CAOS 2025 - 5 | Rovereto, May 7-9 | Stefania Bracci

Value of the Information & Strength of the Argument

The talk provides valuable insights into the organization of the visual cortex, challenging the traditional view that object recognition is the sole goal. The argumentation is solid, based on a series of well-designed fMRI experiments and computational models. The speaker systematically builds the case that action-related properties, such as manipulability and being an action effector, explain the topography of category-selective areas better than visual or semantic features alone. The use of multiple complementary analyses (univariate, ROI, pattern similarity) strengthens the conclusions. The comparison with neural networks is particularly informative, showing limitations of current models in capturing human-like organization. The talk is well-structured and logically progresses from initial findings to current work, making a compelling argument for the role of behavior in shaping visual cortex.

Scientific Rigor, Source Quality, Title Accuracy

The talk demonstrates high scientific rigor with careful experimental controls and appropriate statistical analyses. The speaker references relevant literature and acknowledges limitations. The title accurately reflects the content, focusing on the role of behavior in shaping visual cortex organization. The sources cited are primarily the speaker’s own published work and that of colleagues, which is appropriate for a conference talk. The talk does not overstate claims and clearly indicates work in progress. The adéquation between title and content is excellent.

The title accurately reflects the content, focusing on the role of behavior in shaping visual cortex organization.

8/10

The talk presents original research from a reputable academic group, with clear methodology and data. The speaker is an established researcher in the field. The content is technical and appears rigorous, though the talk is a conference presentation and not peer-reviewed.

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

• 02:00Introduction to the talk and the main computational goal of the ventral pathway.
• 10:00Discussion of the continuum of properties from visual to high-level and the opposing views.
• 20:00Presentation of the first study on hand-selective activation and its overlap with tools.
• 30:00Description of the fMRI experiment with action-related properties and the opposite gradients in lateral and ventral VTC.
• 40:00Testing deep neural networks and self-organizing maps to replicate the topography.
• 50:00Adding behavioral ratings to self-organizing maps to improve capture of action-based topography.
• 1:00:00Investigation of food selectivity and its relation to manipulability and grasping properties.
• 1:10:00Conclusion and summary of findings, emphasizing the role of behavior in shaping visual cortex.

• CAOS 2025 Workshop — Workshop page with information about the conference and the talk.

• Bracci et al. (2016) — Original study on hand-selective activation in lateral occipitotemporal cortex.
• Peelen & Downing (2017) — Review on the neural basis of object perception.

• Grill-Spector & Weiner (2014) — Alternative view emphasizing the role of visual features in organizing ventral temporal cortex.