Save

Saccades and Subjective Time in Seconds Range Duration Reproduction

In: Timing & Time Perception
View More View Less
  • 1 Department of Psychology, National University of Singapore, Singapore
  • | 2 LSI Programme in Neurobiology and Aging, National University of Singapore, Singapore
  • | 3 Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore
  • | 4 Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
  • | 5 Department of Electrical and Computer Engineering, National University of Singapore, Singapore
  • | 6 Singapore Institute for Neurotechnology, National University of Singapore, Singapore
Full Access

A transient suppression of visual perception during saccades ensures perceptual stability. In two experiments, we examined whether saccades affect time perception of visual and auditory stimuli in the seconds range. Specifically, participants completed a duration reproduction task in which they memorized the duration of a 6 s timing signal during the training phase and later reproduced that duration during the test phase. Four experimental conditions differed in saccade requirements and the presence or absence of a secondary discrimination task during the test phase. For both visual and auditory timing signals, participants reproduced longer durations when the secondary discrimination task required saccades to be made (i.e., overt attention shift) during reproduction as compared to when the discrimination task merely required fixation at screen center. Moreover, greater total saccade duration in a trial resulted in greater time distortion. However, in the visual modality, requiring participants to covertly shift attention (i.e., no saccade) to complete the discrimination task increased reproduced duration as much as making a saccade, whereas in the auditory modality making a saccade increased reproduced duration more than making a covert attention shift. In addition, we examined microsaccades in the conditions that did not require full saccades for both the visual and auditory experiments. Greater total microsaccade duration in a trial resulted in greater time distortion in both modalities. Taken together, the experiments suggest that saccades and microsaccades affect seconds range visual and auditory interval timing via attention and saccadic suppression mechanisms.

Abstract

A transient suppression of visual perception during saccades ensures perceptual stability. In two experiments, we examined whether saccades affect time perception of visual and auditory stimuli in the seconds range. Specifically, participants completed a duration reproduction task in which they memorized the duration of a 6 s timing signal during the training phase and later reproduced that duration during the test phase. Four experimental conditions differed in saccade requirements and the presence or absence of a secondary discrimination task during the test phase. For both visual and auditory timing signals, participants reproduced longer durations when the secondary discrimination task required saccades to be made (i.e., overt attention shift) during reproduction as compared to when the discrimination task merely required fixation at screen center. Moreover, greater total saccade duration in a trial resulted in greater time distortion. However, in the visual modality, requiring participants to covertly shift attention (i.e., no saccade) to complete the discrimination task increased reproduced duration as much as making a saccade, whereas in the auditory modality making a saccade increased reproduced duration more than making a covert attention shift. In addition, we examined microsaccades in the conditions that did not require full saccades for both the visual and auditory experiments. Greater total microsaccade duration in a trial resulted in greater time distortion in both modalities. Taken together, the experiments suggest that saccades and microsaccades affect seconds range visual and auditory interval timing via attention and saccadic suppression mechanisms.

1. Introduction

During a saccade, the image on the retina is displaced at a speed of several hundred degrees per second (Tovee, 1996). Although this sensory stimulation is transmitted from the retina to the primary visual cortex (e.g., Watson & Krekelberg, 2011; for review, see Ibbotson & Krekelberg, 2011), the observer is unaware of it, an effect termed saccadic suppression (Zuber & Stark, 1966). Saccadic suppression has been attributed to visual masking and corollary discharge (e.g., Campbell & Wurtz, 1978; Wurtz, 2008) and at the neural level may result from an inhibitory mechanism (Riggs et al., 1974), a reduction in the input to the visual cortex from the lateral geniculate nucleus (Thilo et al., 2004), or a reduction of processing in higher cortical areas (Bremmer et al., 2009).

Previous research has shown an effect of saccades on milliseconds range timing. For example, Morrone et al. (2005) demonstrated that a single saccade made immediately after presentation of a 100 ms standard stimulus resulted in the standard stimulus being judged equivalent to a 50 ms comparison stimulus (i.e., a compression effect). However, visual stimuli were perceived accurately when the saccade occurred 200 to 400 ms after the standard stimulus. Moreover, judgments of auditory stimulus duration were unaffected by saccades (also see Suzuki & Yamazaki, 2010).

In contrast, when the onset of a 1000 ms visual stimulus occurred during a saccade, it was perceived as longer than a stimulus that did not co-occur with a saccade (Yarrow et al., 2001) and larger saccades (55 vs. 22 degrees) resulted in larger overestimations. Yarrow et al. proposed that the duration of a saccade is retrospectively filled in after the eyes land on the next visual target. However, the filled-in duration is longer than the time lost to the saccade, thereby creating the illusion of temporal lengthening. This phenomenon, termed saccadic chronostasis, is sometimes experienced in daily life as the perception that the second hand of an analog clock pauses briefly when the viewer first fixates on it. Chronostasis occurs for highly reflexive saccades like prosaccades (cued voluntary saccades made in response to a sudden and unpredictable peripheral onset), as well as volitional self-timed saccades (Yarrow et al., 2004a).

To date, studies of the effect of saccades on time perception have been limited to paradigms in which the stimuli were in the range of tens of milliseconds to one second and the saccade either occurred after the timing signal (e.g., Morrone et al., 2005; Suzuki & Yamazaki, 2010) or partially overlapped with the timing signal (e.g., Yarrow et al., 2001, 2004a, b). However, in the case of suprasecond timing it is possible for one or more saccades to be fully embedded within the timing signal. Indeed, with normal scanning of the visual environment, several saccades are likely to occur in any period of a few seconds.

