NeuroTech: Neuroscience at NSF

1
NeuroTechNeuroscience at NSF

• D. H. Whalen
• Program Director, Cognitive Neuroscience

2
Current NSF Neuroscience

• Across virtually all directorates.
• About 60 million in FY06.
• Difficult to assess, because neuroscience is not
  coded directly.
• Depends on your definition as well, of course.
• Even this number does not include centers.
• Still not seen as a priority despite this rather
  large outlay.

3
Current Reassessment

• Neuroscience at NSF is being examined
  Foundation-wide.
• Objective Redefine NSFs mission in this area.
• Takes NIH into account.
• Input from most directorates.
• Primary tools three workshops.

4
Workshop 1 July 2006

• Grand Challenges of Mind and Brain.
• Four member Steering Committee.
• Sheila Blumstein (chair).
• Thomas Carew.
• Nancy Kanwisher.
• Terry Sejnowski.
• Twelve more in Workshop Panel.
• Report available on CogNeuro web page.

5
Grand Challenge Areas

• Adaptive Plasticity.
• Conflict and Cooperation.
• Spatial Knowledge.
• Time.
• Language.
• Causal Understanding.

6
Grand Challenge Tools

• Human Brain Circuitry.
• Imaging, of course.
• Also specifically mentions nonhuman species.
• Mathematical Innovations.
• Non-linear varieties in particular.
• Information Databases.
• Molecular Tools.
• Cyberinfrastructure.

7
Workshop 2 August 2006

• Primarily MPS.
• Brain Science as a Mutual Opportunity for the
  Physical and Mathematical Sciences, Computer
  Science, and Engineering.
• Somewhat less grand.
• Chaired by Chris Wood (Santa Fe Inst.).
• Thirteen other participants.
• Report available on CogNeuro website.

8
Three Broad Areas

• A Shift in the Scope and Scale of Experimental
  Investigations.
• Recordings from multiple events and sites.
• A Shift in the Character of Theoretical
  Understanding.
• Simultaneously bottom-up and top-down.
• A Shift in How Knowledge Can Be Used.
• Greater ability to model complex systems.
• More potential for direct brain/machine links.

9
Instrumentation and Measurement

• Functional measurements in neurons and circuits.
• Labeling in neurons and circuits.
• Controlling activity in neurons and circuits.
• The importance of model organisms.

10
Data Analysis, Statistical Modeling and
Informatics.

• Multidisciplinary input to new measurement
  techniques.
• Methods to integrate Diverse Data Sources.
• Statistics, Signal Processing and Machine
  Learning.
• New Tools for Control Theory.
• Analyzing Multiple Levels and Time Scales.
• Inferring Causality in Neural Systems.
• New Approaches to Data Management/Sharing.

11
Conceptual and Theoretical Approaches

• Fundamental Role of Mathematics.
• Dynamical Systems.
• Statistical Physics/Large Degrees of Freedom.
• Engineering Approaches.
• Machine Learning Tools.
• Large-Scale Simulations.

12
Brain-Like Devices and Systems

• Analog Approaches.
• Stochastic Semiconductor Circuits.
• Neural Coding and Functional Biomimetic Systems.
• Brain-Like Robots.
• Biocompatible Neural Interfaces.

13
Now, Four Broad Areas

• Instrumentation and Measurement.
• Data Analysis, Statistical Modeling, and
  Informatics.
• Conceptual and Theoretical Approaches.
• Building Brain-Like Devices and Systems.

14
(No Transcript)
15
Workshop 3 March 2007

• The workshop was organized around seven
  interdisciplinary themes, and led by
• Chris Wood, Vice President, Santa Fe Institute.
• Ted Berger, Dept of Biomedical Engineering, USC.
• Emery Brown, Dept of Brain/Cognitive Sciences,
  MIT.
• Eve Marder, Dept of Biology, Brandeis
  University.
• Tom Mitchell, Machine Learning, Carnegie Mellon
  U.
• Partha Mitra, Cold Spring Harbor Laboratory.
• Marcus Raichle, Dept of Radiology, Washington U.
• Jonathan Sweedler, Dept of Chemistry, U. of
  Illinois.

16
Workshop 3 March 2007

• The workshop was organized around seven
  interdisciplinary themes, and led by
• Chris Wood, Vice President, Santa Fe Institute.
• Ted Berger, Dept of Biomedical Engineering, USC.
• Emery Brown, Dept of Brain/Cognitive Sciences,
  MIT.
• Eve Marder, Dept of Biology, Brandeis
  University.
• Tom Mitchell, Machine Learning, Carnegie Mellon
  U.
• Partha Mitra, Cold Spring Harbor Laboratory.
• Sheila Blumstein, Dept of Linguistics, Brown U.
• Jonathan Sweedler, Dept of Chemistry, U. of
  Illinois.

17
SBE in this Workshop

• Main working group of interest to SBE is
  Raichles/Blumsteins
• Cognitive Systems Neural Bases of Thought and
  Behavior.
• Meant to convey the extensive interaction of
  systems seen in every cognitive domain.
• Many other relevant topics in other working
  groups.

18
Various Promising Topics

• Navigation and Spatial Cognition.
• Speech Perception.
• Face Recognition.
• Reading.
• Perception and Action.
• Number Skills.
• Tools for the Future.

19
The Big Question

• How does the brain create thought and behavior?
• Despite many interesting findings, the
  fundamentals are still not known.
• Converging techniques, in imaging and in theory
  (e.g., complexity), are making new advances
  possible.

20
The Specific Big Question

• As Dr. Bement asked at the workshop, how does
  consciousness arise?
• Nancy Kanwisher, our presenter, pointed out that
  this is still too difficult a question to answer
  directly.
• However, we have interesting components that have
  been located, such as face perception with and
  without awareness.

21
SBE Cognitive Neuroscience

• Most of the SBE-specific topics were in cognitive
  neuroscience, but not necessarily the Cognitive
  Neuroscience program
• Perception, Action and Cognition
• Linguistics
• Social Psychology
• Economics
• Decision, Risk and Management Sciences
• Developmental and Learning Sciences
• Cultural Anthropology
• Physical Anthropology
• Human Origins (HOMINID)

22
Other Potential Areas

• Neural imaging of voting decisions.
• Culturally-specific patterns of perception.
• MRI lie detectors?
• New statistical tools.
• Neural study of ethics decisions.
• Basic questions posed by diseases (natures
  knock-out experiments).

23
Support from Other Programs

• Cyber-enabled Discovery and Innovation
• Data-Net
• Community-based Data Interoperability Networks
  (INTEROP)
• Collaborative Research in Computational
  Neuroscience (CRCNS) (soon)

24
Challenges

• Enormous area, dominated by biology.
• Differentiating NSF from NIH.
• Choosing right level of problem within SBE
  parameters.
• International collaboration (at least, getting
  the word out, if not providing new mechanisms).

25
Opportunities

• Additional funds may be forthcoming.
• Partnering with European funders.
• New tools and techniques may truly make future
  advances happen at a currently unimaginable pace.
• 2010s Decade of the Mind.

26
Thank you.