Specific contributions of basal ganglia and cerebellum to the neural tracking of rhythm

Abstract

How specific brain networks track rhythmic sensory input over time remains a challenge in neuroimaging work. Here we show that subcortical areas, namely the basal ganglia and the cerebellum, specifically contribute to the neural tracking of rhythm. We tested patients with focal lesions in either of these areas and healthy controls by means of electroencephalography (EEG) while they listened to rhythmic sequences known to induce selective neural tracking at a frequency corresponding to the most-often perceived pulse-like beat. Both patients and controls displayed neural responses to the rhythmic sequences. However, these response patterns were different across groups, with patients showing reduced tracking at beat frequency, especially for the more challenging rhythms. In the cerebellar patients, this effect was specific to the rhythm played at a fast tempo, which places high demands on the temporally precise encoding of events. In contrast, basal ganglia patients showed more heterogeneous responses at beat frequency specifically for the most complex rhythm, which requires more internal generation of the beat. These findings provide electrophysiological evidence that these subcortical structures selectively shape the neural representation of rhythm. Moreover, they suggest that the processing of rhythmic auditory input relies on an extended cortico-subcortico-cortical functional network providing specific timing and entrainment sensitivities.

Rhythm in speech and animal vocalizations: a cross‐species perspective

“…) composed of fast, short‐range and slower, long‐range temporal information contributes both to perception and production of auditory−motor sequences, such as found in human speech and music . Empirical evidence confirms that the ascribed temporal properties form the basis of temporal pattern formation found in simple and complex rhythm processing, which also relies on the same neural frontostriatal architecture as temporal processing per se …”

“…5,173,174 Empirical evidence confirms that the ascribed temporal properties form the basis of temporal pattern formation found in simple and complex rhythm processing, which also relies on the same neural frontostriatal architecture as temporal processing per se. 175,176 Shared neural circuitry, but where are cross-species boundaries? While there is now ample evidence that several species of birds and mammals, including some nonhuman primates, rely on comparable frontostriatal circuitry (e.g., see Ref.…”

Predictive rhythmic tapping to isochronous and tempo changing metronomes in the nonhuman primate

“…These findings set the limits in the beat‐entrainment abilities of monkeys quite above the reactive hypothesis previously suggested and emphasize that the beat‐based mechanisms of macaques are not as restricted, as previously thought. Thus, it is crucial that future experiments focus on finding the limits in entrainment capabilities of monkeys, using gradually more complex levels of metrical periodicity in their stimuli, such as rhythms based on nonisochronous temporal patterns or based on grouping of events at slower temporal scales . In addition, the comparison of beat‐entrainment abilities between human and nonhuman primates can be complemented with noninvasive electrophysiological experiments that allow the investigation of the neural correlates of beat perception, a precondition for rhythmic entrainment .…”

Neural bases of rhythmic entrainment in humans: critical transformation between cortical and lower‐level representations of auditory rhythm

“…To determine the frequencies at which the cortical frequency‐tagged responses to the acoustic envelope were expected in the EEG, the temporal envelope of the three 40.8‐s auditory stimuli (rhythm #1, rhythm #1 tempo x4 and rhythm #2) was extracted using the Hilbert transform as implemented in matlab and then transformed into the frequency domain using a discrete Fourier transform, yielding a frequency spectrum of acoustic energy (Cirelli et al ., ; Nozaradan et al ., ,b,c, ,b). The frequencies of interest appeared in the obtained spectrum as 12 peaks of acoustic energy ranging from the frequency corresponding to the period of an entire rhythm cycle to the period of the shortest interval for each rhythm and tempo (Fig.…”

“…Recently, intracranial depth-electrode EEG recordings in humans revealed that processes from the earliest auditory cortical areas (Heschl's gyrus) shape the neural representation of rhythmic inputs in favour of increased relative amplitude of the activity corresponding to meter-related frequencies (Nozaradan et al, 2016c). This selective enhancement of meter-related frequencies may require connections between the auditory cortex and other brain structures such as the premotor cortex, the basal ganglia and the supplementary motor area, which are involved in beat-based timing (Grahn, 2012;Merchant & Honing, 2014;Patel & Iversen, 2014;Ross et al, 2016;Nozaradan et al, 2017b). However, this selective enhancement may also be the product of lower-level properties of auditory neurons that process sound input before the cortex.…”