Is the experience of agency necessarily retrospective? A predictive approach

Abstract

In the present work, I shall defend that although there is undoubtedly a retrospective component as a source of the experience of agency (that is, of goal-directed action), that does not imply that it may have another component that is determined by an actual link in the causal chain that leads from thought to action. Moreover, I shall defend that such a link may actually exist, and I will attempt a brief description of what it could be. Hence, in Section 1 I will begin by examining Daniel Wegner’s Illusion argument as a token argument of the perspective that the experience of agency is retrospective inference, and argue that he does not demonstrate that the experience of agency is not causally connected to intentional action (sections 1.1 and 1.2). In Section 2 I will review the evidence gathered in the research surrounding the intentional binding effect, which will serve as positive proof that a retrospective component is not sufficient to account for the sense of agency.

Section 3 will be dedicated to a brief description of the Predictive Processing framework, and in 3.1 I will analyse the Comparator model for the experience of agency, which supports retrospective inference and is based of Predictive Processing. Finally, in Section 4 I shall discuss Daniel Dennett’s brilliant account on qualia as predictive phenomena (4.1), and in section 4.2 I will explore whether Dennett’s ideas may shed light on the nature of the sense of agency.

As an introductory note, I need to clarify my uses of three expressions: 1) the sense of authorship I the sensation that one is the cause of the action that one performs; 2) by the sense of agency I will denote the hypothesised feeling one might experience when an action has been indeed caused by a conscious thought or intention; and 3) under the term experience of agency I will consider every experience about agency that has a retrospective component, regardless of whether a predictive factor may also be involved. Additionally, it may be necessary to make clear that I do not intend to defend the possibility of the conscious will’s causal efficacy, nor even its existence. My contention is that conscious thought may still have a causal role in the production of behaviour, even in an entirely physical world.