Neuroscience needs to make a transition from small science to large scale systems engineering and analysis, analogous to the changes in 20th century physics that gave rise to particle accelerators, space telescopes, the Standard Model and the transistor. Yet the complexity and diversity of neuroscience has made it difficult to know how to proceed: a different kind of integration is needed in the science of complexity than in the science of simplicity.
Using examples from brain mapping technology and from the neuroscience-AI theoretical interface, Adam Marblestone argues that the achievement of accelerated progress in the field is gated by support (organizational, cultural, financial) for truly unconventional integrations of otherwise disparate ideas and methods. Such approaches sometimes require us to deviate from familiar modes of hypothesis-driven science, as we search for integrative frameworks and build new kinds of observational tools, but there is a chance that they could finally allow us to explore the relevant regions of hypothesis space for understanding how the brain works.
In the future, we plan to post a transcript for this talk, but we haven’t created one yet. If you’d like to create a transcript for this talk, contact Aaron Gertler — he can help you get started.
Adam Marblestone: What sets the exponent of neuroscience progress
Link post
Neuroscience needs to make a transition from small science to large scale systems engineering and analysis, analogous to the changes in 20th century physics that gave rise to particle accelerators, space telescopes, the Standard Model and the transistor. Yet the complexity and diversity of neuroscience has made it difficult to know how to proceed: a different kind of integration is needed in the science of complexity than in the science of simplicity.
Using examples from brain mapping technology and from the neuroscience-AI theoretical interface, Adam Marblestone argues that the achievement of accelerated progress in the field is gated by support (organizational, cultural, financial) for truly unconventional integrations of otherwise disparate ideas and methods. Such approaches sometimes require us to deviate from familiar modes of hypothesis-driven science, as we search for integrative frameworks and build new kinds of observational tools, but there is a chance that they could finally allow us to explore the relevant regions of hypothesis space for understanding how the brain works.
In the future, we plan to post a transcript for this talk, but we haven’t created one yet. If you’d like to create a transcript for this talk, contact Aaron Gertler — he can help you get started.