Keeping the balance of the whole body and of each of its segments during voluntary motor actions requires specific activities in postural muscles, which are called Anticipatory Postural Adjustments (APAs). APAs precede and counterbalance the postural perturbations induced by voluntary movements (Bouisset and Zattara, 1981; Bouisset and Do, 2008) and are programmed in a feed-forward way. Indeed, postural muscles activate prior to the onset of the focal movement, so as to develop one or more fixation chains toward the available support points. APAs are also involved in gait initiation, where their specific role is still debated: classic literature reports that they de-stabilize the body by displacing the center of mass (for a review, see Yiou and Caderby, 2017), while a recent study brings them back to the general role of body stabilization, as they appear to fix the trunk and upper body segments to the moving hip (Farinelli et al., 2021).

The role of APAs in whole-body stabilization is well-documented during movements involving large masses, like raising the arms or flexing a leg (Bouisset and Zattara, 1987; Breniere, 1987; Crenna and Frigo, 1991). These postural activities take the name of inter-limb APAs because they usually spread over several muscles of different limbs (Belen’kii et al., 1967; Friedli et al., 1984; Aruin and Latash, 1995). However, APAs are also produced when the voluntary movement involves only tiny masses, like flexing a finger or the wrist; in these cases, intra-limb APAs are observed in the same limb where the distal segment is moved (Aoki, 1991; Caronni and Cavallari, 2009a; see also Cavallari et al., 2016 for a review). These actions are needed in order to stabilize the proximal segments and to contribute to attain a higher precision of the focal movement (Hopf and Hufschmidt, 1963; Almeida et al., 1995; Bruttini et al., 2016).

Inter- and intra-limb APAs share several properties, like the ability to tune in to the needs of the postural context (Aruin and Latash, 1995; Hall et al., 2010; Bruttini et al., 2014), the adaptation to changes in movement speed (Esposti et al., 2015), and the important link with movement precision (Caronni et al., 2013; Bruttini et al., 2016). Moreover, several neural structures are involved in both inter- and intra-limb anticipatory postural control, like the primary motor cortex, the supplementary motor area (SMA), the sensorimotor areas, the pontomedullary reticular formation, and also subcortical structures such as the basal ganglia and the cerebellum (Viallet et al., 1987; Massion et al., 1999; Schepens and Drew, 2004; Schmitz et al., 2005; Caronni and Cavallari, 2009b; Petersen et al., 2009; Bolzoni et al., 2015, 2018; Bruttini et al., 2015; Farinelli et al., 2020).

With regard to the neural organization of voluntary movement and posture, an important finding suggests that the postural command governing APAs and the focal command that controls prime mover recruitment are functionally joined into a single motor command (Bruttini et al., 2014). However, transcranial Direct Current Stimulation (tDCS) applied on SMA separately modulated APAs without affecting the prime mover recruitment. This supports the idea that the motor command splits into its postural and focal components before entering SMA. It was therefore of interest to investigate hierarchically higher integration centers, maybe involved in supporting the APA adaptation to the postural context, thus being able to process both sensory and motor streams.

Recently, we investigated the role of Parietal Operculum (PO) because this neural structure is an important hub for integrating the proprioceptive and tactile information for the motor control (Milner et al., 2007; Sepulcre, 2014). Moreover, visual, somatosensory, and auditory functional streams, originating from different cerebral areas, converge in PO and are then headed to motor and premotor areas (Felleman and Van Essen, 1991). We started by using anodal or cathodal tDCS to selectively modulate the excitability of the PO contralateral to the moving segment (coPO, Marchese et al., 2019), while assessing intra-limb APAs associated to brisk flexions of the index finger, according to the paradigm of Caronni and Cavallari (2009a). Indeed, tDCS is known to modulate the cortical excitability by increasing or decreasing its activity, depending on the polarity. For example, anodal stimulation on the primary motor cortex transiently increases its excitability, while cathodal tDCS decreases it (Furubayashi et al., 2008; Tatemoto et al., 2013). TDCS also affects motor performance bilaterally, influencing finger sequence movements on both body sides (Vines et al., 2006). In fact, those Authors showed that anodal stimulation of the left motor area improved right-hand performance more than cathodal polarity, whereas opposite effects were induced on the left-hand. Against expectations, tDCS on coPO did not significantly affect the control of APAs associated to index-finger flexion: the intra-limb APA timing and pattern (which muscle showed excitation and which one inhibition) during and after tDCS of either polarity were at all comparable to those observed in the sham condition. These negative results suggested us that coPO may not be involved in the network governing APAs.

