In vivo validation of simulated run-to-run variability in population receptive field estimations
Poster No:
2080
Submission Type:
Abstract Submission
Authors:
Siddharth Mittal1, Michael Woletz1, David Linhardt1, Christian Windischberger1
Institutions:
1Medical University of Vienna, Vienna, Austria
First Author:
Co-Author(s):
Introduction:
Population receptive field (pRF) mapping is a widely used approach in functional neuroimaging to analyze the relationship between visual field stimulation and neural activation in the visual cortex. Despite advancements in modelling, software, and stimulus design, significant run-to-run variability persists in estimated pRF parameters. This study used computational modelling to simulate variability across runs and validate results against in vivo variability observed in 30 empirical retinotopy runs.
Methods:
All computations were performed using our novel pRF mapping software GEM-pRF, which was designed to take full advantage of GPU acceleration in order to minimise computation time (Mittal et al., 2024). Empirical data was collected from a retinotopy experiment on a young healthy female subject across five sessions (six runs each, 30 runs total). A moving bar with a flickering checkerboard served as the stimulus. pRF estimations for each vertex were computed using GEM-pRF, yielding 30 pRF values per vertex based on a 2D isotropic Gaussian model, with parameters including spatial coordinates (μx, μy) and size (σ).
For simulations, receptive fields were uniformly distributed across 51×51 spatial locations (μx, μy) with varying sizes (σ). Using the same stimulus, 5000 simulated BOLD responses per location and size were generated with white noise, achieving a contrast-to-noise ratio (CNR) of 1 to 4. This dataset was processed using GEM-pRF, and the resulting parameters were used to estimate the probability distribution of pRF parameters for each ground-truth position. Maximum likelihood (ML) estimation matched simulated distributions to empirical voxel data. Multivariate Kolmogorov-Smirnov tests (Naaman, 2021) assessed differences between simulated and empirical distributions.
For simulations, receptive fields were uniformly distributed across 51×51 spatial locations (μx, μy) with varying sizes (σ). Using the same stimulus, 5000 simulated BOLD responses per location and size were generated with white noise, achieving a contrast-to-noise ratio (CNR) of 1 to 4. This dataset was processed using GEM-pRF, and the resulting parameters were used to estimate the probability distribution of pRF parameters for each ground-truth position. Maximum likelihood (ML) estimation matched simulated distributions to empirical voxel data. Multivariate Kolmogorov-Smirnov tests (Naaman, 2021) assessed differences between simulated and empirical distributions.
Results:
Empirical data showed that the estimations of pRF results across runs varied depending on the position in the visual field (μx, μy). Variability was lower in the foveal regions compared to the more peripheral areas. In particular, distributions in foveal regions were found circular and Gaussian-like, while peripheral locations exhibited ellipsoidal distributions with increased width along the eccentricity direction. Figure 1 illustrates a matching result for a foveal vertex, while Figure 2 represents the matching of a parafoveal vertex. Both figures highlight the scattering of estimated pRF parameters for a given vertex, overlaid on the KDE distribution of the matched simulated pRF. The matched KDEs of the parafoveal vertex (Figure 2) and the near-foveal region vertex (Figure 1) reveal distinct differences in parameter variability across the spatial dimensions (μx, μy) and pRF size (σ). The parafoveal vertex exhibits broader spreads of 5.58, 6.3, and 1.95 for μx, μy, and σ, respectively, with corresponding standard deviations of 1.18, 1.47, and 0.57. The covariance matrix for the spatial domain (μx, μy) of the parafoveal vertex yields eigenvalues of 0.153 and 3.395, confirming an elliptical distribution with greater variability along one axis. In contrast, the near-foveal vertex shows narrower spreads of 0.54, 0.72, and 0.788, and smaller standard deviations of 0.16, 0.16, and 0.25 for the same parameters. Its covariance matrix produces nearly equal eigenvalues of 0.025 and 0.025, supporting a circular distribution in the spatial domain. These findings suggest that the parafoveal vertex covers a larger and more variable parameter space, while the near-foveal vertex reflects a more constrained and isotropic representation.
Conclusions:
This study shows that run-to-run variability in empirical pRF estimates can be replicated with simulations. This validation opens up the door to examining intrinsic biases in pRF stimulus configurations, quantifying estimation efficiency for arbitrary visual stimulation patterns, and developing novel layouts to optimize pRF estimation experiments. As a result, pRF runs will become shorter and/or more reliable, benefiting clinical applications.
·Fig. 1: Foveal vertex results showing empirical pRF distribution along with best-fit simulated data KDE (marginalized over sigma) and its ground truth spatial parameters (μx, μy).
·Fig. 2: Parafoveal vertex results showing empirical pRF distribution along with best-fit simulated data KDE (marginalized over sigma) and its ground truth spatial parameters (red dot).
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Perception, Attention and Motor Behavior:
Perception: Visual 1
Keywords:
Computational Neuroscience
Cortex
Data analysis
FUNCTIONAL MRI
Open-Source Code
Open-Source Software
Spatial Normalization
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.
Naaman, M. (2021). On the tight constant in the multivariate Dvoretzky–Kiefer–Wolfowitz inequality. Statistics & Probability Letters, 173, 109088.