Seconds range timing (i.e., interval timing) is critical for making decisions about stimuli/events and the relationships among them (e.g., Allman et al., 2014; Buhusi & Meck, 2005). For example, continuing to converse with someone at a party is partially a function of how many interesting things the person says over a period of seconds to minutes. We were interested in determining whether saccades that are fully enveloped by the timing signal also impact the perceived duration of that signal. In addition, we also examined the influence on reproduced duration of small eye movements (i.e., microsaccades) made during fixation in conditions that did not require saccades. Microsaccades are involuntary and in both human and nonhuman primates microsaccade direction and rate are modulated by covert visual attention allocation (Engbert & Kliegl, 2003; Hafed et al., 2011; for reviews, see Engbert, 2006; Rolfs, 2009).

In an initial training phase, we asked participants to memorize a 6 s target duration demarcated by a visual (Experiment 1) or an auditory (Experiment 2) stimulus and to reproduce that target duration in the subsequent test phase. In one condition, participants made saccades during the timing signal and had to complete a secondary cue discrimination task. Saccades were not included for two conditions, but the secondary discrimination task was included, and for the final condition neither saccades nor the discrimination task were included. In these No Saccade conditions, we measured microsaccades. We reasoned that if time that elapses during a saccade is unavailable to the interval timing system, then reproductions of the target duration would be longer when saccades were made during reproduction as compared to when they were not. In a similar manner, we expected a positive correlation between total microsaccade duration within a trial and reproduced duration on that trial.

2. Experiment 1 – Visual Modality

2.1. Methods

2.1.1. Participants

Twenty-two undergraduate students from the National University of Singapore participated. The mean age of the participants was 22.1 years (SD = 2.02) and all had normal or corrected-to-normal vision.

2.1.2. Apparatus

An Eye-Link 1000 SR-Research system (0.15 to 0.5 degrees average accuracy; 2000 Hz sampling rate) was used to non-invasively track the position of the participant’s dominant eye. All visual stimuli were presented on a 22 inch Samsung TFT-LCD Monitor (1680 × 1050). Participants were seated in front of the monitor at an eye-to-screen distance of 57 cm. Visual stimuli were produced by a computer program written in MATLAB (The Mathworks) using the Psychophysics Toolbox (Pelli, 1997) and responses were collected via computer keyboard.

2.1.3. Design and Procedure

The experiment comprised two sessions, which were conducted on different days. Session A included Saccade–Discrimination, No Saccade–Central Discrimination, and No Saccade–No Discrimination conditions, whereas Session B included No Saccade–Peripheral Discrimination, No Saccade–Central Discrimination, and No Saccade–No Discrimination conditions. Each session comprised a training phase and a test phase and the order of sessions was counterbalanced across participants.

In both sessions, participants reproduced a 6 s target interval in the visual modality as the primary task. In the Saccade–Discrimination and No Saccade–Central Discrimination conditions of Session A and the No Saccade–Peripheral Discrimination and No Saccade–Central Discrimination conditions of Session B participants also performed a secondary task. This was a simple discrimination task in which they indicated whether a ‘3’ or an ‘E’ was the final stimulus presented before target duration termination. The sequence of ‘3’ and ‘E’ characters presented on each trial was random, so the discrimination task required participants to carefully monitor the computer screen for characters and to continuously update their memory representation of the most recent character. The Saccade–Discrimination and No Saccade–Peripheral Discrimination conditions occurred in separate test sessions to ensure that participants would not be confused about the saccade requirements on a given trial. The various conditions are described in detail below and presented in schematic form in Fig. 1.

Figure 1.
Figure 1.

Schematic illustration of the trial types in Experiment 1. See the Methods section for detailed descriptions of each condition.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

2.1.4. Training Phase

The training phase comprised two components. In the first component, participants learned a simple discrimination task in which they had to indicate via the keyboard whether the presented character was a ‘3’ or an ‘E’. Participants were asked to press the ‘z’ key if they saw a ‘3’ and to press the ‘/’ key if they saw an ‘E’. Five practice trials were provided as training. In the second phase of training, participants were taught to reproduce a 6 s duration. Duration reproduction training consisted of one example trial and ten practice trials. During the example trial, participants first saw a grey dot (1.5 cm in diameter) superimposed on a white fixation cross (1.5 cm in length and width) at the center of the computer screen. One second later, a blue rectangle (4 cm × 6 cm) was presented at screen center for 6 s. At the end of 6 s, the blue rectangle changed color to magenta for 1.5 s. All stimuli were presented against a black background. Participants were told to memorize the duration of the blue rectangle, but without counting or using any other method of duration subdivision (e.g., foot tapping, humming, etc.). A blank, black screen followed the offset of the magenta rectangle and participants were told to press the spacebar when they were ready to continue to the ten practice trials.

The practice trials used the same stimuli as the duration example trial. However, in the practice trials, participants were required to respond by pressing the ‘z’ key when they believed that the target duration had elapsed. The blue rectangle remained onscreen until the 6 s target duration had elapsed and was replaced by the magenta rectangle even when the participant had pressed the ‘z’ key earlier than 6 s. Hence, the color change to magenta served as feedback that participants could use to help form an accurate representation of the target duration. Each practice trial terminated after offset of the magenta rectangle and was followed by an onscreen prompt to press the spacebar to proceed to the next trial. Participants did not make saccades during target duration presentation in the training phase.

