Cortical feedback signals generalise across different spatial frequencies of feedforward inputs

Revina, Yulia; Petro, Lucy S.; Muckli, Lars

doi:10.1016/j.neuroimage.2017.09.047

Cited by 35 publications

(43 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As previously argued (Muckli et al, 2015), this observation could suggest that the feed-back signals from higher-level regions with larger receptive fields carry information that is more abstract and therefore spatially coarser than its feed-forward counterpart. While this claim is supported by a recent study directly investigating the contribution of spatial frequencies to feedback signal (Revina et al, 2018), as indicated by the size of the error bars, this finding could simply be related to noisier signal rather than an apparent coarser resolution. Moreover, as previously mentioned, we observed that decoding accuracy is differentially modulated by misalignment for feedback and feedforward signals.…”

Section: Differences Between Volume and Surface Based Misalignmentsmentioning

confidence: 71%

Multivoxel Pattern of Blood Oxygen Level Dependent Activity can be sensitive to stimulus specific fine scale responses

Vizioli

Martino

Petro

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

1.AbstractAt ultra-high field, fMRI voxels can span the sub-millimeter range, allowing the recording of blood oxygenation level dependent (BOLD) responses at the level of fundamental units of neural computation, such as cortical columns and layers. This sub-millimeter resolution, however, is only nominal in nature as a number of factors limit the spatial acuity of functional voxels. Multivoxel Pattern Analysis (MVPA) may provide a means to detect information at finer spatial scales that may otherwise not be visible at the single voxel level due to limitations in sensitivity and specificity. Here, we evaluate the spatial scale of stimuli specific BOLD responses in multivoxel patterns exploited by linear Support Vector Machine, Linear Discriminant Analysis and Naïve Bayesian classifiers across cortical depths in V1. To this end, we artificially misaligned the testing relative to the training portion of the data in increasing spatial steps, then investigated the breakdown of the classifiers’ performances. A one voxel shift led to a significant decrease in decoding accuracy (p<.05) across all cortical depths, indicating that stimulus specific responses in a multivoxel pattern of BOLD activity exploited by multivariate decoders can be as precise as the nominal resolution of single voxels (here .8 mm isotropic). Our results further indicate that large draining vessels, prominently residing in proximity of the pial surface, do not, in this case, hinder the ability of MVPA to exploit fine scale patterns of BOLD signals. We argue that tailored analytical approaches can help overcoming limitations in high-resolution fMRI and permit studying the mesoscale organization of the human brain with higher sensitivities.

show abstract

Section: Differences Between Volume and Surface Based Misalignmentsmentioning

confidence: 71%

Multivoxel Pattern of Blood Oxygen Level Dependent Activity can be sensitive to stimulus specific fine scale responses

Vizioli

Martino

Petro

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our results could be justified by the findings in de-Wit et al (2012), where top-down prediction was shown to be topologically inaccurate as it led to activity reduction in the whole of V1, rather than at the predicted location. Revina et al (2017) also found that brain patterns due to top-down modulation did not share information with the corresponding bottom-up signals. In the prediction error realm, successful prediction would lead to zero error in the higher visual areas, and thus feature gain would decrease.…”

Section: Discussionmentioning

confidence: 95%

“…The reverse hierarchy theory also suggests that top-down modulation serves to fine-tune sensory signals by means of predictions initially made using lower spatial frequency features (Hochstein and Ahissar, 2002;Ahissar and Hochstein, 2004). Furthermore, Revina et al (2017) showed that blurred stimuli can generate top-down processes that generalize to higher spatial frequencies.…”

Section: Discussionmentioning

confidence: 97%

Sharpening of hierarchical visual feature representations of blurred images

Abdelhack

Kamitani

2017

Preprint

View full text Add to dashboard Cite

The robustness of the visual system lies in its ability to perceive degraded images. This is achieved through interacting bottom-up and top-down pathways that process the visual input in concordance with stored prior information. The interaction mechanism by which they integrate visual input and prior information is still enigmatic. We present a new approach using deep neural network (DNN) representation to reveal the effects of such integration on degraded visual inputs. We transformed measured human brain activity resulting from viewing blurred images to the hierarchical representation space derived from a feedforward DNN. Transformed representations were found to veer towards the original non-blurred image and away from the blurred stimulus image. This indicated deblurring or sharpening in the neural representation, and possibly in our perception, from top-down modulation. We anticipate these results will help unravel the interplay mechanism between bottom-up and top-down pathways, leading to more comprehensive models of vision.

show abstract

“…Because of its latency and its polarity reversal for top versus bottom stimulation, this component is believed to reflect feedbacks from higher-level visual areas to the early visual cortex (Miller et al 2015). It was previously suggested that these feedbacks might provide information about low and high-level scene features, such as the category or the depth of the stimuli (Morgan et al 2016;Revina et al 2018). That top-bottom feedback signals are thought to modulate sensory input, providing information about the global scene structure.…”

Section: Discussionmentioning

confidence: 99%

Dynamics of the straight-ahead preference in human visual cortex

et al. 2019

View full text Add to dashboard Cite

The objects located straight-ahead of the body are preferentially processed by the visual system. They are more rapidly detected and evoke stronger BOLD responses in early visual areas than elements that are retinotopically identical but located at eccentric spatial positions. To characterize the dynamics of the underlying neural mechanisms, we recorded in 29 subjects the EEG responses to peripheral targets differing solely by their locations with respect to the body. Straight-ahead stimuli led to stronger responses than eccentric stimuli for several components whose latencies ranged between 70 and 350 ms after stimulus onset. The earliest effects were found at 70 ms for a component that originates from occipital areas, the contralateral P1. To determine whether the straight-ahead direction affects primary visual cortex responses, we performed an additional experiment (n = 29) specifically designed to generate two robust components, the C1 and C2, whose cortical origins are constrained within areas V1, V2 and V3. Our analyses confirmed all the results of the first experiment and also revealed that the C2 amplitude between 130 and 160 ms after stimulus onset was significantly stronger for straight-ahead stimuli. A frequency analysis of the pre-stimulus baseline revealed that gaze-driven alterations in the visual hemi-field containing the straight-ahead direction were associated with a decrease in alpha power in the contralateral hemisphere, suggesting the implication of specific neural modulations before stimulus onset. Altogether, our EEG data demonstrate that preferential responses to the straight-ahead direction can be detected in the visual cortex as early as about 70 ms after stimulus onset.

show abstract

Cortical feedback signals generalise across different spatial frequencies of feedforward inputs

Cited by 35 publications

References 41 publications

Multivoxel Pattern of Blood Oxygen Level Dependent Activity can be sensitive to stimulus specific fine scale responses

Multivoxel Pattern of Blood Oxygen Level Dependent Activity can be sensitive to stimulus specific fine scale responses

Sharpening of hierarchical visual feature representations of blurred images

Dynamics of the straight-ahead preference in human visual cortex

Contact Info

Product

Resources

About