Passive listening to preferred motor tempo modulates corticospinal excitability

Abstract

Rhythms are an essential characteristic of our lives, and auditory-motor coupling affects a variety of behaviors. Previous research has shown that the neural regions associated with motor system processing are coupled to perceptual rhythmic and melodic processing such that the perception of rhythmic stimuli can entrain motor system responses. However, the degree to which individual preference modulates the motor system is unknown. Recent work has shown that passively listening to metrically strong rhythms increases corticospinal excitability, as indicated by transcranial magnetic stimulation (TMS). Furthermore, this effect is modulated by high-groove music, or music that inspires movement, while neuroimaging evidence suggests that premotor activity increases with tempos occurring within a preferred tempo (PT) category. PT refers to the rate of a hypothetical endogenous oscillator that may be indicated by spontaneous motor tempo (SMT) and preferred perceptual tempo (PPT) measurements. The present study investigated whether listening to a rhythm at an individual's PT preferentially modulates motor system excitability. SMT was obtained in human participants through a tapping task in which subjects were asked to tap a response key at their most comfortable rate. Subjects listened a 10-beat tone sequence at 11 log-spaced tempos and rated their preference for each (PPT). We found that SMT and PPT measurements were correlated, indicating that preferred and produced tempos occurred at a similar rate. Crucially, single-pulse TMS delivered to left M1 during PPT judgments revealed that corticospinal excitability, measured by motor-evoked potentials (MEPs), was modulated by tempos traveling closer to individual PT. However, the specific nature of this modulation differed across individuals, with some exhibiting an increase in excitability around PT and others exhibiting a decrease. These findings suggest that auditory-motor coupling induced by rhythms is preferentially modulated by rhythms occurring at a preferred rate, and that individual differences can alter the nature of this coupling.

Keywords: corticospinal excitability; individual differences; rhythm perception; tempo and timing; transcranial magnetic stimulation.

Figure 1

(A) Schematic of combined behavioral/TMS task for a representative subject with an SMT of 0.531 s. Subjects passively listened to 10-beat tone sequences traveling at a range of tempos relative to their SMT. Tempos were created by multiplying the subject’s SMT by 10 logarithmic modifiers to create 5 faster tempos and 5 slower tempos. While subjects listened to each tone sequence, single-pulse TMS was delivered over left M1 100 ms before the onset of the eighth beat. (B) Behavioral ratings to determine PPT. Subjects listened to tone sequences traveling at 11 different tempos, one at the subject’s SMT and 10 tempos spaced logarithmically above and below SMT (1 = SMT). At the end of each tone sequence, subjects rated the tempo according to how “right” the speed felt to them using the scale pictured above. (C) Calculation of PPT for a single subject. Ratings were averaged across trials and fit with a linear regression. The zero crossing indicates the subject’s PPT.

Figure 2

Behavioral results of PPT and SMT for all subjects. SMT and PPT values are displayed as seconds between tones/taps. (A) PPT and SMT were highly correlated. However, subjects consistently rated PPT as slower than SMT, with larger deviations between PPT and SMT for subjects with slower SMTs. (B) Variability in behavioral results of PPT and SMT. On the x-axis of the graph on the right, “SMT Modifier” refers to the values that each subject’s SMT was multiplied by to create the log-spaced tempos of the tone sequences. This means that 1 = SMT, smaller values designate faster tempos, and higher values designate slower tempos. These modifier values were the same for each subject while the actual tempos changed depending on the subjects’ SMTs. Subjects with slower SMTs showed greater variability in tap rate. Similarly, subjects were more variable in their perceptual ratings of slower tempos.

Figure 3

Calculation of effects of PT on MEP modulation. Subjects were analyzed based on the fit of a quadratic model to EMG data and segregated according to the direction of the maximal change in MEP amplitude. (A) Eight subjects were best fit by a negative curve (Excitation Pattern), meaning MEP amplitudes were maximal closest to SMT (1 = SMT). (B) Six additional subjects exhibited a positive curve (Suppression Pattern), meaning MEP amplitudes were minimal closest to SMT. Together, these graphs demonstrate that listening to PT modulates MEP amplitude, but that the direction of change differs across individuals. As in Figure 2B, “SMT modifier” refers to the values that each subjects’ SMT was multiplied by to create the log-spaced tempos of the tone sequences. This means that 1 = SMT, smaller values designate faster tempos, and higher values designate slower tempos. These modifier values were the same for each subject while the actual tempos changed depending on the subjects’ SMTs.

Figure 4

Relationship between PT and MEP amplitude modulation. PPT and SMT were each highly correlated with the tempo at which changes in MEP amplitude were maximal (EMG Peak), and subjects with slower SMT and PPT values exhibited greater deviations from EMG Peak. Subjects are color coded according to whether they showed an increase in MEPs when listening to PT (Excitation Pattern, shown in red) or a decrease in MEPs when listening to PT (Suppression Pattern, shown in blue). SMT, PPT and EMG Peak values are displayed as seconds between tones/taps.