Andrea Campos

B.Sc. in Neurosciences by the National Autonomous University of Mexico (UNAM), Faculty of Medicine. Honors Thesis: “Bimodal Encoding in a Neuronal Population of the Dorsal Premotor Cortex during Working Memory”.

M.Sc. in Neurosciences by the International Max-Planck Research School for Neurosciences, University of Göttingen. Thesis: “Reshaping Representational Spaces through Statistical Learning Investigated with Intracranial Electroencephalography in Humans”.

Neural and Network Mechanisms Underlying Statistical Learning, Predictive Coding, and Adaptive Flexibility during Visual Cognitive Tasks across Different Species

This project investigates the neural and network mechanisms underlying statistical learning, predictive coding, and adaptive flexibility during visual cognitive tasks, with a primary focus on object and face recognition. By examining how the brain processes prediction errors and adapts to uncertainty, the research aims to uncover how neural circuits implement learning and prediction. Using advanced techniques such as fMRI, ECoG, and invasive electrophysiology, combined with advanced computational modeling, machine learning, and multivariate analyses, this work will compare humans, non-human primates, and potentially mice to identify shared and species-specific principles of cognition. The overarching goal is to advance our understanding of how brains across species enable flexible, predictive, and efficient cognitive processes, with implications for both theoretical neuroscience and computational model development, and providing insights into disorders involving impaired learning and adaptability.

My research interests focus on how the brain efficiently and optimally processes information from the environment. The brain is not just a passive receptacle or filter of information; it is also an energy-efficient agent that processes input in ways that minimize resource expenditure. One theoretical framework for understanding this is predictive coding, which suggests that the brain builds models of the environment based on prior experiences or stimuli and then tests these models against incoming sensory information. In this way, we do not need to encode the full complexity of the world, but rather only those features that signal whether our predictions were correct or not. This active sampling of information, guided by uncertainty, is what fascinates me most, as it reflects a hallmark of perceptual curiosity.

• Ojeda-Alonso, J., Bégay, V., Garcia-Contreras, J. A., Campos-Pérez, A. F., Purfürst, B., & Lewin, G. R. (2022). Lack of evidence for participation of TMEM150C in sensory mechanotransduction. Journal of General Physiology, 154(12), e202213098.
• Nigam T, Campos-Pérez AF, Mégevand P, Mehta AD, Vidal JR, Perrone-Bertolotti M, Kahane P, Lachaux J-P, Thesen T, Doyle W, Devinsky O, Melloni L, & Schwiedrzik CM. Multiple task-demands flexibly optimize neural geometry in human temporal cortex. In revision. 2025.

• Poster participation at the 14th Primate Neurobiology Meeting, German Primate Center – Leibniz Institute for Primate Research. Göttingen, DE.
• Poster participation at the IV Neurobiology Meeting, Mexican Society for Biochemistry. Oaxaca, MX.
• Laboratory Animal Science Course on Non-Human Primates. German Primate Center, Göttingen, DE.
• Intensive Systems Biology Workshop by Prof. Dr. Robert Endres, Imperial College of London. Institute of Biomedical Research, UNAM, Mexico City, MX.
• Electrophysiology & Imaging, virtual Plymouth Workshop. Marine Biological Association (MBA), England, GB.