In a universe that never collapses, a measurement can leave behind two observers who remember being the same person. Quantum mechanics describes the branching of their physical records. Whether either future observer is uniquely the person who existed before the split is a different problem, one that reaches into the nature of identity, causal structure, and experience itself.
Imagine that you are preparing to measure a quantum system. A detector will register one of two outcomes, and you will remember whichever result appears on its screen. Before the experiment, you know the possible readings, the quantum state of the apparatus, and the probabilities predicted by the Born rule. You also know what it will feel like to await a measurement, watch a number appear, and remember having watched it. Ordinary laboratory practice treats the last of these facts as unremarkable: you perform an experiment, observe a result, and continue living as the person who performed it. Yet one of the most serious interpretations of quantum mechanics makes that apparent continuity philosophically complicated.
In the Everett or many-worlds approach, the Schrödinger equation applies to the detector, the laboratory, and the observer just as it applies to the particle. Measurement need not terminate in a fundamental physical collapse. An initial superposition becomes correlated with distinct detector readings and distinct records in the observer. Decoherence makes the resulting macroscopic alternatives largely incapable of interfering with one another. Each future observer remembers entering the same laboratory; each sees a definite reading; and neither has access to the other's records by ordinary means. The universal quantum state contains the correlations that give each observer a world with a coherent history. [1][2][3]
There is a familiar way to describe this situation: two versions of you now exist. It is memorable, but philosophically imprecise. What makes either observer you? What makes them observers rather than merely complicated physical configurations? Does the continuity of memory establish continuity of subjectivity, and can the same earlier subject continue into two distinct futures? These questions do not disappear when one adopts the mathematics of quantum mechanics. The attempt to answer them may be more revealing than another argument about whether parallel worlds are real.
A measurement that leaves the observer inside the equation
Hugh Everett's 1957 relative-state formulation begins with a methodological demand that remains radical: no physical observer should have to stand outside the theory that describes the universe. A laboratory instrument can be modeled quantum mechanically; so can the scientist who later reads the instrument. Everett's move was to retain unitary dynamics, allowing observer records to become correlated with the systems they measure. A world with definite remembered outcomes appears relative to those records, even though the encompassing state is a superposition. [1]
For a simple two-outcome experiment, consider a quantum system initially in a normalized state and an observer ready to read it:
Here the states and are possible values of the measured system, represents an appropriate physical record state of the observer, and describes the initial environmental degrees of freedom included in the model. An idealized measurement and subsequent environmental correlation can evolve this into
Each term contains a consistent relation between particle, memory, and environment. The observer state records one result; records the other. The equation does not assume that consciousness causes the measurement, nor does it claim that a human brain must maintain coherent quantum superpositions of distinct experiences. The relevant observer states are macroscopic physical record states, which can be treated effectively as classical within their decohered sectors. The quantum description is global; the records inside each sector are definite relative to the corresponding correlated states.
The language of 'worlds' is not a literal instruction for finding a sharply bounded moment at which the universe splits into exactly two countable copies. In contemporary Everettian accounts, branching is an emergent, approximate description of a continuously evolving quantum state. The stability of certain record states and the suppression of interference are physical facts; the number and identity of branches may depend on the level of description. This matters when we ask how many observers exist, because the answer cannot simply be read off from the number of terms in whichever mathematical expansion happens to be written on the page. [2][3]
Where the classical observer comes from
The mathematics above does not by itself explain why our memories behave as stable, mutually consistent records rather than as coherent mixtures of incompatible readings. That part of the account requires decoherence. A detector becomes entangled with an immense environment of scattered photons, vibrating atoms, emitted radiation, and other degrees of freedom. Environmental states correlated with different macroscopic outcomes can rapidly become almost orthogonal, so that
When the environmental degrees of freedom are not accessible, a reduced description of system and observer loses the interference terms between the two recorded alternatives. Schematically, for the idealized states used above,
This explains why local experimental predictions take the form of an effectively classical mixture. It does not, by itself, establish that only one outcome exists fundamentally; that further step belongs to an interpretation or modified dynamics. A global superposition and a locally decohered mixture can be compatible descriptions of different aspects of the same experiment. [2][4]
Wojciech Zurek's work on quantum Darwinism adds a particularly interesting dimension. The environment is not merely a place where interference disappears. It can store many redundant imprints of certain stable, or pointer, states. Multiple observers who interrogate different environmental fragments can therefore recover consistent information about a system without each having to disturb it directly. This gives an account of why physical events become publicly recordable and why many observers can agree about a world. However, the existence of many mutually reinforcing records is not equivalent to the existence of a conscious subject. The environment can preserve a reading even when nobody is looking at it. [4][5]
This creates an important distinction. Quantum mechanics can help explain the emergence of stable records and apparently definite experiences of outcomes, conditional on observers existing and functioning as record-forming systems. But the theory of decoherence does not on its own explain why a given record-forming process should possess a first-person point of view. The difference is easy to miss because the word 'observer' is routinely used in both physics and philosophy. In a physics calculation it may mean any interacting apparatus. In the philosophy of consciousness it names something far harder to identify: a locus for which there is something it is like to undergo a state.
