Marta Isabel Garrido

The paper uses longitudinal dMRI data to show that subcortical visual pathways weaken while cortical pathways strengthen during early adolescence, with both linked to faster visual processing.

This study investigates how predictability and controllability affect aversive learning and stress responses through distinct computational mechanisms.

The paper investigates whether human Bayesian behavior stems from explicit Bayesian strategies or likelihood-sensitive strategies by using metamers to test prior-likelihood integration.

This study reanalyzes MEG data using dynamic causal modeling to test predictions of global neuronal workspace theory and integrated information theory during face perception, finding support for GNWT but mixed results for IIT.

The paper investigates how spatial frequency affects expectation effects in object-scene associations using EEG data, finding no impact on classification accuracy but effects on event-related potentials.

This study uses fMRI and DCM to show that insula and pulvinar connectivity patterns mediate precision-weighted prediction in visual tasks, revealing a dual-route mechanism for predictive processing.

The paper identifies pulvinar-STS and pulvinar-amygdala pathways as critical for non-conscious affective processing in affective blindsight.

This study examines how adolescent brains process threat and safety cues, finding a shift from hippocampus-dependent to hippocampus-vmPFC network-dependent safety discrimination as adolescents age.

The paper uses computational modeling to show that both sensory and frontal theories of consciousness can explain EEG data, with a slight preference for frontal theories.

Neurodesk provides a portable and reproducible neuroimaging analysis platform using software containers to enable accessible and flexible data analysis pipelines.

The paper tests whether humans perform Bayesian computations by measuring how well they transfer learned priors to new likelihoods, finding sub-optimal Bayesian behavior.

This study uses Bayesian modeling to show that psychotic-like experiences are linked to reduced precision in prior information and increased noise in sensory processing, supporting both weaker bottom-up and top-down perceptual processes.

This paper introduces a salience-based model of reward prediction errors that simulates addictive behaviors, including reduced sensitivity to negative consequences and steeper delay discounting.

The paper reconstructs subcortical visual pathways in humans and finds that the LGN-MT pathway is more heavily myelinated, suggesting faster signal transmission.

This study shows that people with tobacco use disorder misattribute non-drug cues as weaker predictors of positive outcomes compared to drug cues, supporting some but not all hypotheses of the salience misattribution model.

This study examines how anxiety affects learning and neural responses to stable versus volatile environments using fMRI and behavioral tasks.

The paper investigates context adjustment in autism using auditory oddball paradigms, finding no group differences in early responses but larger evoked responses in autism associated with auditory sensitivities.

This study uses fMRI and computational modeling to show that fearful faces are perceived faster than neutral faces, involving accelerated evidence accumulation in fronto-parietal and visual areas.

This study compares neural activity during inhibitory control tasks between people with nicotine use disorder and non-smokers, finding differences in prefrontal and anterior cingulate cortex activity.

This study investigates sensory-perceptual differences in autism using a decision-under-uncertainty paradigm, finding intact priors, precise likelihood, reduced context updating, and heightened variability in sensory learning.

This study uses EEG data and machine learning to predict subclinical psychotic-like experiences on a continuum between healthy individuals and schizophrenia patients.

This study investigates how the brain processes visible and invisible prediction errors using EEG and DCM, revealing distinct neural mechanisms for each.

This study uses machine learning to predict the severity of specific psychotic symptoms from fMRI and behavioural data in schizophrenia patients.

The paper investigates how practice-induced plasticity between the putamen and pre-SMA affects multitasking limitations in humans.

This study investigates differences in auditory precision encoding connectivity between schizophrenia spectrum disorder patients, non-psychotic psychiatric inpatients, and healthy controls, finding group-specific and continuum-related connectivity changes.

The paper investigates how feature-based attention interacts with prediction, finding that attention's effect is stronger for surprising stimuli than predicted ones.

This study replicates findings that sensory learning and inference impairments are linked to psychotic-like experiences in a non-clinical population.

This paper introduces Porthole and Stormcloud, visualization tools for spatiotemporal M/EEG data, offering enhanced graphical capabilities for print and screen.

The paper investigates whether healthy individuals with psychotic experiences show reduced auditory prediction errors and white matter microstructure, using EEG and MRI data.

This study investigates how attention affects neural encoding of prediction errors using EEG data from a visual task, finding that attention enhances the encoding of mismatch information in response to surprising stimuli.