2.1.5. Test Phase

Each experimental condition in the test phase comprised 25 trials. These trials did not include duration feedback (i.e., color change) in order to reduce the possibility that participants would update their target duration representation with values potentially distorted by the presence of saccades. Instead, the participant’s key press response merely terminated the blue rectangle. However, to ensure that participants maintained an accurate representation of the target duration throughout the test phase, every tenth trial, for a total of eight trials, was a color change feedback trial. These feedback trials were the same as the training phase practice trials and were not included in the data analysis. To begin each trial, participants were prompted to press the spacebar. As such, the experiment was self-paced by the participants and they could rest their eyes in between trials without informing the experimenter.

Saccade–Discrimination. Participants first fixated on the cross at screen center. One second later, a grey dot appeared and was accompanied by a blue rectangle. The grey dot then disappeared from the screen center and randomly reappeared at one of the four corners of the computer monitor at random intervals drawn from a uniform distribution between 1 and 4 s. The corner dot location was 20 degrees of visual angle from screen center. A ‘3’ or an ‘E’ was randomly selected to appear in the center of the grey dot, but only when it was in a corner position. One second after the grey dot appeared at a corner position, it disappeared and reappeared in its original position at the screen center fixation cross, but without the character in the center. Participants were instructed to make a saccade to the new dot position each time it moved and the movement sequence was repeated until the participants terminated the trial by pressing the ‘z’ key when they believed that the blue rectangle had been on the screen for the target duration. After the trial termination response they were asked to report the identity of the last character they saw in the grey dot by pressing the ‘z’ key if it was a ‘3’ and ‘/’ if it was an ‘E’. Completion of the discrimination task marked the end of the trial and a prompt appeared asking participants to press the spacebar when they were ready to begin the next trial.

No Saccade–Peripheral Discrimination. As in the Saccade–Discrimination condition, the grey dot disappeared from the screen center and reappeared at one of the four corners of the computer monitor. A ‘3’ or an ‘E’ was randomly selected to appear in the center of the grey dot, but only when it was in a corner position. One second after the grey dot appeared at a corner position, it disappeared and reappeared in its original position at the screen center fixation cross, but without the character in the center. Participants were required to report the identity of the final character presented during the timing signal, but unlike in the Saccade–Discrimination condition they were instructed to maintain fixation at screen center and not make saccades (i.e., a covert attention shift was required). Presenting the Saccade–Peripheral Discrimination condition and No Saccade–Peripheral Discrimination condition in separate test sessions was necessary to eliminate confusion over whether to make a saccade to the dot when it moved.

No Saccade–Central Discrimination. This condition was similar to the Saccade–Discrimination condition with the exception that the grey dot remained at screen center and the discrimination characters (i.e., ‘3’ or ‘E’) appeared within the grey dot at that location. Hence, participants did not make saccades or overtly shift attention during the timing signal, but they still had to attend to the discrimination task and reproduce the target duration.

No Saccade–No Discrimination. As in the No Saccade–Central Discrimination condition, the grey dot remained at screen center. However, in this condition, the character discrimination task was not included and participants merely were required to press the ‘z’ key when they believed the target duration had elapsed.

2.1.6. Data Analysis

To ensure that only trials in which participants successfully completed the task were included in the data analysis several performance-based data exclusion rules were applied. First, trials including character discrimination in which the participant failed to correctly report the final character were excluded from the analyses. Second, for those conditions where participants were explicitly instructed not to move their eyes and to fixate the center of the screen (i.e., No Saccade–Peripheral Discrimination, No Saccade–Central Discrimination in Sessions A and B, No Saccade–No Discrimination in Sessions A and B) all trials containing at least one saccade with an amplitude exceeding 5° were excluded. Finally, trials with a reproduced duration three standard deviations above or below the participant’s mean duration for a given condition were removed from the data analysis. The percentage of excluded trials was larger for the No Saccade–Peripheral Discrimination, No Saccade–Central Discrimination, and No Saccade–No Discrimination conditions as compared to the Saccade–Discrimination condition (Table 1). However, this was a consequence of more trials being excluded due to incorrect reporting of the final character in the discrimination task and presence of saccades, rather than a difference in the number of reproduced duration outliers.

tab1

Microsaccades were identified in the No Saccade conditions using the Engbert and Mergenthaler (2006) algorithm. Horizontal and vertical fixation position time series were transformed into a 2D velocity space with a 19-point moving average window in order suppress noise. Horizontal and vertical velocity thresholds were calculated with λ = 5 and microsaccades were characterized as outliers with high velocity samples and minimal duration of 6 ms. The duration of microsaccades within each trial was then summed. Engbert and Kliegl (2003) treated microsaccades as binocularevents. Hence, an additional criterion for microsaccade detection was the temporal overlap of microsaccade events in the left and right eyes. Here, we tracked the dominant eye only, introducing the possibility of misidentification of microsaccades. However, we minimized this possibility by manually screening identified microsaccades for their velocity and distance trajectory.

