CCN 2026 | Tutorial: In silico neuroscience: an emerging paradigm for brain discovery

CCN 2026 | Tutorial: In silico neuroscience: an emerging paradigm for brain discovery

by Alessandro Gifford, Domenic Bersch, Gemma Roig, Radoslaw Cichy

Life & Natural Systems Neuroscience PSANNeurosciencesPSAN5Cognitive and behavioural neuroscience
🎙 Alessandro Gifford, Domenic Bersch, Gemma Roig, Radoslaw Cichy 👥 4K 📅 August 12, 2026 ⏱ 52 min 👁 53 📄 tutorial 🧭 2026-08-15
Available in: English (current) Français

Keywords

in silicoencoding modelsfMRIvisual cortexBERG

Summary

This tutorial from the Cognitive Computational Neuroscience conference (CCN 2026) introduces the paradigm of in silico neuroscience, which uses encoding models to predict neural responses to stimuli, offering a scalable and cost-effective alternative to in vivo experiments. The presenters, Alessandro Gifford and Domenic Bersch, guide attendees through the use of the Brain Encoding Response Generator (BERG), a toolbox that provides pretrained encoding models and a Python package for generating in silico neural responses. The tutorial is structured in three stages: first, generating and visualizing in silico fMRI responses; second, validating these responses by checking prediction accuracy and reproducing known organizing principles of the visual cortex, such as retinotopy and category selectivity; and third, running independent experiments using the toolbox. The presenters emphasize the importance of model validation and the potential of in silico approaches to accelerate neuroscientific discovery. The tutorial includes hands-on exercises using Google Colab and provides access to model cards and metadata for transparency. The content is technical and aimed at researchers familiar with computational neuroscience, offering practical guidance for adopting this emerging methodology.

176 words

Critical Evaluation

Value of the Information & Strength of the Argument

The tutorial provides valuable information on a novel methodological approach, demonstrating how encoding models can be used to generate neural responses in silico, which can significantly reduce the cost and time of neuroscientific experiments. The argumentation is solid, as the presenters support the paradigm with theoretical advantages, empirical examples, and methodological tools. They also address limitations and emphasize the need for validation, which strengthens the credibility of the approach. The hands-on nature of the tutorial allows participants to directly engage with the tools, enhancing the practical value.

Scientific Rigor, Source Quality, Title Accuracy

The tutorial demonstrates scientific rigor by providing access to model cards with detailed metadata, including encoding accuracies and noise ceilings, and by encouraging validation through prediction accuracy and replication of known neural phenomena. The sources cited are primarily the BERG toolbox and the associated website, which are directly relevant to the content. The title accurately reflects the tutorial’s focus on in silico neuroscience as an emerging paradigm. The presenters are affiliated with the CCN conference, which adds credibility. However, the video does not provide a comprehensive literature review or detailed methodological validation, relying instead on the toolbox’s documentation.

200 words

Title / Content Match

The title accurately reflects the content, which is a tutorial on in silico neuroscience as an emerging paradigm for brain discovery.

Quality & Reliability

8/10

The tutorial is presented by researchers with expertise in computational neuroscience, and it introduces a well-documented toolbox (BERG) with model cards and metadata. The content is based on published encoding models and includes validation steps, but the presentation is primarily a tutorial and does not provide full methodological details or peer-reviewed evidence within the video itself.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

Concurring Sources

Contribution & Novelties

The tutorial introduces BERG, a novel toolbox that provides pretrained encoding models and a Python package for generating in silico neural responses, which is a significant contribution to the field. It demonstrates the feasibility of using in silico responses for large-scale experimentation, potentially accelerating brain discovery. The tutorial also emphasizes the importance of validation and provides practical guidance for researchers to adopt this paradigm.

Pour aller plus loin :

  • Encoding models in vision — Overview of encoding models and their use in neuroscience.
  • Natural Scenes Dataset (NSD) — A large-scale fMRI dataset used for training encoding models.
  • Algonauts Challenge — A benchmark for predicting brain responses to visual stimuli, relevant to the models used in BERG.

116 words

Radar Profile

The radar profile shows high scores across all dimensions, indicating a well-rounded tutorial with substantial information, technical depth, and reliability. The balance suggests that the content is both informative and practical, with a strong emphasis on methodological rigor.

Reliability 8/10