The Sense of Meaning: From Ancient Instincts to Artificial Minds

Annotation 

Alastair Waterman’s The Sense of Meaning: From Ancient Instincts to Artificial Minds offers an interdisciplinary exploration of consciousness, weaving together evolutionary biology, neuroscience, philosophy, and artificial intelligence (AI) to propose a groundbreaking conceptual framework. The book introduces three key concepts that reframe the nature of consciousness, its evolutionary development, and its potential replication in artificial systems. Written for a broad audience, including neuroscientists, evolutionary biologists, philosophers, AI researchers, cognitive psychologists, and anyone intrigued by the fundamental questions of consciousness and its role in biological and technological evolution.

The learning hypothesis posits that consciousness emerged approximately 500 million years ago in early vertebrates as a “sense of meaning”—a functional mechanism that enhances adaptive learning. Consciousness is viewed as an active facilitator of conditioned reflexes, long-term memory, and motivations, enabling organisms to transform sensory input into adaptive behaviour. This hypothesis bridges the philosophical divide between Chalmers’s “hard problem of consciousness” (1995, 1996) and materialist perspectives, arguing that subjective experience serves an evolutionary function by statistically enhancing adaptive traits through learning.

The irritability gradation model traces the evolution of consciousness across five stages, from molecular reflexes in single-celled organisms to the reflective awareness of mammals. This model redefines consciousness as an “additional” form of sensitivity, distinct from the basic irritability of simpler organisms. Waterman suggests that consciousness arose with the development of the forebrain in early vertebrates, such as cartilaginous fish, enabling symbolic processing of stimuli to form conditioned reflexes. This model emphasizes the gradual complexification of consciousness, avoiding anthropocentric biases, and draws on behavioural studies.

Phenomenological neuroplasticity is presented as the biological foundation of consciousness, where dynamic neural reorganization through synaptic, structural, and functional plasticity shapes subjective experience. The author proposes that consciousness emerges from the integration of sensory data and memory in thalamo-cortical systems, creating a “sense of meaning” that guides learning. Neuroplasticity enables the brain to prioritize meaningful stimuli, strengthening adaptive neural connections. Waterman advocates modelling these processes in AI using neuroevolutionary algorithms and neuromorphic architectures to replicate rudimentary forms of subjective experience, raising questions about the feasibility of artificial consciousness.

Mental space-time is introduced as a concept explaining the topological complexity of consciousness, which transcends four-dimensional spacetime. Waterman suggests that consciousness forms an internal reality where neural states create multidimensional topological structures, allowing a single neural state to correspond to multiple mental states. This idea frames consciousness as an adaptive mechanism for navigating complex environments through learning, linking the learning hypothesis and phenomenological neuroplasticity. It proposes that consciousness enables organisms to construct internal models of the world, enhancing adaptability, and could be modelled in AI to create systems with proto-subjectivity.

Consciousness as a reward is an idea of the book, viewing consciousness as an evolutionary mechanism that reinforces learning through intrinsic rewards. Waterman argues that subjective experiences, such as the pleasure of solving a problem or the joy of success, act as “rewards” that motivate learning and consolidate adaptive behaviour. For example, the joy of mastering a new task, like playing a musical instrument, reinforces skills through phenomenological experience, positioning consciousness as an active catalyst for learning rather than a mere byproduct.

The Sense of Meaning Test (SMT), proposed by the author, offers a novel approach to assessing consciousness in biological and artificial systems. Unlike traditional tests like the Turing Test, which focus on external behaviour, the SMT evaluates internal processes related to the “sense of meaning,” measuring behavioural and neural correlates such as enhanced conditioned reflexes or neuroplastic changes. This test is applicable to both animals and AI, providing a framework to detect the presence of consciousness.

The book also explores the prospects and challenges of modelling consciousness in AI, suggesting that architectures mimicking phenomenological neuroplasticity could create adaptive systems capable of rudimentary subjective experience. This opens possibilities for autonomous systems, personalized educational platforms, and immersive virtual realities but raises ethical questions about AI’s moral status, safety, and societal impact. Waterman compares his approach to existing AI paradigms, such as deep learning and neuromorphic computing, highlighting the unique potential of phenomenological neuroplasticity to model internal experience.

By integrating philosophical insights, neurobiological evidence, and evolutionary research, the book offers a testable, interdisciplinary framework for studying consciousness. It challenges traditional theories like global workspace theory and integrated information theory, opening new avenues for research, including the study of neural correlates of consciousness in simple organisms and the development of AI with elements of subjectivity. The work invites readers to rethink the nature of consciousness and its role in shaping a future where biological and artificial minds coexist in a new civilization focused on the pursuit of meaning.