Two participants in Experiment 1 were removed from the microsaccade analysis, but not the saccade analysis, due to missing data caused by technical errors during recording. In addition, one participant was removed from each of Experiments 1 and 2 for having fewer than seven trials containing microsaccades.

Linear mixed effects analyses, using R (R Core Team, 2015) and the R nlme package (Pinheiro et al., 2015), were conducted to examine the effect of experimental condition on reproduced duration (Model 1), the relationship between saccade duration and reproduced duration (Model 2), and the relationship between microsaccade duration and reproduced duration (Model 3).

In Model 1, experimental condition was a fixed effect and subject intercept was a random effect. Planned contrasts were conducted using multcomp (Hothorn et al., 2008) and p-values were adjusted for multiple comparisons using the Westfall method (Westfall, 1997). In Model 2, saccade duration was a fixed effect and intercepts and slopes for the effect of saccade duration by subject were random effects. In Model 3, microsaccade duration was a fixed effect and intercepts and slopes for the effect of microsaccade duration by subject were random effects. For all models, the maximum likelihood framework was used for parameter estimation.

Finally, the standard errors of the mean (SE) reported for reproduced durations were calculated using the method for within-subjects designs described by Morey (2008).

2.2. Results

Inspection of the condition means presented in Fig. 2 suggests that reproduced duration differed between the Saccade–Discrimination condition and a subset of the No Saccade conditions. Indeed, the mixed effects model revealed a significant effect of condition on reproduced duration, F(3, 2740) = 93.09, p < 0.001, with planned comparisons indicating that reproduced duration was greater in the Saccade–Discrimination (M = 7.13, SE = 0.10) condition relative to the No Saccade–No Discrimination (M = 6.32, SE = 0.14) and No Saccade–Central Discrimination (M = 6.92, SE = 0.05) conditions, t(2740) = 14.02, p < 0.001 and t(2740) = 3.59, p < 0.001, respectively, but not the No Saccade–Peripheral Discrimination condition, t(2740) = 0.58, p = 0.56. Moreover, reproduced duration was also larger for the No Saccade–Peripheral Discrimination (M = 7.11, SE = 0.15) condition relative to the No Saccade–No Discrimination condition, t(2740) = 10.69, p < 0.001, and the No Saccade–Central Discrimination condition, t(2740) = 2.31, p = 0.021.

Figure 2.
Figure 2.

For each condition of Experiment 1, the group mean reproduced duration is indicated by a horizontal black bar and the individual participant mean reproduced durations are indicated by filled black circles.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

For the Saccade–Discrimination condition, the mixed effects model revealed that saccade duration predicted reproduced duration at the subject level, b = 2.5, t(505) = 3.73, p < 0.001. That is to say, a 1 ms increase in saccade duration predicted a 2.5 ms increase in reproduced duration (see Fig. 3). The mean number of saccades per trial in the Saccade–Discrimination condition of Experiment 1 was 4.52 (SD = 1.23).

Figure 3.
Figure 3.

The relationship between saccade duration and reproduced duration in the Saccade–Discrimination condition of Experiment 1 is illustrated. The estimated fixed effect of saccade duration on reproduced duration is indicated by the solid black line and the shaded area demarcates the 95% pointwise confidence interval around this effect. The circles represent individual trial data from all participants contributing to the mixed effects analysis.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

Finally, mixed effects models revealed that microsaccade duration predicted reproduced duration at the subject level for both the No Saccade–No Discrimination (b = 0.0061, t(874) = 3.47, p < 0.001) and No Saccade–Central Discrimination conditions (b = 0.0067, t(728) = 3.91, p < 0.001) (Fig. 4). However, the relationship between microsaccade duration and reproduced duration was not statistically significant (b = 0.0061, t(236) = 1.75, p < 0.082) for the No Saccade–Peripheral Discrimination condition.

Figure 4.
Figure 4.

The relationships between microsaccade duration and reproduced duration in the various No Saccade conditions of both Experiment 1 (left) and Experiment 2 (right) are illustrated. For each plot, the estimated fixed effect of microsaccade duration on reproduced duration is indicated by the solid black line and the shaded area demarcates the 95% pointwise confidence interval around this effect. The circles represent individual trial data from all participants contributing to the mixed effects analysis.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

2.3. Discussion

As described in the Introduction, if saccadic suppression causes a loss of sensitivity to the passage of time, then target duration reproduction should be longer when saccades are made during reproduction as compared to when they are not. Consistent with this view, the presence of multiple saccades during test trials comprising a visual duration reproduction task and a character discrimination task (Saccade–Discrimination condition) significantly lengthened reproduced durations as compared to test trials in which participants kept their eyes fixated at screen center and completed the duration reproduction task only (No Saccade–No Discrimination condition). More importantly, reproduced durations in Saccade–Discrimination condition trials were also significantly longer than in trials where saccades were not made and the character discrimination occurred at central fixation (No Saccade–Central Discrimination condition). Taken together, these results suggest that the longer reproduced durations were not merely a consequence of participants completing a character discrimination task in conjunction with a timing task (i.e., a dual task effect), but were related to the saccades and perhaps due to a saccadic suppression mechanism. This interpretation is supported by the mixed effects analysis showing that the larger the summed duration of the saccades in a trial in the Saccade–Discrimination condition the longer the reproduced duration on that trial.