What survives when a person has two futures?
The question of identity under branching predates quantum physics. Derek Parfit's work on personal identity considered thought experiments involving psychological continuity and fission: cases in which one person's mental organization has two equally strong continuations. If each continuation preserves the earlier person's memories, projects, character, and causal history, it seems arbitrary to declare that only one is the true successor. Yet ordinary numerical identity is one-to-one. If future persons A and B are distinct, they cannot both be numerically identical to the same earlier person in the unrestricted classical sense without collapsing their distinctness. [6]
Everettian branching gives that problem a precise physical setting. Before decoherence, the observer has a common past. Afterward, branch-relative records diverge. Each successor may correctly say, in an everyday sense, that they remember conducting the experiment. Neither successor needs to be a counterfeit. But the fact that memory is duplicated cannot determine a unique privileged continuation, and quantum dynamics does not label a single future observer with a metaphysical marker saying 'the original consciousness'.
A more productive picture may be to treat the earlier observer as a temporally extended physical process whose possible continuations are not necessarily unique. The process has a causal past; it can possess several later causal descendants. If identity is defined purely through continuity of functioning and organization, branching can yield a one-to-many continuation relation. Yet it would be a further assumption to infer that personal identity just is this continuation relation. Parfit's insight is not that the traditional question of identity is trivially solved, but that we may have been asking it to do philosophical work better performed by a richer account of psychological and causal connectedness. [6]
The same problem becomes acute for concern about the future. Suppose two successors experience very different outcomes. If both are continuations, whom should the earlier person care about? Everettian attempts to reconstruct probability often appeal to rational choice over branch-relative outcomes or to self-locating uncertainty. Sebens and Carroll, for instance, developed an account in which an observer can be uncertain which post-branch location describes them before learning the recorded outcome. The squared-amplitude weights enter their proposed treatment of credence. Those proposals are substantive philosophical arguments, not settled consequences of the bare claim that many worlds exist; branch counting alone is not a reliable route to the Born rule. [3][7][8]
For an initial state with amplitudes , the familiar weight is
These weights determine operational quantum statistics and underpin familiar probability assignments. They do not, without additional premises, measure the amount of subjective experience in a branch. A low-weight observer is not thereby shown to experience a diminished or incomplete consciousness, just as multiplying a term in the universal state by a coefficient does not itself tell us what it is like to live through that term. The relationship between physical measure, rational expectation, and moral concern remains philosophical territory even when the physics is exceptionally well understood.
From memory to the identity of a subject
Thomas Nagel's question about what it is like to be a bat exposed a gap that becomes sharper, rather than softer, in the Everettian setting. An external physical description may specify the dynamics of an organism with extraordinary precision and still leave a question about the character of its experience. In a branching universe we encounter a second gap: even if a theory describes how memories and internal dynamics are copied into separate branches, what determines whether those continuations are subjects, and in what sense they are continuations of the same subject? The first question concerns subjectivity; the second concerns persistence and individuation. A satisfactory theory of consciousness must be careful not to silently identify them. [9]
One approach is to look for an organized causal process instead of a privileged atom, a particular spatial position, or a snapshot of neural activity. The motivation is straightforward. A brain changes continuously at the level of molecular constituents and neural firing patterns, yet an ordinary episode of experience has temporal organization. Memory, perception, action, and recurrent interaction are processes spread over time. If a subject has a physical basis, the basis might be better characterized by how a system makes differences to its own future and integrates information across time than by the exact inventory of matter present at one instant.
