Retrospective Attention Can Trigger Visual Perception Without Dependence on Either Cue Awareness or Target Reporting

  1. Ye Li
  2. Xingbei Wan
  3. Mingxing Mao
  4. Lingying Jin
  5. Xilin Zhang

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

The interplay between attention and visual consciousness remains pivotal in cognitive neuroscience, framed by three hypotheses: dual dissociation (independent mechanisms), single dissociation (consciousness requires attention but not vice versa), and interdependence (bidirectional coupling). We tested these through retrospective attention—a bottom-up post-stimulus mechanism—to clarify its role in visual awareness.

Three EEG experiments combined backward masking and no-report paradigms to assess retrospective attention’s effects on Gabor patch discrimination. Experiments 1&2 showed invisible retro-cues enhanced orientation discrimination at 66.67 ms SOA (via N2pc modulation), while visible cues sustained effects up to 100 ms. Experiment 3 revealed explicit reporting amplified P3b and attenuated Visual Awareness Negativity (VAN), whereas no-report conditions retained contrast-driven VAN without P3b modulation. Multivariate decoding linked report-dependent attention to frontoparietal networks with temporally generalized activity, while no-report effects localized to parieto-occipital regions, enhancing awareness decoding at high contrast.

Results support the single dissociation hypothesis: retrospective attention enhances perception independently of cue awareness and reporting, yet flexibly modulates both access (report-linked P3b) and phenomenal consciousness (contrast-dependent VAN). This study provides the first neural evidence for unconscious retrospective attention, extends the global neuronal workspace theory by dissociating report-dependent/independent processes, and clarifies temporal dynamics in awareness formation. By distinguishing phenomenal and access consciousness, we establish attention’s critical role in mediating visual awareness, advancing cognitive neuroscience frameworks.

Significance Statement

This study reveals that unconscious retrospective attention enhances visual perception independently of explicit reporting, challenging traditional views that attention requires consciousness. By dissociating neural mechanisms of phenomenal (subjective experience) and access consciousness (reportable awareness), we demonstrate attention’s dual role in modulating both processes. Our findings extend the global neuronal workspace theory, provide the first neural evidence for unconscious retroactive attentional effects, and clarify how attentional mechanisms temporally shape awareness. Methodologically, integrating no-report paradigms with multivariate decoding advances consciousness research by minimizing reporting biases. These insights bridge theoretical debates in cognitive neuroscience and inform clinical approaches to consciousness disorders and AI models of visual processing.

Article activity feed

  1. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/17117530.

    Summary

    The article "Retrospective Attention Can Trigger Visual Perception Without Dependence on Either Cue Awareness or Target Reporting" by Li et al 2024 examined visual consciousness through the usage of retrospective attention experimentation. The overall goal was elucidating retrospective attention effects as an attention or sensory based mechanism, whether it persists in an unconscious paradigm, and characterizing its associated neural activity. 

    Strengths

    • Introduction

      • Introduces research topics very well by using some existing studies to build upon their credibility. 

      • Makes strong connections with existing theories, such as the Global Neuronal Workspace Theory (GNWT). 

      • Establishes the research gap being studied and its context.

    •  Results

      • Data is anonymized and available. 

      • This section highlights both significant and non-significant outcomes, which improves the transparency of the paper.

      • Figures are well made and labeled.

    • Discussion

      • The discussion is well-rounded, tying back to the concepts mentioned in the introduction and addressing limitations.

    • Methodology

      • The control conditions such as visible and invisible cues with some manipulation help the study make comparisons of conscious vs. unconscious cue processing and validating the results.

      • The 2AFC staircase method helps adjust the task to the participants ability making the difficulty fair for all conditions.

      • Threshold tests were implemented before the main task which helped ensure that each of the participants' performances were measured which reduced variability.

      • Sessions were counterbalanced to reduce order effects. 

    Suggestions/Improvements

    • Introduction and Discussion

      • Although this may be intentional, the paper seems to be written for an expert audience because of its extensive usage of technical terms. Providing context or definitions of vocabulary would serve a wider audience. 

      • The introduction and discussion could benefit from well-elaborated real world examples connected to the data. 

      • Discussion lacks clarity on key takeaways and has unnecessary repetition, like when talking about how retroactive attention boosts conscious perception.

    • Methodology and Results

      • The results section heavily summarizes information already available from the figures. It would be beneficial to instead emphasize the main takeaway. Having a better and clearer narrative over the trends would make the results easier to interpret. 

      • Error bars in the figures are often covered in data points. They should be visually enhanced for clarity. 

      • All the participants fall into the range of 18-29 years old which could make the results less applicable to people from other age groups.

    Competing interests

    The authors declare that they have no competing interests.

  2. Version published to 10.1101/2024.11.09.622275 on bioRxiv