Skip to main content

Rethinking Conscious Agency

Periodic Reporting for period 1 - ReConAg (Rethinking Conscious Agency)

Reporting period: 2018-05-01 to 2019-10-31

The Rethinking Conscious Agency project explores the nature and significance of conscious agency along multiple fronts.
First, we study the nature and structure of the conscious experiences agents have in different contexts of action. We are asking what kinds of evidence reports about conscious experience in action should have for the science of action control.
Second, we study models of action control. We are interested in how expertise is built up over time, in how this changes cognitive architecture via influences on memory and concepts, and in what roles conscious experience has to play in action control across differences of context and differences of skill level.
Third, we study the practical and moral significance of conscious agency. We are asking whether consciousness is important for free will, and why people think that it is. We are asking whether models of skilled action can be transferred to philosophical reflection on morally responsible action, and what the implications of different views of moral skill might be for our moral practices.
Progress along all three fronts is important to society for a number of reasons. First, conscious agency plays an intimate role in our conception of ourselves as free, responsible agents, capable of pursuing goals and effecting change in the world. It is important to map the structure of conscious agency to illuminate our self-conception further, and to understand ways this self-conception might change in response to scientific progress. Second, the nature of consciousness is one of the great outstanding scientific problem areas. Progress on the science of consciousness requires collaborative efforts from philosophers as well as mind scientists, and our project is a part of a global network seeking to advance knowledge in this foundational area. Third, moral and legal frameworks make commitments and assumptions to implicit models of agency and action control. It is important for the viability and trustworthiness of these frameworks that these commitments be made explicit and held up to the light of the best available philosophy and science regarding agency and action control.
We conceive of ourselves as contributing to important conversations along all of the main areas of our research. We wish to articulate foundational questions along all three fronts, and to contribute to philosophical and scientific progress by offering clear conceptual frameworks for further investigation into these issues.
The PI of this project, Joshua Shepherd, worked alone for the first year of the project. He conducted foundational research along all three wings of the project, organized workshops at the University of Barcelona (images attached are from our first workshop on Metacognition, in May of 2019), and delivered seven talks across the world.

See a summary on twitter of our first workshop here:

This work has resulted in the publication of two important papers, and the completion of a book manuscript that we believe will be highly influential in the philosophy of mind and action.
The first paper, ‘Intelligent action guidance and the use of mixed representational formats,’ has been published in Synthese. This paper engages a debate about the structure of action control, and brings a number of innovative ideas to bear on this debate. These ideas are about the representational format of the psychological states that combine to help explain and produce intelligent, flexible behavior.
The second paper, ‘Why does the mind wander?’ has been published in Neuroscience of Consciousness. This paper examines a puzzle for accounts of our conscious control, namely, the fact that the mind frequently unintentionally turns away from whatever task the agent is pursing to engage in a more free-form process of mind wandering. Shepherd offers a novel proposal that integrates work on the neuroscience of mind wandering with work on the neuroscience of cognitive control. He offers the cognitive control proposal: mind wandering is a form of non-conscious guidance due to the cognitive control system. Shepherd outlines ways that this proposal could be tested, and why this would be important. Given the prevalence of unintentional mind wandering, and the potential costs to many of doing so in various contexts, a greater understanding of why mind wandering happens, and thus of how we might influence our tendencies to do so, is sorely needed.
The book manuscript is titled The Shape of Agency and it will constitute a major contribution. It is slated to be published by Oxford University Press, probably in 2020. In the book Shepherd articulates a novel conceptual foundation for a number of hot-button issues in the philosophy and science of agency. He offers interlinked accounts of control, non-deviant causation, intentional action, skill, and knowledgeable action. The book thus touches on a wide range of issues, and has direct relevance to those working in the philosophy and science of action.
Recently the first post-doc, Adrian Alsmith, joined the project. Alsmith’s work is crucial in mapping the structure of the representations underlying action control, and the phenomenology of agency. Alsmith is also contributing to the project’s experimental goals by developing a collaboration with scientists working in relevant areas. In his six months on the project, Alsmith has been focussed on two topics concerning the role of the body in structuring perceptually guided actions. The first of these topics concerns the nature and targets of the representations of the body underlying coordinated bodily action. The second topic concerns the extent to which the relative rotation of body parts (specifically the head and torso) on the horizontal plane affects the spatial organisation of perceptual experience. This can be tested by using a “misalignment paradigm,” and Alsmith is developing an experimental protocal with researchers at Birkbeck and Royal Holloway, University of London.
Shepherd’s work has led to three major outputs so far. These outputs progress beyond the state of the art in offering novel proposals regarding {a} the psychological processes underlying intelligent action (output 1), {b} the mechanisms that give rise to unintentional mind wandering (output 2), and {c} the entire conceptual framework via which philosophers understand agency (output 3).

Soon the project’s second post-doc, Wayne Christensen, will join the team. We expect the next years of the project to bring increased collaboration, as well as a wide range of papers on the project’s key themes. Some of the expected results include:

-Offering a viable model of components of ‘sense of agency’
-Offering testable proposals regarding the representational format of motor representation
-Offering models of the place of different kinds of experiences, such as the experiences of effort, boredom, and curiosity, in cognitive control
-Developing an integrative framework that explains how skill is built up in the human brain, the place of different representations – including those that underlie bodily self-consciousness (Alsmith), those that underlie cognitive control (Shepherd), and those that underlie the work of memory (Christensen) – in the exercise of skill, and the epistemic role of conscious experiences for the sciences of skill
-Making theoretical progress on how to think about the functions of consciousness
Speaker at Metacognition Workshop, May 2019