Abstract
Stimulus-driven actions are preceded by preparatory brain activity that can be expressed by event-related potentials (ERP). Literature on this topic has mainly focused on simple actions, such as the finger keypress, finding activity in frontal, parietal, and occipital areas detectable up to two seconds before the stimulus onset. However, little is known about the preparatory brain activity when the action complexity increases, and specific brain areas designated to achieve movement integration intervene. The main aim of this paper is to identify the time course of preparatory brain activity associated with actions of increasing complexity using ERP analysis and a visuomotor discrimination task. Motor complexity was manipulated by asking nineteen volunteers to provide their response by simply pressing a key or by adding to the keypress arm extensions (reaching) alone, or in combination with a standing step (involving the whole body). Results showed that these actions of increasing levels of complexity appear to be associated with different patterns of preparatory brain activity. Specifically, the simple keypress was characterized by the largest motor excitatory preparation in premotor areas paralleled by the largest prefrontal inhibitory/attentional control. Reaching presented a dominant parietal preparation confirming the role of these integration areas in reaching actions toward a goal. Stepping was characterized by localized activity in the bilateral dorsomedial parieto-occipital areas attributable to sensory readiness, for the approaching stimulus. In conclusion, the brain is able to optimally anticipate any stimulus-driven action modulating the activity in the brain areas specialized in the preparation of that action type.
Keywords : action preparation, action complexity, brain activity, pre-stimulus ERP
Introduction
Voluntary actions can be defined as the materialization of thought through movement and massively involve the brain from sensory to cognitive and motor areas (e.g., Flanders 2009). Most of these cortical involvements happen well before the movement onset during the anticipatory/preparatory stages of action processing. Understanding the neural activity that supports this action-planning process is a long-standing challenge in neuroscience (e.g., Nguyen, Breakspear & Cunnington, 2014). To study the brain dynamics preceding voluntary actions, the event-related potential (ERP) methods greatly contributed thanks to their high temporal resolution and to the possibility of extracting the specific brain activity related to a movement onset or stimuli requiring motor responses (e.g., Kornhuber & Deecke, 1964; Shibasaki & Hallett, 2006).
Voluntary actions can be executed in response to either external stimuli (here called externally-triggered or stimulus-driven actions) or internal decisions (here called internally-triggered or self-paced actions). While preparatory ERP associated with self-paced actions has been widely studied for many types of movements, from simple finger flexion to complex praxic movements, (e.g., Shibasaki & Hallett, 2006; Wheaton et al., 2005), ERP preceding stimulus-driven actions has been less investigated and, if any, only for tasks requiring just a finger flexion (a keypress) as responses (for a review, Di Russo et al. 2017).
The time course of brain activity preceding any voluntary movement is a slow-rise negative ramp that culminates at movement onset for self-paced action or at stimulus onset for stimulus-driven actions. In both cases, the main ERP component, named Bereitschaftspotential (BP) or Readiness Potential (RP), has been localized in premotor brain areas reflecting the necessary motor preparation (Shibasaki & Hallett, 2006) for any movement, including jumping into an abyss (Nan et al., 2019). The timing of the BP component has been described as having a bi-phasic trend, an early “unconscious” phase, and a later “conscious” phase (e.g., Shibasaki & Hallett, 2006). While the early phase has been described as an unspecific preparation associated with the intention to move (e.g., Lau et al., 2004), consciousness is deployed in the later phase to allow last-minute changes and enable learning, in order to improve performance (e.g., Deecke 2012; Dennet 2005; Mele 2007). For externally triggered actions, larger late BP amplitude has been associated with faster response time in speeded tasks (e.g., Di Russo et al., 2019).
For more complex movements such as reaching, grasping, or pantomimes, additional activity in the posterior parietal cortex (PPC) has been found in few studies (Wheaton et al., 2005a, b; Inouchi et al., 2013; Bozzacchi et al., 2012) and named posterior BP (pBP). The pBP has been described for self-paced, but not for externally-triggered movements.
In addition to the BP and the pBP, there are other two ERP components preceding externally-triggered simple movements that were found in visuomotor tasks. The prefrontal negativity (pN) and the visual negativity (vN). The pN has been localized in the inferior prefrontal gyrus and has been associated with cognitive preparation intended as top-down cognitive control (mainly attentional and inhibitory) for the upcoming actions (e.g., Berchicci et al., 2020). The BP and the pN have been described as a form of accelerator/braking system based on predictive internal models able to modulate the response accuracy/speed trade-off in performance (e.g., Di Russo et al., 2017) and to regulate the behavior proactively before the event (e.g., Aron, 2011).
