, 2008; Lerner et al., 2011). In regions nearer to the sensory periphery, cortical activity is reliably modulated by instantaneous physical parameters (e.g., the acoustics of a word), DNA Damage inhibitor but processing is largely independent of temporal context (e.g., whether that word occurs in a meaningful sentence). These more peripheral regions have been said to have short “temporal receptive windows” (TRWs). Further up the processing hierarchy, more and more of the sensory history is found to affect processing in the present moment. In areas with especially “long TRWs,” such as the temporoparietal junction, the cortical activity at each moment may depend on information that arrived over prior tens

of seconds. In this study, we aimed to map the large-scale topography of TRWs using electrocorticographic (ECoG) recording of the human brain. We further CB-839 in vivo asked whether regions with longer TRWs have distinctive properties in their population dynamics, which may be important for their capacity to accumulate information over long timescales. In particular, we hypothesized that slow components of neuronal dynamics would be more evident

in regions with long TRWs, relative to regions with short TRWs. We tested this hypothesis by performing ECoG recordings from the cerebral cortex of humans watching intact and scrambled audiovisual movie clips (Figure 1A). In quantifying local neuronal dynamics, we measured multiple signal components, but focused on fluctuations of power within the broad high-frequency range of 64–200 Hz. Human and monkey electrophysiology suggest that power fluctuations in the 64–200 Hz band are a distinct phenomenon from the γ oscillations found in visual cortices, and that shifts in this nonrhythmic broadband component index the population spike rate near an electrode (Crone et al., 2011; Manning et al., 2009; Miller, 2010; Nir et al., 2007; Ray and Maunsell, 2011; Whittingstall and Logothetis, 2009). Thus, when we mention fast or slow components of neuronal population dynamics, we are referring to

faster and slower Thymidine kinase fluctuations of broadband high-frequency power, which indexes the population spike rate. By measuring the ECoG responses to intact and scrambled movie clips, we confirmed, first, the presence of shorter TRWs in more sensory areas, and longer TRWs in higher order perceptual and cognitive cortices. Second, we observed that regions with long TRWs exhibit relatively more slow (<0.1 Hz) fluctuations of high-frequency power for both intact and scrambled movie clips. Third, we observed that these slow fluctuations of power were modulated with reliable time courses across repeated presentations of the movie. The slow fluctuations were more reliable for the intact than for the scrambled movie, suggesting that they may be connected to the processing of information over long timescales. We measured neural responses to stimuli with intact information and with scrambled information structure.