(A) Alpha-Gamma Nested Coding (AGNC) hypothesis, adapted and modified from Jensen et al.72: Five stimuli have neuronal representations with decreasing levels of excitability, decreasing from a to e, which interact with an inhibitory oscillatory alpha drive (8–13 Hz). As a result, representations discharge in sequence a-b-c, and the less excited representations d and e are prevented from discharge. The discharging representations are kept apart in time by fast recurrent inhibition from an interneuronal network.
(B) Left plots: X- and y-position of the attentional spotlight, decoded from multiple simultaneous recordings in macaque frontal eye field (FEF), as a function of time from attentional cue onset in one exemplary trial. Right plots: Corresponding power spectra, showing peaks in the theta-frequency range. Adapted and modified from Gaillard et al.87.
(C) Difference between spectrograms of the envelopes of the LFP gamma-band (45–100 Hz) oscillations in gamma-modulated versus non-modulated recording sites. Recordings were performed in PFC of monkeys performing a working memory task. During the task, two samples (indicated here as S1 and S2) are shown and are kept in working memory until they can be compared with the test (indicated here as T1). Adapted and modified from Lundqvist et al.92.