The 2019 summer session...
June 5th - July 10th
Predictive Processing, Embodiment and Conscious Experience
The brain as a multi-level prediction engine...
This summer we will be exploring the emerging image of the brain as a multi-level prediction machine. According to such accounts, incoming sensory signals are processed relative to a multi-level, multi-area neuronal organization that seeks to predict the evolving flow of sensory information. These, mostly unconscious, predictions prepare us to deal rapidly and efficiently with the stream of signals coming from the world. If the sensory signal is as expected, we see and hear the things that we have already started to prepare to see and hear, or launch behaviours that we have already started to organize. But if all is not as expected, then a distinctive signal results: a so-called ‘prediction-error’ signal. These signals, calculated in every area and at every level of neuronal processing, highlight what the prediction engine - or 'generative model' - got wrong, and invite the brain to try again armed with specific information concerning current errors. Brains like this are forever trying to guess the shape and evolution of the current sensory signal, using stored knowledge about the world.
The vision of the brain as a predictive engine is increasingly being applied to the investigation of a wide variety of mental, behavioural and emotional phenomenon. In this six-week course we will meet a number of prominent researchers working on this framework today (see below), and hear about what they think about this vision and its applicability to a variety of topics such as embodiment, experience of pain, the nature of consciousness, self control and addiction, affect and attention, and more...
Some good background watching and reading...
Andy Clark's Brains Blog series is a good place to start, it can be found HERE
Andy Clark's lecture series entitled Only Predict? On the Nature, Scope and Limits of Predictive Processing, found HERE
Anil Seth's article in AEON on Predictive Processing and consciousness, can be found HERE
Anil Seth's TED talk on How Your Brain Hallucinates Your Reality, found HERE
WIRED's article on Karl Friston and Predictive Processing and AI, found HERE
And for books - Andy Clark's Surfing Uncertainty (HERE) and Jakob Hohwy's The Predictive Mind (HERE)
This summer we will be exploring the emerging image of the brain as a multi-level prediction machine. According to such accounts, incoming sensory signals are processed relative to a multi-level, multi-area neuronal organization that seeks to predict the evolving flow of sensory information. These, mostly unconscious, predictions prepare us to deal rapidly and efficiently with the stream of signals coming from the world. If the sensory signal is as expected, we see and hear the things that we have already started to prepare to see and hear, or launch behaviours that we have already started to organize. But if all is not as expected, then a distinctive signal results: a so-called ‘prediction-error’ signal. These signals, calculated in every area and at every level of neuronal processing, highlight what the prediction engine - or 'generative model' - got wrong, and invite the brain to try again armed with specific information concerning current errors. Brains like this are forever trying to guess the shape and evolution of the current sensory signal, using stored knowledge about the world.
The vision of the brain as a predictive engine is increasingly being applied to the investigation of a wide variety of mental, behavioural and emotional phenomenon. In this six-week course we will meet a number of prominent researchers working on this framework today (see below), and hear about what they think about this vision and its applicability to a variety of topics such as embodiment, experience of pain, the nature of consciousness, self control and addiction, affect and attention, and more...
Some good background watching and reading...
Andy Clark's Brains Blog series is a good place to start, it can be found HERE
Andy Clark's lecture series entitled Only Predict? On the Nature, Scope and Limits of Predictive Processing, found HERE
Anil Seth's article in AEON on Predictive Processing and consciousness, can be found HERE
Anil Seth's TED talk on How Your Brain Hallucinates Your Reality, found HERE
WIRED's article on Karl Friston and Predictive Processing and AI, found HERE
And for books - Andy Clark's Surfing Uncertainty (HERE) and Jakob Hohwy's The Predictive Mind (HERE)
The speakers...
Kathryn Nave
University of Edinburgh Chris Burr, Abby Tabor & Max Jones University of Oxford, Bath & Bristol Julian Kiverstein University of Amsterdam Jelle Bruineberg University of Amsterdam Jose Araya Instituto de Filosofía y Ciencias de la Complejidad Madeleine Ransom & Sina Fazelpour University of British Colombia |
The schedule...
Week 1: Predictive Processing and Enactivism
Kathryn Nave (University of Edinburgh)
June 5th 6pm - 8pm (BST)
Predictive Processing is commonly traced back to Hermann von Helmholtz’s account of perception as a process of inference (Clark, 2013; Hohwy, 2013). Seen through this lens the brain's predictive operations are understood as dedicated towards constructing an accurate model of its distal environment - achieved by using prior knowledge to disambiguate between hypotheses that are typically underdetermined by current proximal evidence alone. Action is important, but instrumental, its purpose being merely to test hypothesis in service of the ultimate aim of constructing this accurate model.