However, the absence of a significant difference between the Saccade–Discrimination condition and the No Saccade–Peripheral Discrimination condition appears to challenge the interpretation that the increased duration reproduction is due to a saccadic suppression mechanism. Instead, attention shifts may at least partially underlie the effect obtained here. Although the Saccade–Discrimination condition required participants to overtly shift attention from the central fixation point to the periphery and back again every time the grey circle moved, the No Saccade–Peripheral Discrimination condition required participants to covertly shift attention from central fixation to the periphery. Even so, there is significant overlap in brain areas activated in tasks involving overt and covert shifts of attention and mechanisms related to saccade preparation may underlie covert attention shifts as well (Corbetta et al., 1998; Moore et al., 2003). If so, then suppressive mechanisms may be active on the No Saccade–Peripheral Discrimination trials also.

Furthermore, Brown (1997) found that an increase in the difficulty of a secondary non-temporal task increased reproduced duration. Hence, the absence of a difference in reproduced durations between the Saccade–Discrimination and the No Saccade–Peripheral Discrimination conditions could be due to the greater difficulty of the latter condition, as indicated by the higher trial exclusion rate (see Table 1), and not legislate against a saccadic suppression mechanism.

Interestingly, in two of the No Saccade conditions an increase in microsaccade duration significantly predicted an increase in the reproduced duration, whereas in the third No Saccade condition this relationship approached, but did not reach significance. We speculate that this result may be a consequence of a mechanism similar to that driving aspects of the increased reproduction duration effect in the Saccade condition (i.e., saccadic suppression-like and/or attention shift effects). Indeed, microsaccadic suppression has been suggested in several studies (e.g. Murakami & Cavanagh, 1998; Ölveczky et al., 2003) and Engbert and Kliegl (2003) reported that microsaccade rate is modulated by covert shifts of visual attention.

In Experiment 2, we addressed whether the effects obtained for seconds range timing of visual stimuli would extend to auditory stimuli. We reasoned that effects due to secondary task difficulty as well as overt and covert shifts of attention should be amodal, whereas effects due to saccadic suppression mechanisms may be limited to the visual modality. Indeed, as mentioned in the Introduction, Morrone et al. (2005) reported that saccades did not affect duration judgments for auditory stimuli in their paradigm.

3. Experiment 2 – Auditory Modality

3.1. Methods

3.1.1. Participants

Twenty undergraduate students from the National University of Singapore participated in the study. The mean age of the participants was 22.2 years (SD = 2.18) and all had normal or corrected-to-normal vision.

3.1.2. Apparatus

The apparatus was the same as in Experiment 1.

3.1.3. Design and Procedure

The design and procedure were the same as in Experiment 1 with the exception that instead of a blue rectangle, the timing signal was a 6 s 350 Hz pure tone presented over computer speakers and the end of the timing signal was indicated by a 1.5 s presentation of a 440 Hz tone. Schematic diagrams of the experiment conditions are presented in Fig. 5.

Figure 5.
Figure 5.

Schematic illustration of the trial types in Experiment 2. See the Methods section for detailed descriptions of each condition.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

3.1.4. Data Analysis

Data exclusion procedures were the same as in Experiment 1. Table 1 summarizes the mean percentage of trials eliminated from each of the experimental conditions based on the performance criteria described in Section 2.1.6.

3.2. Results

Inspection of the condition means presented in Fig. 6 suggests that reproduced duration differed between the Saccade–Discrimination condition and the various No Saccade conditions. The mixed effects model revealed a significant effect of condition on reproduced duration, F(3, 2420) = 241.32, p < 0.001. Reproduced duration was greater in the Saccade–Discrimination (M = 8.39, SE = 0.22) condition relative to the No Saccade–Peripheral Discrimination (M = 7.56, SE = 0.15), No Saccade–No Discrimination (M = 6.64, SE = 0.15), and No Saccade–Central Discrimination (M = 7.54, SE = 0.06) conditions (t(2420) =8.49, p < 0.001; t(2420) = 26.08, p < 0.001; t(2420) = 12.36, p < 0.001, respectively). Moreover, reproduced duration was also larger for the No Saccade–Peripheral Discrimination condition relative to the No Discrmination-No Saccade condition, t(2420) = 11.62, p < 0.001. Finally, reproduced duration in the No Saccade–Peripheral Discrimination condition did not differ from that in the No Saccade–Central Discrimination condition, t = 0.86, p = 0.39.

Figure 6.
Figure 6.

For each condition of Experiment 2, the group mean reproduced duration is indicated by a horizontal black bar and the individual participant mean reproduced durations are indicated by filled black circles.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

For the Saccade–Discrimination condition, the mixed effects model revealed that saccade duration predicted reproduced duration at the subject level, b = 4.7, t(451) = 4.62, p < 0.001. That is to say, a 1 ms increase in saccade duration predicted a 4.7 ms increase in reproduced duration (see Fig. 7). The mean number of saccades per trial in the Saccade–Discrimination condition of Experiment 2 was 5.29 (SD = 1.88).

Figure 7.
Figure 7.

The relationship between saccade duration and reproduced duration in the Saccade–Discrimination condition of Experiment 2 is illustrated. The estimated fixed effect of saccade duration on reproduced duration is indicated by the solid black line and the shaded area demarcates the 95% pointwise confidence interval around this effect. The circles represent individual trial data from all participants contributing to the mixed effects analysis.

