In the present study, we examined variations in corticomotor excitability associated with unimanual pattern recognition at the fingertip in young and older adults. In Experiment I, we showed that haptic enhancement in corticomotor excitability in the working hemisphere was similar in magnitude in young and older adults, when task difficulty was adjusted for age difference. In Experiment II, we showed that haptic enhancement in corticomotor excitability could also be reliably detected in the "resting" hemisphere opposite to the working hemisphere and that this cross-modulation was similar in magnitude between the two age groups and regardless of which hemisphere (i.e., right or left) was engaged in haptic sensing. We will first address the significance of the haptic enhancement observed in Experiment I in the context of aging. Then, we will examine issues pertaining to bilateral corticomotor facilitation as assessed by TMS, in the context of aging with an emphasis on the importance of task demands, in line with the haptic enhancement observed in Experiment II in the "resting" hemisphere.
Experiment II. Corticospinal excitability in the resting hemisphere
As stressed before, increases in ipsilateral sensorimotor activity have been observed at different levels in the motor system (i.e. peripheral and central) during a variety of unimanual tasks [18–20]. In the present study, we observed that engaging one hemisphere and one hand in haptic sensing was associated with enhanced excitability of the opposite hemisphere associated with the resting hand. In addition, this enhancement was comparable in magnitude between young and old participants, regardless of which hemisphere was engaged in the task. In this regard, our observations are consistent with many TMS reports in young adults, wherein enhanced corticomotor excitability on the "resting" hemisphere has been described when the other was engaged in various unimanual tasks [4, 19, 21–25]. The present findings may also be interpreted in light of neuroimaging data showing increases in the activity of the ipsilateral sensorimotor cortex during various unimanual tasks, particularly more complex tasks [1, 26]. This raises the question as to why smooth trials, which produced similar pattern of background EMG as that seen with grating trials in the task hand, failed to elicit any detectable changes in MEP amplitude in the resting hand; even when compared to baseline values in the resting state. One possible reason for this may reside in the nature of the finger movement, which required only minimal effort on the part of the participant and was simple and highly predictable from trial to trial. In line with this, Tinazzi and Zanette  observed a gradual decrease in corticomotor facilitation in the resting hand when the task performed with the opposite hand became more automatic after sequential training. Evidently, grating trials were not susceptible to such training effects with repetitions, because participants were required to deploy tactile attention and use cognitive resources at each trial to discriminate the grating orientation.
Concerning the neural mechanism for the observed haptic enhancement in the "resting" hemisphere, as indicated before, the selectivity of the effect for grating trials points to the parietal-frontal network recruited during tactile attention associated with haptic sensing [15, 16] as the primary source of facilitatory influences. In principle, such central facilitation could be exerted at two sites, i.e. either indirectly through the active hemisphere or directly through the "resting" hemisphere. With regard to the first possibility, one possible source is descending ipsilateral facilitation coming from the activated motor cortex through uncrossed corticospinal projections reaching spinal motoneurons. Such a possibility is very unlikely, however, given that ipsilateral corticospinal projections destined to distal hand muscles are rare in primates , and were shown recently to have only very weak effects in modulating ipsilateral motoneuronal activity during hand movements . Another related source is irradiation of facilitatory influences or a release from inhibition from the activated primary motor cortex to the homologous cortex through transcallosal projections. Such a possibility is also unlikely, however, for three major reasons. First, most TMS studies looking at interhemispheric influences between motor cortices have emphasized the inhibitory nature of these interactions [see  for review] such that conditioning stimuli applied to one hemisphere lead to suppression of MEPs elicited from the opposite hemisphere over a wide range of inter-stimulus intervals. Interhemispheric facilitation can also be elicited but the phenomena has been described as less reliable and remains highly susceptible to variations in experimental paradigm [e.g., . Second, the fact that no mirror activity was detected in the resting FDI homologous to the one engaged in haptic sensing, and that smooth trials elicited no sign of MEP facilitation, argue against the notion of a simple spread of excitation from one motor cortex to the other as the primary source for the observed corticomotor facilitation. Finally, both anatomical tracer work in non-human primates  and tractography analyses of diffusion tensor imaging in humans  converge to show that primary motor and somatosensory cortices have relatively scant callosal connections when compared to secondary motor or somatosensory areas. This leaves the "resting" hemisphere ipsilateral to the task hand as the most probable site of origin for the observed MEP facilitation in the resting hand. In this respect, a likely source of ipsilateral facilitation during trials with grating discrimination is the recruitment of secondary motor areas, such as the pre-SMA, SMA proper and ventral premotor cortex; which all have been shown to be activated to some degree in tasks involving haptic sensing with the finger [15, 16]. As stressed earlier, these secondary motor areas have dense callosal connections and are also densely interconnected with M1 intra-cortically [31, 33]. In addition, both SMA and premotor neurons are known to exhibit pre-movement activity during ipsilateral unimanual movement, which contrasts to MI neurons where such premovement activity is seldom found [34, 35]. Therefore, bilateral recruitment of secondary motor areas associated with cognitive demand and tactile attention in the context of unilateral haptic sensing seems the most likely source of facilitation to raise motor excitability in the "resting" hemisphere. Yet, one can ask as to why a large increase in corticomotor excitability in the resting hemisphere was not accompanied by overt motor overflow in the resting hand. In this respect, it is possible that interhemispheric inhibitory influences from the active hemisphere to the "resting" hemisphere likely balanced out the facilitation mediated by the secondary motor areas recruited during haptic sensing, reducing it to a sub-threshold level so as to prevent overt mirror movements of the non task hand . To summarize, although we cannot totally rule out the possibility that irradiation from the active hemisphere contributed to the observed ispsilateral MEP enhancement, both anatomical and physiological evidence point to the "resting" hemisphere as the most likely site for the facilitation in link with the recruitment of secondary motor areas in the context of haptic processing.
