This study investigated time to APA as a measure of motor planning and its relationship to cognitive and motor function across young adults, older adults, and individuals with PD. The results suggest that time to APA is not merely a simple motor response but also relies on executive function and response inhibition, particularly in older adults and individuals with PD.

Time to APA was not associated with upper extremity SRT. Since both time to APA and upper extremity SRT are, putatively, forms of a reaction timed test, one might have anticipated observing correlations between these two tests; but we did not (recall Table 2A; p > 0.25). SRT is a simple visuomotor task most likely generated by the premotor cortex which activates the basal ganglia and inhibits the globus pallidus, leading to excitation of the motor cortex (Aron et al. 2007). Notably, upper limb SRT does not appear to involve a large degree of prefrontal cortical resources, possibly explaining why no association was found between it and time to APA. An additional factor that may explain the lack of an association is that the SRT test includes the beginning of a motor task, while time to APA does not.

We found, instead, that time to APA was longer than SRT in all groups, reinforcing the idea that it is not a simple motor response. Interestingly, among young adults, time to APA was shorter than CRT duration. One interpretation of this finding is that time to APA is a relatively automatic task in healthy young adults, perhaps more analogous to a SRT than a CRT, involving minimal cognitive resources, especially as compared to the older adults and people with PD. In the young adults, time to APA was, nonetheless, longer than SRT, perhaps due to differences related to upper versus lower function or specifics of the two tasks (Pfister et al. 2014).

In contrast, among the older adults and individuals with PD, time to APA was comparable to CRT. Some have posited that the CRT reflects short latency inhibition performance (Antonelli et al. 2011; Yang et al. 2018). The present findings would suggest that in aging and PD, time to APA increasingly resembles a cognitively demanding task rather than an automatic motor reaction and that time to APA engages higher-level planning and decision-making mechanisms.

Furthermore, time to APA was significantly associated with performance on cognitive tests, including the MoCA test and the Color Trails Test (CTT). Longer time to APA correlated with worse cognitive performance, supporting the idea that this phase of gait initiation requires executive function, rather than merely reflecting motor readiness. In contrast, APA duration was not related to cognitive measures but was associated with PD motor severity, reinforcing its role as a marker of motor impairment rather than cognitive processing.

As shown in Table 2, the correlations between time to APA, on the one hand, and cognitive function, gait, and mobility, on the other hand, were strongest when examining all subjects as one group and weaker or not significant when examining correlations within each group. For example, time to APA was significantly (p < 0.001) correlated with CTT-A among all subjects, but this correlation was not observed in the young adults and was less strong in the older adults (p = 0.015) and the people with PD (only a trend). One interpretation of this and related findings (e.g., recall Table 2C and D) is that the association between CTT-A and time to APA was driven by the differences across groups, rather than within homogeneous groups individually. This underscores the importance of considering population heterogeneity. Another, complementary possibility that could also contribute to these explanations relates to the sample size. By definition, the sample size was largest when we evaluated all subjects as one group and much smaller when studying within-group correlations. Future work is needed to test these different possibilities.

Our findings align with prior research on motor initiation deficits in PD and aging (Antonelli et al. 2011; Schlenstedt et al. 2017). Individuals with PD and, to a lesser extent, older adults showed lower Total Inhibition scores and a smaller CRT-SRT cost, suggesting difficulties in suppressing prepotent motor responses. These deficits may contribute to prolonged time to APA, as successful gait initiation likely requires inhibitory control to transition from standing to walking. Indeed, people with PD, and to some extent older adults, have deficiencies in movement initiation and motor impulsivity (Antonelli et al. 2011; Doridam et al. 2019; Levin et al. 2014; Wylie et al. 2012). Those impairments may be associated with a longer time to APA (Gazit et al. 2020; Schlenstedt et al. 2017), perhaps due to inefficient ability to initiate desired motor responses and inhibition of the prepotent motor response (Cohen et al. 2017).

A key finding was the significant group × time interaction, which indicates that aging and PD do not uniformly affect all reaction times. If bradyphrenia (slowed cognitive processing in PD) were the sole explanation, we would expect parallel increases in time to APA, SRT, and CRT across groups. However, we observed that time to APA was disproportionately prolonged in older adults and individuals with PD, suggesting distinct underlying mechanisms. The greater reliance on frontal cortical resources and executive function may explain why time to APA is particularly affected in these populations, as suggested previously (Richard et al; 2017). Studies of other aspects of gait have also reported an increased reliance on executive function and frontal cortical areas in aging and PD, putatively to compensate for impairments and deficits in motor function (Bayot et al. 2023; Richard et al; 2017; Yogev-Seligmann et al. 2008; Kahya et al. 2019). Nonetheless, the cross-sectional nature of this study limits interpretations regarding cause and effect.