However, tDCS literature reports several cases in which the simultaneous stimulation of a given area in both hemispheres with opposite polarities (dual-hemisphere, dh-tDCS) elicited stronger effects than the unilateral tDCS. Given these premises, a new question opens: what will occur if the tDCS involves both POs? The dh-tDCS method interferes with the inter-hemispheric inhibition (Curtis, 1940), by which one hemisphere inhibits the contralateral one during the generation of voluntary unimanual movements (Boddington and Reynolds, 2017). This interhemispheric mechanism is revealed in healthy subjects by the inhibition of the contralateral motor cortex during movement initiation (see Beaulé et al., 2012, for a review); it was also proposed that interhemispheric inhibition would suppress mirror movements that may ruin the task performance, however it remains an unproven hypothesis.

Several Authors tried to exploit the inter-hemispheric balance in the cerebral network for increasing hand motor performance in healthy subjects and to recover motor function after brain injury (Fregni and Pascual-Leone, 2006; Williams et al., 2010). In this perspective, dh-tDCS has been proposed as a tool used to improve cerebral functionality. For example, Vines et al. (2008) tested the non-dominant hand and found that the simultaneous application of cathodal tDCS over the dominant motor cortex and anodal tDCS over the non-dominant motor cortex facilitated motor performance, with larger effects than the uni-hemisphere tDCS. Dh-tDCS has also been applied to stroke patients, trying to improve performance of the paretic hand by enhancing motor skill learning and reducing the spasticity (Lindenberg et al., 2010; Vandermeeren et al., 2012; Lefebvre et al., 2013; Vandermeeren and Lefebvre, 2015). However, the balance of results in the literature as well as the contribution of interhemispheric inhibition to motor deficits after stroke is still debated. Nevertheless, recent studies have shown that dh-tDCS improved cognitive functions (Cohen Kadosh et al., 2010; Kasahara et al., 2013) and tactile discrimination on the dominant hand of healthy subjects (Fujimoto et al., 2014, 2017). In particular, in Fujimoto et al. (2017) highlighted that dh-tDCS on PO was significantly more effective compared with uni-hemisphere tDCS.

Based on these findings, the aim of this study is to test whether the more powerful dh-tDCS set-up applied over PO could elicit significant effects on intra-limb APAs, which were not observed with unilateral tDCS. A positive result would suggest a PO involvement in APA neural control. Should also this protocol fail in affecting APAs and focal movement, we could definitely exclude such structure from the neural network governing APAs.

Twenty subjects were enrolled in this two-groups, mixed-design, experimental study (13 males and seven females, mean age 26.7 ± 9.2 SD). Subjects were randomly assigned and equally distributed between the two stimulation polarities (see Neuronavigation and Tdcs for details). All volunteers were right-handed, as ascertained by the Oldfield handedness questionnaire. No subject had any history of neurological or orthopedic diseases, as well as of intake of drugs acting on the Central Nervous System. Participants provided their informed consent but were kept completely unaware of the stimulation condition. The experimental protocol complied with the policies and principles contained in the Declaration of Helsinki and were approved by the Ethical Committee of the University of Milan (counsel 6/19).

Subjects were sitting on a chair with both arms at rest. When performing the required movements (see below), the subject actively kept the right upper arm along the body with the elbow flexed at 90° and the hand prone, lined-up with the forearm. The index-finger was kept extended and aligned with the hand, while all other fingers were hanging. Throughout the experiment, subjects had to keep their back supported and both feet on the ground (Figure 1A). The experimenters supervised the subject’s position during the whole experimental session.

Panel A of Figure 2 shows the mean traces of the integrated EMG and of finger kinematics signals recorded from a representative subject of the group who underwent dh-tDCS with the anode placed on the left PO and the cathode on the right (ANODE LEFT). Each trace is the average of the 30 movement trials that were recorded immediately before tDCS application (Pre), in the last minutes of full-current stimulation (Dur 20′), and after 10 and 30 min of recovery (Post 10′ and Post 30′). Confirming Caronni and Cavallari (2009a), in Pre, the FDS onset (solid vertical line) was preceded by an inhibition in AD and BB, and by a burst of activity in TB. Notably, these EMG changes preceded movement onset (dashed vertical line) and stabilized the arm against the perturbation produced by the finger flexion; for these reasons, such actions are classified as APAs. All the traces recorded in Dur 20', Post 10', and Post 30' were comparable to those recorded in Pre. Figure 2B refers to population data and shows that the mean values of APA amplitude and latency obtained during and after ANODE LEFT dh-tDCS (including Dur 10′ and Post 20′) were at all comparable to those recorded in Pre. Experiments conducted in the subjects who underwent ANODE RIGHT stimulation (Figures 3A,B) led to results that actually replicated those obtained in the anode left group, confirming that either polarity had no effect on APA amplitude or latency.

These findings were also supported by statistics: two-way ANOVAs did not find any significant interaction (in all muscles, F_5,90 ≤ 1.077, p ≥ 0.38, η²_p ≤ 0.056). The same was true for the main effects of time (F_5,90 ≤ 1.713, p ≥ 0.14, η²_p ≤ 0.087) and polarity (F_1,18 ≤ 2.657, p ≥ 0.12, η²_p ≤ 0.129). Finally, tDCS had no effect on amplitude of FDS recruitment and index-finger kinematics (two-way ANOVAs: interaction F_5,90 ≤ 1.778, p ≥ 0.13, η²_p ≤ 0.090; time F_5,90 ≤ 1.486, p ≥ 0.20, η²_p ≤ 0.076; and polarity F_1,18 ≤ 1.506, p ≥ 0.24, η²_p ≤ 0.077).

