behavior Research Methods, 53, 2576–2590, https://doi.org/10.3758/s13428-021-01587-5. [PubMed]
Lenth, R. (2023). emmeans: Estimated marginal means, aka least-squares means (version 1.8.5), https://CRAN.R-project.org/package=emmeans.
Lo, S., & Andrews, S. (2015). To transform or not to transform: Using generalized linear mixed models to analyse reaction time data. Frontiers in Psychology, 6, 1171, https://doi.org/10.3389/fpsyg.2015.01171. [PubMed]
Lumer, E. D., Friston, K. J., & Rees, G. (1998). Neural correlates of perceptual rivalry in the human brain. Science, 280, 1930–1934. [PubMed]
Maffei, A., & Angrilli, A. (2018). Spontaneous eye blink rate: An index of dopaminergic component of sustained attention and fatigue. International Journal of Psychophysiology, 123, 58–63, https://doi.org/10.1016/j.ijpsycho.2017.11.009.
Matsuo, K., Ban, R., Hama, Y., & Yuzuriha, S. (2015). Eyelid opening with trigeminal proprioceptive activation regulates a brainstem arousal mechanism. PLoS ONE, 10, 1–19, https://doi.org/10.1371/journal.pone.0134659.
Mifsud, N. G., Beesley, T., Watson, T. L., Elijah, R. B., Sharp, T. S., & Whitford, T. J. (2018). Attenuation of visual evoked responses to hand and saccade-initiated flashes. Cognition, 179, 14–22, https://doi.org/10.1016/j.cognition.2018.06.005. [PubMed]
Murali, S., & Händel, B. (2024). Spontaneous eye blinks map the probability of perceptual reinterpretation during visual and auditory ambiguity. Cognitive Science, 48(2): e13414, 1–24, https://doi.org/10.1111/cogs.13414. [PubMed]
Nakano, T. (2015). Blink-related dynamic switching between internal and external orienting networks while viewing videos. Neuroscience Research, 96, 54–58, https://doi.org/10.1016/j.neures.2015.02.010. [PubMed]
Nakano, T., Kato, M., Morito, Y., Itoi, S., & Kitazawa, S. (2013). Blink-related momentary activation of the default mode network while viewing videos. Proceedings of the National Academy of Sciences, 110, 702–706, https://doi.org/10.1073/pnas.1214804110.
Nakano, T., & Kitazawa, S. (2010). Eyeblink entrainment at breakpoints of speech. Experimental Brain Research, 205, 577–581, https://doi.org/10.1007/s00221-010-2387-z. [PubMed]
Nakano, T., & Kuriyama, C. (2017). Transient heart rate acceleration in association with spontaneous eyeblinks. International Journal of Psychophysiology, 121, 56–62, https://doi.org/10.1016/j.ijpsycho.2017.09.003.
Nakano, T., Kuriyama, C., Himichi, T., & Nomura, M. (2015). Association of a nicotinic receptor gene polymorphism with spontaneous eyeblink rates. Scientific Reports, 5, 8658, https://doi.org/10.1038/srep08658. [PubMed]
Nakano, T., Yamamoto, Y., Kitajo, K., Takahashi, T., & Kitazawa, S. (2009). Synchronization of spontaneous eyeblinks while viewing video stories. Proceedings of the Royal Society B: Biological Sciences, 276, 3635–3644, https://doi.org/10.1098/rspb.2009.0828.
Nakatani, H., Orlandi, N., & van Leeuwen, C. (2011). Precisely timed oculomotor and parietal EEG activity in perceptual switching. Cognitive Neurodynamics, 5, 399–409, https://doi.org/10.1007/s11571-011-9168-7. [PubMed]
Oh, J., Jeong, S.-Y., & Jeong, J. (2012). The timing and temporal patterns of eye blinking are dynamically modulated by attention. Human Movement Science, 31, 1353–1365, https://doi.org/10.1016/j.humov.2012.06.003. [PubMed]