Citation: Timing & Time Perception 4, 2 (2016) ; 10.1163/22134468-00002066

Finally, mixed effects models revealed that microsaccade duration predicted reproduced duration at the subject level for the No Saccade–No Discrimination, No Saccade–Central Discrimination and the No Saccade–Peripheral Discrimination conditions (b = 0.0049, t(846) = 3.70, p < 0.001; b = 0.0060, t(724) = 4.61, p < 0.001 and b = 0.0054, t(224) = 4.43, p < 0.001 respectively, Fig. 4).

3.3. Discussion

The results of Experiment 2 replicated the increase in reproduced duration in the Saccade–Discrimination condition as compared to the No Saccade–No Discrimination and No Saccade–Central Discrimination conditions obtained in Experiment 1, but in the auditory modality. Importantly, but in contrast to Experiment 1, making a saccade and overt shift of attention (i.e., Saccade–Discrimination condition) resulted in longer reproduced durations as compared to making a covert shift of attention (No Saccade–Peripheral Discrimination) condition. As in Experiment 1, the interpretation that saccades and/or overt attention shifts impact interval timing also received support from a mixed effects analysis showing that the larger the summed duration of the saccades in a trial in the Saccade–Discrimination condition the longer the reproduced duration on that trial.

Although reproduced durations were longer in the No Saccade–Peripheral Discrimination condition than in the No Saccade–No Discrimination condition they did not differ between the No Saccade–Peripheral Discrimination and the No Saccade–Central Discrimination conditions. The latter effect, which is inconsistent with the results of Experiment 1, suggests that covert shifts of attention did not alter the perceived duration of the auditory timing signal beyond the effect of engaging in a secondary discrimination task.

Taken together, these results suggest that for auditory stimuli the longer reproductions in the Saccade–Discrimination condition were not solely a consequence of shifting attention and/or secondary task difficulty, but were also influenced by the saccades themselves (i.e., a saccadic suppression mechanism played a role).

Finally, as in Experiment 1, an increase in microsaccade duration significantly predicted an increase in the reproduced duration, although here the effect manifested across all three No Saccade conditions. In light of the apparent effect of saccades on auditory timing outlined above, it is possible that the microsaccade on auditory timing also could be due to saccadic suppression-like and/or attention shift effects.

4. General Discussion

Surprisingly, the data from Experiment 1 suggest that in the visual modality the critical element underlying the effect of saccades on seconds range interval timing may not be the saccade per se, but rather the shift in attention that co-occurs with a saccade. We obtained a similar lengthening of the reproduced duration when participants merely had to covertly shift attention in the absence of a saccade as compared to when they had to overtly shift attention via a saccade. In contrast, for the auditory modality the effect of overtly shifting attention (i.e., making a saccade) was larger than the effect of covertly shifting attention even though the secondary task difficulty differences between the Saccade–Discrimination and the No Saccade–Peripheral Discrimination conditions were as salient in Experiment 1 as in Experiment 2 (see Table 1). This effect of saccades on auditory timing is particularly interesting because, as mentioned above, Morrone et al. (2005) failed to obtain a saccade effect on the perception of a 100 ms auditory stimulus. Of course, the neural substrates underlying milliseconds and seconds range timing in different tasks may be at least partially distinct (e.g., Buhusi & Meck, 2005), which could account for the difference between the present and earlier studies. In addition, various phenomena (e.g., the McGurk effect) indicate that auditory and visual information is integrated, so an effect of saccades on the timing of auditory stimuli is not entirely surprising.

Reasons for the lack of a difference between the Saccade–Discrimination and the No Saccade–Peripheral Discrimination conditions in the visual experiment are less obvious, but it is worth noting that there is evidence for differential sensitivity to the effects of saccades even within visual system functions. For example, dorsal stream (where) cognitive functions, such as mental rotation (Irwin & Brockmole, 2000) and identification of object orientation (Irwin & Brockmole, 2004), are affected by saccades, whereas ventral stream (what) functions, such as word identification and recognition (Irwin, 1998), are not.

The microsaccade data from both experiments also suggest that saccadic suppression-like mechanisms or attention effects, or perhaps both, modulate the subjective experience of time when relatively small eye movements are made during a timing signal. Of course, the effect sizes obtained in microsaccade analyses were comparatively smaller than the effect sizes obtained in saccade analyses. However, it is worth noting that microsaccades are much smaller in magnitude than saccades and that Yarrow et al. (2001) observed a larger time distortion as saccade amplitude increased. Hence, the small amplitude of microsaccades may account for their small, but statistically significant, effects on time perception.

Examining the effect of attention on timing has been a recurrent theme in the interval timing literature over the past few decades (e.g., Brown, 1997; Coull et al., 2004; Penney et al., 2014). In general, reducing the amount of attention allocated to time reduces the accuracy of time judgments with a given duration appearing subjectively shorter than when greater attention is allocated to timing. Within the information processing framework of Scalar Timing Theory (STT; Gibbon et al., 1984), the present findings can be interpreted as a consequence of the impact of attention on the clock stage of the internal clock. Specifically, when attention is reduced due to overt and covert shifts of attention fewer pacemaker pulses pass through the mode switch into the accumulator per unit of objective time. Such an effect could be additive in that it merely affects switch closure and opening at timing onset and offset, or it could be multiplicative by either altering the pacemaker rate or causing the switch to oscillate between the closed and opened states during the timing signal (e.g., Penney et al., 2000, 2005). In either case, it would take longer to reach the duration criterion, assuming the duration representation was created under conditions of full (or at least greater) attention. Here, we found a systematic relationship between the total saccade duration within a trial and the magnitude of the reproduced duration in both experiments with larger total saccade durations within a trial being associated with longer reproduced durations and a similar pattern held for the microsaccade analyses. These findings provide tentative support for a multiplicative effect of overt and covert attention shifts on timing.

