An important feature of our day-to-day existence is the way in which humans attribute causality between their own actions, the actions of others’ and the corresponding sensory consequences. This causal attribution is commonly referred to as the Sense of Agency, which is the understanding that “I” caused an event to occur through my actions. It is still not fully understood the ways in which we attribute agency between actions and effects, and in particular the way in which we distinguish agency for consequences that are self-made or caused by another agent. One-way to indirectly measure sense of agency involve investigations into the phenomenon of intentional binding. In these tasks a voluntary, self-made action is followed in time by a sensory consequence (effect). When asked, individuals perceive the timing of the action and effect as occurring much closer together in time, than when presented in isolation. It remains to be determined whether this phenomenon also occurs when watching others’ execute goal-directed actions. In this thesis I present a series of experiments that use intentional binding as a measure to investigate the cognitive processes and neural correlates involved in the attribution of agency for our own and others’ actions.
In Chapter 2 of this thesis I investigated the intentional binding phenomenon when observing others’ actions using a novel interval estimation paradigm. I found evidence that participants perceived the interval between an action, either self-made or observed, and the corresponding effect as significantly shorter than compared to the interval between two sensory stimuli. These findings suggest that an implicit sense of agency for others’ actions is formed to a similar extent as has been found for our own actions.
The cognitive processes involved in the binding of actions and effects for our own and others’ actions were investigated in Chapters 3 and 4. I used a traditional intentional binding paradigm, which was modified to place the video of the observed action centrally at fixation while participants performed the task. Chapter 3 investigated whether perceived shifts of the action and perceived shifts of the tone occurred to the same extent for our own and others’ actions. It was found that self-made actions resulted in greater backwards shifts of the effect, whereas observing another agents’ actions resulted in shifts of both the action and the effect towards each other. In Chapter 4, I directly investigated the role feed-forward (predictive) and retrospective (postdictive) processes have on action binding in self-made and others’ actions. In this task the expectation of whether a consequent tone occurred after a goal-directed action was manipulated. It was found that for self-made actions, binding was influenced by the expectation that a sensory consequence will follow an action. In contrast, the presence of a sensory consequence, indicative of retrospective processes, influenced the binding for others’ actions. From these chapters I have concluded that differential weightings are used to influence the binding of our own actions, which use more internal predictions, and others’ actions, where external events play an influential role.
In Chapter 5 the neural processes underlying the attribution of agency for our own and others’ actions were investigated during an EEG task. Using an interval estimation paradigm, the neural activity prior to the onset of an action as well as the activity related to the presentation of a sensory event was measured. Decreases in amplitude of an early component of the auditory event related response were found for effects caused by a self-made or observed action compared to a control tone. I concluded that this suppression of activity provides evidence for the role of predictive processes for both our own and others’ actions.
The final experimental chapter, Chapter 6 investigated the anatomical regions involved in the predictive processes involved in binding for our own and others’ actions. Using an fMRI paradigm the expectation that an action would cause a specific consequence was manipulated. It was found that a medial area of the anterior PFC was activated more during tasks involving self-made actions than observed actions. I have concluded that this area is involved in directly matching action-effect expectations to the incoming effect for self-made actions but not for the actions of others’.
In the concluding chapter I combine the findings of the experiments and discuss the implications of their conclusions in relation to sense of agency, predictive models, intentional binding and schizophrenia. I propose that the inference of causality over actions and effects lies on a continuum where agency over self-made actions involves a greater weighting of internal cues; whereas attributing agency over the actions of others’ involves a combination of predictive and external cues. In this way we are able to process sensory events in the world as the consequence of our own or another agents’ behaviour.