Quantum Agency Conflicts with No-Cloning, Preventing World-Model Creation and Reliable Action

The fundamental requirements for agency, the capacity to act purposefully, remain a profound question, and recent work by Emily C. Adlam, Kelvin J. McQueen, and Mordecai Waegell from the Institute for Quantum Studies at Chapman University investigates whether this capacity can arise from purely quantum systems. The team demonstrates that building a world-model and evaluating potential actions, both essential components of agency, fundamentally conflict with the laws of quantum mechanics, specifically the no-cloning theorem which prevents the perfect copying of unknown quantum states. Their analysis reveals that even approximate copying strategies lack the necessary precision for viable agency within a purely quantum framework, and furthermore, the inherent linearity of quantum dynamics prevents reliable action selection. These findings establish clear constraints on the physical basis of agency, offering new insights into how classical agents emerge and challenging existing assumptions about the nature of free will and consciousness.

Scientists investigated the physical requirements for agency, specifically whether a purely quantum system could exhibit agential behaviour, defined by the ability to construct a world-model, evaluate action consequences, and reliably implement optimal actions. The research demonstrates a fundamental conflict between these agency conditions and the no-cloning theorem, which prohibits the copying of unknown quantum states, a critical requirement for both world-model construction and deliberation. Attempts to circumvent this limitation, such as extracting environmental states or assuming access to multiple identical copies, were found to be insufficient for sustaining agential behaviour. Analysis of quantum agency circuits revealed that the linearity of quantum dynamics prevents a unitary operation from reliably selecting and implementing the ‘best’ action based on simulated outcomes.

Microtubules, Quantum Computation, and Conscious Agency

This extensive research explores the intersection of consciousness, quantum mechanics, and the concept of agency, critically examining theories like Orch OR and Integrated Information Theory (IIT). The central question concerns how subjective experience arises, and whether quantum phenomena play a role. Orch OR proposes that consciousness emerges from quantum computations occurring within microtubules inside brain neurons, governed by a process called objective reduction, a form of quantum wave function collapse linked to gravity. While intriguing, Orch OR faces challenges, including maintaining quantum coherence within the warm, wet environment of the brain.

IIT proposes that consciousness is directly related to the amount of integrated information a system possesses, measured by a value called Φ, with higher Φ equating to higher consciousness. IIT leans towards panpsychism, suggesting consciousness isn’t limited to brains but can exist in any system with non-zero Φ. Understanding agency also requires considering the neural mechanisms underlying the sense of control over one’s actions. Efference copy, a copy of motor commands sent to the brain, is crucial for this feeling, as is the ability to imagine alternative scenarios, known as counterfactual thinking.

Formal models of decision-making and the concept of rational agents are also relevant, but agents need world-models, internal representations of the environment, to make informed decisions. Robust agents learn causal world-models, allowing them to predict outcomes and guide their actions. The possibility that quantum effects, specifically involving nuclear spins, might play a role in cognitive processes is also being explored, alongside the potential for quantum computing to enhance machine learning.

Quantum Agency Conflicts with No-Cloning Theorem

This research demonstrates fundamental limitations to achieving agency within purely quantum systems, revealing that constructing a world-model, evaluating actions, and reliably executing choices are incompatible with the laws of quantum mechanics. Specifically, the no-cloning theorem prevents the necessary copying of quantum states required for both world-model creation and deliberation, effectively blocking these essential steps for agency. While approximate cloning strategies and alternative approaches were explored, the team found these insufficient to overcome the core limitations and sustain reliable agential behaviour. These findings have significant implications, establishing principled constraints on the physical basis of agency and clarifying how agents might emerge within a quantum universe.

The work suggests that classical resources, such as a preferred basis for information copying, are necessary preconditions for agency, and highlights a technological limitation for quantum computers attempting to simulate agential behaviour without substantial classical control. Furthermore, the research challenges existing quantum theories of agency, free will, and consciousness, requiring these proposals to explicitly account for the origin of necessary classical resources and justify the role of quantum components. The authors acknowledge that a complete theory of mind and agency remains an open question, but this work narrows the possibilities by definitively excluding a purely quantum basis for these phenomena.

Quantum Agency Limited by No-Cloning Theorem

Scientists established that a purely quantum system cannot fulfill all three conditions for agency, highlighting a critical constraint on the physical basis of agency itself. The research demonstrates that constructing a world-model, evaluating actions, and reliably executing choices are fundamentally incompatible with the no-cloning theorem, which prohibits the copying of unknown quantum states. Attempts to circumvent this limitation, such as extracting environmental states or assuming access to multiple identical copies, proved insufficient for sustaining agential behaviour. These findings have four key consequences: they place principled constraints on the physical possibilities for agency, narrowing the scope for potential theories of mind; they clarify how agents could emerge within a quantum universe, suggesting that the emergence of classical features, such as a preferred basis for information copying, is essential for agential behaviour; they establish a technological limitation for quantum computers, demonstrating that simulating agency requires significant classical external control; and they challenge existing quantum theories of agency, free will, and consciousness, suggesting that these concepts may require classical foundations. The authors acknowledge that a complete theory of mind and agency remains an open question, but this work narrows the possibilities by definitively excluding a purely quantum basis for these phenomena.