The aim of this work was to re-test with a new approach whether the PO is engaged in the organization of intra-limb APAs. We addressed our studies to this structure in search of an area in which the postural and the voluntary components of the motor command, which split before entering SMA (Bolzoni et al., 2015), may be still processed together. We focused on PO as an important hub able to integrate multisensory streams within the motor organization (Milner et al., 2007; Sepulcre, 2014). Even though our recent work (Marchese et al., 2019) demonstrated that anodal or cathodal tDCS limited to the PO contralateral to the moving finger (coPO) did not disturb the APAs, here, we bilaterally stimulated both POs with opposite currents (dh-tDCS) as this approach has been proved to produce stronger effects than the unilateral stimulation in several neuromodulation studies. Dh-tDCS has been described to be a proficient tool to improve cerebral functionality (Vines et al., 2008) and to be more effective with respect to the uni-hemisphere approach. In particular, Fujimoto et al. (2017) reported that dh-tDCS over the POs, with the same 2 mA intensity used in the present study, actually produced a greater improvement in tactile orientation discrimination than the uni-hemisphere tDCS. Nevertheless, we did not observe any significant effect of dh-tDCS on amplitude or latency of intra-limb APAs associated with the index-finger flexion, in this way mimicking the results and confirming the conclusion of our previous work (Marchese et al., 2019).

Being confident, on the above premises, that dh-tDCS actually modulated PO excitability, it remains to exclude that our negative findings stem from experimental flaws. In this regard, the sample size seems adequate, as witnessed by the 80% power of ANOVA for the time effect and the polarity × time interaction; the two meaningful effects to be evaluated in our design (see Materials and Methods). It could be also argued that we did not perform sham dh-tDCS. However, Marchese et al. (2019) reported that sham on the coPO did not elicit any effect. Considering that in sham the 60 s fade-in period was immediately followed by the 30 s fade-out, so that current exposure was practically null, it is apparently unrealistic that repeating such protocol with the dual-hemisphere electrodes arrangement would lead to different results. The last important aspect is that in both Marchese et al. (2019) and in the present study, all finger flexions occurred with the arm kept in the same position (actively supported by the subject, with elbow at 90° and hand prone in axis with the forearm). It could then be debated that such experimental protocol does not allow to test the well-known ability of APAs to tune to the needs of the postural context in which the movement occurs (Aruin and Latash, 1995; Caronni and Cavallari, 2009a; Hall et al., 2010; Bruttini et al., 2014) and, consequently, it is unsuitable to highlight a possible involvement of PO in such function. However, the subjects had resting periods between each movement sequence and, in such periods; they laid their right arm on a support, thus implying a cyclic change between the experimental posture and the resting posture. Considering that dh-tDCS always started during a resting period, if PO were important for either perceiving the mechanical context or in integrating such information in the motor flow, one should expect a disruption of APA pattern or timing as soon as the subject assumed the experimental posture. Present results actually disproved such expectation.

On the other hand, the conclusion that PO is not involved in the control of APAs does not contrast with current literature. In fact, such area seems more influential on the earlier strategic phase of selecting the motor goal rather than in the planning phase of the motor act (Tunik et al., 2008; Woods et al., 2014; Valyear and Frey, 2015), while the commands for the prime mover and the related APA chains are defined in the latter phase. Moreover, the contribution of PO may concern more specific motor actions and learning-memory rather than the planning of motor acts. As evidenced by several works, PO network may be involved in other motor functions like working memory and tactile learning (Jäncke et al., 2001) and it might be more important for object-directed motor behavior (Maule et al., 2015). Additionally, the PO may modulate auditory-sensorimotor control, by mediating multimodal integration (Tanaka and Kirino, 2018), as well as being involved in orofacial muscles activities during phonation (Grabski et al., 2012).

On the bases of the present results and of the above considerations, we feel confident in concluding that our search for an area in which the motor command to prime mover and postural muscles are still functionally unique will have to address other structures, such as the premotor cortices.

The datasets generated for this study are available on reasonable request to the corresponding author.

The studies involving human participants were reviewed and approved by Ethical Committee of the University of Milan (counsel 6/19). The patients/participants provided their written informed consent to participate in this study.

PC conceived the study and raised the funds. RE and SM conducted the experiments with the contribution of FB and VF and analyzed the results. PC, RE, and SM drafted the manuscript, which was approved, in the final version, by all authors. All authors contributed to the article and approved the submitted version.

This study was supported by internal funds from “Laboratorio Analisi dell’Università degli Studi di Milano” (TARIFFARIO09_CAVALLARI).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.