Neuroelectromagnetic signatures of the reproduction of supra-second durations

When participants are asked to reproduce an earlier presented duration, EEG recordings typically show a slow potential that develops over the fronto-central regions of the brain and is assumed to be generated in the supplementary motor area (SMA). This contingent negative variation (CNV) has been linked to anticipation, preparation and formation of temporal judgment (Macar, Vidal, and Casini, 1999, Experimental Brain Research, 125(3), 271-80). Although the interpretation of the CNV amplitude is problematic (Kononowicz and Van Rijn, (2011), Frontiers in Integrative Neuroscience, 5(48); Ng, Tobin, and Penney, 2011, Frontiers in Integrative Neuroscience, 5(77)), the observation of this slow potential is extremely robust, and thus one could assume that magnetic recordings of brain activity should show similar activity patterns. However, interval timing studies using durations shorter than one second did not provide unequivocal evidence as to whether CNV has a magnetic counterpart (CMV). As interval timing has been typically associated with durations longer than one second, participants in this study were presented intervals of 2, 3 or 4s that had to be reproduced in setup similar to the seminal work of Elbert et al. (1991, Psychophysiology, 28(6), 648-55) while co-recording EEG and MEG. The EEG data showed a clear CNV during the standard and the reproduction interval. In the reproduction interval the CNV steadily builds up from the onset of interval for both stimulus and response locked data. The MEG data did not show a CNV-resembling ramping of activity, but only showed a pre-movement magnetic field (preMMF) that originated from the SMA, occurring approximately 0.6s before the termination of the timed interval. These findings support the notion that signatures of timing are more straightforwardly measured using EEG, and show that the measured MEG signal from the SMA is constrained to the end of reproduction interval, before the voluntary movement. Moreover, we investigated a link between timing behavior and the early iCNV and late CNV amplitudes to evaluate the hypothesis that these amplitudes reflect the accumulation of temporal pulses. Larger iCNV amplitudes predicted shorter reproduced durations. This effect was more pronounced for the 2s interval reproduction, suggesting that preparatory strategies depend on the length of reproduced interval. Similarly to Elbert et al. (1991, Psychophysiology, 28(6), 648-55), longer reproductions were associated with smaller CNV amplitudes, both between conditions and across participants within the same condition. As the temporal accumulation hypothesis predicts the inverse, these results support the proposal by Van Rijn et al. (2011, Frontiers in Integrative Neuroscience, 5) that the CNV reflects other temporally driven processes such as temporal expectation and preparation rather than temporal accumulation itself.