Twenty-one healthy older adults (age: 62.8 ± 4.2 years, 55–68 years; six women) were recruited to participate in this cross-sectional study through poster advertising at the university campus. Additionally, 21 young adults (age: 28.1 ± 3.2 years, 21–33 years; seven women) were recruited from the university. The inclusion criterion was right-handedness according to the Edinburgh Handedness Inventory (Veale, 2014). Individuals were excluded from the study if they had any known history of neurological or psychiatric diseases, were taking neuropsychiatric drugs, or had contraindications to TMS (Rossi et al., 2011). This study was approved by the Human Research Ethics Sub-Committee of The Hong Kong Polytechnic University (Reference Number: HSEARS20200621001) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all the participants.

All TMS procedures were performed on the dominant and non-dominant hemispheres over an EEG cap placed on the participant’s head. A figure-eight cooling coil (Cooling B-65, external diameter of each wing: 75 mm) connected to a magnetic stimulator (model X100 option, MagVenture A/S, Denmark) was used to deliver the biphasic TMS pulses. To effectively stimulate M1, the coil was placed approximately 45° away from the midline with the handle pointed backward and laterally (Lazzaro et al., 2008). Coil positioning on the scalp was continuously monitored using a frameless stereotactic neuronavigation system (Localite, Bonn, Germany). A motor hotspot was defined as the position at which the largest and most reliable MEPs could be obtained from the first dorsal interosseous muscle contralateral to the stimulation site. Surface EMG activity was recorded using disposable Ag-AgCl electrodes positioned in a belly tendon montage, and a ground electrode was placed on the ulnar styloid process. The raw EMG signals were digitized at 5 kHz, and a Butterworth bandpass filter (fourth-order, 10 Hz–2 kHz) was applied for offline analysis.

A battery of TMS-EMG measurements was performed. The resting motor threshold (RMT) was defined as the minimum intensity (measured as the % of the maximal stimulator output or MSO) that could elicit peak-to-peak MEP amplitudes higher than 50 μV in at least five out of 10 trials. The intensity of the test pulses was set at 120% of the RMT. Single-pulse MEPs were used to measure corticospinal excitability at rest. Short-interval intracortical inhibition (SICI) and intracortical facilitation (ICF) were assessed by delivering a test pulse after a subthreshold conditioning pulse at 80% of the RMT with inter-pulse intervals of 2 and 10 ms, respectively. The ICF and SICI values were calculated as the ratio of MEPs produced by paired-pulse protocols to those produced by single pulses at rest. Cortical silent period (cSP) was the disruption of background EMG activity by a suprathreshold test pulse while sustaining 30% of the maximal voluntary strength of thumb-index finger contraction (Werhahn et al., 1999). The duration of cSP was defined as the time from TMS pulse onset to the first point of a 5-ms window, at which 50% of the EMG signal data samples returned to a level of at least three-fold standard deviations from the silent period. We also quantified MEP amplitudes of active MEPs (aMEPs) in the cSP measurement. Eight trials were recorded for each protocol (inter-trial intervals: 4–5 s) and averaged to obtain the grand means for further statistical analyses.

Concurrent TMS-EEG recordings were performed using a TMS-compatible DC EEG system (SynAmps, NeuroScan) with 64 Ag/AgCl electrodes mounted according to the international 10–10 system. The raw data were referenced online to FCz, grounded to AFz, digitized at a sampling rate of 5 kHz, and filtered online below 2 kHz (Bai et al., 2021). The impedance between the scalp and the electrodes was maintained below 5 kΩ. During recording, 90 TMS pulses at 110% of RMT were applied to M1, with inter-trial intervals of 4–5 s. To suppress the auditory-evoked potentials produced while the coil was discharged, all participants wore an inserted earphone and white noise was played (Tremblay et al., 2019). To minimize the TMS-decay artifacts, a thin piece of foam (3 mm thick) was placed underneath the coil to prevent direct contact with the electrodes (Tremblay et al., 2019), and the direction of the lead wires near the coil was rearranged so that they were perpendicular to the coil (Sekiguchi et al., 2011).

The TMS-EEG signals were pre-processed offline using EEGLAB (Delorme and Makeig, 2004), the TESA extension (Rogasch et al., 2017), FieldTrip (Oostenveld et al., 2011), and custom-made MATLAB scripts. Continuous signals were segmented into individual trials (−2000 to 1999 ms) and baseline-corrected (−500 to −10 ms). Poor trials and channels were excluded from the analysis. The numbers of TEP trials in young (89.6 ± 1.3) and older (87.2 ± 3.1) participants were comparable. The data around the TMS pulses (−2 to 15 ms) were removed and interpolated using a cubic method, followed by a down-sampling procedure to 1 kHz. Thereafter, two rounds of FastICA (systematic approach and tanh contrast function) were performed. The first round aimed to remove the largest TMS-decay artifact detected by a semi-automated component classification algorithm implemented in the TESA. The signals were bandpass-filtered (1–80 Hz) and bandstop-filtered (48–52 Hz) using a fourth-order Butterworth filter, followed by another segmentation from −1,000 to 999 ms. FastICA was conducted again to remove the remaining physiological artifacts. The excluded channels were interpolated back and the reference channel was recovered. Finally, the EEG signals were referenced to a common average and the TEPs were obtained by averaging across trials.