In line with a previous TMS study comparing right-left differences in the excitability of corticomotor projections to the resting hand when the other hand performed simple movements of varying intensities , no difference was found between the two hemispheres with regard to task-related enhancement in corticomotor excitability. However, ipsilateral cortical activity during unimanual performance has often been reported to be more pronounced in the left hemisphere and more important for left hand task performance, particularly for complex tasks [e.g., [1, 38]. The present observation that haptic-related corticomotor facilitation in the resting hemisphere was unaffected by hand could be related to the nature of our task, as we have argued before. The fact that our finger task consisted of a simple action requiring minimal effort at the muscular level could have contributed to accentuate differences arising from central influences at the motor cortical level between smooth and grating trials; leading to similar facilitation irrespective of the hand performing the task. Similar to observation between hemispheres, no differences was also observed between young and old participants in the level of MEP facilitation elicited in the "resting" hemisphere when the opposite one was engaged in haptic sensing. This is somewhat in contrast to previous neuroimaging and behavioural studies reporting age-dependant increase in bilateral sensorimotor activation and in peripheral muscle activity during simple unimanual tasks [20, 39–41]. Seniors also exhibited less interhemispheric motor inhibition than young adults during performance of unimanual tasks, suggesting that attending to a task may affect inhibitory modulation of ipsilateral motor cortical activity with aging . However, in the present study, the possibility of motor overflow affecting the status of the "resting" hand was greatly reduced by asking participants to refrain from contracting their muscles in the relaxed state and by constant monitoring of EMG level. Such active strategy has been shown to be highly effective in suppressing mirror movements in the non-task hand when complex movements were performed with the other hand in both young and old adults . Thus, any age-dependant increases in motor overflow might have been mitigated by our active control over spreading muscle activity in the resting hand as senior and young participants performed the finger task. Another factor that might have contributed to attenuate age differences is the fact that we controlled for the level of task difficulty. In general, older adults tend to exhibit more widespread cortical activation than younger subjects when motor task demands are increased in terms of complexity [e.g., [2, 20, 39–41, 44]. On this basis, one could have predicted greater haptic-related motor facilitation in the older group owing to the recruitment of a more extensive bilateral frontal and parietal network associated with discrimination of the grating orientation; but this was not the case. As suggested before, such compensatory over-recruitment might have been largely cancelled out in this study by the combined effect of using a simple finger action task and adjusting the level of perceptual difficulty for age for the haptic component. Overall, these observations with regard to the effect of age would be compatible with the compensation-related utilization of neural circuits hypothesis (CRUNCH) theory . This theory suggests that aging results in a loss of hemispheric specialization only in situations in which performance deterioration is evident, to increase compensatory recruitment of available cognitive resources and help improve task performance in seniors . Indeed, poorer unimanual motor task performance in seniors was associated with greater recruitment of ipsilateral M1, decreased callosal size, and integrity, indicating a shift towards excitation in the balance of excitatory and inhibitory interhemispheric interactions [36, 39, 46]. In the present study, given the careful design of task demands and stimuli to facilitate good performance in seniors in the haptic sensing trials, there would be no reason to expect a compensatory over-recruitment of ipsilateral or contralateral motor areas.