The Role of Interaction and Embodied Cognition in Computational Ontology

The Role of Interaction and Embodied Cognition in Computational Ontology

by Marco Schorlemmer

Formal & Physical Sciences AI — News & Applications UYComputer scienceUYQArtificial intelligence (AI)
🎙 Marco Schorlemmer 👥 2K 📅 January 18, 2019 ⏱ 41 min 👁 140 📄 expert opinion 🧭 2026-08-18
Available in: English (current) Français

Keywords

ontologyembodied cognitionconceptual blendinginteractionsemantic alignment

Summary

Dr. Marco Schorlemmer’s plenary talk challenges two assumptions in computational ontology: that categorization is best captured by set theory and symbolic logic, and that successful interaction requires a pre-agreed shared ontology. He argues that meaning is embodied and dynamic, grounded in sensory-motor experience and image schemas, and that interaction itself creates meaning. He illustrates the first point with the Buddhist monk riddle, showing how conceptual blending offers a more intuitive solution than formal axiomatization. He then presents a model for conceptual blending using category theory, developed in the COINVENT project. For the second point, he discusses how meaning is interaction-specific, citing Wittgenstein’s language games, and presents a finite-state automaton model for semantic alignment through interaction, developed in the ESSENCE project. He concludes by outlining the WeNet project, which aims to develop diversity-aware AI that empowers human interactions rather than constraining them within fixed conceptualizations.

144 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights by questioning mainstream assumptions in ontology engineering and offering alternative perspectives grounded in cognitive science. The argumentation is coherent and well-structured, using concrete examples like the monk riddle and bargaining interaction to illustrate abstract concepts. The speaker draws on established theories (Lakoff & Johnson, Wittgenstein) and his own research projects, lending credibility. However, the presentation is more conceptual than empirical, and the proposed models are simplified demonstrations rather than fully validated systems.

86 words

Title / Content Match

The title accurately reflects the content, which focuses on the role of interaction and embodied cognition in computational ontology.

Quality & Reliability

7/10

The talk is given by a recognized researcher in AI, presenting a well-argued critique of classical ontology approaches and proposing alternatives based on embodied cognition and interaction. The arguments are supported by references to established theories (e.g., Lakoff & Johnson, Wittgenstein) and the speaker's own research projects. However, the talk is an opinion piece rather than a systematic review or empirical study, and the proposed methods are illustrated with simple examples rather than validated at scale.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • COINVENT project — EU-funded project on computational conceptual blending
  • ESSENCE network — Marie Curie training network on evolution of shared semantics
  • WeNet project — EU-funded project on diversity-aware AI

Concurring Sources

Contribution & Novelties

The talk offers a novel perspective by integrating embodied cognition and interaction into computational ontology, challenging the traditional static view. It proposes concrete models, such as category theory for conceptual blending and finite-state automata for semantic alignment, which are innovative in the field. The emphasis on diversity and dynamic meaning creation points to future research directions.

Pour aller plus loin :

89 words

Radar Profile

The radar profile shows high scores in quality of information and technical level, reflecting the speaker's expertise and the depth of the content. The quantity of information is moderate, as the talk is a conceptual overview rather than a detailed exposition. The global reliability is solid, though the talk is opinion-based.

Reliability 7/10

💬 No comments were provided for analysis.