Investigating the effects of memory on contextual feedback signals in layers of V1
Poster No:
2082
Submission Type:
Abstract Submission
Authors:
Zirui Zhang1,2, Clement Abbatecola1,2, Lucy Petro1,2, Angus Paton2, Yulia Lazarova1,2, Lars Muckli1,2
Institutions:
1School of Psychology and Neuroscience, University of Glasgow, Glasgow, United Kingdom, 2Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow, Glasgow, United Kingdom
First Author:
Zirui Zhang
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
Co-Author(s):
Clement Abbatecola
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
Lucy Petro
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
Angus Paton
Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom
Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom
Yulia Lazarova
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
Lars Muckli
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
School of Psychology and Neuroscience, University of Glasgow|Imaging Centre for Excellence, Queen Elizabeth University Hospital, University of Glasgow
Glasgow, United Kingdom|Glasgow, United Kingdom
Introduction:
Cortical feedback processing plays a crucial role in visual processing. We isolate feedback signals in V1 using visual scene occlusion (Smith & Muckli, 2010). Recent studies recording in mice V1 have shown that responses of L2/3 pyramidal cells selective for occluded images are strengthened upon familiarization with the images, and multivariate decoding analyses showed that these responses contain image-specific information (Seignette et al., 2024). Previous occlusion paradigms using single-specific images demonstrated that contextual feedback information is predominantly presented in superficial layers of V1 (Muckli et al., 2015). Here, we used unrepeated multi-image sets from two categories and manipulated the familiarization of these images. We investigated how varying contextual surround familiarity influences cortical feedback and hypothesized that memory enhances the decoding of image categories in feedback signals in V1.
Methods:
Eighteen participants (mean age = 24.81 years) completed a 7T fMRI study at the Imaging Centre of Excellence, University of Glasgow investigating the effects of memory and familiarity on cortical feedback during natural scene perception. Using 192 grayscale images from two categories (City and Office) across three familiarity conditions (Novel, Familiar, Very Familiar), the experiment included: (1) Pre-training, where participants viewed Very Familiar images seven times and Familiar images once; (2) Scanning, where we presented all images with a lower-right occluder while recording brain activity and (3) Post-testing, where we checked familiarity with recognition task (Fig 1a). To define visual areas (V1-V3) we performed retinotopic mapping (polar angle) with high-contrast flickering checkerboards and localized regions of interest (ROIs) within V1, V2, and V3 by mapping the occluded target region, the border and the surround of the ROIs (Target, Surround, Border) (Fig 1b&c). A multivoxel pattern analysis (MVPA) using a linear SVM compared category and memory decoding between feedback and feedforward V1 regions, further stratified into six equidistant cortical depth layers (0.1 to 0.9 from deep to superficial) (Fig 1d).
·Fig1: a) Experimental Procedure & stimuli example: 1. pre-train 2. scanning 3. post-test. b - c) ROI definition: processing of V1 feedback and feedforward areas and layers
Results:
We applied SVM classification to fMRI data from the V1-visually stimulated feedforward regions (including feedback in these regions) and the visually occluded feedback-only stimulated regions to investigate image category decoding (City vs. Office). The classifier accuracy was significantly above chance in the feedback and feedforward regions (Fig 2a). Separating data by familiarity levels, decoding was significant only for the Very Familiar condition in the feedback region (Fig 2b) and for Familiar and Very Familiar images in the feedforward region (Fig 2c).
Layer-wise analyses revealed that in the feedback region, decoding for City vs. Office was successful only in deeper layers (10% - 26%) and only for very familiar images. In the feedforward region, image category decoding occurred in superficial and middle layers across all 3 familiarity conditions, but in the deep layers only for familiar images (Fig 2c).
Layer-wise analyses revealed that in the feedback region, decoding for City vs. Office was successful only in deeper layers (10% - 26%) and only for very familiar images. In the feedforward region, image category decoding occurred in superficial and middle layers across all 3 familiarity conditions, but in the deep layers only for familiar images (Fig 2c).
·Fig2: Categorical City vs Office decoding in V1 (left plots: layers) a) Overall results at feedback and feedforward b) feedback: 3 memory conditions c) feedforward: 3 memory conditions
Conclusions:
Previous studies in human and mice found image related feedback to superficial layers of visual cortex in regions processing occluded image regions (Muckli et al 2015; Seignette et al. 2024). Here, we found image categorisation in deep layers, that was dependent on memory. In follow-up studies, we are investigating these layer profiles and their relation to cognitive task.
Our findings suggest that V1 receives categorical information via cortical feedback to deep layers, with memory effects enhancing categorical discriminability through this feedback mechanism. Preliminary results suggest memory-driven feedback influences categorical decoding in deeper cortical layers.
Our findings suggest that V1 receives categorical information via cortical feedback to deep layers, with memory effects enhancing categorical discriminability through this feedback mechanism. Preliminary results suggest memory-driven feedback influences categorical decoding in deeper cortical layers.
Learning and Memory:
Learning and Memory Other 2
Modeling and Analysis Methods:
Multivariate Approaches
Novel Imaging Acquisition Methods:
BOLD fMRI
Perception, Attention and Motor Behavior:
Perception: Visual 1
Keywords:
Learning
Memory
Vision
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Seignette, K., de Kraker, L., Papale, P., Petro, L. S., Hobo, B., Montijn, J. S., Self, M. W., Larkum, M. E., Roelfsema, P. R., Muckli, L., & Levelt, C. N. (2024). Experience-dependent predictions of feedforward and contextual information in mouse visual cortex. bioRxiv. https://doi.org/10.1101/2024.06.10.598181
Smith, F. W., Muckli, L. (2010). Nonstimulated early visual areas carry information about surrounding context. Proceedings of the National Academy of Sciences - PNAS, 107(46), 20099-20103. https://doi.org/10.1073/pnas.1000233107