Hume and Cognitive Architectures - Hume's Conception of the Mind Through the Lens of Cognitive Modeling

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23330%2F25%3A43976573" target="_blank" >RIV/49777513:23330/25:43976573 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Hume and Cognitive Architectures - Hume's Conception of the Mind Through the Lens of Cognitive Modeling

Popis výsledku v původním jazyce

This paper examines the connection between David Hume’s conception of the mind, cognitivescience, and artificial intelligence. It proposes a thesis that Hume’s account of mind and itsfaculties, as presented in A Treatise of Human Nature (Hume, D. 2007), can be coherentlyreinterpreted as a form of cognitive architecture - and further as a viable blueprint forpotential computational implementation in a system of artificial cognition. Within theproposed interpretation, the epistemological and skeptical features of Hume’s thinking aresuppressed, while emphasis is put on cognitive-scientific and cognitive-psychological aspectsof his work.The first part of the paper discusses the notion of cognitive architectures, their typology, andtheir defining features. (Ye, P. &amp; Wang, T. 2018, Kotseruba, I. &amp; Tsotsos, J. 2020, Gudivada,V. 2018). Particular attention is given to two crucial components of cognition - representationsand operations (Schlicht, T., Smortchkova, K. &amp; Dołęga, K. 2020, Thagard, P. 2012,Frankish, K. &amp; Ramsey, W. 2012) and to three key features of artificial cognitive systems asdefined by Fodor and Pylyshyn, i.e. compositionality, systematicity, and productivity (Fodor,J. &amp; Pylyshyn, Z. 1988, Fodor, J. 1983).The second part of the paper examines Hume’s account of the mind, mental faculties, andprinciples of association, while arguing that:(1) Humean perceptions can be understood as a species of a non-linguistic mentalrepresentations.(2) Hume’s distinction between simple perceptions (qua mental atoms) and complexperceptions (qua composite mental entities), together with Hume’s conception of associativeprinciples and imaginative faculty, provides sufficient ground for the ascription ofcompositionality to the Humean conception of the mind.(3) To this conception of mind also, productivity can be attributed in virtue of theimagination’s capacity to create novel perceptions by recombining mental content alreadyavailable to it.(4) Humean associative principles also imply the presence of a weak (or improper) type ofsystematicity.(Cf. Fodor, J. 2003, Buckner, C. 2023).The last section of the paper presents two examples of cognitive architectures inspired by thehistory of philosophy, offering further context and support for the case of Humean conceptionof the mind qua cognitive architecture. The first example is C. Buckner’s DoGMA, anarchitecture based on principles of moderate empiricism that incorporates mental faculties in aform of domain-general cognitive mechanisms (See Buckner, C. 2023). The second is R.Evans’s Aperception engine, derived from Kant’s a priori psychology in the Critique of PureReason (Kant, I. 1998), an architecture that was recently implemented in a real-world systemof artificial cognition (Evans, R. 2022, cf. Schlicht, T. 2022).Taken together, these considerations should support the relevance of the thesis that Hume’sconception of mind can be coherently interpreted as a cognitive architecture with possiblecomputational implementation, especially when similar strategies as those employed byBuckner and Evans would be used. They also highlight the relevance of Hume’s thinking forcontemporary research in cognitive sciences and artificial intelligence, showing how hisaccount of mind aligns with contemporary computational models of cognition.Event: (En)active Minds and AIVenue: PortoDate: 05.09.2025 - 05.09.2025

Název v anglickém jazyce

Hume and Cognitive Architectures - Hume's Conception of the Mind Through the Lens of Cognitive Modeling

Popis výsledku anglicky

This paper examines the connection between David Hume’s conception of the mind, cognitivescience, and artificial intelligence. It proposes a thesis that Hume’s account of mind and itsfaculties, as presented in A Treatise of Human Nature (Hume, D. 2007), can be coherentlyreinterpreted as a form of cognitive architecture - and further as a viable blueprint forpotential computational implementation in a system of artificial cognition. Within theproposed interpretation, the epistemological and skeptical features of Hume’s thinking aresuppressed, while emphasis is put on cognitive-scientific and cognitive-psychological aspectsof his work.The first part of the paper discusses the notion of cognitive architectures, their typology, andtheir defining features. (Ye, P. &amp; Wang, T. 2018, Kotseruba, I. &amp; Tsotsos, J. 2020, Gudivada,V. 2018). Particular attention is given to two crucial components of cognition - representationsand operations (Schlicht, T., Smortchkova, K. &amp; Dołęga, K. 2020, Thagard, P. 2012,Frankish, K. &amp; Ramsey, W. 2012) and to three key features of artificial cognitive systems asdefined by Fodor and Pylyshyn, i.e. compositionality, systematicity, and productivity (Fodor,J. &amp; Pylyshyn, Z. 1988, Fodor, J. 1983).The second part of the paper examines Hume’s account of the mind, mental faculties, andprinciples of association, while arguing that:(1) Humean perceptions can be understood as a species of a non-linguistic mentalrepresentations.(2) Hume’s distinction between simple perceptions (qua mental atoms) and complexperceptions (qua composite mental entities), together with Hume’s conception of associativeprinciples and imaginative faculty, provides sufficient ground for the ascription ofcompositionality to the Humean conception of the mind.(3) To this conception of mind also, productivity can be attributed in virtue of theimagination’s capacity to create novel perceptions by recombining mental content alreadyavailable to it.(4) Humean associative principles also imply the presence of a weak (or improper) type ofsystematicity.(Cf. Fodor, J. 2003, Buckner, C. 2023).The last section of the paper presents two examples of cognitive architectures inspired by thehistory of philosophy, offering further context and support for the case of Humean conceptionof the mind qua cognitive architecture. The first example is C. Buckner’s DoGMA, anarchitecture based on principles of moderate empiricism that incorporates mental faculties in aform of domain-general cognitive mechanisms (See Buckner, C. 2023). The second is R.Evans’s Aperception engine, derived from Kant’s a priori psychology in the Critique of PureReason (Kant, I. 1998), an architecture that was recently implemented in a real-world systemof artificial cognition (Evans, R. 2022, cf. Schlicht, T. 2022).Taken together, these considerations should support the relevance of the thesis that Hume’sconception of mind can be coherently interpreted as a cognitive architecture with possiblecomputational implementation, especially when similar strategies as those employed byBuckner and Evans would be used. They also highlight the relevance of Hume’s thinking forcontemporary research in cognitive sciences and artificial intelligence, showing how hisaccount of mind aligns with contemporary computational models of cognition.Event: (En)active Minds and AIVenue: PortoDate: 05.09.2025 - 05.09.2025

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

60301 - Philosophy, History and Philosophy of science and technology

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů