Elsevier

Cognition

Volume 115, Issue 1, April 2010, Pages 186-191
Cognition

Brief article
Rescuing stimuli from invisibility: Inducing a momentary release from visual masking with pre-target entrainment

https://doi.org/10.1016/j.cognition.2009.11.010Get rights and content

Abstract

At near-threshold levels of stimulation, identical stimulus parameters can result in very different phenomenal experiences. Can we manipulate which stimuli reach consciousness? Here we show that consciousness of otherwise masked stimuli can be experimentally induced by sensory entrainment. We preceded a backward-masked stimulus with a series of rapid visual events presented at 12 Hz for 800 ms. Peaks in visual sensitivity (d′) were observed when the target appeared at the time that the next entraining stimuli would have occurred. Observers’ sensitivity for identical masked near-threshold stimuli increased by factors as large as 55%, but only at this precise moment in time. These data thus reveal that awareness of near-threshold stimuli can be manipulated by entrainment to rhythmic events, supporting the functional role of induced oscillations in underlying cortical excitability, and suggest a plausible mechanism of temporal attention.

Introduction

When stimuli are masked to be at the threshold of consciousness, identical stimulus presentations can result in very different states of awareness. Psychophysical laws interpret these variations as stochastic phenomena and do not address why any given stimulus reaches awareness. However, given that the stimulus parameters are identical, the source of this variability in awareness must reflect variability in the state of the visual system, which is typically not under experimental control. Here we ask whether these differing states can in fact be manipulated, inducing predictable variations in awareness via rhythmic visual signals in the environment.

Prior work has shown modulation of auditory processing by a rhythmic series of tones, resulting in enhanced sample matching at times in-phase with the preceding tones (Jones, Moynihan, MacKenzie, & Puente, 2002). Jones et al. presented participants with a target tone at the beginning of a trial, followed by a series of irrelevant tones with an embedded rhythmicity. A probe tone was presented at variable intervals following the end of the rhythmic sequence. Highest matching-task performance was found on trials when the probe tones were presented in-phase with the rhythmic sequence, compared to those presented out-of-phase. The authors interpreted this finding in terms of “tuning” of temporal attention by predictable events; periodic tones (even if irrelevant) had the effect of narrowing the temporal selection window of attention (more precise time-based selection), and increasing the attentional gain (better signal-to-noise ratio). Similarly, it has been shown that oddball targets in a series of attended standard tones elicit larger mismatch negativity (MMN) in a constant rate train than those in a variable rate train of tones (Erickson & Hetrick, 2009), again indicating enhanced auditory processing for environmental stimuli occurring at predictable times.

Following Jones et al.’s work in the auditory modality, a related effect has been documented in temporal attention in the visual modality (the “attentional awakening” effect, see Ariga & Yokosawa, 2008; Ambinder and Lleras, 2009, Ariga & Yokosawa, 2008). In those experiments, participants were asked to identify a white letter embedded in a rapid stream of visually presented (RSVP) blue letters. Ariga and Yokosawa observed that performance in this task was poor for targets appearing early on in the RSVP stream, but improved to asymptote as the RSVP unfolded. Much like Jones et al.’s auditory temporal tuning account, this effect was interpreted as evidence of temporal attention tuning-into the timing of the series of visual events with the goal of better processing each individual item.

Moreover, it has recently been shown that sensory processing and awareness vary with respect to the phase of the ongoing EEG alpha rhythm (Busch et al., 2009, Lakatos et al., 2008, Mathewson et al., 2009). Given that rhythmic visual stimuli have been shown to induce comparable oscillations in neural firing (Herrmann, 2001, Lakatos et al., 2008, Pastor et al., 2002, Williams et al., 2004), the present study also sought to corroborate the findings of concomitant oscillations in neural firing and awareness, investigating for the first time the effect of the phase of these induced oscillations on visual awareness.

