[Paper Review] Types of Cognition and its Implications for future High-Level Cognitive Machines

Camilo Miguel Signorelli|arXiv (Cornell University)|Jun 5, 2017

Neural dynamics and brain function40 references3 citations

TL;DR

This paper explores the relationship between high-level cognitive processes and biological neural systems, identifying paradoxes in designing future artificial cognitive machines—particularly the tension between achieving human-like cognition and preserving machine-like accuracy. It argues that creating advanced cognitive AI may require sacrificing deterministic precision, challenging traditional machine design principles.

ABSTRACT

This work summarizes part of current knowledge on High-level Cognitive process and its relation with biological hardware. Thus, it is possible to identify some paradoxes which could impact the development of future technologies and artificial intelligence: we may make a High-level Cognitive Machine, sacrificing the principal attribute of a machine, its accuracy.

Motivation & Objective

• To analyze the interplay between high-level cognition and biological neural substrates in natural intelligence.
• To identify fundamental paradoxes in designing artificial cognitive machines that emulate human-level cognition.
• To explore the implications of these paradoxes for future AI systems, especially regarding reliability and accuracy.
• To challenge the assumption that machine accuracy is a prerequisite for advanced cognitive functionality.
• To propose a reevaluation of core design principles in artificial intelligence to accommodate cognitive complexity.

Proposed method

• Synthesizes existing knowledge on high-level cognitive processes and their neural underpinnings in biological systems.
• Analyzes the structural and functional parallels between biological cognition and artificial intelligence architectures.
• Identifies conceptual contradictions in applying machine-like precision to systems designed for flexible, adaptive cognition.
• Uses insights from neuroscience and AI to frame the trade-off between cognitive flexibility and computational accuracy.
• Draws on the AAAI Spring Symposium Series to contextualize the work within current research in science of intelligence.
• Applies a comparative framework to assess how biological cognition achieves high-level functions without strict determinism.

Experimental results

Research questions

• RQ1 How do high-level cognitive processes in biological systems differ from traditional machine computation?
• RQ2 What paradoxes arise when attempting to design artificial cognitive machines that mimic human-level cognition?
• RQ3 To what extent can machine accuracy be sacrificed in favor of cognitive flexibility and adaptability?
• RQ4 What are the implications of biological cognition's inherent uncertainty for artificial intelligence system design?
• RQ5 How can future AI systems balance reliability with the capacity for creative, context-sensitive reasoning?

Key findings

• The paper identifies a core paradox: achieving high-level cognition may require accepting imprecision, contradicting the foundational principle of machine accuracy.
• Biological cognition operates with inherent uncertainty and approximation, yet achieves robust, adaptive behavior—challenging the assumption that accuracy is essential for intelligence.
• Artificial systems aiming for human-level cognition may need to adopt probabilistic, non-deterministic architectures, even at the cost of predictability.
• The integration of cognitive flexibility into AI may necessitate a redefinition of 'correctness' in machine behavior.
• Current AI design paradigms, which prioritize precision and determinism, may be fundamentally incompatible with the emergence of true high-level cognition.
• The study suggests that future cognitive machines may need to embrace 'controlled error' as a feature, not a flaw, to achieve human-like reasoning.