As the global population ages, understanding and addressing age-related changes in the brain, particularly cognitive decline and mood disorders, has become increasingly crucial (Amanzio and Palermo, 2021; Di Fazio et al., 2024; Ferreira et al., 2016; Amanzio et al., 2021; Stern, 2012). Promoting brain health is essential not just for extending life, but also for improving quality of life and maintaining cognitive function and resilience in older age (Amanzio and Palermo, 2021; Morese and Palermo, 2022; Palermo and IntechOpen, 2021). Cognitive resilience, defined as the mind’s capacity to adapt and maintain functionality in the face of stressors and changes, is a key aspect of protection against decline with aging (Joshi and Galvin, 2022; de Vries et al., 2024; Stern et al., 2023). Research on cognitive resilience offers valuable information regarding how older adults are able to sustain independence and general quality of life (Prince et al., 2024).

Identifying reliable biomarkers to assess cognitive resilience and predict cognitive decline is a significant priority in aging research. Cortical excitability (CE), which reflects the tendency of the cerebral cortex to generate electrical activity in response to stimuli, holds great potential as an important and less invasive biomarker for the comprehension of cognitive health (Menardi et al., 2018; Jannati et al., 2023). CE is related to the brain’s adaptive capacity and reflects the balance of excitatory and inhibitory processes. While CE generally declines with age, studies suggest that the relationship between CE and cognitive performance can be complex, potentially following an inverted-U hypothesis where both excessive and diminished excitability can impair function (Menardi et al., 2018). Lower CE has been observed in older adults with preserved robust cognitive performance compared to those with less cognitive robustness (Menardi et al., 2018; Palermo et al., 2025). Measuring CE non-invasively can provide insights into neuronal network communication and how aging affects different networks (Cespón et al., 2022; Berns et al., 2020).

Non-invasive brain stimulation (NIBS) techniques, such as transcranial magnetic stimulation (TMS), have emerged as tools to assess and potentially modulate CE, offering a promising approach to understanding and influencing cognitive aging (Buch et al., 2011; Rossi et al., 2009; Rossini and Rossi, 2007; Di Fazio et al., 2025a; Di Fazio et al., 2025b). Repetitive transcranial magnetic stimulation (rTMS) is a type of NIBS that employs magnetic fields to modulate neuronal activity in targeted brain regions. rTMS can influence neurophysiological, affective, and cognitive brain functions (Buch et al., 2011; Pascual-Leone and Walsh, 2001; Santarnecchi et al., 2018; Silvanto et al., 2007). Prior evidence suggests that rTMS can modulate cortical plasticity, which is crucial for learning and memory, and may facilitate neurophysiological adaptations associated with cognitive resilience and cognitive reserve (Fitzsimmons et al., 2024; Zhang et al., 2025). rTMS has been studied for its effects on various cognitive functions, including executive functions, working memory, attention, and processing speed, which are often affected by age (Di Fazio et al., 2025a; Turriziani et al., 2012; Asgarinejad et al., 2024; Sharbafshaaer et al., 2023; Xu et al., 2024). The left dorsolateral prefrontal cortex (DLPFC) is a frequent target in rTMS applications due to its central role in executive functions, cognitive control, and mood regulation (Amanzio and Palermo, 2021; Asgarinejad et al., 2024; Cui et al., 2020; Taylor et al., 2019; Amanzio et al., 2011). Modulating the DLPFC with rTMS has shown potential for improving cognitive performance and mood in older adults (Palermo et al., 2025; Di Fazio et al., 2025b; Zhang et al., 2025; Cui et al., 2022).

Building on our previous theoretical framework (Palermo et al., 2025) and earlier single-case evidence (Di Fazio et al., 2025a), this protocol expands the research scope by introducing a group-based, cross-over design. The study will involve healthy older adults, carefully screened and enrolled in alternating active and sham high-frequency rTMS conditions. This within-subject approach enables direct intra-individual comparisons and improves the interpretability of potential longitudinal changes.

