Curious Inferences: Reply to Sun and Firestone on the Dark Room Problem

Seth, Anil K.; Millidge, Beren; Buckley, Christopher L.; Tschantz, Alexander

doi:10.1016/j.tics.2020.05.011

Cited by 16 publications

(15 citation statements)

References 8 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We suggest the answer to the dark room problem is that the value of a generative model depends on how well the model does on average, over the course of the agent's life, at keeping the agent adapted to the dynamical environment in which it is situated. To optimise the value of a generative model it pays to sometimes be curious in ways that lead to temporary increases in error, so long as in doing so the agent makes progress in learning (see Kiverstein, Miller and Rietveld 2019;Van de Cruys, Friston & Clark 2020;Seth et al 2020).…”

Section: Discussionmentioning

confidence: 99%

The Predictive Dynamics of Happiness and Well-Being

2021

View full text Add to dashboard Cite

We offer an account of mental health and well-being using the predictive processing framework (PPF). According to this framework, the difference between mental health and psychopathology can be located in the goodness of the predictive model as a regulator of action. What is crucial for avoiding the rigid patterns of thinking, feeling and acting associated with psychopathology is the regulation of action based on the valence of affective states. In PPF, valence is modelled as error dynamics—the change in prediction errors over time . Our aim in this paper is to show how error dynamics can account for both momentary happiness and longer term well-being. What will emerge is a new neurocomputational framework for making sense of human flourishing.

show abstract

Section: Discussionmentioning

confidence: 99%

The Predictive Dynamics of Happiness and Well-Being

2021

View full text Add to dashboard Cite

show abstract

“…The present perspective may inform controversies surrounding the explanatory scope of PP in cases where utility and accuracy conflict with one another, such as reward-biased perception [73], affect-biased attention [74], and more general issues linked to the Dark Room Problem [75] (cf. [39,76,77]). Future open questions include whether utility-biased expectations differ…”

Section: Discussionmentioning

confidence: 99%

Useful misrepresentation: perception as embodied proactive inference

Martin

Solms

Sterzer

2021

Trends in Neurosciences

View full text Add to dashboard Cite

“…In some cases, this concern has also targeted mAI, although perhaps inappropriately once mAI is understood as purely a theory of low-level motor control after decisions have been made. 3 One prominent example concern is the "dark room problem" (Friston, Thornton, & Clark, 2012;A. K. Seth, Millidge, Buckley, & Tschantz, 2020;Sun & Firestone, 2020;S.…”

Section: Addressing Concerns About the (Apparent) Purely Doxastic Ontology Of Active Inferencementioning

confidence: 99%

“…As the reader may already gather, this concern is somewhat misplaced in the context of dAI (Badcock, Friston, Ramstead, Ploeger, & Hohwy, 2019;A. K. Seth et al, 2020;S.…”

Section: Addressing Concerns About the (Apparent) Purely Doxastic Ontology Of Active Inferencementioning

confidence: 99%

Active inference models do not contradict folk psychology

Smith¹,

Ramstead²,

Kiefer³

2021

Preprint

View full text Add to dashboard Cite

Active inference offers a unified theory of perception, learning, and decision-making at computational and neural levels of description. In this article, we address the worry that active inference may be in tension with folk psychology because it does not include terms for desires (or other conative constructs) at the mathematical level of description. To resolve this concern, we first provide a brief review of the historical progression from predictive coding to active inference, enabling us to distinguish between active inference formulations of motor control (which need not have desires under folk psychology) and active inference formulations of decision processes (which do have desires within folk psychology). We then show that, despite a superficial tension when viewed at the mathematical level, the active inference formalism contains terms that are readily identifiable as encoding both the objects of desire and the strength of desire at the psychological level. We demonstrate this with simple simulations of an active inference agent motivated to leave a dark room for different reasons. Despite their consistency, we further show how active inference may increase the granularity of folk-psychological descriptions by highlighting distinctions between drives to seek information vs. reward – and how it may also offer more precise, quantitative folk-psychological predictions. Finally, we consider how the implicitly conative components of active inference may have partial analogues (i.e., “as if” desires) in other systems describable by the broader free energy principle to which it conforms.

show abstract

Curious Inferences: Reply to Sun and Firestone on the Dark Room Problem

Cited by 16 publications

References 8 publications

The Predictive Dynamics of Happiness and Well-Being

The Predictive Dynamics of Happiness and Well-Being

Useful misrepresentation: perception as embodied proactive inference

Active inference models do not contradict folk psychology

Contact Info

Product

Resources

About