With reference to previous literature (Tremblay et al., 2019), we first predefined five peaks in the temporal domain: P30 (29–35 ms), N45 (42–48 ms), P60 (57–63 ms), N100 (70–110 ms for young participants, 90–130 ms for older participants), and P180 (180–230 ms for young participants, 180–250 ms for older participants). Second, based on the topographical plot of grand-averaged TEPs of all participants, several electrodes were selected to predefine the region of interest for each TEP peak with right stimulation, for example, P30 (C2, C4, CP2, CP4), N45 (AF3, F1, F3, FC1, FC3), P60 (CP4, CP6, P4, P6), N100 (FC1, FCz, FC2, C1, Cz, C2), and P180 (F1, Fz, F2, FC1, FCz, FC2). Third, the amplitude of the TEP peak was obtained by averaging over its predefined time window and region of interest.

Statistical analyses were performed using SPSS22 (IBM, NY, United States). The alpha threshold was set to 0.05 (two-tailed). The peak-to-peak amplitudes of single-pulse MEPs and aMEPs were log-transformed to decrease the inter-participant variability. The normality of variables prior to parametric tests was checked using both one-sample Kolmogorov–Smirnov tests and histogram plots. The TMS-EMG measures and TEP peaks were subjected to two-way repeated measures analysis of variance (rmANOVA) with two main effects (lateralization and aging) and a lateralization-by-aging interaction effect.

A two-way rmANOVA showed that the RMT of older participants was significantly higher than that of younger participants (F = 8.31, p = 0.006) (Figure 1). In contrast, older participants showed significantly lower amplitudes of aMEPs (F = 15.90, p < 0.001) and MEPs (F = 9.12, p = 0.004) than younger participants. Regarding the comparison of SICI, there was a significant aging effect (F = 4.07, p = 0.050), suggesting that older participants tended to present with a reduction in GABA-A receptor-mediated intracortical inhibition within M1.

The two-way rmANOVA showed a significant lateralization effect in the comparison of cSP (F = 4.63, p = 0.038) (Figure 2), suggesting that the cSP of the non-dominant M1 was significantly longer than that of the dominant M1; the aging (F = 0.06, p = 0.807) and lateralization-by-aging effects (F = 0.09, p = 0.771) were not significant.

The comparison of ICF failed to yield any significant effects (all p > 0.05). There was no significant hemisphere or lateralization-by-aging effect in the comparison of RMT, aMEPs, MEPs, or SICI (all p > 0.05).

The grand-averaged TEP waveforms from bilateral M1 are shown in Figure 3. The comparison of the TEP amplitudes is presented in Figure 4. Two-way rmANOVA failed to show any significant aging (F = 1.97, p = 0.168) or lateralization (F = 0.57, p = 0.457) effects on P30. There was no significant lateralization effect on N45 (F = 3.82, p = 0.058), but a significant lateralization effect on P60 (F = 9.41, p = 0.004), N100 (F = 7.41, p = 0.010), and P180 (F = 5.10, p = 0.029), suggesting that the non-dominant hemisphere had larger amplitudes in these peaks than the dominant hemisphere did.

Regarding the aging effect, two-way rmANOVA found statistical significance on N100 (F = 5.80, p = 0.021), but not on N45 (F = 3.66, p = 0.063), P60 (F = 1.29, p = 0.263) or P180 (F = 3.44, p = 0.071). These findings suggest that older participants tend to present smaller amplitudes on N100. There was no significant lateralization-by-aging effect in any comparison of the five TEPs (all p > 0.05). As shown in Figure 3, N100 and P180 peaked at later latencies in older participants than in younger participants. The two-way rmANOVA showed a significant aging effect on the latencies of N100 (F = 6.28, p = 0.016) and P180 (F = 9.25, p = 0.004), suggesting that both peaks of older participants occurred significantly later than those of younger participants (Figure 5). There were no significant hemisphere or lateralization-by-aging effects (all p > 0.05).

On the one hand, we found that the dominant and non-dominant M1 was not significantly different in RMT, aMEPs, MEPs, or SICI, in line with previous studies (Brouwer et al., 2001; Livingston et al., 2010). On the other hand, our study replicated the previous finding that the cSP of the non-dominant M1 was significantly longer than that of the dominant M1, which is also in line with previous studies (Priori et al., 1999; Reid and Serrien, 2012). The initial inhibition of cSP occurs at the level of spinal cord, while the later one is due to GABA-B receptor-mediated intracortical inhibition (Chen et al., 1999). Therefore, researchers have proposed that asymmetry of cSP might occur in the spinal cord or cortex (Priori et al., 1999). TMS-EEG can be used to probe cortical responses to TMS pulses without contamination from spinal cord activity, and the N100 amplitude is associated with GABA-B receptor-mediated intracortical inhibition (Premoli et al., 2014). In the present study, the N100 amplitude derived from non-dominant M1 stimulation was significantly larger than that derived from dominant M1 stimulation, suggesting that the difference in bilateral cSP likely occurs in the cortex rather than in the spinal cord.

