dariozancaThe following is a summarized and AI-curated English transcript of the podcast episode. While it aims to capture the key points and essence of the conversation, this is not a verbatim transcription. Some sections may be rephrased for clarity and readability.
Host: Dario Zanca
Guest: Francesco Cavarretta
Francesco: My work focuses on biophysically detailed neuron modeling. That means we simulate the brain not only structurally, but also in terms of the actual biological and physical properties of neurons—down to how ion channels behave.
Using real neuron morphologies reconstructed in 3D and equations like the Hodgkin-Huxley model, we can simulate neural behavior with high accuracy. This helps us understand complex patterns, such as those seen in pathological conditions like Alzheimer’s or Parkinson’s.
Dario: So even though these simulations are built from individual neurons, they can scale to help us understand more complex brain functions?
Francesco: Exactly. For instance, I’ve simulated 10% of the olfactory bulb. By assembling neuron models and their connectivity, we can study how large-scale brain functions emerge. The bottlenecks are mainly computational power and the availability of experimental data—like realistic neuron shapes and measurements to fine-tune the models.
Dario: And what could we do if we had a full brain model?
Francesco: We could explore the root causes of neurodegenerative diseases in much more detail. More than unlocking consciousness, I think this path could help us simulate therapeutic interventions, test treatments virtually, and accelerate medical advances.
Dario: We also talked about your work on the olfactory system. Why does smell have such a deep emotional impact?
Francesco: The olfactory pathway is unique—it bypasses the thalamus and goes straight to areas like the amygdala, tied to instinct and memory. That’s why smells can trigger intense, emotional memories. But simulating the full olfactory pathway is still out of reach.
Dario: So if we wanted to build an AI that could interpret smells, like we do for vision or sound—what’s stopping us?
Francesco: Sensors are improving, but the real challenge is replicating the brain’s deep, unconscious processing of smell. The olfactory cortex lacks the modular organization we see elsewhere, making it harder to model or emulate computationally.
Dario: As someone who studies real neural networks, how do you see current artificial neural networks?
Francesco: There’s very little in common. AI neural networks are inspired by brain structure, but they miss the complexity of how biological neurons process information. A single neuron in the brain can perform far more nuanced computations than an artificial one.
Dario: So could better biological understanding improve AI?
Francesco: Definitely. If we could incorporate more accurate models of neurons and synapses, we might build more powerful and versatile AI. But then the challenge becomes how to use and interpret that complexity.
Dario: Final question—what still amazes you about the brain?
Francesco: That gap between physiology and behavior. We understand so many neural mechanisms, yet still can’t fully link them to what we think, feel, or do. There’s a mysterious “grey zone” between biology and psychology, and that fascinates me. Even if we don’t see the full picture yet, I believe the knowledge we’re building will one day make a difference.
Dario: Francesco, thank you for sharing your ideas with us. Your work is inspiring.
Francesco: Thank you, Dario.