This is close to the central concern of Beyond ϕ: Invariant Causal Process Structures as the Substrate of Subjectivity, a computational and philosophical exploration published on Wolfram Community in September 2026. Its proposal, called ICPS, investigates whether candidates for subjecthood should be understood through temporally extended causal processes and structural features stable under legitimate changes of description. It asks how one can identify processes, their boundaries, persistence, and irreducibility without presupposing that a single number automatically reports consciousness. The work contains executable models and visualizations, but it does not establish that measuring persistence proves the presence of experience. That remains the philosophical bridge to be examined. [10]
The importance of invariance can be stated without deciding the metaphysics in advance. A physical system may be represented at different temporal resolutions, in different coordinates, or with different choices of descriptive variables. Many changes preserve the system's relevant causal relations; others erase precisely the information needed to distinguish one process from another. A candidate subject should not appear and disappear merely because a researcher renamed variables or drew the same network in another orientation. But the theory also cannot be completely indifferent to scale, boundary, or causal intervention, because these changes can alter which physical process actually exists and what it can do.
As an illustrative notation rather than an established consciousness equation, suppose represents a temporally extended causal organization and is an admissible redescription. An identity-relevant structural property could be required to obey
The hard work is hidden in the phrase admissible redescription. It is too easy to define the transformations so broadly that meaningful differences vanish, or so narrowly that the supposed invariant is merely a disguised choice of coordinates. In a quantum setting the challenge is greater still. A decomposition of a Hilbert space into subsystems, a choice of coarse-grained observables, and the physically preferred structure selected through decoherence do not always coincide with arbitrary mathematical factorizations. A physical process theory must articulate its admissible transformations in relation to interventions, dynamics, and observable consequences. [2][10][11]
Can a single causal past support more than one subject?
Now apply this style of reasoning to Everettian measurement. Before an event, an observer's recent causal history is shared. After the measurement, the two successor organizations contain different outcome records and rapidly become embedded in distinct decohered environmental histories. We might sketch their relation as
Here denotes a putative causally individuated process, not a literal quantum branch projector or a consciousness meter. If the two successors preserve the organization required by some account of subjecthood, each might qualify as a successor subject. They have a common causal ancestry without being one continuing numerically unique system after their histories diverge. On that picture, asking 'which branch gets the real me?' presupposes a unique identity relation that the physical situation may not possess.
But three limitations must be kept distinct. First, a causal-descendant relation does not by itself show that its endpoints are conscious. Second, two highly similar structures are not necessarily one physical subject: distinct instantiations can have different causal environments and independent futures. Third, identical memories are evidence of shared information, not evidence of an occult channel along which first-person awareness travels. An account of observers should be able to say what exactly is duplicated, what is conserved, and what is irreversibly different, without smuggling a migrating consciousness into the mathematical notation.
A further subtlety concerns time. Human experience is not obviously made of instantaneous points. There are ongoing perceptual processes, integration windows, memory formation, and action-selection dynamics. A theory that assigns a subject to an instant-by-instant graph might mistake a representation of a process for the process itself. On the other hand, extending the analysis too far in time can blur genuinely separate experiences or distinct causal organizations. A physically meaningful grain and temporal duration must be argued for and, where possible, empirically constrained. ICPS's focus on process identity and persistence is relevant here precisely because the continuity problem survives even when the history is not branching. [10]
What integrated information can, and cannot, settle
Integrated Information Theory approaches consciousness through a more explicit identification of phenomenological and physical structure. Its fourth major formulation aims to characterize a system's intrinsically irreducible cause–effect power. It supplies mathematical operations for determining candidate substrates, their causal distinctions and relations, and their integrated-information structure. This is more specific than the general claim that complex brains process information. Yet it is also a theory whose connection between formal postulates and phenomenal experience is contested, especially when applied beyond the neural systems used to motivate it. [12]
In the Everettian context, IIT introduces questions that are not answered merely by selecting a branch. If a candidate substrate is defined by its intrinsic causal powers, should those powers be assessed relative to a particular decohered physical history or to the universal state? When the same causal architecture appears in two decohered successors, do they instantiate distinct complexes? How should the relevant system boundaries be chosen when a memory-bearing brain is entangled with parts of its environment? Existing approaches to quantum integrated information and extensions of IIT provide different proposals, but no universally accepted quantum consciousness criterion answers all of these questions.