Although STT remains the dominant model of timing in the cognitive literature, it has been challenged by more physiologically realistic models (e.g., Buhusi & Meck, 2005; Coull et al., 2011; Meck et al., 2008; Merchant et al., 2013). For example, according to the Striatal Beat Frequency model (Matell & Meck, 2000, 2004) a cortico–striatal–thalamic–cortical loop underlies temporal processing. Neurons in the cortex firing at different frequencies (i.e., oscillatory patterns) converge onto the striatal medium spiny neurons, which serve as coincidence detectors for patterns of activity that correspond to particular durations. As noted in Section 2.2, mechanisms related to saccade preparation may underlie both overt and covert attention shifts (Corbetta et al., 1998; Moore et al., 2003). Saccadic suppression itself has been attributed to effects at the level of the lateral geniculate nucleus (Thilo et al., 2004), which reduces the visual input from the retina to the visual cortex and other cortical areas, as well as suppression in dorsal stream areas, specifically the middle temporal, medial superior temporal, ventral intraparietal, and lateral intraparietal areas, that is independent of saccade related changes in visual input (Bremmer et al., 2009). Hence, it is plausible that components of a core timing network comprising cortical–striatal loops are affected by saccadic suppression mechanisms, although whether this occurs and a specific mode of action underlying such an effect remain to be determined.

In sum, the two experiments reported here suggest that saccades and microsaccades affect seconds range visual and auditory interval timing via attention and saccadic suppression mechanisms.

References

  • Allman M. J., , Teki S., , Griffiths T. D., , & Meck W. H. (2014). Properties of the internal clock: First-and second-order principles of subjective time. Annu. Rev. Psychol., 65, 743771.

    • Search Google Scholar
    • Export Citation
  • Bremmer F., , Kubischik M., , Hoffmann K. P., , & Krekelberg B. (2009). Neural dynamics of saccadic suppression. J. Neurosci., 29, 1237412383.

  • Brown S. W. (1997). Attentional resources in timing: Interference effects in concurrent temporal and nontemporal working memory tasks. Percept. Psychophys., 59, 11181140.

    • Search Google Scholar
    • Export Citation
  • Buhusi C. V., , & Meck W. H. (2005). What makes us tick? Functional and neural mechanisms of interval timing. Nat. Rev. Neurosci., 6, 755765.

    • Search Google Scholar
    • Export Citation
  • Campbell F. W., , & Wurtz R. H. (1978). Saccadic omission: Why we do not see a grey-out during a saccadic eye movement. Vision Res., 18, 12971303.

    • Search Google Scholar
    • Export Citation
  • Corbetta M., , Akbudak E., , Conturo T. E., , Snyder A. Z., , Ollinger J. M., , Drury H. A., , Linenweber M. R., , Petersen S. E., , Raichle M. E., , Van Essen D. C., , & Shulman G. L. (1998). A common network of functional areas for attention and eye movements. Neuron, 21, 761773.

    • Search Google Scholar
    • Export Citation
  • Coull J. T., , Vidal F., , Nazarian B., , & Macar F. (2004). Functional anatomy of the attentional modulation of time estimation. Science, 303, 15061508.

    • Search Google Scholar
    • Export Citation
  • Coull J. T., , Cheng R. K., , & Meck W. H. (2011). Neuroanatomical and neurochemical substrates of timing. Neuropsychopharmacology, 36, 325.

    • Search Google Scholar
    • Export Citation
  • Engbert R. (2006). Microsaccades: A microcosm for research on oculomotor control, attention, and visual perception. Prog. Brain Res., 154, 177192.

    • Search Google Scholar
    • Export Citation
  • Engbert R., , & Kliegl R. (2003). Microsaccades uncover the orientation of covert attention. Vision Res., 43, 10351045.

  • Engbert R., , & Mergenthaler K. (2006). Microsaccades are triggered by low retinal image slip. Proc. Natl Acad. Sci. U.S.A., 103, 71927197.

    • Search Google Scholar
    • Export Citation
  • Gibbon J., , Church R. M., , & Meck W. H. (1984). Scalar timing in memory. Ann. N. Y. Acad. Sci., 423, 5277.

  • Hafed Z. M., , Lovejoy L. P., , & Krauzlis R. J. (2011). Modulation of microsaccades in monkey during a covert visual attention task. J. Neurosci., 31, 1521915230.

    • Search Google Scholar
    • Export Citation
  • Hothorn T., , Bretz F., , & Westfall P. (2008). Simultaneous inference in general parametric models. Biom. J., 50, 346363.

  • Ibbotson M., , & Krekelberg B. (2011). Visual perception and saccadic eye movements. Curr. Opin. Neurobiol., 21, 553558.

  • Irwin D. E. (1998). Lexical processing during saccadic eye movements. Cogn. Psychol., 36, 127.

  • Irwin D. E., , & Brockmole J. R. (2000). Mental rotation is suppressed during saccadic eye movements. Psychon. Bull. Rev., 7, 654661.

