Skip to main content
. Author manuscript; available in PMC: 2011 Oct 27.
Published in final edited form as: J Neurosci. 2011 Apr 27;31(17):6414–6420. doi: 10.1523/JNEUROSCI.5773-10.2011

Figure 5. Summary of predictability results.

Figure 5

(A) Predictability of each cell from noise envelope, for traditional linear STA and 2D STA methods. Red and blue symbols correspond to individual cells from desynchronized and synchronized states, respectively. Symbol shape indicates different recordings, for 3 experiments giving sufficient numbers of cells recorded in both states. For the 2D STA method, the time-lag giving optimal performance was used. (B) Predictability of each cell from LFP. Prediction from LFP is better with the nonlinear method, for both brain states. (C) Prediction from stimulus envelope is much better in the desynchronized than synchronized state. (D) LFPs typically predict spiking better in the synchronized than desynchronized state. (E) In the synchronized state, prediction is much better from the LFP than from the stimulus envelope. (F) Even in the desynchronized state, the best prediction is from the LFP.