Differences in motor learning success are associated with differences in M1 excitability

Abstract

Primary motor cortex (M1) plays a role in motor learning, although the exact nature of that involvement remains unclear. The present study examined the relationship between motor learning and cortical plasticity by manipulating augmented feedback during motor training. Two groups of 10 participants performed a wrist flexion–extension waveform-tracking task with either concurrent and terminal augmented feedback after every trial (100% FB) or only terminal feedback after every alternate trial (50% FB). Single- and paired-pulse transcranial magnetic stimulation (TMS) was used to assess cortical excitability short-interval intracortical inhibition (SICI) and intracortical facilitation (ICF) before, after, and 24 h following (retention) motor training. The 50% FB group performed better at retention than the 100% FB group, indicative of enhanced learning due to reduced FB scheduling. Cortical excitability did not change during acquisition for either group, however, the 50% FB group had elevated M1 excitability at retention, suggesting M1 involvement in the consolidation of learning. Reduced SICI following practice suggests a reduction of intracortical inhibition during motor skill acquisition. ICF was unchanged. It is concluded that the nature of M1 modulation associated with the acquisition and retention of a novel motor skill appears to vary with the nature and complexity of task requirements.

Introduction

Functional and structural cortical change in the human adult brain, or cortical plasticity, is a process of importance to both the learning of new motor skills, and to functional recovery following brain injury. Although research has indicated that primary motor cortex (M1) plays a role in motor learning, the exact nature of that involvement has not yet been clarified. Transcranial magnetic stimulation (TMS) has been used to non-invasively observe changes in M1 associated with motor training. Use-dependent plasticity, such as increases in motor map representations of muscles involved in motor tasks have commonly been observed (e.g., Karni et al., 1998, McDonnell and Ridding, 2006, Nudo et al., 1996, Pascual-Leone et al., 1995). TMS has also been used to observe modulations of measures of cortical excitability, such as motor evoked potentials (MEPs), following simple ballistic (e.g., Muellbacher et al., 2001, Stefan et al., 2006) and complex (e.g., Garry et al., 2004, McDonnell and Ridding, 2006) motor training.

The factors determining M1 plasticity are not yet clear, however it seems that the process of acquiring a motor skill, or practicing motor tasks with a high degree of processing difficulty, are more likely to be associated with cortical change, compared to non-skilled or simple motor behavior (e.g., Carey et al., 2005, Perez et al., 2004, Plautz et al., 2000). The nature of M1 change associated with motor skill training and performance gain has generally been in the form of increases in excitability (e.g., Muellbacher et al., 2001, Perez et al., 2004), a state which may facilitate or be a result of motor learning. For example, Perez et al. (2004) found increases in MEPs from the leg cortical area following skilled motor training. Importantly, no change was observed for non-skill or passive motor training. Interestingly, McDonnell and Ridding (2006) recently found MEP suppression following training on a complex sensorimotor task, and suggested that the modulation of cortical excitability may vary according to the nature and complexity of the motor task. This implies that both increases and decreases in MEPs may be meaningfully associated with motor skill learning; a finding consistent with recent fMRI data showing that experts have an overall lower volume of task-related brain activation compared to novices (Milton, Solodkin, Hlustik, & Small, 2007). To identify a role for task complexity in shaping the nature of M1 involvement in motor skill learning, complexity needs to be manipulated systematically within a task.

Manipulating practice conditions to influence the success with which a motor task is learnt may be one way of further clarifying the role of M1 in motor learning and performance. The behavioral motor learning literature reveals that the provision of feedback of performance on all practice trials can lead to poorer learning (as measured by performance at a delayed retention test) compared with when feedback is provided on a reduced schedule (on alternate trials) (Lee, Swinnen, & Serrien, 1994). Learning on the reduced feedback schedule is likely to have a higher degree of processing difficulty (Winstein & Schmidt, 1990), and be more demanding of cognitive and neural resources compared to the frequent feedback schedule during the process of skill acquisition. This is because the learner must rely on developing an internal representation of the task on which to base performance when feedback is absent (Schmidt, 1982). As such, practicing a motor task on a reduced feedback schedule is of increased cognitive complexity, and leads to more successful learning.

Overall, aspects of the learning process with which M1 is concerned are not well understood (Jensen et al., 2005, Stefan et al., 2006). There is a wealth of research demonstrating an association between motor learning and changes in cortical excitability (Muellbacher et al., 2001, Perez et al., 2004), however it is not yet known whether differences in the success of motor learning (as measured behaviorally at a delayed retention test) are associated with differences in M1 excitability change. The purpose of the present study was to examine the nature of the M1 response, in terms of both intracortical and corticospinal excitability, to motor training under different feedback schedules. It was expected that varying the frequency of feedback during skill acquisition would lead to differences in task complexity, and the success with which the motor task is learnt. If M1 is sensitive to these aspects of learning, then differences in measures of M1 excitability should be apparent.