Age interferes with sensorimotor timing and error correction in the supra-second range

Abstract

Introduction: Precise motor timing including the ability to adjust movements after changes in the environment is fundamental to many daily activities. Sensorimotor timing in the sub-and supra-second range might rely on at least partially distinct brain networks, with the latter including the basal ganglia (BG) and the prefrontal cortex (PFC). Since both structures are particularly vulnerable to age-related decline, the present study investigated whether age might distinctively affect sensorimotor timing and error correction in the supra-second range.

Methods: A total of 50 healthy right-handed volunteers with 22 older (age range: 50-60 years) and 28 younger (age range: 20-36 years) participants synchronized the tap-onsets of their right index finger with an isochronous auditory pacing signal. Stimulus onset asynchronies were either 900 or 1,600 ms. Positive or negative step-changes that were perceivable or non-perceivable were occasionally interspersed to the fixed intervals to induce error correction. A simple reaction time task served as control condition.

Results and discussion: In line with our hypothesis, synchronization variability in trials with supra-second intervals was larger in the older group. While reaction times were not affected by age, the mean negative asynchrony was significantly smaller in the elderly in trials with positive step-changes, suggesting more pronounced tolerance of positive deviations at older age. The analysis of error correction by means of the phase correction response (PCR) suggests reduced error correction in the older group. This effect emerged in trials with supra-second intervals and large positive step-changes, only. Overall, these results support the hypothesis that sensorimotor synchronization in the sub-second range is maintained but synchronization accuracy and error correction in the supra-second range is reduced in the elderly as early as in the fifth decade of life suggesting that these measures are suitable for the early detection of age-related changes of the motor system.

Keywords: healthy; sensorimotor; sensorimotor synchronization; sub-second and supra-second; tapping; timing.

Figure 1

Paradigm. (A) In a series of 10 pacing signals a step-change was randomly interspersed between the 4th and 7th pacing signal. The 3 taps preceding the step-change served as a measure of baseline synchronization performance. (B) In two separate blocks baseline SOAs of either 900 or 1,600 ms were adopted. (C) Step-changes were either positive or negative and either small (i.e., non-perceivable) or large (i.e., perceivable). Data from a pilot study suggest asymmetrical detection thresholds with smaller deviations being detected in positive as compared to negative deviations. Step-changes were adapted with respect to those findings.

Figure 2

Experimental setup. The participants were seated at a table with standard speakers placed about 50 cm in front of them. Tap-onsets of the right index finger disrupted a photo-electric barrier mounted on a board (Elekta Neuromag^®, Helsinki, Finland). Timing of tone presentation as well as registration of tap onsets were realized by E-Prime 3 (Psychology Software Tools Inc., Sharpsburg, PA, United States).

Figure 3

Synchronization accuracy in trials preceding the step-change. (A) Mean tap-to-tone asynchrony. Significant group differences emerged in trials preceding positive step-changes only. Prior to small step-changes the asynchrony was significantly smaller in the old group and changed to positive values prior to large step-changes. (B,C) Synchronization variability. While in trials with sub-second intervals no significant group-effect emerged (B), variability was significantly larger in the old group in trials with supra-second intervals (C). This was particularly evident in trials with small positive step-changes. A trend toward significantly larger variability in this group was found in trials with large negative step-changes. Error bars indicate the standard error of the mean (*) p ≤ 0.10; *p ≤ 0.05; **p ≤ 0.001.

Figure 4

Phase correction response (PCR) in percent of the base SOA. (A) In trials with sub-second intervals no significant group differences emerged. (B) In trials with supra-second intervals the PCR was significantly smaller in the old as compared to the young group. (C) Small step-changes were not associated with significant group differences, (D) while large positive step-changes yielded significantly smaller PCR in the old as compared to the young group. Error bars indicate the standard error of the mean *p ≤ 0.05; **p ≤ 0.001.