Approaching PP from an enactivist standpoint, however, flips this picture around (Bruineberg, 2017). Here, neither action nor perception are taken as primary but rather, as advocated by Hurley (2001), as constitutively interdependent – locked in an ongoing process of circular causality, tending towards mind-body-world attunement.
Integrating PP and enactivism, I argue, is mutually advantageous. By understanding the processes of model construction and inference as derivative of a more fundamental imperatives to adaptivity and control we gain a more biologically plausible PP. A PP that incorporates the normativity that is the hallmark of living systems. In turn PP can aid the enactivist by 1) clarifying how to operationalise the notion of sensorimotor mastery and 2) provide a framework for scaling up from simple autopoietic living systems to the complex internal and external loops of cognition and adaptation that are characteristic of conscious minds like us.
Suggested Readings:
Kathryn Nave (University of Edinburgh)
June 5th 6pm - 8pm (BST)
Predictive Processing is commonly traced back to Hermann von Helmholtz’s account of perception as a process of inference (Clark, 2013; Hohwy, 2013). Seen through this lens the brain's predictive operations are understood as dedicated towards constructing an accurate model of its distal environment - achieved by using prior knowledge to disambiguate between hypotheses that are typically underdetermined by current proximal evidence alone. Action is important, but instrumental, its purpose being merely to test hypothesis in service of the ultimate aim of constructing this accurate model.
Approaching PP from an enactivist standpoint, however, flips this picture around (Bruineberg, 2017). Here, neither action nor perception are taken as primary but rather, as advocated by Hurley (2001), as constitutively interdependent – locked in an ongoing process of circular causality, tending towards mind-body-world attunement.
Integrating PP and enactivism, I argue, is mutually advantageous. By understanding the processes of model construction and inference as derivative of a more fundamental imperatives to adaptivity and control we gain a more biologically plausible PP. A PP that incorporates the normativity that is the hallmark of living systems. In turn PP can aid the enactivist by 1) clarifying how to operationalise the notion of sensorimotor mastery and 2) provide a framework for scaling up from simple autopoietic living systems to the complex internal and external loops of cognition and adaptation that are characteristic of conscious minds like us.
Suggested Readings:
- Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and brain sciences, 36(3), 181-204 HERE
- Ward, D., Silverman, D., & Villalobos, M. (2017). Introduction: The varieties of enactivism. Topoi, 36(3), 365-375 HERE
- Thompson, E. (2017). The Enactive Approach. The Brains Blog HERE
- Bruineberg, J., Kiverstein, J., & Rietveld, E. (2018). The anticipating brain is not a scientist: the free-energy principle from an ecological-enactive perspective. Synthese, 195(6), 2417-2444 HERE
Week 2: Painful Possibilities
Chris Burr (University of Oxford), Abby Tabor (University of Bath) and Max Jones (University of Bristol)
June 12th 6pm - 8pm (BST)
In this lecture for the IMINDS 2019 session, we explore the nature of pain experience from the perspective of embodied predictive processing.
Pain experience is often cited by philosophers as a paradigmatic case of an irreducibly subjective experience. There is a sense in which we all know what pain experience is. Despite this, there are many discrepancies between the folk conception of pain and the findings from pain science. The aim of this lecture is to provide an overview of some of the ways that an embodied account of PP can begin to resolve these discrepancies.
Building on our work in the philosophy of predictive processing (Burr & Jones 2016) and the science of embodied pain experience (Tabor et al. 2017a, 2017b; Tabor & Burr 2019), we have been developing a novel approach to pain experience that changes the traditional understanding of both its function and the nature of the processes that give rise to it.
Pain is often seen as representing bodily damage that has already occurred. However, we argue that this is both conceptually flawed and inconsistent with a range of evidence from the scientific study of pain. Instead, we see pain experience as essentially anticipatory, action-oriented, and emergent from active exploration of one’s own body. We suggest that, given these features, the emerging predictive processing framework, combined with various insights from embodied cognition, is best-placed to explain pain experience, as well as the underlying mechanisms that give rise to it.
Suggested Readings:
Chris Burr (University of Oxford), Abby Tabor (University of Bath) and Max Jones (University of Bristol)
June 12th 6pm - 8pm (BST)
In this lecture for the IMINDS 2019 session, we explore the nature of pain experience from the perspective of embodied predictive processing.