The primary aim of this protocol is to describe the methodological and analytical framework for investigating whether cortical excitability (CE), increasingly discussed as a potential biomarker of cognitive resilience and neuroplasticity, is sensitive to neuromodulatory intervention in aging. Neuromodulatory interventions in aging (Palermo et al., 2025; Menardi et al., 2022; Motta et al., 2018; Ferreri et al., 2021). Motor-evoked potentials (MEPs), elicited by single-pulse TMS over the primary motor cortex, are used as a well-established, non-invasive proxy for CE (Spampinato et al., 2023; Huerta and Volpe, 2009). By integrating neurophysiological and cognitive assessments, the study aims to characterize the feasibility and within-subject sensitivity of CE to rTMS-induced modulation, rather than to establish diagnostic validity or clinical efficacy. Importantly, CE is conceptualized here as a state-sensitive, within-subject marker of neuroplastic responsiveness, reflecting changes relative to each individual’s own baseline.

This protocol outlines the rationale, methodological framework, and technical implementation of a 12-week, high-frequency rTMS intervention targeting the left dorsolateral prefrontal cortex (DLPFC), a region critically involved in executive functions, cognitive control, and mood regulation (Myers-Schulz and Koenigs, 2012; Toepper et al., 2014). Active stimulation consists of 10-Hz rTMS delivered over the left DLPFC at an intensity of 120% of the individual resting motor threshold (rMT). Each session includes 800 pulses, administered in 20 trains of 40 pulses, and is conducted three times per week for a total of 36 sessions over 12 weeks. The present work is conceived as a protocol-focused contribution, aimed at describing and justifying the feasibility, assessment strategy, and methodological rationale of a crossover rTMS design in healthy aging. The study design includes pre- and post-intervention assessments at multiple time points to evaluate changes in cortical excitability (CE), indexed by motor-evoked potentials (MEPs), along with a comprehensive neuropsychological battery assessing processing speed, executive functioning, mood, and quality of life. By systematically examining the relationship between neuromodulation, cortical excitability, and cognitive outcomes, this protocol aims to examine the feasibility and within-subject sensitivity of CE as a potential biomarker of neuroplasticity and cognitive resilience in aging and to provide a methodological foundation for future large-scale investigations of rTMS-based interventions for healthy cognitive aging.

This study protocol outlines a 12-week clinical trial with 36 rTMS sessions, administered three times per week, each lasting about one hour, with at least one rest day between consecutive sessions. The trial uses a cross-over design to reduce interindividual variability, with participants randomized to begin with either active or sham stimulation before switching to the other condition. The active and sham stimulation phases are administered consecutively, without a formal washout period between conditions, reflecting the feasibility-oriented nature of the pilot phase. The protocol includes a multimodal assessment framework to capture both functional and neurophysiological outcomes.

Specifically, participants will undergo:

Observations of individual cases can help verify procedural feasibility and methodological coherence of the protocol prior to larger-scale implementation. They allow us to evaluate tolerability, adherence, and measurement characteristics, and to illustrate inter-individual variability in a feasibility and exploratory framework. Given the small number of participants, these case descriptions are provided for context and protocol refinement rather than for drawing clinical or causal conclusions.

This work presents a study protocol designed to investigate whether cortical excitability, assessed through TMS-derived MEPs, may serve as a sensitive and reliable biomarker of neuroplastic changes following rTMS in older adults, and how such physiological modulation relates to neuropsychological outcomes. The accompanying piloting phase was instrumental in testing the feasibility, safety, and sensitivity of the procedures, while also providing preliminary observations in a small group of older adults. This within-subject design allowed each individual to serve as their own control, thereby increasing the internal validity of the findings despite the small sample size. Assessments were carried out at two timepoints: baseline (T0) and end of treatment (T36), to evaluate cumulative changes in cognitive functioning and cortical excitability observed following the full intervention. Although behavioral assessments were limited to T0 and T36, the crossover design allowed condition-specific comparisons by contrasting within-subject changes following active versus sham stimulation. This approach strengthened interpretability despite the reduced number of assessment timepoints. Moreover, the protocol was intentionally designed to capture cumulative, longer-term effects of repeated rTMS rather than acute modulation following single sessions; therefore, intermediate or immediate post-session assessments were not included to limit participant burden and repeated testing effects. By combining this crossover design with both neurophysiological and neuropsychological measures, the pilot data support the feasibility and methodological rigor of the protocol, while offering preliminary descriptive observations of measurable cortical and behavioral changes following high-frequency rTMS targeting the left DLPFC.

