Final Report Summary - HUVAC (Neurophysiological and functional mechanisms of human voluntary action control)
Human action serves two complementary purposes. On the one hand, actions are meant to achieve effects in the environment. On the other hand, people act as a consequence of external events. Research on action control has focused primarily on the reactive type of action. The current project, by contrast, picked up new paradigmatic and theoretical developments emerged in the domain of action control, all stressing the importance of internal factors. It investigated the question as to what are the functional and neurophysiological underpinnings of voluntary actions, notably the anticipation of action effects, along 6 axes: 1. The first axis studied the neuronal and functional underpinnings of two phenomena related to action effect anticipation: sensory attenuation (the observation that self-generated stimuli are perceptually attenuated compared to the same stimuli when generated externally) and intentional binding. These phenomena are normally attributed to forward action models, such that when action prediction is consistent with changes in our environment, our experience of these effects is altered. Although much progress has been made in recent years in understanding sensory attenuation and intentional binding, a number of important questions regarding the nature of the mechanisms involved remain unanswered. We systematically investigated and reviewed the role of several processes that might be the cause of the two phenomena at hand, namely temporal prediction, temporal control, identity prediction, and motor prediction. By isolating the individual processes that have previously been contrasted and incorporating these experiments with research in the related fields of temporal attention and stimulus expectation, we assessed the degree to which existing and our own data provide evidence for the role of forward action models in these phenomena. 2. A second line of research studied the effects of attention upon motor predictive processes. We conceived a paradigm, using EEG (electroencephalography), to manipulate attention and motor prediction orthogonally. In this paradigm, tones could be presented at the attended or unattended ear, as well as being congruent or incongruent with prior action-effect learning. The results suggest attention and motor prediction to be not opposing processes but to operate in tandem to modulate prediction. 3. Furthermore, we investigated whether sensory attenuation is modulated by prior beliefs of action authorship. Our results suggest that sensory attenuation is not only inducing authorship believe, it is also a consequence of prior belief about the causal link between an action and a sensory change in the environment. 4. We modelled potential interactions between intentional and stimulus-driven action preparation. We presented and tested a two-stage model in which an intentional and a stimulus-driven system interact only in the second stage. In the first stage of the model, the intentional and stimulus-driven processes race independently to reach a transition threshold between the two stages. This model is consistent with some physiological results that indicate that both parallel and interactive processing take place between intentional and stimulus-driven information.
5. We conceived a model of action effect prediction based on the notion that, according to ideomotor theory, humans do not only predict the sensory effects that their actions have on the environment, they even selected actions with respect to desired sensory effects by means of an inverse model. We developed a new account of how this prediction might be implemented in the brain. This account supposes that action selection involves the pre-activation of perceptual representations of learnt action-effects. Our account can explain a number of phenomena concerning the perception of anticipated sensory action effects. 6. We studied the temporal dynamics of action-effect anticipation by tracing the time course of its perceptual consequences. To do so, we presented motion stimuli that were congruent or incongruent to previously learned action effects at different intervals before or after action execution. We observed higher sensitivity (d') to motion discrimination in congruent vs. incongruent trials only when stimuli were presented from about 220 ms before the action to 280 ms after the action. The temporal dynamics of our effect suggest that action-effect prediction modulates perception at later stages of motor preparation.
5. We conceived a model of action effect prediction based on the notion that, according to ideomotor theory, humans do not only predict the sensory effects that their actions have on the environment, they even selected actions with respect to desired sensory effects by means of an inverse model. We developed a new account of how this prediction might be implemented in the brain. This account supposes that action selection involves the pre-activation of perceptual representations of learnt action-effects. Our account can explain a number of phenomena concerning the perception of anticipated sensory action effects. 6. We studied the temporal dynamics of action-effect anticipation by tracing the time course of its perceptual consequences. To do so, we presented motion stimuli that were congruent or incongruent to previously learned action effects at different intervals before or after action execution. We observed higher sensitivity (d') to motion discrimination in congruent vs. incongruent trials only when stimuli were presented from about 220 ms before the action to 280 ms after the action. The temporal dynamics of our effect suggest that action-effect prediction modulates perception at later stages of motor preparation.