Skip to main content
. 2014 Apr 29;8:138. doi: 10.3389/fnbeh.2014.00138

Figure 7.

Figure 7

Computational modeling of the neural circuits underlying context-driven OB activation. (A) Odor representations in granule cells after training on a randomly chosen odor [mean activity over 200 ms; odor and context (Odor+ctxA), trained context only (No-odor+ctxA), novel, untrained context (No-odor+ctxB)]. (B) Correlation between post-training activation patterns (Odor+ctxA/No-odor+ctxtA and No-odor+ctxA/No-odor+ctxB). (C) Spikes per seconds evoked in granule cells after training in response to the trained odorant and context (Odor+ctxA), the trained context only (No-odor+ctxA), and a novel context (No-odor+ctxB). (D) Spikes per second evoked in pyramidal cells in a naive model exposed to context only (before training; odor naive, No-Odor_naive), to the trained odorant/context (Odor+ctxA), or to the trained context only (No-odor+ctxA). (E) Odor representations in granule cells in response to stimulation with the trained context only with the intact (No-odor+ctxA) or lesioned (No-odor+ctxA-no feedback) feedback. (F) Correlation between Odor+ctxA/No-odor+ctxtA and Odor+ctxA/No-odor+ctxA+no-feedback. (*p < 0.05).