Javier González-Castillo

This paper introduces pBOLD, a new fMRI quality metric that quantifies BOLD dominance in multi-echo scans, validates it on datasets, and shows its utility in predicting phenotypes.

This study investigates the neural mechanisms of afterimages using fMRI, finding that they engage widespread brain activity and specific cortical layers, challenging previous views on their origins.

This study shows that rapid neural information can be decoded from ultra-fast fMRI signals using multivariate pattern decoders, achieving significant accuracy within 2 seconds of stimulus onset.

This paper introduces rtPupilPhase, an open-source software for real-time detection of pupil size fluctuation trends across species.

The paper shows that in-scanner experiences significantly affect resting-state functional connectivity measures in fMRI data.

The paper presents a network model that explains both concordant and discordant results in human neuroimaging studies of facial viewpoint selectivity by incorporating biological constraints and analyzing low-level stimulus biases.

The paper investigates the relationship between visual imagery vividness and afterimage perception, finding a positive correlation between imagery vividness and afterimage contrast and sharpness.

This commentary critiques Pang et al.'s 2023 study, arguing that their geometric mode approach to brain dynamics lacks specificity to brain shape and orientation, and does not sufficiently establish geometry as a more fundamental constraint than connectivity.

The paper shows that during decreased alertness, active task engagement increases frontoparietal and long-range brain synchronization, while passive listening increases thalamocortical synchronization.

The paper argues against hiding subthreshold results in neuroimaging studies and proposes a 'highlighting' approach to improve reproducibility and understanding.

This paper evaluates manifold learning techniques for visualizing time-varying functional connectivity in fMRI data, finding that intrinsic dimension varies and methods differ in capturing subject and task information.

This paper introduces a novel hemodynamic deconvolution algorithm for multi-echo fMRI that improves reliability and provides statistical certainty without requiring prior knowledge of experimental paradigms.

This paper identifies ultra-slow fMRI fluctuations in the fourth ventricle as a marker of drowsiness and shows their impact on functional connectivity during sleep.

This paper uses TBS and resting-state fMRI to show that disrupting the right posterior superior temporal sulcus reduces functional connectivity across the face-processing network, including between non-stimulated regions.

The paper analyzes dynamic functional connectivity during rest and finds that distinct cognitive periods influence resting-state brain activity, which can be inferred from data without behavioral input.

This paper introduces ME-SPFM, a deconvolution algorithm for multiecho fMRI that estimates transverse relaxation changes without prior event timing knowledge, showing high concordance with standard analysis.

This paper uses real-time fMRI neurofeedback to train brain networks in ASD participants, resulting in significant long-term changes in connectivity patterns correlated with behavioral improvements.