  • Irwin D. E., , & Brockmole J. R. (2004). Suppressing where but not what: The effect of saccades on dorsal- and ventral-stream visual processing. Psychol. Sci., 15, 467473.

    • Search Google Scholar
    • Export Citation
  • Matell M. S., , & Meck W. H. (2000). Neuropsychological mechanisms of interval timing behavior. BioEssays, 22, 94103.

  • Matell M. S., , & Meck W. H. (2004). Cortico-striatal circuits and interval timing: Coincidence-detection of oscillatory processes. Cogn. Brain Res., 21, 139170.

    • Search Google Scholar
    • Export Citation
  • Meck W. H., , Penney T. B., , & Pouthas V. (2008). Cortico-striatal representation of time in animals and humans. Curr. Opin. Neurobiol., 18, 145152.

    • Search Google Scholar
    • Export Citation
  • Merchant H., , Harrington D. L., , & Meck W. H. (2013). Neural basis of the perception and estimation of time. Annu. Rev. Neurosci., 36, 313336.

    • Search Google Scholar
    • Export Citation
  • Moore T., , Armstrong K. M., , & Fallah M. (2003). Visuomotor origins of covert spatial attention. Neuron, 40, 671683.

  • Morey R. D. (2008). Confidence intervals from normalized data: A correction to Cousineau (2005). Tutor. Quant. Methods Psychol., 4, 6164.

    • Search Google Scholar
    • Export Citation
  • Morrone M. C., , Ross J., & Burr D. (2005). Saccadic eye movements cause compression of time as well as space. Nat. Neurosci., 8, 950954.

  • Murakami I., , & Cavanagh P. (1998). A jitter after-effect reveals motion-based stabilization of vision. Nature, 395, 798801.

  • Ölveczky B. P., , Baccus S. A., , & Meister M. (2003). Segregation of object and background motion in the retina. Nature, 423, 401408.

  • Pelli D. G. (1997). The VideoToolbox software for visual psychophysics:

  • Transforming numbers into movies. Spat. Vis., 10, 437–442.

  • Penney T. B., , Gibbon J., , & Meck W. H. (2000). Differential effects of auditory and visual signals on clock speed and temporal memory. J. Exp. Psychol. Hum. Percept. Perform., 26, 17701787.

    • Search Google Scholar
    • Export Citation
  • Penney T. B., , Meck W. H., , Roberts S. A., , Gibbon J., , & Erlenmeyer-Kimling L. (2005). Attention mediated interval timing deficits in individuals at high risk for schizophrenia. Brain Cogn., 58, 109118.

    • Search Google Scholar
    • Export Citation
  • Penney T. B., , Yim E. N. K., & Ng., K. K. (2014). Distractor expectancy effects on interval timing. Timing Time Percept., 2, 119.

  • Pinheiro J., , Bates D., , DebRoy S., , Sarkar D., , & Core Team R. (2015). nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-122, http://CRAN.R-project.org/package=nlme.

    • Search Google Scholar
    • Export Citation
  • Core Team R (2013). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.

    • Search Google Scholar
    • Export Citation
  • Riggs L. A., , Merton P. A., , & Morton H. B. (1974). Suppression of visual phosphenes during saccadic eye movements. Vision Res., 14, 9971011.

    • Search Google Scholar
    • Export Citation
  • Rolfs M. (2009). Microsaccades: Small steps on a long way. Vision Res., 49, 24152441.

  • Suzuki M., , & Yamazaki Y. (2010). Distortion of space and time during saccadic eye movements. Intell. Inf. Manag., 2, 9094.

  • Thilo K. V., , Santoro L., , Walsh V., , & Blakemore C. (2004). The site of saccadic suppression. Nat. Neurosci., 7, 1314.

  • Tovee M. J. (1996). An introduction to the visual system. Cambridge, UK: Cambridge University Press.

  • Watson T., , & Krekelberg B. (2011). An equivalent noise investigation of saccadic suppression. J. Neurosci., 31, 65356541.

  • Westfall P. H. (1997). Multiple testing of general contrasts using logical constraints and correlations. J. Am. Stat. Assoc., 92, 299306.

    • Search Google Scholar
    • Export Citation
  • Wurtz R. H. (2008). Neuronal mechanisms of visual stability. Vision Res., 48, 20702089.

  • Yarrow K., , Haggard P., , Heal R., , Brown P., , & Rothwell J. C. (2001). Illusory perceptions of space and time preserve cross-saccadic perceptual continuity. Nature, 414, 302305.

    • Search Google Scholar
    • Export Citation
  • Yarrow K., , Johnson H., , Haggard P., , & Rothwell J. C. (2004a). Consistent chronostasis effects across saccade categories imply a subcortical efferent trigger. J. Cogn. Neurosci., 16, 839847.

    • Search Google Scholar
    • Export Citation
  • Yarrow K., , Haggard P., , & Rothwell J. C. (2004b). Action, arousal, and subjective time. Conscious. Cogn., 13, 373390.

  • Zuber B. L., , & Stark L. (1966). Saccadic suppression: Elevation of visual threshold associated with saccadic eye movements. Exp. Neurol., 16, 6579.

    • Search Google Scholar
    • Export Citation

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 308 49 5
Full Text Views 232 2 0
PDF Views & Downloads 15 0 0