Pain experience is often cited by philosophers as a paradigmatic case of an irreducibly subjective experience. There is a sense in which we all know what pain experience is. Despite this, there are many discrepancies between the folk conception of pain and the findings from pain science. The aim of this lecture is to provide an overview of some of the ways that an embodied account of PP can begin to resolve these discrepancies.
Building on our work in the philosophy of predictive processing (Burr & Jones 2016) and the science of embodied pain experience (Tabor et al. 2017a, 2017b; Tabor & Burr 2019), we have been developing a novel approach to pain experience that changes the traditional understanding of both its function and the nature of the processes that give rise to it.
Pain is often seen as representing bodily damage that has already occurred. However, we argue that this is both conceptually flawed and inconsistent with a range of evidence from the scientific study of pain. Instead, we see pain experience as essentially anticipatory, action-oriented, and emergent from active exploration of one’s own body. We suggest that, given these features, the emerging predictive processing framework, combined with various insights from embodied cognition, is best-placed to explain pain experience, as well as the underlying mechanisms that give rise to it.
Suggested Readings:
- Burr, C., & Jones, M. (2016). The body as laboratory: Prediction-error minimization, embodiment, and representation. Philosophical Psychology, 29(4), 586-600 HERE
- Tabor, A., Thacker, M. A., Moseley, G. L., & Körding, K. P. (2017a). Pain: a statistical account. PLoS computational biology, 13(1), e1005142 HERE
- Tabor, A., Keogh, E., & Eccleston, C. (2017b). Embodied pain—negotiating the boundaries of possible action. Pain, 158(6), 1007-1011 HERE
- Tabor, A., & Burr, C. (2019). Bayesian Learning Models of Pain: A Call to Action. Current Opinion in Behavioral Sciences, 26, 54-61 HERE
Week 3: Determining the Boundaries of the Mind: A Markov Blanket Story
Julian Kiverstein (University of Amsterdam)
June 19th 6pm - 8pm (BST)
Abstract coming soon.
Suggested Readings:
Julian Kiverstein (University of Amsterdam)
June 19th 6pm - 8pm (BST)
Abstract coming soon.
Suggested Readings:
Week 4: Predictive Processing and Synthetic Philosophy
Jelle Bruineberg (University of Amsterdam)
June 26th 6pm - 8pm (BST)
In combining Bayesian inference with natural selection, predictive-processing and its systems-theoretic cousin, the free-energy principle, have become the perfect tools for what Eric Schliesser calls a “synthetic philosophy”: a style of philosophy that brings together insights from the special sciences in order to provide a coherent account of complex systems like the human mind. Over the last decade, several of such coherent accounts have been developed, each being based on predictive-processing. These accounts range from being neo-Cartesian to radial embodied, and, needless to say, are not all mutually compatible. I will introduce a number of conceptual issues such as Edelman’s distinction between selectionist and instructionist learning and between epistemic and motivational accounts of inference to make sense of the radical differences between these frameworks.
Suggested Readings:
Jelle Bruineberg (University of Amsterdam)
June 26th 6pm - 8pm (BST)
In combining Bayesian inference with natural selection, predictive-processing and its systems-theoretic cousin, the free-energy principle, have become the perfect tools for what Eric Schliesser calls a “synthetic philosophy”: a style of philosophy that brings together insights from the special sciences in order to provide a coherent account of complex systems like the human mind. Over the last decade, several of such coherent accounts have been developed, each being based on predictive-processing. These accounts range from being neo-Cartesian to radial embodied, and, needless to say, are not all mutually compatible. I will introduce a number of conceptual issues such as Edelman’s distinction between selectionist and instructionist learning and between epistemic and motivational accounts of inference to make sense of the radical differences between these frameworks.
Suggested Readings:
Week 5: Predictive Processing and Self Control
Jose Araya (Instituto de Filosofía y Ciencias de la Complejidad)
June 26th 6pm - 8pm (BST)
Abstract coming soon...
Suggested Readings:
Jose Araya (Instituto de Filosofía y Ciencias de la Complejidad)
June 26th 6pm - 8pm (BST)
Abstract coming soon...
