Exploring the internal forward model: action-effect prediction and attention in sensorimotor processing

Abstract

Action-effect predictions are believed to facilitate movement based on its association with sensory objectives and suppress the neurophysiological response to self- versus externally generated stimuli (i.e. sensory attenuation). However, research is needed to explore theorized differences in the use of action-effect prediction based on whether movement is uncued (i.e. volitional) or in response to external cues (i.e. stimulus-driven). While much of the sensory attenuation literature has examined effects involving the auditory N1, evidence is also conflicted regarding this component's sensitivity to action-effect prediction. In this study (n = 64), we explored the influence of action-effect contingency on event-related potentials associated with visually cued and uncued movement, as well as resultant stimuli. Our findings replicate recent evidence demonstrating reduced N1 amplitude for tones produced by stimulus-driven movement. Despite influencing motor preparation, action-effect contingency was not found to affect N1 amplitudes. Instead, we explore electrophysiological markers suggesting that attentional mechanisms may suppress the neurophysiological response to sound produced by stimulus-driven movement. Our findings demonstrate lateralized parieto-occipital activity that coincides with the auditory N1, corresponds to a reduction in its amplitude, and is topographically consistent with documented effects of attentional suppression. These results provide new insights into sensorimotor coordination and potential mechanisms underlying sensory attenuation.

Keywords: Action-effect contingency; event-related potential; motor preparation; sensory attenuation; volition.

Fig. 1

A. Participants pressed a keyboard button with the index finger on either their left or right hand, based on the direction indicated by an arrow at their point of fixation. In motor-stimulus blocks (i.e. trial types indicated in violet text on the top row), each press with one hand elicited a tone with 100% probability. Each press with the other hand had a 50% chance of eliciting a different tone and a 50% chance of triggering a silent audio track to mark the event. In motor blocks (i.e. trial types indicated in orange text on the bottom row), button presses with each hand elicited the silent audio track. Motor and motor-stimulus trial types are denoted “M” and “M-S,” respectively, and will be referred to in this manner henceforth. B. In uncued blocks, participants were presented with a series of white line fragments across four rows. While the outer two rows were adjacent, a gap that was equal in height to the arrow and each line fragment separated the inner two rows. These fragments moved from right to left at a constant rate and were randomly distributed with a density that corresponded to four fragments (i.e. one per row) every three seconds. Participants fixed their gaze on an arrow at the Center of the screen, which was positioned within a small gap in a vertical red (fixation) line. They were instructed to press the required button every two to four seconds, at will and with unpredictable timing. In cued conditions, participants were presented with a series of white (stimulus) lines that moved from right to left at a constant rate. The spacing of these lines was based on the interval between participants’ button presses in the preceding uncued motor-stimulus block. Participants were instructed to press using the hand indicated at the precise moment that each stimulus line intersected with the fixation line. Uncued and cued blocks were matched in terms of the order of button presses, as well as whether each trial elicited a tone or silent audio track.

Fig. 2

A. Grand-averaged recordings at Cz, demonstrating mean amplitude and 95% CI by uncued condition (left) and cued condition (right). B. Within-subject differences in late RP amplitude between 50% and equivalent 100% conditions (i.e. cued or uncued), with mean differences and 95% CIs. C. Topographic voltage maps demonstrating mean amplitude recordings by 50 and 100% conditions, with p-value and Bayes factor representing the contrast effect of probability. D. Difference in grand-averaged recordings at C3 and C4 (i.e. contralateral minus ipsilateral), 95% CIs and LRP topographic voltage maps by cued and uncued condition (i.e. collapsing across probabilities). Note that, for consistency, electrodes were inverted along the sagittal plane for selected conditions by counterbalancing group. This was done such that topographic maps demonstrate lateralized effects as if each trial had involved a button press with the right hand. To remove activity not lateralized relative to the effector hand, unadjusted grand-averages (i.e. from all participants) were subtracted from unadjusted averages for each counterbalancing group prior to collation in the manner described. This had the effect of removing non-lateralized components, as well as unrelated lateralized activity (e.g. activity associated with visual attention) from topographic maps. E. Difference in grand-averaged voltage recordings at C3 and C4 (i.e. contralateral minus ipsilateral) for cued and uncued variants by probability condition, including 100% [M-S] (top), 50% [M-S] (middle) and 50% [M] (bottom). F. Topographic voltage maps by cued and uncued condition, representing mean voltage recordings at latencies corresponding to LEP (top) and LPP (bottom). Note that the same adjustments were applied based on counterbalancing group as described for panel D.

Fig. 3

A. Difference in grand-averaged voltage recordings at P5 and P6 (i.e. contralateral minus ipsilateral) for cued and uncued variants by probability condition, including 100% [M-S] (top), 50% [M-S] (middle) and 50% [M] (bottom). B. Difference in grand-averaged voltage recordings at P5 and P6 (i.e. contralateral minus ipsilateral) and 95% CIs by cued and uncued condition (i.e. collapsing across probabilities; top). Difference between collapsed cued and uncued conditions (bottom). C. Legend for panels A, E, and F. D. Topographic voltage maps representing mean amplitude recordings by cued and uncued condition, with p-values and Bayes factors representing the main effect of cueing at latencies and electrodes corresponding to LEP (top) and LPP (bottom). Note that these were adjusted in the same manner as described for Fig. 2D. E. Within-subject differences in LEP amplitude between cued and uncued condition, with mean difference and 95% CIs. F. Within-subject differences in LPP amplitude between cued and uncued condition, with mean difference and 95% CIs.

Fig. 4

A. Motor-corrected auditory evoked potentials (top), representing pooled mean amplitudes at Fz, FCz and Cz by condition, as well as 95% CIs. Difference between collapsed cued and uncued conditions (bottom), demonstrating sustained attenuation of cued conditions between approximately 100 and 200 ms post-stimulus B. Within-subject contrasts of N1 amplitude with mean difference and 95% CIs, as well as p-values representing the results of paired samples Student’s t-tests. C. Topographic voltage maps for N1 components with corresponding condition labels and legend for panels A, B, and D. D. Mean voltages and 95% CIs for N1 amplitudes by condition, as well as results reflecting the main effect of cueing on N1 amplitude.