Paffen, C. L. E., Alais, D., & Verstraten, F. A. J. (2006). Attention speeds binocular rivalry. Psychological Science, 17, 752–756, https://doi.org/10.1111/j.1467-9280.2006.01777.x. [PubMed]
Poe, G. R., Foote, S., Eschenko, O., Johansen, J. P., Bouret, S., Aston-Jones, G., ... Sara S.J. (2020). Locus coeruleus: A new look at the blue spot. Nature Reviews Neuroscience, 21, 644–659, https://doi.org/10.1038/s41583-020-0360-9. [PubMed]
Rainwater, G., & Cogan, R. (1975). Simultaneous disappearances of both images during binocular rivalry of contours and effect on eye-blink rate. Perceptual and Motor Skills, 40, 71–77, https://doi.org/10.2466/pms.1975.40.1.71. [PubMed]
Samuels, E. R., & Szabadi, E. (2008). Functional neuroanatomy of the noradrenergic locus coeruleus: Its roles in the regulation of arousal and autonomic function part I: Principles of functional organisation. Current Neuropharmacology, 6, 235–253, https://doi.org/10.2174/157015908785777229. [PubMed]
Shimizu, M., & Kimura, E. (2020). Eye swapping temporally modulates potency of continuous flash suppression. Journal of Vision, 20(3):1, 1–10, https://doi.org/10.1167/jovi.20.3.1. [PubMed]
Shimizu, M., & Kimura, E. (2023). Afterimage duration depends on how deeply invisible stimuli were suppressed. Journal of Vision, 23(8):1, 1–13, https://doi.org/10.1167/jov.23.8.1. [PubMed]
Siegle, G. J., Ichikawa, N., & Steinhauer, S. (2008). Blink before and after you think: Blinks occur prior to and following cognitive load indexed by pupillary responses. Psychophysiology, 45, 679–687, https://doi.org/10.1111/j.1469-8986.2008.00681.x. [PubMed]
Stein, T. (2019). The breaking continuous flash suppression paradigm: Review, evaluation, and outlook. In Hesselmann, G. (Ed.), Transitions between consciousness and unconsciousness (pp. 1–38). New York, NY: Routledge/Taylor & Francis Group.
Stenner, M. P., Bauer, M., Haggard, P., Heinze, H. J., & Dolan, R. (2014). Enhanced alpha-oscillations in visual cortex during anticipation of self-generated visual stimulation. Journal of Cognitive Neuroscience, 26, 2540–2551, https://doi.org/10.1162/jocn_a_00658. [PubMed]
Stern, J. A., Walrath, L. C., & Goldstein, R. (1984). The endogenous eyeblink. Psychophysiology, 21, 22–33, https://doi.org/10.1111/j.1469-8986.1984.tb02312.x. [PubMed]
Tsuchiya, N., & Koch, C. (2005). Continuous flash suppression reduces negative afterimages. Nature Neuroscience, 8, 1096–1101, https://doi.org/10.1038/nn1500. [PubMed]
van Dam, L. C. J., & van Ee, R. (2005). The role of (micro)saccades and blinks in perceptual bi-stability from slant rivalry. Vision Research, 45, 2417–2435, https://doi.org/10.1016/j.visres.2005.03.013. [PubMed]
Van Opstal, F., De Loof, E., Verguts, T., & Cleeremans, A. (2016). Spontaneous eyeblinks during breaking continuous flash suppression are associated with increased detection times. Journal of Vision, 16(14):21, 1–10, https://doi.org/10.1167/16.14.21.
Wagenmakers, E.-J. (2007). A practical solution to the pervasive problems of p values. Psychonomic Bulletin & Review, 14, 779–804, https://doi.org/10.3758/BF03194105. [PubMed]
Waterhouse, B. D., & Navarra, R. L. (2019). The locus coeruleus-norepinephrine system and sensory signal processing: A historical review and current perspectives. Brain Research, 1709, 1–15, https://doi.org/10.1016/j.brainres.2018.08.032. [PubMed]
Figure 1.
 