Our goal in the current study was to ask whether the visual system could be entrained to produce temporally well defined peaks in sensitivity at predictable times in a sequence. Moreover, we asked whether such entrainment could be used to modulate awareness of near-threshold masked stimuli. That is, given attention to a single location in space, can predictable events in the environment improve visual sensitivity to better process brief periods in time, such that we can manipulate which stimuli will reach consciousness? We hypothesized that the threshold of visual awareness could be entrained or synchronized to the expected occurrence of rapid, rhythmic events; with maximal sensitivity occurring for targets presented in-phase with the preceding entraining stimulation, and reduced sensitivity for out-of-phase targets. The experimental task was to detect the presence of a luminance disk at fixation. We used a metacontrast mask (at optimal stimulus-onset-asynchrony; SOA) to reduce target visibility to near-threshold levels and minimize contributions from trace processing of post-target visible persistence activation by diminishing re-entrant activity (Di Lollo et al., 2000, Enns, 2004, Ro et al., 2003).

Section snippets

Participants

Sixteen students from the University of Illinois at Urbana-Champaign received course credit in a Psychology class for participation and signed informed consent.

Materials and design

Subjects were seated 57 cm away from a 21 in CRT monitor set to 85 Hz in a dimly lit room. Fig. 1 shows the stimulus dimensions and task sequence. All stimuli were presented at fixation. The target consisted of a 1° disk. The mask annulus was 2°, with a 1° center cut out, and was also used as the entraining stimulus. All stimulus

Results

The average detection rate in the no-entrainers condition (M = 0.41) indicated considerable metacontrast masking. It should be noted, however, that this condition is not directly comparable to the remaining conditions because it lacks a forward masking stimulus: that is to say, in all entraining conditions (as well as in the control condition), the last entraining annulus in the sequence served as a forward masking stimulus to the target, further reducing target visibility.

Fig. 2 plots the

Discussion

Here we showed for the first time that a momentary release from the effects of visual masking can be elicited by rapid presentation of visual stimuli, causing a temporally-precise peak in visual sensitivity. When targets were presented in-phase with preceding entrainers, target detection and sensitivity were substantially improved. In contrast, targets presented out-of-phase with the previous entrainment were less likely to reach awareness, although they were no less detectable than targets in

Conclusion

Our results show that in the presence of rapid, periodic visual events, visual processing appears exogenously synchronized to the frequency of visual stimulation to produce phase-locked peaks of sensitivity. This entrainment may represent a mechanism by which temporal attention is tuned to produce temporally-precise peaks in visual sensitivity, to both anticipate and optimize visual processing of brief visual events. The phase effects here mirror those related to endogenously generated brain

Acknowledgements

Supported by a Natural Science and Engineering Research Council of Canada Scholarship to K.E.M.

References (30)

  • Ariga et al.

    Attentional awakening: Gradual modulation of temporal attention in rapid serial visual presentation

    Psychological Research

    (2008)
  • F. Bauer et al.

    Gamma flicker triggers attentional selection without awareness

    Proceedings of the National Academy of Science

    (2009)
  • K.M. Bausenhart et al.

    Temporal preparation improves temporal resolution: Evidence from constant foreperiods

    Perception & Psychophysics

    (2008)
  • B.G. Breitmeyer

    Visual masking: An integrative approach

    (1984)
  • N.A. Busch et al.

    The phase of ongoing EEG oscillations predicts visual perception

    The Journal of Neuroscience

    (2009)
  • Cited by (140)

    • Occipital alpha-TMS causally modulates temporal order judgements: Evidence for discrete temporal windows in vision

      2021, NeuroImage
      Citation Excerpt :

      The LGN is hypothesized to project not only to V1, but also directly to higher cortical areas like V2 and V3 (Schmid et al., 2010). Strictly speaking every possible target of visual entrainment could serve as a potential source for the behavioral observations previously reported (Mathewson et al., 2010; Ronconi et al., 2018; Spaak et al., 2014; Chota and VanRullen, 2019). TMS allows for a direct and relatively localized interaction with endogenous cortical rhythms and is especially efficient at 10 Hz, even when individual α-peak frequencies differ (Romei et al., 2010; Thut et al., 2011).

    • Neural Oscillations and Multisensory Processing

      2024, Advances in Experimental Medicine and Biology
    View all citing articles on Scopus
    View full text