Recent studies suggest that other factors may also contribute to time to APA nulli. For example, De Waele et al. (2024) reported that time to APA was related to weight and that age played a significant role in the second phase of gait initiation in people with PD. In contrast, in a study of unmedicated people with PD, Palmisano et al. (2020) found that changes in the APA related to the “imbalance phase” (initial APA phase) were correlated with the dopaminergic innervation of the putamen and substantially improved with levodopa but were not influenced by anthropometric parameters. Some authors also suggest differences in APA among patients with PD related to the presence or absence of freezing of gait (Palmisano et al 2020, 2022; Lencioni et al. 2024; de Lima-Pardini et al. 2020; Taximaimaiti and Wang 2021). On that note, it is also interesting to consider the suggestion that APA errors (i.e., alterations in timing, amplitude, sequence, and the need for multiple APAs) are larger in patients with PD and freezing of gait (Bayot et al. 2023). In the present study, patients with PD and freezing were not included. Follow-up studies directly comparing time to APA and potential explanatory factors (Gerard et al. 2024; Jacobs et al. 2009; Silva-Batista et al 2024) in early and more advanced PD and in PD patients with and without freezing of gait can help shed light on these open questions and the role of these additional potential mechanisms.

A shorter time to APA was associated with better gait and mobility performance, including faster gait speed and longer step length. This suggests that efficient motor planning contributes to smoother gait transitions and overall mobility. Importantly, these relationships were strongest in individuals with PD, where longer time to APA correlated with slower performance on the Timed Up and Go test. Given that impaired gait initiation is a predictor of falls, time to APA may serve as a valuable marker for fall risk and functional decline in PD and aging, consistent with previous suggestions (Hu et al. 2024).

This study has several limitations. First, we assessed cued gait initiation using auditory stimuli rather than self-initiated movement. The basal ganglia, which are affected in PD, may be more involved in self-paced movement than in externally triggered tasks (Toyomura et al. 2012). Moreover, external cues are known to facilitate the generation of rhythmical stepping and to improve step preparation in people with PD (Toyomura et al. 2012; Delval et al. 2014b; Russo et al. 2022). The cued gait initiation allowed us to extract the specific planning stage of gait initiation, before any movement occurred, but it may interfere, to some degree, with the interpretation of study results. This may contribute to the lack of differences between the older adults and adults with PD in gait initiation stage duration and reaction time duration. Future research should examine whether similar patterns emerge in self-initiated gait.

Second, the number of trials per participant was relatively low (three repetitions), which may have reduced the reliability of time to APA measurements. Some suggest that at least ten trials may be necessary for robust gait initiation analysis (Seuthe et al. 2021). Increasing the number of trials in future studies could provide more stable estimates of time to APA. More trials could also allow us to study across trial variability, a factor that has been associated with PD in some, but not all, studies (De Waele et al. 2024; Lin et al. 2016; Roemmich et al. 2012). Nonetheless, among the patients with PD, the time to APA was relatively strongly associated (rs = 0.654; p < 0.001) with the Timed Up and Go scores, a test previously associated with cognitive function and fall risk (Çekok et al. 2020; Mirelman et al 2014). Still, the magnitude of the correlation indicates that other factors are needed to fully explain the variance of the time to APA and/or that physiologic noise due to the small number of repetitions may have been a factor. Additionally, participants with PD were tested in their ON medication state, meaning that our findings reflect the influence of dopaminergic therapy. Future research should investigate time to APA in the OFF state to determine whether it is a stable marker of disease severity.

We note, too, that we examined upper limb and lower limb reaction times separately. In the future, it would be interesting to explore the neural mechanisms linking motor planning across different effectors. Methods such as EEG and EMG, the simultaneous study of medial–lateral stability, and the assessment of a lower extremity SRT and CRT could also provide further insight into the cortical dynamics underlying time to APA (Jacob & Nora 2022; Watanabe & Higuchi 2022). In addition, it is important to keep in mind that while several meaningful and statistically significant associations were observed, the relatively small sample size, especially within each group, may limit the ability to detect more subtle differences or effects. Without an a prior power analysis, it is difficult to definitively state whether the power was fully satisfactory. Future studies should examine these questions using larger sample sizes. We also point out that we examined associations using correlations; an alternative approach is to use mixed-effect models or regression analyses. We opted to use correlation analyses as this does not make any assumptions about cause-and-effect and the direction of the associations; however, this may have come with some negative costs. Our choice of using a random delay, rather than a fixed delay to signal the cue to start, also has implications since some studies have indicated that APA duration is longer in the random delay condition in people with PD (Lu et al. 2017). In the future, it will be interesting to compare time to APA in fixed and random delays. Finally, as mentioned above, we note that the present study focused on patients with PD who did not experience FOG. Given the importance of start hesitation to FOG and the previously reported differences in APA characteristics in those PD patients who have FOG, as compared to those who do not (Palmisano et al. 2022; Lencioni et al. 2024; de Lima-Pardini et al. 2020), it would also be interesting to examine time to APA in patients with PD and FOG.