Item

Abstract

Oscillations prevail in encephalographic (EEG, MEG) signals and supposedly reflect cognitive processes such as sensory representations or the routing of sensory information [1]. Typical MEG/EEG studies focus on the relation between oscillation amplitude (power) and the sensory-cognitive variables, making power an important marker for studying the brain [2]. However, recent studies have begun to also consider the dynamic signature of EEG/MEG signals, such as characterized by the phase of slow oscillations [3]. Several studies have shown that the precise temporal structure (phase) of slow encephalographic oscillations can be informative about sensory stimuli or details of the cognitive task. Noteworthy, in some studies the phase proved to be more informative about the presented stimuli than the same signal’s power [4,5,6]. However, the neural correlates underlying the information carrying capacity of the phase of slow oscillations remain unclear. We here directly tested whether the stimulus selectivity of low frequency EEG phase patterns indeed reflects the selectivity of neuronal firing in the underlying cortical areas. We employed the same naturalistic acoustic stimuli in two experiments, one recording scalp EEG in human subjects and another recording intracortical field potentials and single neurons in macaque auditory cortex (see e.g. [7]). Using stimulus decoding techniques we found that stimulus selective patterns of neural firing imprint on the phase of slow (theta band) oscillations rather than on their amplitudes. We found that sets of stimuli (sampled from the long acoustic stimulation sequence) that can be discriminated by the oscillatory phase pattern slow oscillations can also be discriminated by neural firing rates and vice versa. Importantly, no such relation was found between oscillatory power and firing rates. Our results demonstrate a level of interrelation between scalp EEGs and neural firing that pertains to stimulus selectivity (preference) and which goes beyond known correlations between the strength of neural firing and EEG oscillatory amplitude. Thereby our findings enhance the link between the activity of sensory cortical neurons and non-invasively measured field potentials, and improve the interpretation of EEG-based studies and their implications towards understanding the neural dynamics of sensory perception.