CS295J/Project ideas from week 4

brain-science workflows

• describe and critique a brain-scientist software workflow
  • which interactive software?
  • workflow (which can include non-software parts, like wet or dry experiments
  • how much of their time is spent on this workflow?
  • what are positive elements of the workflow?
  • what are negative elements?
    • wasted time, lost opportunities, annoyances, etc. --
    • quantify negatives, if possible
  • how could our proposed ideas improve this?

(50)-day projects

• initial brain scientist agenda
  • we're pondering how to build analysis-assistiance software
    • integrates primary (imaging?) data
    • connectivity (cells, cell groups/types, brain regions)
    • literature pointers re connections and cell types and brain regions
  • what would you want to see in a tool like this?
  • what would you use it for and how?
• Make an agenda of questions for Schnitzer and then have a conf call
  • some input on what Mark Schnitzer wants to do
    • look at primary microscope data at level of groups of cells
    • access literature related to connections that they see (e.g., specific database I can't remember)
    • generate hypotheses (about?)
    • identify experiments to do
    • pose or test models
• other local brain scientists
  • Win Gongvatana/Ron Cohen
  • Steve Correia (VA)
  • David Badre
  • Michael Frank
  • Steve Sloman (reasoning, decision making) (mind, not brain)
  • Thomas Serre (modeling specific vision-related circuits)
  • everyone in CLiPS
• some preliminary demo that goes beyond correlative analysis to a causal system
  • this might be a completely different proposal...
• Observe brain scientists analyze with diagrams & summarize
  • Paper diagrams,
  • Something they actually use
  • Commercial (Tableau) or other existing tools
  • Postits “interface” (data & analysis results)
  • build on Agrawal et al and have students design a system
    • then pull out design principles, perhaps cognitively motivated
    • use proxy task based on social network analysis that would be student friendly
• automatically detect cognitive load from voice analysis (or galvanic skin response? or facial expression?)
• diagram recommender
• open-source repository
  • Caputured video of index card experiments
  • library of brain imagery
  • Codings of video
  • Way for others to contribute codings (incentivized?)
• taxtonomy of tasks (or design space) for #1
• cognitive skills of brain scientists (spatial, visual/textual, WM size, prob solving, ... )