Position representations of moving objects align with real-time position in the early visual response

  1. Philippa Anne Johnson
  2. Tessel Blom
  3. Simon van Gaal
  4. Daniel Feuerriegel
  5. Stefan Bode
  6. Hinze Hogendoorn

  • Curated by eLife

    eLife logo

    eLife assessment

    This paper is of potential interest to any neuroscientist, given it asks how the brain compensates for its own neural transmission delays. This is a problem that runs across neuroscientific disciplines. The authors use a clever and simple design where they study this question in the context of decoding from EEG signals during visual motion processing. They robustly show evidence that the brain can indeed compensate for these delays, although all compensation appears to be afforded by early processing. The manuscript is well-written but can be strengthened by outlining its significance for the broader community as well as some further analyses.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

When interacting with the dynamic world, the brain receives outdated sensory information, due to the time required for neural transmission and processing. In motion perception, the brain may overcome these fundamental delays through predictively encoding the position of moving objects using information from their past trajectories. In the present study, we evaluated this proposition using multivariate analysis of high temporal resolution electroencephalographic data. We tracked neural position representations of moving objects at different stages of visual processing, relative to the real-time position of the object. During early stimulus-evoked activity, position representations of moving objects were activated substantially earlier than the equivalent activity evoked by unpredictable flashes, aligning the earliest representations of moving stimuli with their real-time positions. These findings indicate that the predictability of straight trajectories enables full compensation for the neural delays accumulated early in stimulus processing, but that delays still accumulate across later stages of cortical processing.

Article activity feed

  1. Version published to 10.7554/elife.82424 on eLife
  2. eLife

    eLife assessment

    This paper is of potential interest to any neuroscientist, given it asks how the brain compensates for its own neural transmission delays. This is a problem that runs across neuroscientific disciplines. The authors use a clever and simple design where they study this question in the context of decoding from EEG signals during visual motion processing. They robustly show evidence that the brain can indeed compensate for these delays, although all compensation appears to be afforded by early processing. The manuscript is well-written but can be strengthened by outlining its significance for the broader community as well as some further analyses.

  3. eLife

    Reviewer #1 (Public Review):

    The present study was concerned with examining the crucial question of how the brain compensates for its own neural transmission delays, such that representations of moving objects bare some resemblance to their real position rather than their position several hundred ms ago (the time for the retinal input to be transmitted to various points in the hierarchy). It asked whether such compensation may all be seen in early visual nodes of the hierarchy (e.g., V1), with any apparent compensation in subsequent nodes (e.g., V4) generated by simply receiving its information from earlier regions, or whether there is evidence for compensation generated in later nodes.

    The authors used a decoding approach to examine this question, where they trained binary classifiers on static locations within a grid and tested on moving stimuli that involved wave-like motion across this grid. One element that makes this study interesting is that it provides a novel demonstration that one can effectively train on static locations and exhibit above-chance performance at tests with moving stimuli. This suggests a common representation of static and moving events. When training and testing on matching timepoints (training=static and test=moving) the classifier performed above chance between 102 and 180 ms after stimulus onset.

    They used different epochs to train (static events) to approximate different stages of neural processing and different test epochs (moving events) to examine when the static location representation was predominantly active during motion. The position of a moving object was represented at a time 70 ms shorter than a static object in the same location. For training times 70-80 ms (likely corresponding to V1-V3 activation) the latency of neural activation for moving stimuli approximately corresponded to its real-time position, unlike on static trials. The authors state that subsequent cortical areas do not implement further compensation for neural delays. The claims appear supported by the data.

    I found this manuscript exceptionally well-written - well-tuned, clear, and interesting. I also find the patterns in the data interesting and believe this view would be shared by other visual neuroscientists as well as neuroscientists from other fields - where neural delays are likely to prove a universal sticking point to many theories.

  4. eLife

    Reviewer #2 (Public Review):

    In this study, the authors set out to decode the latency of position representations of static and moving stimuli using EEG multivariate pattern analysis. Linear classifiers were trained on the positions of static stimuli and then generalized to the positions of moving objects in a time-resolved manner. The authors find that the early neural representations of the position of moving stimuli are close to positions in the real world. As neural delays from the retina to the early visual cortex should theoretically induce a latency of ~70 ms their findings suggest that these delays are compensated very early in the visual hierarchy. Furthermore, they find that delays that are accumulated during subsequent processing stages of the visual hierarchy are not compensated, supporting the interpretation of an early compensation mechanism.

    I congratulate the authors on this excellent scientific work. I believe its major strength lies in the successful attempt to generalize neural representations of static objects to moving objects. This is made possible due to the large amount of collected EEG data as well as smart task design. Effectively this allows the authors to track which location is currently represented in the brain and how this compares to the actual physical location, all in a time-resolved manner. The approach is remarkably robust against biases due to its relative simplicity, both in task design and analysis.

    One of the few limitations of the study is their inability to generalize very early location signals from static to moving objects. This might be indicative of differences in neural codes/mechanisms and in turn, limits the interpretation of which stages of the visual hierarchy are involved in motion extrapolation. That being said, I agree with the authors that this is a fundamentally difficult problem to solve, and importantly it does not negatively impact the main conclusions of this paper.

    The current work provides significant methodological and theoretical utility. I am certain that the classification method and principal task design will be used by future studies investigating motion perception due to their effectiveness in tracking internally represented locations. On a theoretical level, the authors' results provide strong evidence that motion compensation processes occur very early in the visual hierarchy. There has been an ongoing debate about how and where this is achieved in the visual system and fMRI studies have only provided limited evidence to solve this issue due to the sluggish nature of the BOLD signal. In addition, the present results challenge previous theories on the role of feed-forward and feedback signals in neural delay compensation and provide concrete directions for future research.

  5. eLife

    Reviewer #3 (Public Review):

    This paper demonstrates neural mechanisms important for the representation of moving stimuli. Specifically, using EEG, the authors investigated the temporal profiles of visual activities that correspond to changes in positions of moving stimuli.

    Strengths:
    The authors examined an interesting question of how moving stimuli can be smoothly represented and perceived by using a neural recording modality with high temporal resolution. To my knowledge, the temporal dynamics of the neural correlates of successful motion perception are not well understood, and the study provides evidence for a plausible mechanism for this process. Additionally, the paper is well-written where the results are clearly communicated, and the figures are clearly presented.

    Weaknesses:
    The findings reported are derived from a specific case of motion perception which may not reflect the general mechanisms optimized for motion perception. The limitations related to task designs and the neural readouts should be discussed as they affect the way that the reported results will be interpreted.

  6. Version published to 10.1101/2022.06.26.496535v1 on bioRxiv