Evoked potential, MUA and theta latency profiles and theta/gamma activity changes after learning. (A) Typical average latency profile for theta, MUA and VEP (P100 and N300) over 40 trials post learning (face-pair shown at time 0). The change in MUA in response to the stimulus is mainly restricted to the 500 ms time window. (B) Mean ± sem % change in theta amplitude (C) Gamma amplitude (D) Theta/gamma ratio (E) Theta gamma coherence and (F) Theta phase tightening (G) Mean ± sem z-scores for theta phase synchronization and (H) Theta phase reset overall and in each of the 3 different animals (A, B and C), and for both the right and left IT, during sessions where discrimination learning performance had yet to reach > 80% (NL = not learned) compared with those where it had (L = learned - corrected t-tests, *p < 0.025 - left IT, and < 0.017 for right IT, **p < 0.01, ***p < 0.001 vs NL and ##p < 0.01, ###p < 0.001 for left IT vs right IT). For overall changes across the sheep p < 0.05, **p < 0.01 and ***p < 0.001 from ANOVA tests. For both NL and L blocks, comparisons were made between data averaged across electrodes and individual trials for the 500 ms immediately prior to stimulus presentation and the first 500 ms during it.