Matteo Grasso
I am a postdoctoral research scholar at UW-Madison (Tononi lab) working on philosophy and computational neuroscience of consciousness.
Supervisors: Giulio Tononi , Anna Marmodoro , and Mario De Caro
Supervisors: Giulio Tononi , Anna Marmodoro , and Mario De Caro
less
Uploads
In order to address these issues, we propose to adopt Predictive Processing (PP), an emerging theoretical framework for cognitive science that aims to unify perception, action and cognition under a single mechanism (Clark 2013, in press; Hohwy 2013). The core idea is that brains are predictive machines with a hierarchical structure, continuously in the business of predicting their own sensory inputs. Applied to the study of dreaming (Hobson & Friston 2012, Hobson, Hong & Friston 2014), PP highlights the continuity and the differences of dreaming with other waking mental states, grounding them on the very same cognitive architecture. However, PP doesn't precisely account for the phenomenological aspects and how we come to experience anything in the first place (the “hard problem” of consciousness).
We argue that if a second theory proves to be efficient in explaining the phenomenal aspect of dreaming, then PP could be integrated with it for a more comprehensive explanation of cognition. Our theory of choice is Integrated Information Theory (IIT) (Tononi 2012, Oizumi, Albantakis & Tononi 2014). According to IIT the quantity of consciousness of a system is equal to the amount of integrated information generated by its elements, while the quality of experience is defined in relation to maximally irreducible conceptual structures (MICS), i.e. constellation of concepts in the “qualia space”. Phenomenal consciousness is hence defined on the basis of the informational relationships generated by the system's repertoire of internal states, which characterizes conscious experience in both the waking and dreaming state.
After the introduction on PP, IIT and the specific issues surrounding dreaming, we procede to explain how PP and IIT can be merged in order to explain the cognitive mechanism behind the emergence of dream phenomenology, focusing on the conceptual similarity between the two theories' vocabularies. Finally, we illustrate a few critical points and some implications of a more general merging of the theories.
In order to address these issues, we propose to adopt Predictive Processing (PP), an emerging theoretical framework for cognitive science that aims to unify perception, action and cognition under a single mechanism (Clark 2013, in press; Hohwy 2013). The core idea is that brains are predictive machines with a hierarchical structure, continuously in the business of predicting their own sensory inputs. Applied to the study of dreaming (Hobson & Friston 2012, Hobson, Hong & Friston 2014), PP highlights the continuity and the differences of dreaming with other waking mental states, grounding them on the very same cognitive architecture. However, PP doesn't precisely account for the phenomenological aspects and how we come to experience anything in the first place (the “hard problem” of consciousness).
We argue that if a second theory proves to be efficient in explaining the phenomenal aspect of dreaming, then PP could be integrated with it for a more comprehensive explanation of cognition. Our theory of choice is Integrated Information Theory (IIT) (Tononi 2012, Oizumi, Albantakis & Tononi 2014). According to IIT the quantity of consciousness of a system is equal to the amount of integrated information generated by its elements, while the quality of experience is defined in relation to maximally irreducible conceptual structures (MICS), i.e. constellation of concepts in the “qualia space”. Phenomenal consciousness is hence defined on the basis of the informational relationships generated by the system's repertoire of internal states, which characterizes conscious experience in both the waking and dreaming state.
After the introduction on PP, IIT and the specific issues surrounding dreaming, we procede to explain how PP and IIT can be merged in order to explain the cognitive mechanism behind the emergence of dream phenomenology, focusing on the conceptual similarity between the two theories' vocabularies. Finally, we illustrate a few critical points and some implications of a more general merging of the theories.
In order to address these problems we propose to integrate PP with a leading theory of consciousness: Integrated Information Theory (IIT), as proposed by Tononi (2012) and Oizumi, Albantakis & Tononi (2014). IIT proposes an explicit account of phenomenal consciousness, defined as integrated information. According to IIT the quantity of consciousness of a system is equal to the amount of integrated information (ΦMAX) generated by its elements, while the quality of experience is defined in relation to maximally irreducible conceptual structures (MICS), i.e. constellation of concepts in the “qualia space”. Phenomenal consciousness is hence defined on the basis of the informational relationships generated by the system's repertoire of internal states, which characterizes conscious experience in both the waking and dreaming state.
After the introduction of PP, IIT and the specific issues surrounding dreaming, we procede to explain how PP and IIT can be merged in order to explain the cognitive mechanism behind the emergence of dream phenomenology. We highlight the conceptual similarity in the two theories' vocabularies and argue that, if IIT proves to be efficient in explaining the phenomenal aspect of dreaming, then PP could be integrated with it for a more comprehensive explanation of cognition. We then move on to illustrate a few critical points in this merging operation. In conclusion, we speculate that if dreaming consciousness is a positive case study for the integration of PP and IIT, it can also be a springboard for a more general merging of the theories, in which IIT is complementary to PP in explaining consciousness as well as cognition.
Innanzitutto si definiscono le condizioni di un'azione libera (possibilità alternative e autodeterminazione) e si considera il rapporto fra libertà e determinismo illustrando pregi e difetti delle posizioni più rappresentate: le teorie della famiglia libertaria, il compatibilismo, lo scetticismo. Questa indagine conduce ad una conclusione scettica per cui il libero arbitrio risulta incoerente sia con una concezione deterministica che con una concezione indeterministica della realtà.
Viene discusso poi il rapporto fra aspetti fenomenici (soggettivi) della libertà e descrizione neuroscientifica: nello specifico viene indagato il rapporto fra decisione di compiere un'azione motoria e preparazione dell'azione a livello corticale. Alcuni recenti esperimenti presi in esame (sulla scia di quelli condotti da Benjamin Libet) mostrano che la consapevolezza cosciente della decisione di agire compare “in ritardo” rispetto all'attivazione corticale preparatoria al movimento. Grazie a questo ritardo sembra possibile prevedere in anticipo il momento in cui il soggetto agisce e l'esito della decisione. A ciò segue un'analisi di alcuni stati psicopatologici rilevanti (come la depersonalizzazione) in cui il senso dell'agency è compromesso.
Alla luce di tutto ciò viene illustrata l'ipotesi (sostenuta da D. Wegner e D. Dennett) che la volontà non rivesta un ruolo causale nel determinare le azioni, ma sia una sensazione frutto di un'autoattribuzione a posteriori del soggetto stesso: sarebbe dunque il prodotto di un'inferenza a posteriori rispetto ai correlati neuronali delle decisioni ed un meccanismo di attribuzione i cui deficit provocano una perdita del senso dell'agency.
Nelle conclusioni vengono delineate le teorie più recenti sulla volontà cosciente in relazione al rapporto fra filosofia della mente, neuroscienze, fenomenologia e psicologia.