The vN has been localized in the extrastriate visual areas and has been associated with the sensory readiness reflecting top-down signals for the allocation of preparatory sensory resources for upcoming events (e.g., Bianco et al., 2020). The pN and the vN were just described in tasks requiring simple movements (finger flexion) and have not yet been studied for more complex movements performed by upper or lower limbs, such as reaching or walking/stepping, respectively. For normative data on these components see Di Russo et al. (2019).
Overall, all this evidence suggests that the brain activity preceding any voluntary movement involves a cortical network focused not only on anterior premotor and prefrontal areas but spanning also on more posterior parietal and occipital regions. The contribution of the PPC preparatory activity for stimulus-driven actions has been widely investigated in both human and animal models (Teixeira et al., 2014; Pilacinsky et al., 2018; Sulpizio et al., 2023). In macaque, several studies (Fattori et al., 2001; Galletti et al., 1997, 1999) have shown that neurons in the PPC become active well before the appearance of any electromyographic movement-related activity, thus confirming the role played by these areas in the planning of reaching (Batista & Andersen, 2001; Battaglia-Mayer et al., 2000; Andersen & Buneo, 2002; Connolly et al., 2003; Cui & Andersen, 2007; Fattori et al., 2005; Snyder et al., 1998, 2000). More recent evidence showed that the macaque medial parietal reaching-related V6A area (located on the parieto-occipital sulcus, POc; Fattori et al., 2005, 2009) decodes the position of the object towards which the arm movement is directed even before the movement begins (Filippini et al., 2018). It follows that the early representation of the action that takes place in the macaque PRR (Parietal Reach region, which includes the V6A; Andersen & Buneo, 2002) is not based exclusively on the position of the target or on action patterns present in memory but contains proprioceptive and kinematic information regarding the body movement that has yet to occur (Kuang et al., 2016).
As repeatedly demonstrated, in humans the PPC plays a crucial role in forming movement intentions (Desmurget et al., 2009) and visual motor-imagery (Crammond, 1997; Sirigu et al., 1996; Sulpizio et al., 2020). Importantly, the medial portion of human PPC has been considered a main substrate for reach planning (e.g., Lindner et al., 2010; Pitzalis et al., 2013; Tosoni et al., 2015), while the more anterior and dorsal portion of human PPC is instead considered a main substrate for reach and walk planning, that is whole body actions performed using both upper and lower limbs (Heed et al., 2011, 2016; Leonè et al., 2014; Pitzalis et al., 2019).
Overall, combining animal and human data, the general idea is that the PPC integrates the contribution from both the visual areas and the motor areas and uses both the visual information on the surrounding environment and the context-independent motor programs to create action plans useful for reaching the target using different effectors (eye, arm, leg). This implies anticipating the consequences of the movement and therefore choosing the most suitable action sequence (e.g., Teixeira et al., 2014).
Based on this background, the present study was aimed at verifying which pattern of brain activity is associated with the prediction and preparation of stimulus-driven actions requiring increasing levels of motor complexity. The motor complexity was manipulated by asking subjects to provide their response by a simple keypress (finger flexion) or by adding additional movements such as long-range reaching movement alone or in combination with a standing stepping movement. Thus, we designed three task conditions (Keypress , Reaching ,Reaching-stepping ) with identical visual and cognitive complexity (50/50 Go/No-go task) but with a different motor engagement. The experiment is based on the general hypothesis that the greater the movement complexity, the greater the need to integrate information of different nature to perform it, so the greater the activation in a specific functional circuit, mainly located in the parietal region, delegated to joint inputs coming both from the occipital visual areas and the frontal/prefrontal motor control areas. Using this experimental design and specific comparison across conditions, it is possible to identify the cortical activity mainly involved in motor control and the ones combining somatosensory and visual information useful for complex movement, involving larger body districts, and implying the correct management of the surrounding space (Piserchia et al., 2017; Whitlock et al., 2014).
The expectation is to identify modulation in amplitude and timing of the BP, pBP, pN, and vN components as a function of the required involvement of premotor, parietal, prefrontal, and visual areas preparatory activity required by the three conditions, respectively. We specifically expect that at the least the early BP should be similar for all movements since the intention to move should be common (e.g., Lau et al., 2004), The late BP should be larger in the simplest task likely requiring less time to be accomplished. For this reason, the pN should be also larger in the simplest task (just requiring a keypress) because the mentioned accelerator/braking system that modulates the response accuracy/speed trade-off (e.g., Di Russo et al., 2017). In this vein, finer movements, requiring inhibition of the rest of the body, should be more controlled by the pN. The reaching preparation should be characterized by the pBP specifically involved in the kind of actions (e.g., Teixeira et al., 2014). Lower limb movements should be dominated by medial PPC activity and visual-related readiness (the vN) for the management of the front space (Berchicci et al., 2020; Piserchia et al., 2017; Whitlock et al., 2014).
Methods