Dissociable neural oscillatory mechanisms underlying unconscious priming of externally and intentionally initiated inhibition

Highlights

• No-Go prime induced a prolonged "Go" response times compared to Go prime.

• The theta-band oscillation may be involved in unconscious priming of externally triggered inhibition.

• The alpha/low-beta band oscillation may be implicated in unconscious priming of intentionally generated inhibition.

Abstract

Externally and intentionally initiated inhibitory processes, which are fundamental for human action control, can be unconsciously launched. However, the neural oscillatory mechanisms underlying unconscious priming of externally and intentionally generated inhibition remain unclear. This study aimed to explore this issue by extracting oscillatory power dynamics from electroencephalographic data with participants performing an unconscious version of the Go/No-Go/Choose task involving subliminally presented primes. The participants presented prolonged response times upon being instructed or intentionally deciding to commit a “Go” response following a No-Go prime compared with those following a Go prime. This indicates that unconscious inhibitory processes can be externally and intentionally initiated. Time-frequency analysis indicated increased theta band oscillatory power on the forced Go response following a No-Go prime compared with that following a Go prime. Contrastingly, there was pronounced alpha/low-beta band oscillatory power on the free-choice Go response following a No-Go prime compared with that following a Go prime. Moreover, there was a positive correlation of theta and alpha/low-beta band oscillations with human behavior performance related to the two distinct unconscious inhibitory processes. Our findings delineate dissociable neural oscillatory mechanisms underlying the unconscious priming of externally and intentionally initiated inhibition. Moreover, they might provide complementary neural oscillatory evidence supporting the discrepancy between instructed and voluntary human action control.

Introduction

Suppressing externally and intentionally initiated impulsive movements is a hallmark of human action control (Hommel and Wiers, 2017). This contributes to physical survival (e.g., braking immediately when a cyclist swerves into a lane to avoid a traffic accident), as well as social interdependency (e.g., stopping the urge to rebuke a work partner to refrain from interpersonal problems) (Aron et al., 2007; Aron and Poldrack, 2006; Haggard, 2008). Increasing evidence indicates that the neural substrates of externally and intentionally initiated inhibition are substantially distinct (Filevich et al., 2012; Schel et al., 2014a). Specifically, externally initiated inhibition, which is termed as “response inhibition”, denotes the active suppression of imperative movements induced by exogenous No-Go or Stop signals (Aron et al., 2014; Bokura et al., 2001; Verbruggen and Logan, 2008). Moreover, it is associated with a lateral premotor frontal network, including the inferior frontal cortex (IFC) and the pre-supplementary motor area (pre-SMA) (Blasi et al., 2006; Chikazoe et al., 2009; Simmonds et al., 2008; van Gaal et al., 2010). Contrastingly, intentionally initiated inhibition, which is termed as “intentional inhibition”, involves voluntarily withholding pre-potent actions at the last moment (Brass and Haggard, 2007, Brass and Haggard, 2008; Kuhn et al., 2009). Further, it is related to a medial frontal system, including the dorsal fronto-median cortex, pre-SMA, rostral cingulated zone, and anterior insular cortex (Brass and Haggard, 2007, Brass and Haggard, 2010; Kuhn and Brass, 2009; Kuhn et al., 2009; Schel et al., 2014b).

However, recent electrophysiological research has reported an association of the N2 component over the prefrontal cortex (PFC) with externally and intentionally generated inhibition (Parkinson and Haggard, 2015). The inconsistency between high temporal- and spatial-resolution electrophysiological (e.g., event-related potentials [ERPs]) and neuroanatomical findings (e.g., functional magnetic resonance imaging [fMRI]), respectively, indicates the need to identify effective neural markers for distinguishing externally and intentionally initiated inhibition. Analyzing neural oscillatory dynamics is a potential candidate. Compared with ERPs based on evoked cortical responses to task-relevant stimuli, neural oscillatory dynamics, which are considered to be characterized by induced cortical responses, are thought to present more extensive information and delineate more realistic brain activity patterns (Cohen, 2017; Engel and Fries, 2010). Furthermore, compared with fMRI, analyzing neural oscillatory dynamics allows for the detection of transient cognitive process variations (Walsh et al., 2010). Neuroscience studies on the neural oscillatory features of externally or intentionally initiated inhibition have yielded convincing evidence suggesting that the different inhibitory processes may involve distinct neural oscillatory mechanisms. For example, studies on externally triggered inhibitory processes have reported more pronounced theta band oscillatory power (4–8 Hz) over the PFC in response to No-Go or Stop signals (compared with the Go signal) (Barry, 2009; Harper et al., 2014; Yamanaka and Yamamoto, 2010). This is consistent with the notion that theta band oscillation is indicative of the need for high-level control processes and can be employed to implement such control (Cavanagh & Frank, 2014). Contrastingly, heightened event-related synchronization/desynchronization (ERS/ERD) of alpha/low-beta band oscillation (8–24 Hz) occurs when an individual intentionally cancels the keypress compared with when the keypress is committed in a time-judgment task, suggesting that alpha/low-beta band oscillation may be involved in intentionally generated inhibition (Walsh et al., 2010).

Notably, these studies explored the inhibitory processes only when task-relevant stimuli appeared above the subjective consciousness threshold, which is consistent with the conventional perspective that inhibitory processes require consciousness (Dehaene and Naccache, 2001; Eimer and Schlaghecken, 2003). However, this has been refuted by recent findings showing that unconsciously presented primes can activate inhibitory processes without accessing conscious awareness (Hughes et al., 2009; Parkinson and Haggard, 2014; van Gaal et al., 2008; van Gaal et al., 2010). For example, van Gaal et al., 2008, van Gaal et al., 2010 reported that the “unconscious priming of externally initiated inhibition” using an unconscious version of Go/No-Go tasks. Specifically, individuals presented a more prolonged response on the (forced) Go target following a No-Go prime compared with following a Go prime. Furthermore, the neural substrates of conscious and unconscious externally initiated inhibition involved similar spatiotemporal patterns of brain activity (e.g., the IFC, pre-SMA, and P3 component). Additionally, Parkinson and Haggard (2014) reported “unconscious priming of intentionally initiated inhibition” using an unconscious version of the Go/No-Go/Choose task. Here, participants showed a slower response when deciding to commit a “Go” response and an increased likelihood of withholding the keypress following a No-Go prime compared with a Go prime. However, the neural mechanisms underlying the unconscious priming of intentionally initiated inhibition remain unclear. Moreover, it is unknown whether there is a discrepancy between the unconscious priming of externally and intentionally initiated inhibition.

To address these issues, we developed an unconscious version of the Go/No-Go/Choose task adapted from Parkinson and Haggard (2014) and van Gaal et al. (2010). This comprised of forced and free-choice Go/No-Go targets following unconsciously presented Go, Neutral, and No-Go primes. We aimed to delineate the neural oscillatory dynamics of unconscious priming of externally and intentionally initiated inhibition using EEG technology. Based on previous findings (Barry, 2009; Harper et al., 2014; Walsh et al., 2010; Yamanaka and Yamamoto, 2010), we hypothesized that theta band oscillation may represent prime-induced modulation of externally initiated inhibition, whereas alpha/low-beta band oscillation may represent prime-induced modulation of intentionally generated inhibition.