Neural coding
What is the best way to test the relevance of a neural code? How should
features of neural responses be linked to downstream decoders, ethology
and/or behavior? How important is this, and how can we do this correctly?
Silencing and stimulating genetically-specified neurons
What's the best way of designing and interpreting experiments where we
either silence or microstimulate a neuron or population of neurons? Various
techniques make these experiments increasingly feasible (e.g., laser
ablation, synaptic silencing, channelrhodopsin stimulation, etc.). What are
their limitations AND how might we work around them?
"The Connectome"
How important is it simply to discover which neurons in a circuit connect to
each other? Do we need the resolution of the C. elegans "connectome"? Is it
important to know connectivity at nanometer-scale resolution, given the
enormous challenges in storing and analyzing these types of images? How can
we design powerful experiments to establish connectivity on different
scales?
Electrophysiology and functional imaging
Electrophysiology permits fine sensitivity and temporal precision, while
functional imaging allows simultaneous recording from many neurons. What
kinds of questions can best be approached by these techniques? What are
disadvantages of both approaches? How can we use these approaches to
complement each other?
Using behavior to test circuit hypotheses
How can we use behavioral paradigms to test specific hypotheses about neural
circuits? Can we use behavior to place constraints on what neural codes are
plausible? How can we create tight links between behavioral approaches and
electrophysiology/imaging?
Neuroethology of the genetic model organism
What questions are best addressed using naturalistic behaviors? When are
highly constrained behaviors more useful? Is it problematic that all genetic
model organisms are highly inbred? By choosing to study genetic model
organisms, are we inevitably rejecting the notion that we should
preferentially study virtuosic "specialists" like the barn owl?
Handling complexity
Is the brain's complexity best tackled by reducing a problem to its most
basic elements (using genetic manipulations/simplified stimuli/constrained
behaviors)? What, in your experience, are the pros and cons of the
reductionist approach? How should be think about the alternatives to
reductionism?
Scale and sufficiency
What do we learn by observing the activity of a very large population of
neurons, as opposed to just a few? What do we learn by trying to define a
“minimal” set of “sufficient” neurons for a particular behavior? What do we
learn by trying to catalog “all” the neurons that are somehow involved in a
particular neural code or behavior?
Neural identity
What does it mean to identify a neuron? If we use genetic tools, do we bias
ourselves in favor of a genetic definition of identity? How does the meaning
of “identifiability” change as the numerical scale of the circuit changes?
How is the concept of “identity” helpful? In technical terms, how do we
identify neurons in extracellular recordings in complex networks? How should
we interpret variability in the properties of an “identified” neuron across
time points and across individuals?