An invariance-centered process approach such as ICPS can sharpen the problem of individuating physical candidates and relating them across changes in description and time. IIT can sharpen the problem of how irreducible cause–effect structure might be formally specified. Neither approach has established that these structures exhaust phenomenal consciousness. It matters whether a formal result is a proven theorem within an assumed model, a measured property of a physical process, or evidence that the process is an experiencing subject. Moving among these claims requires substantive arguments that should be visible rather than hidden in terminology. [10][12]
Is quantum consciousness necessary?
Many-worlds and quantum consciousness are often fused in popular discussion because both appear to involve an observer and the nature of reality. The connection is far weaker than the vocabulary suggests. Everettian mechanics is a claim about how quantum states evolve and how observers should be included in a physical description. It does not require the brain to be a fault-tolerant quantum computer, and it does not make consciousness the agent of collapse. Conversely, one can investigate the physical basis of experience while remaining agnostic about whether the correct interpretation is Everettian, Bohmian, collapse-based, or another consistent account of quantum mechanics.
The Penrose–Hameroff Orch OR program represents a distinct and much stronger proposal: biologically organized coherent processes in neuronal microtubules are hypothesized to terminate in objective reductions associated with moments of consciousness, with the reduction mechanism connected to gravitational considerations. Its proponents have developed the idea at length, and critics have challenged both the physical decoherence times and the proposed bridge to phenomenal experience. Max Tegmark famously estimated rapid decoherence for selected proposed brain degrees of freedom, while Hameroff and collaborators disputed the modeling assumptions and their relevance to Orch OR. Those exchanges show why a serious quantum theory of consciousness needs explicit dynamics, feasible biological timescales, and discriminating observations. [13][14]
A process-structural account has no obligation to identify consciousness with long-lived microscopic superposition. Quantum theory may constrain the ontology in which biological processes exist without making mesoscopic neural computation depend upon experimentally detectable entanglement for its ordinary operation. Similarly, decoherence does not imply that biological processes cease to be quantum in the fundamental physical sense. It explains why selected collective observables admit an effectively classical description. The interesting empirical question is whether any specifically quantum degree of freedom provides a distinct, necessary causal contribution to the kind of processing implicated in conscious experience. That question has to be settled by experiments, not by the mere existence of a wavefunction. [2][13][14]
Where a serious test might begin
Branching thought experiments are powerful because they expose conceptual assumptions about persistence, duplication, and identity. They do not, by themselves, provide a laboratory assay of subjectivity. A more tractable scientific program would begin by distinguishing structural conditions needed to identify a candidate subject from measurable conditions that covary with the presence or absence of conscious experience. One could ask whether a putatively relevant temporally integrated causal organization persists through wakefulness, changes through sleep or anesthesia, or fragments under interventions that disrupt recurrent communication. The requirement is not simply to find a signal that tracks behavioral responsiveness. It is to determine whether the proposed causal invariant survives alternative grains and descriptions while responding to genuine changes in the physical interactions that maintain the process.
Such a program would need demanding controls. Different network representations should not generate arbitrarily different subject counts from the same physical data. Correlation alone should not be mistaken for causal integration. An intervention that preserves outward stimulus-response behavior while disrupting internal recurrent structure would be especially informative if a theory predicts a difference in experience; the practical and ethical limitations of obtaining independent reports would need to be acknowledged. Computational models can test internal coherence, counterexamples, and the sensitivity of candidate criteria long before any claim about actual consciousness is justified. [10][12]
There is also a theoretical test for an Everettian extension. A proposed criterion should make clear whether it is defined using operationally accessible branch-relative states, fully universal states, or a physical coarse-graining justified by decoherence. It should remain stable under harmless changes of basis while changing appropriately when the relevant physical causal relations change. Finally, it should not assert that splitting amplitudes into more mathematical terms creates more conscious subjects. The model must distinguish an arbitrary expansion of a state vector from the emergence of dynamically autonomous records. A construction that fails this test has confused representational multiplicity with physical multiplicity.
The hardest test remains Nagel's: suppose all these structural criteria were established. Would they explain why the process is accompanied by experience, or would they merely identify a class of physical systems already suspected of being conscious? A research program can make progress on causal individuation and the conditions under which physical organization persists while leaving that explanatory question open. Intellectual progress is not diminished by acknowledging where a derivation ends. On the contrary, the open boundary between structural description and phenomenology is exactly where stronger theories must eventually do their work. [9][10]
The observer is not written in the branching diagram
Quantum mechanics supplies an extraordinary description of how a single laboratory history can become correlated with mutually inaccessible future records. Decoherence explains the stability of those records, and Everettian theories explore what it would mean to take their universal quantum description literally. Philosophical accounts of identity then ask whether continuation can be one-to-many. Causal and structural theories of consciousness ask which organized physical processes are plausible bearers of experience. These are connected questions, but none can substitute for the others.