All participants underwent both active and sham stimulation within a cross-over design, and all comparisons were made relative to each participant’s own baseline and condition-specific outcomes. Within this protocol framework, the piloting results provide illustrative, hypothesis-generating within-subject trajectories. In the two participants who received active stimulation first, TMT-B completion times decreased from baseline to post-treatment, whereas changes following sham were smaller or inconsistent. These individual-level observations are not generalizable and may reflect practice, expectancy, or carryover effects; nevertheless, they show that the assessment battery can capture within-subject variability in executive performance across conditions, a key target of DLPFC stimulation. Indeed, the TMT-B is widely regarded as a sensitive index of attentional control, cognitive flexibility, and the efficiency of executive networks (Kortte et al., 2002; Varjacic et al., 2018; MacPherson et al., 2019; Linari et al., 2022; Seibert et al., 2025; Lazari et al., 2022; Chu et al., 2021; Xie et al., 2021; Zhang et al., 2022). In the context of this feasibility study, changes on this task are interpreted conservatively as descriptive indicators of protocol sensitivity rather than as evidence of cognitive enhancement.

From a neurocognitive perspective, contemporary models emphasize that higher-order cortical regions such as the lateral prefrontal cortex do not implement isolated cognitive operations, but instead shape adaptive control through top–down modulation of distributed networks (Martínez-Molina et al., 2024; Momi et al., 2021). Recent computational accounts suggest that prefrontal regions contribute to processes such as conflict expectation, contextual prediction, and flexible adjustment of control policies, dynamically influencing downstream sensorimotor and associative systems (Martínez-Molina et al., 2024; Momi et al., 2021). Within this framework, neuromodulation of the left DLPFC is unlikely to exert purely focal effects, but may bias the state of large-scale control networks involved in executive regulation.

Accordingly, changes in cortical excitability should not be interpreted as direct correlates of specific cognitive computations, but rather as state-dependent physiological reflections of broader network reconfiguration. In this context, MEP-derived cortical excitability is conceptualized as a global index of system-level responsiveness, sensitive to changes in distributed circuits engaged by prefrontal control (Martínez-Molina et al., 2024; Momi et al., 2021). Crucially, here we do not imply a focal or site-specific causal relationship between stimulation of the left DLPFC and motor cortex output. Instead, observed MEP modulation is interpreted as an indirect consequence of network-level dynamics elicited by prefrontal neuromodulation, in line with evidence that rTMS effects are more consistently expressed at the level of large-scale networks than at isolated cortical sites (Martínez-Molina et al., 2024; Momi et al., 2021; Menardi et al., 2022).

By contrast, outcomes related to self-reported fatigue and mood appeared more variable across conditions. While one participant in the active phase reported reductions in global fatigue and mood symptoms, similar decreases were also present in the sham condition, and another participant in the active phase showed only minimal changes. These mixed patterns highlight the inherent variability of subjective measures in small pilot samples and the need for cautious interpretation. Within this pilot, some changes appeared to co-occur with the active condition; however, practice, expectancy, and potential carryover effects cannot be excluded. Neurophysiological recordings also demonstrated that the protocol can acquire stable MEP measures and characterize within-individual changes across sessions in the piloting phase. In the two participants who received active stimulation first, MEP amplitudes varied across post-session time points (T0–T2) in the exploratory within-subject models; in contrast, MEP amplitudes were comparatively stable during sham. These descriptive patterns are compatible with stimulation-related modulation but cannot establish mechanism or efficacy in this feasibility sample (Xu et al., 2024; Avenanti et al., 2018; Turrini et al., 2023).

We therefore interpret them primarily as evidence of measurement feasibility and signal detectability, which supports the choice of MEP-derived cortical excitability as the primary physiological outcome in the planned study (Battaglia et al., 2024a; Paracampo et al., 2017; Avenanti et al., 2018; Borgomaneri et al., 2015; Borgomaneri et al., 2020; Borgomaneri et al., 2015; Borgomaneri et al., 2021). In one participant, the largest MEP change coincided with the largest descriptive cognitive change, suggesting a potential within-individual covariation that warrants systematic testing in adequately powered studies. In this protocol, cortical excitability is considered a state-sensitive physiological readout that may help stratify responsiveness; however, any association with behavioral outcomes remains a hypothesis and should be interpreted cautiously in the present pilot (Petitet et al., 2021; Yan et al., 2024).

