Biology subjects

Eickenberg, M.

Publications and source records attributed to Eickenberg, M..

3 recordsLinked to original sources

bioRxiv2024

A biologically-inspired hierarchical convolutional energy model predicts V4 responses to natural videos

V4 is a key area within the visual processing hierarchy, and it represents features of intermediate complexity. However, no current computational model explains V4 responses under natural conditions. To address this, we developed a new hierarchical convolutional energy (HCE) model reflecting computations thought to occur in areas V1, V2, and V4, but which consists entirely of simple- and complex-like units like those found in V1. In contrast to prior models, the HCE model is trained end-to-end on neurophysiology data, without relying on pre-trained network features. We recorded 313 V4 neurons during full-color nature video stimulation and fit the HCE model to each neuron. The models predicted optimal patterns (POPs) revealed complex spatiotemporal pattern selectivity in V4, supporting its role in representing space, time, and color. These findings indicate that area V4 is crucial for image segmentation and grouping operations that are essential for complex vision. Thus, responses of V4 neurons under naturalistic conditions can be explained by a hierarchical three-stage model where each stage consists entirely of units like those found in area V1.

neuroscience↗

bioRxiv2022

Feature-space selection with banded ridge regression

Encoding models provide a powerful framework to identify the information represented in brain recordings. In this framework, a stimulus representation is expressed within a feature space and is used in a regularized linear regression to predict brain activity. To account for a potential complementarity of different feature spaces, a joint model is fit on multiple feature spaces simultaneously. To adapt regularization strength to each feature space, ridge regression is extended to banded ridge regression, which optimizes a different regularization hyperparameter per feature space. The present paper proposes a method to decompose over feature spaces the variance explained by a banded ridge regression model. It also describes how banded ridge regression performs a feature-space selection, effectively ignoring non-predictive and redundant feature spaces. This feature-space selection leads to better prediction accuracy and to better interpretability. Banded ridge regression is then mathematically linked to a number of other regression methods with similar feature-space selection mechanisms. Finally, several methods are proposed to address the computational challenge of fitting banded ridge regressions on large numbers of voxels and feature spaces. All implementations are released in an open-source Python package called Himalaya.

neuroscience↗

bioRxiv2021

The Cost of Untracked Diversity in Brain-Imaging Prediction

Brain-imaging research enjoys increasing adoption of supervised machine learning for singlesubject disease classification. Yet, the success of these algorithms likely depends on population diversity, including demographic differences and other factors that may be outside of primary scientific interest. Here, we capitalize on propensity scores as a composite confound index to quantify diversity due to major sources of population stratification. We delineate the impact of population heterogeneity on the predictive accuracy and pattern stability in two separate clinical cohorts: the Autism Brain Imaging Data Exchange (ABIDE, n=297) and the Healthy Brain Network (HBN, n=551). Across various analysis scenarios, our results uncover the extent to which cross-validated prediction performances are interlocked with diversity. The instability of extracted brain patterns attributable to diversity is located preferentially to the default mode network. Our collective findings highlight the limitations of prevailing deconfounding practices in mitigating the full consequences of population diversity.

neuroscience↗