Suggested Readings:
- Sripada (2014). How is willpower possible? The puzzle of synchronic self-control and the divided mind. Nous, 48(1), 41–74 HERE
Week 6: Attention in the Predictive Mind
Madeleine Ransom (University of British Colombia) & Sina Fazelpour (University of British Colombia)
July 10th 6pm - 8pm (BST)
It has recently become popular to suggest that cognition can be explained as a process of Bayesian prediction error minimization. Some advocates of this view propose that attention should be understood as the optimization of expected precisions in the prediction-error signal (Clark, 2013, 2016; Feldman & Friston, 2010; Hohwy, 2012, 2013). This proposal successfully accounts for several attention-related phenomena. We claim that it cannot account for all of them, since there are certain forms of voluntary attention that it cannot accommodate. We therefore suggest that, although the theory of Bayesian prediction error minimization introduces some powerful tools for the explanation of mental phenomena, its advocates have been wrong to claim that Bayesian prediction error minimization is ‘all the brain ever does’.
Suggested Readings:
Madeleine Ransom (University of British Colombia) & Sina Fazelpour (University of British Colombia)
July 10th 6pm - 8pm (BST)
It has recently become popular to suggest that cognition can be explained as a process of Bayesian prediction error minimization. Some advocates of this view propose that attention should be understood as the optimization of expected precisions in the prediction-error signal (Clark, 2013, 2016; Feldman & Friston, 2010; Hohwy, 2012, 2013). This proposal successfully accounts for several attention-related phenomena. We claim that it cannot account for all of them, since there are certain forms of voluntary attention that it cannot accommodate. We therefore suggest that, although the theory of Bayesian prediction error minimization introduces some powerful tools for the explanation of mental phenomena, its advocates have been wrong to claim that Bayesian prediction error minimization is ‘all the brain ever does’.
Suggested Readings:
- Ransom, M., Fazelpour, S. & Mole, C. (2017). Attention in the predictive mind. Consciousness and cognition, 47, 99-112 HERE
- Clark (2017). Predictions, precision, and agentive attention. Consciousness and cognition, 56, 115-119 HERE
BONUS WEEK: Dissolving the Self: Active Inference, Psychedelics, and Ego-dissolution
George Deane (University of Sussex)
July 17th 6pm - 8pm (BST)
Psychedelic drugs are known to induce radical disruptions in self-experience, including a ‘dissolving’ of the usual sense of self. Recent research has garnered increasing interest in the therapeutic value of such ‘peak’ experiences. In the following paper, ego-dissolution is accounted for within an active inference framework as a collapse in the ‘temporal thickness’ of an agent’s deep temporal model, as a result of lowered precision on high-level priors in the psychedelic state. The argument is composed of three moves: first, a view of the self–model is proposed as arising within a temporally deep generative model of an embodied organism navigating an affordance landscape in the service of allostasis. Next, a view of the action of psychedelics as lowering the precision on high-level priors within the generative model is unpacked in terms of a high Bayesian learning rate. Finally, the relaxation of high-level priors is argued to cause a ‘collapse’ in the temporal thickness of the generative model, resulting in a collapse in the self-model. This account has implications for our understanding of ordinary experience of self and disruptions of self-experience in psychosis and autism. The philosophical, theoretical and therapeutic implications of this account of ego-dissolution are touched upon.
Suggested Readings:
George Deane (University of Sussex)
July 17th 6pm - 8pm (BST)
Psychedelic drugs are known to induce radical disruptions in self-experience, including a ‘dissolving’ of the usual sense of self. Recent research has garnered increasing interest in the therapeutic value of such ‘peak’ experiences. In the following paper, ego-dissolution is accounted for within an active inference framework as a collapse in the ‘temporal thickness’ of an agent’s deep temporal model, as a result of lowered precision on high-level priors in the psychedelic state. The argument is composed of three moves: first, a view of the self–model is proposed as arising within a temporally deep generative model of an embodied organism navigating an affordance landscape in the service of allostasis. Next, a view of the action of psychedelics as lowering the precision on high-level priors within the generative model is unpacked in terms of a high Bayesian learning rate. Finally, the relaxation of high-level priors is argued to cause a ‘collapse’ in the temporal thickness of the generative model, resulting in a collapse in the self-model. This account has implications for our understanding of ordinary experience of self and disruptions of self-experience in psychosis and autism. The philosophical, theoretical and therapeutic implications of this account of ego-dissolution are touched upon.
Suggested Readings:
- Carhart-Harris, R. L., & Friston, K. J. (2019). REBUS and the Anarchic Brain: Toward a Unified Model of the Brain Action of Psychedelics. Pharmacological Reviews, 71(3), 316-344. HERE.
- Friston, K. (2018). Am I self-conscious?(or does self-organization entail self-consciousness?). Frontiers in psychology, 9. HERE.
- Pezzulo, G., & Cisek, P. (2016). Navigating the affordance landscape: feedback control as a process model of behavior and cognition. Trends in cognitive sciences, 20(6), 414-424. HERE.