To date, few studies have investigated hemispheric differences in peak amplitudes of TEPs. In the present study, significantly larger amplitudes of P60 and P180 were observed in the non-dominant M1 stimulation, but the mechanisms of P60 and P180 are not clearly enucleated. A recent pharmacological study demonstrated that AMPA receptor antagonists could significantly reduce the amplitude of P60 over the non-stimulated hemisphere, suggesting that AMPA receptors are involved in interhemispheric signal propagation (Belardinelli et al., 2021). However, the larger P60 amplitudes found in the current study were over the stimulated hemisphere, which meant that AMPA receptor-mediated glutamatergic excitation may not be the source of the hemispheric difference. GABAergic drugs were found to have no modulatory effects on the amplitude of P180 (Premoli et al., 2014), but a single oral dose of carbamazepine (a voltage-gated sodium channel blocker) decreased the amplitude of P180 (Darmani et al., 2019). Although these hemispheric differences have been confirmed, their implications in terms of motor performance and/or learning are not clear and further studies are needed.

In line with previous studies using TMS-EMG, we observed reduced corticospinal excitability in older participants (Brouwer et al., 2001; Livingston et al., 2010). Regarding the effect of aging on intracortical inhibition, largely inconsistent findings have been reported, and no consensus has been reached. This discrepancy may be caused by methodological variations among studies (Opie and Semmler, 2014), which are vital for the inhibitory effect of SICI. For instance, the greatest SICI is yielded by a test pulse at an intensity of 110–120% RMT, and higher test intensities reduce the inhibition of SICI (Garry and Thomson, 2008). The intensity of a conditional pulse and inter-pulse interval also have interaction effects on SICI (Vucic et al., 2009). In the present study, older participants presented with reduced SICI. Simultaneously, we also found that the N45 amplitude of older participants tended to be smaller than that of younger participants, in line with a previous study (Noda et al., 2021), providing further evidence that aging reduces GABA-A receptor-mediated intracortical inhibition, which can contribute to the deterioration of motor function (Levin et al., 2014).

The N100 amplitude and duration of cSP are believed to be mediated by GABA-B receptors (Stetkarova and Kofler, 2013; Premoli et al., 2014). However, dissociation of the two measures was found in the current study. Specifically, the N100 amplitude decreased in older participants, but the duration of cSP was not significantly changed by aging, suggesting that N100 and cSP represent the magnitude and duration of GABA-B receptor-mediated intracortical inhibition, respectively. N100 has not been extensively investigated in older adults; however, many previous studies have shown that long-interval intracortical inhibition, having a similar mechanism to that of N100 (Premoli et al., 2014; Darmani et al., 2016), is reduced in older adults (Opie and Semmler, 2014; Hermans et al., 2018, 2019). Therefore, these multiple-model data support the notion that the magnitude of GABA-B receptor-mediated intracortical inhibition is reduced in older adults, which may be a compensatory adaptation for reduced corticospinal excitability (Kallioniemi et al., 2022). A contradictory aging effect on the duration of cSP has been reported in the literature, and we found a non-significant aging effect on it. One possible explanation for this discrepancy is the different intensities used, which influence the duration of the cSP (Säisänen et al., 2008). Further studies should employ multiple levels of suprathreshold intensity to comprehensively test how aging modulates cSP. In addition to the significantly modulated TEP amplitudes due to aging, our analysis further demonstrated later latencies of N100 and P180, which is partially consistent with a previous study (Opie et al., 2018). The mechanism of TEP latency has not been fully elucidated and is probably associated with nerve conduction velocity and the reactivity of specific neural transmitters to TMS pulses (Noda et al., 2021). Further studies should investigate how these neurophysiological biomarkers are related to behavioral changes in older adults.

This study had some limitations. First, all participants were right-hand dominant, but appropriate measures were not carried out to evaluate the behavioral asymmetry of the bilateral hands, which may weaken the implications of the current study. Second, the average age of the older participants was lower than that in previous studies, decreasing the comparability between the current study and others. Finally, our analysis was conducted for predefined ROIs, in which the largest amplitude of a certain peak was. Whether there were hemispheric and/or aging-related differences in TEPs beyond predefined ROIs is not clear. Fourth, white nose masking was used to suppress auditory-evoked potentials during TMS-EEG recording, but a sham control condition was not employed to rule out the contamination of somatosensory-evoked potentials.

In conclusion, our study found that lateralization and aging have significant effects on intracortical excitation and inhibition, and such lateralization effects are stable across young and older adults. Longer cSP and larger amplitudes of P60, N100, and P180 were observed in the non-dominant M1. Older participants presented lower corticospinal excitability and N100 than young participants, suggesting that the function of excitatory and inhibitory circuits in M1 might be reduced due to aging. The latencies of both N100 and P180 were later than those of young participants, which requires further studies on how these neurophysiological biomarkers are associated with behavioral changes during normal aging.