The appeal of thinking in terms of invariant causal processes is that it places the burden on what a physical organization does through time, rather than on a mysterious label attached to matter or an arbitrary coordinate in a quantum state. It allows the possibility that a single past can have several future continuations without insisting that a single private witness must travel into only one. But it must also explain what distinguishes merely persistent machinery from an experiencing subject. A persistent record is not automatically a mind, and the duplication of a record is not automatically the duplication of a point of view.
If a universe really branches, it may leave the question 'Which version of me survives?' without the kind of unique answer the question assumes. That would be a challenge to a particular picture of personal identity. It would not solve the problem of consciousness. The more fundamental question is what makes any physically describable continuation a place where there is something it is like to exist. Until that bridge is established, the many-worlds observer remains both a precise quantum system and a profound philosophical unknown.
References and further reading
[1] Everett, H. III, “Relative State” Formulation of Quantum Mechanics, Reviews of Modern Physics 29, 454–462 (1957). https://doi.org/10.1103/RevModPhys.29.454
[2] Wallace, D., The Emergent Multiverse: Quantum Theory according to the Everett Interpretation (Oxford University Press, 2012). https://doi.org/10.1093/acprof:oso/9780199546961.001.0001
[3] Vaidman, L., Why the Many-Worlds Interpretation?, Quantum Reports 4, 264–271 (2022). https://doi.org/10.3390/quantum4030018
[4] Zurek, W. H., Quantum Theory of the Classical: Einselection, Envariance, Quantum Darwinism and Extantons, Entropy 24, 1520 (2022). https://doi.org/10.3390/e24111520
[5] Zurek, W. H., Quantum Darwinism, Nature Physics 5, 181–188 (2009). https://doi.org/10.1038/nphys1202
[6] Parfit, D., Reasons and Persons (Clarendon Press, 1984), Part Three: Personal Identity. https://academic.oup.com/book/12484
[7] Sebens, C. T. and Carroll, S. M., Self-Locating Uncertainty and the Origin of Probability in Everettian Quantum Mechanics, The British Journal for the Philosophy of Science 69, 25–74 (2018). https://doi.org/10.1093/bjps/axw004
[8] Barrett, J. A., Everett's Relative-State Formulation of Quantum Mechanics, Stanford Encyclopedia of Philosophy (2018 revision). https://plato.stanford.edu/archives/win2018/entries/qm-everett/
[9] Nagel, T., What Is It Like to Be a Bat?, The Philosophical Review 83, 435–450 (1974). https://philosophy.uconn.edu/wp-content/uploads/sites/365/2020/03/Nagel-What-is-it-like-to-be-a-bat.pdf
[10] Shah, R., Beyond ϕ: Invariant Causal Process Structures as the Substrate of Subjectivity, Wolfram Community (2026). https://community.wolfram.com/t/beyond-invariant-causal-process-structures-as-the-substrate-of-subjectivity/28167
[11] Ladyman, J. and Ross, D., with Spurrett, D. and Collier, J., Every Thing Must Go: Metaphysics Naturalized (Oxford University Press, 2007). https://doi.org/10.1093/acprof:oso/9780199276196.001.0001
[12] Albantakis, L. et al., Integrated Information Theory (IIT) 4.0: Formulating the Properties of Phenomenal Existence in Physical Terms, PLOS Computational Biology 19, e1011465 (2023). https://doi.org/10.1371/journal.pcbi.1011465
[13] Hameroff, S. and Penrose, R., Consciousness in the Universe: A Review of the ‘Orch OR’ Theory, Physics of Life Reviews 11, 39–78 (2014). https://doi.org/10.1016/j.plrev.2013.08.002
[14] Tegmark, M., Importance of Quantum Decoherence in Brain Processes, Physical Review E 61, 4194–4206 (2000). https://doi.org/10.1103/PhysRevE.61.4194
Cover art from the author's original Beyond ϕ / ICPS Wolfram Community post. Research figures are reproduced with attribution under their respective publishers' CC BY 4.0 licenses, as credited in each figure caption. The distinctions developed here are philosophical analysis, not empirical confirmation of a particular quantum mechanism of consciousness.