Just how do we retrieve vivid recollections upon encountering a straightforward cue? Computational versions claim that this feat can be accomplished by design completion processes relating to the hippocampus. discovered that gamma power (50C90 Hz) increases C in conjunction with alpha power (8C12 Hz) decreases not only distinguish AR from IR, but also correlate with the level of hippocampal reinstatement. These results link single-shot hippocampal pattern completion to episodic recollection and reveal how oscillatory dynamics in the gamma and alpha bands orchestrate these mnemonic order AZD5363 processes. DOI: http://dx.doi.org/10.7554/eLife.17397.001 Trial-by-trial Pearson correlations of gamma and alpha timecourses were calculated for each participant for 0.5 to 2 s (spanning the time windows where gamma and alpha effects were observed for AR vs. IR, Figure order AZD5363 3), separately for AR and IR trials. The resulting correlation maps were then collapsed across AR and IR, Fisher z transformed and compared against 0 to reveal positive or negative correlations reliable across participants. Statistical map represents t values for the group-level t test against 0. Values above the diagonal reflect earlier alpha power correlating with later gamma power and values below the diagonal reflect earlier gamma power correlating with later alpha power. Black contour highlights a significant cluster (P 0.05, corrected) in which earlier gamma power from ~.8 to 1 1.5 s correlates negatively with later alpha power from ~1.3 to 1 1.5 s. em Right /em : Across-participant Pearson order AZD5363 correlation of 0.5 to 1 1.3 s gamma power with 1 to 2 2 s alpha power (i.e. the time-frequency clusters showing condition differences between AR vs. IR, Figure 3). Power values were derived separately for AR and IR trials and then collapsed prior to calculating the correlation. DOI: http://dx.doi.org/10.7554/eLife.17397.009 Figure 3figure supplement 3. Open in a separate window No condition differences in the baseline period.Gamma and alpha retrieval time courses (log transformed) are shown without baseline correcting the data, ensuring that condition differences are not driven by pre-stimulus baseline effects (see main text for statistics). DOI: http://dx.doi.org/10.7554/eLife.17397.010 Figure 3figure supplement 4. Open in a separate window Encoding data.(A) Time-frequency representation of the contrast subsequent associative recognition (AR) subsequent non-associative item recognition (IR). Note that the transition in resolution at 30 Hz reflects the switch from Hanning-taper for lower frequencies to multitaper for higher frequencies (Materials?and?methods). (B) Same as A, after statistical thresholding (P 0.05, corrected). Note the pronounced cluster centred on the alpha frequency band (8C12 Hz). order AZD5363 (C) em top /em : Encoding power time courses of AR and IR for 8C12 Hz. em bottom /em : For comparison, alpha power (8C12 Hz) timecourses of AR and IR at retrieval are plotted. When dividing the time axis into an early and a late interval (0 to 1 1 s vs. 1 to 2 2 s), there was a significant interaction of time window (early, late) x memory stage (encoding, retrieval) x condition (AR, IR) (F(1,10) = 10.27, P =?0.009). This interaction reflected the actual fact that there is Ctsk a significant boost of IR versus. AR in the first time home window at encoding (t(10) = 4.25, P =?0.002) however, not in retrieval (t(10) = 1.82, P =?0.099), whereas the enhance of IR vs AR in the past due time window was more pronounced at retrieval (t(10) = 8.47, P 0.001) than in encoding (t(10) = 2.30, P =?0.044). DOI: http://dx.doi.org/10.7554/eLife.17397.011 To make sure that these results are not powered by condition differences in the pre-stimulus baseline period, we omitted baseline correction and log transformed the raw power data instead (Cohen, 2014) (Body 3figure supplement 3). We after that compared AR versus. IR power both in the pre-stimulus baseline home window (?0.5 to 0?s) and in the post-stimulus home window where we observed the consequences above (0.5 to at least one 1.3?s for gamma and one to two 2?s for alpha). Initial, a significant home window x condition conversation for gamma power (F(1,10) = 22.15, P =?0.001) reflected order AZD5363 a substantial boost for AR vs. IR in the post-stimulus home window (t(10) = 2.65, P =?0.024) however, not in the pre-stimulus baseline home window (t(10) = 1.72, P 0.1). Also, a significant home window x condition conversation for alpha power (F(1,10) = 32.62, P 0.001) reflected a substantial boost for IR vs. AR in the post-stimulus home window (t(10) = 4.84, P =?0.001) however, not in the pre-stimulus baseline.