A central aim of this protocol was to explore whether cortical excitability could function as a reliable biomarker of neuroplastic responsiveness within a structured study framework. Observations from the piloting phase suggest that CE captures both cortical responsiveness and meaningful inter-individual variability in potential for functional change. For example, MF, whose baseline profile included mild functional complaints, showed the largest modulation of CE alongside cognitive changes, whereas AM exhibited marked CE modulation with minimal behavioral variation. Although these observations are descriptive and not generalizable, they illustrate the protocol’s ability to detect differential responsiveness across individuals (Borgomaneri et al., 2020; Borgomaneri et al., 2015). Interpreting such variability requires consideration of the growing emphasis on personalization in non-invasive brain stimulation. Inter-individual differences in baseline cortical excitability, network integrity, cognitive reserve, and anatomical features are known to critically shape responsiveness to rTMS, particularly in aging populations (Figueroa-Vargas et al., 2024; Soleimani et al., 2025). In this protocol, a standardized stimulation approach was intentionally adopted to establish tolerability, procedural robustness, and baseline variability prior to introducing additional layers of individualization. Within this framework, heterogeneous or null responses are not interpreted as lack of neuromodulatory potential, but as reflecting meaningful individual differences that motivate subsequent personalized implementations (Figueroa-Vargas et al., 2024; Soleimani et al., 2025). Accordingly, low-cost personalization strategies (such as individualized frequency tuning or baseline excitability–informed dosing) represent natural extensions of the present design and will be systematically explored in future studies.

Although formal predictive analyses were not conducted due to the small sample size, qualitatively, participants with lower baseline cognitive performance or higher fatigue or mood symptoms showed greater CE and neuropsychological improvements, while high-functioning participants demonstrated robust CE enhancement with minimal behavioral change. These pilot observations suggest the protocol’s potential to detect baseline-dependent responsiveness (Figueroa-Vargas et al., 2024; Soleimani et al., 2025), a hypothesis that should be systematically tested in larger studies.

Taken together, these pilot findings indicate the feasibility of collecting MEP-derived cortical excitability and neuropsychological measures within the planned crossover schedule. The observed condition-related patterns are hypothesis-generating and consistent with the conceptual framework motivating cortical excitability as a candidate physiological marker for future confirmatory work, while underscoring the need for larger, adequately powered trials to test efficacy and longer-term impact (Di Fazio et al., 2025a; Sabesan et al., 2015; Hsu et al., 2016; Pavon et al., 2019; Miniussi et al., 2013; Cappon et al., 2022). Although the study involved only four participants, it was designed as an initial feasibility phase embedded within a structured longitudinal cross-over protocol. This methodological choice allowed us to test recruitment and adherence procedures, stimulation tolerability, and the acquisition of neurophysiological and neuropsychological measures under the planned schedule. The pilot observations should therefore be interpreted as descriptive and hypothesis-generating, and they primarily inform operational refinement, variance estimation, and the design of larger, adequately powered trials.

Several methodological limitations of the pilot phase should be acknowledged when interpreting these findings. The absence of a formal washout period between active and sham phases represents a methodological limitation of the pilot phase and may have introduced potential carryover effects. However, as the primary aim of this feasibility study was not to assess efficacy but to test procedural implementation, tolerability, and data acquisition reliability, this limitation does not undermine the methodological objectives of the protocol. In addition, potential sex-related differences in cortical excitability and responsiveness to rTMS could not be examined due to the very limited sample size and should be explicitly addressed in future studies with adequately powered and sex-balanced samples.

In conclusion, this manuscript presents a methodologically grounded rTMS protocol designed to examine cortical excitability modulation in aging within a crossover design. The feasibility phase indicates that the protocol allows reliable acquisition of neurophysiological and cognitive measures and captures within-individual variability across conditions. In this pilot, the clearest condition-related pattern was observed in MEP-derived cortical excitability; however, these observations are descriptive and may be influenced by carryover or expectancy effects, and they are not intended as evidence of efficacy. Within a network-level perspective, these observations motivate the use of cortical excitability as a system-level physiological outcome rather than as a direct marker of localized effects. The protocol is positioned to support future investigations explicitly testing neurocognitive and computational models of adaptive control and plasticity in aging, with larger trials required to confirm efficacy and long-term impact.