Stimulus sequence (left side) and typical percept (right side) during the b-CFS paradigm used in the present study. The suppressor was a dynamic Mondrian pattern that was refreshed at 10 Hz and presented to the right eye. The target was a Gabor patch (oriented clockwise or counterclockwise) and presented to the left eye. The observer's task was to indicate the target orientation as soon as possible.
Figure 1.
 
Stimulus sequence (left side) and typical percept (right side) during the b-CFS paradigm used in the present study. The suppressor was a dynamic Mondrian pattern that was refreshed at 10 Hz and presented to the right eye. The target was a Gabor patch (oriented clockwise or counterclockwise) and presented to the left eye. The observer's task was to indicate the target orientation as soon as possible.
Figure 2.
 
Mean discrimination response time (DRT) for different classified trials in Experiment 1: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 2.
 
Mean discrimination response time (DRT) for different classified trials in Experiment 1: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 3.
 
Frequency of the target discrimination response after blinking in Experiment 1. The histogram shows the observed discrimination response frequency (ODF) as a function of time after the blink in (a) spontaneous-blink trials and (b) voluntary-blink trials. The shaded area in each panel represents the 95% interval of the random discrimination response frequency (RDF), and the upper and lower limits are shown by dotted lines (RDF limits). Uniformly colored bars illustrate instances where the ODF exceeds or falls short of the limits, indicating a significant increase or decrease in the frequency of target discrimination at those times.
Figure 3.
 
Frequency of the target discrimination response after blinking in Experiment 1. The histogram shows the observed discrimination response frequency (ODF) as a function of time after the blink in (a) spontaneous-blink trials and (b) voluntary-blink trials. The shaded area in each panel represents the 95% interval of the random discrimination response frequency (RDF), and the upper and lower limits are shown by dotted lines (RDF limits). Uniformly colored bars illustrate instances where the ODF exceeds or falls short of the limits, indicating a significant increase or decrease in the frequency of target discrimination at those times.
Figure 4.
 
Blink frequency after target discrimination in Experiment 1. The histogram shows observed blink frequency (OBF) as a function of time after target discrimination in (a) no-blink trials and (b) spontaneous-blink trials. The shaded area in each panel represents the 95% interval of the random blink frequency (RBF), and the upper and lower limits (RBF limits) are shown by dotted lines (the lower limit for the spontaneous-blink trials was very close to zero). Uniformly colored bars illustrate instances where the OBF exceeds the upper RBF limit, indicating a significant increase in blink frequency at those times.
Figure 4.
 
Blink frequency after target discrimination in Experiment 1. The histogram shows observed blink frequency (OBF) as a function of time after target discrimination in (a) no-blink trials and (b) spontaneous-blink trials. The shaded area in each panel represents the 95% interval of the random blink frequency (RBF), and the upper and lower limits (RBF limits) are shown by dotted lines (the lower limit for the spontaneous-blink trials was very close to zero). Uniformly colored bars illustrate instances where the OBF exceeds the upper RBF limit, indicating a significant increase in blink frequency at those times.
Figure 5.
 
Mean DRT for different classified trials in Experiment 2: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 5.
 
Mean DRT for different classified trials in Experiment 2: (a) Session 1 and (b) Session 2. Error bars show ±1 SD across different observers. The asterisks indicate p < 0.05 for multiple comparisons using Bonferroni correction.
Figure 6.
 
Frequency of the target discrimination response after blinking in Experiment 2. Other aspects are the same as those in Figure 3.
Figure 6.
 
Frequency of the target discrimination response after blinking in Experiment 2. Other aspects are the same as those in Figure 3.
Figure 7.
 
Blink frequency after target discrimination in Experiment 2. Other aspects are the same as those in Figure 4.
Figure 7.
 
Blink frequency after target discrimination in Experiment 2. Other aspects are the same as those in Figure 4.
Table 1.
 
Results of Bayes factor analysis of the DRT in Session 1 using GLMMs in Experiment 2.
Table 1.
 
Results of Bayes factor analysis of the DRT in Session 1 using GLMMs in Experiment 2.
Table 2.
 
Results of Bayes factor analysis of the DRT in Session 2 using GLMMs in Experiment 2.
Table 2.
 
Results of Bayes factor analysis of the DRT in Session 2 using GLMMs in Experiment 2.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×