Summary: weekly mtg 20160420 (Byron, Steve, Matt, me) - mobeets/nullSpaceControl GitHub Wiki

What we discussed

• we go through my hypothesis cartoons, and they give advice on making them clearer
• I show my best evidence for/against various hypotheses: minimum/baseline/unconstrained are out, the others are close

The plan

• will definitely shoot for an SfN abstract
• aim to have manuscript written by mid-summer, submitted by fall

To do

second panel for cartoons of cursor

size of cloud and habitual on average, for a given date (how much are you cutting out by intersecting them?) account for resampling

bounds of factors vs bounds of mapping 2 activity

chart to show what info used by each hypothesis

new IME thetas

compare size of null activity under IME vs true decoder

choose best two thetas, worst two, for each day

find ways to discriminate close hypotheses