Effects of an external compared to an internal focus of attention on the excitability of fast and slow(er) motor pathways

Abstract

The neurophysiological mechanisms underlying the behavioural improvements usually associated with an external (EF) compared with an internal focus of attention (IF) remain poorly investigated. Surround inhibition in the primary cortex has been shown to be more pronounced with an EF, indicating a more spatial restriction of the motor command. However, the influence of different foci on the temporal aspect of the motor command, such as the modulation of fast versus slow(er) motor pathways, remains unknown and was therefore investigated in this study. Fourteen participants were asked to press on a pedal with the right foot to match its position with a target line displayed on a screen. The deviation of the pedal from the target line was used as a behavioural parameter and compared between both foci (EF vs IF). Additionally, conditioned H-reflexes were evoked during the motor task to assess the excitability of fast (direct) and slower (more indirect) motor pathways when adopting an EF or IF. With an EF compared to an IF, the motor performance was enhanced (P = .001; + 24%) and the activation of slow(er) motor pathways was reduced (P < 0.001, - 11.73%). These findings demonstrate for the first time that using different attentional strategies (EF and IF) has an influence on the excitability of slow(er) motor pathways. Together with the increased intracortical inhibition and surround inhibition known from previous studies, the diminished activation in the slow(er) motor pathways further explains why using an EF is a more economic strategy.

Figure 1

(A) Shown is the experimental setup. The TMS coil was placed over the left primary motor cortex (1) as the participants were seated in an upright position facing a monitor placed in front of them (2). In order to evoke H-reflexes in the soleus muscle (SOL), an anode was attached to the anterior aspect of the knee and a cathode was placed in the popliteal fossa (3). To record electrical muscular activity, bipolar surface electrodes were placed on the right SOL (4). Participants were asked to place the right foot on a pedal (5) with a knee angle between 120° and 130°. The position of the pedal which the foot controlled was continuously displayed on the monitor in front of them (2). Participants were asked to match the position of the pedal with the target line displayed on the screen adopting either an external (EF) or an internal focus of attention (IF). The verbal instructions inducing an EF or an IF were given before the motor task. (B) The H-reflex conditioning technique allows investigating the excitability of the cortico-spinal tract in further details: it allows differentiating between slow and fast corticospinal motor pathways emerging and projecting from M1. By stimulating the primary motor cortex (M1) with transcranial magnetic stimulation (TMS) and the tibialis nerve with peripheral nerve stimulation (PNS) at the same time, the synaptic input from afferent volleys at the spinal level (H-reflex loop: 1a afferent pathway—spinal motoneuron) will coincide with the descending cortico-spinal volley evoked by TMS. The final responses are conditioned H-reflexes in the SOL muscle, which can be recorded with surface EMG and then compared between conditions in terms of amplitudes. (C) The descending volley evoked by TMS over M1 is dispersed. The early waves are attributed to fast corticomotor pathways whereas later waves are attributed to slower pathways. The PNS stimulation can be shifted in order to coincide with different parts of the TMS-induced descending volley. The first response observed in the muscle is called the early facilitation and represents the earliest arriving cortical input evoked by TMS that coincides with the input evoked by PNS. The early facilitation was identified and adjusted for each participant between interstimulus intervals (ISIs) between − 6 and − 1 ms (PNS sent before TMS). All other ISIs (+ 4, + 8, + 12, + 16 ms) were identical for all participants (PNS after TMS).

Figure 2

Shown are individual values (n = 14) and boxplots comparing both attentional foci (EF and IF) of the deviation from the target position of the pedal in degrees (°). Participants could better match the target line by moving the pedal with their foot when adopting and EF contrasted to an IF (around + 24% improvement). Left boxplot (black) represents the external focus of attention condition (EF) and the right one (white) the internal focus of attention condition (IF). ** P ≤ 0.01.

Figure 3

In order to normalise the conditioned H-reflexes in percent (%) for each participant, the unconditioned test H-reflex (not represented) was considered as the baseline. In all graphs, black boxplots/shapes represent the external (EF), the white ones the internal focus of attention (IF) and the grey ones the REST condition (no contraction and no focus). On the abscissa of (A) and (C), the early facilitation was adapted individually for each participant as the ISI that resulted in the first facilitation peak followed by an inhibition of the test H-reflex 1–2 ms later. (A) Shown are mean values and SEM (n = 14) of the peak-to-peak amplitudes of the conditioned H-reflexes at five different interstimulus intervals (ISIs : early facilitation, + 4, + 8, + 12 and + 16 ms) under both attentional strategies (EF and IF) in the right soleus muscle. On the abscissa, the ‘early facilitation’ was adapted individually for each participant as the ISI that resulted in the first facilitation peak followed by an inhibition of the test H-reflex 1–2 ms later. When comparing ISI curves, post hoc comparisons (see C for all interactions) showed that adopting an EF resulted in reduced conditioned H-reflexes at ISIs + 4, + 8 ms and + 16 ms, when compared with an IF. (B) Shown are individual values and boxplots of the of the peak-to-peak amplitudes of the conditioned H-reflexes comparing the three conditions (EF, IF and REST). The main effect of condition revealed that conditioned H-reflexes at REST were significantly smaller when compared with H-reflexes at EF and IF. *** P ≤ 0.001. (C) Shown are individual values (n = 14) and boxplots of the peak-to-peak amplitudes of the conditioned H-reflexes at five different interstimulus intervals (ISIs : early facilitation, + 4, + 8, + 12 and + 16 ms) under both attentional strategies (conditions: EF and IF) and the REST condition (no contraction and no specific focus of attention) in the right soleus muscle. When comparing the early facilitation, no significant difference was found between the conditions (EF, IF and REST). When comparing the ISI curves between the EF and IF condition, post hoc comparisons showed that adopting an EF resulted in reduced conditioned H-reflexes at ISIs + 4 (around − 16%) + 8 ms (around − 25%) and + 16 ms (around − 10%), when compared with an IF. The H-reflexes measured at the + 4 ms ISI were significantly smaller during the REST condition than the EF and or IF conditions. In order to normalise the conditioned H-reflexes in percent (%) for each participant, the unconditioned test H-reflex (not represented) was considered as the baseline.

Figure 4

In all graphs, shown are individual values (n = 14) and boxplots comparing of (A) the test single-pulse MEPs (unconditioned) under both attentional strategies (EF and IF), (B) background activity of the SOL muscle 100 ms before TMS and/or PNS (bEMG) under both attentional strategies (EF and IF) and (C) the test H-reflexes (unconditioned) under both attentional strategies and the REST condition (no contraction and no specific focus of attention). No significant difference was found in all these parameters between conditions (all P > 0.05 In all graphs, black boxplots represent the external (EF), the white ones the internal focus of attention (IF) and the grey ones the REST condition (no contraction and no focus).