that function less as engrams than as “weights” assigned to synaptic connections, that we may reinterpret in terms of probabilities of future activations. A particular configuration of connection weights corresponds practically to a disposition for producing certain activation patterns in response to specific stimuli, sensory or otherwise. To repeat, conservation doesn’t involve encoding discrete traces that correspond one-to-one with well-defined memory contents. Rather, there are multiple patterns deposited across different brain organizational levels, following different pathways or networks of superimposed dispositions or activation prob-abilities in a non-bijective relationship that theoretically allows multiple possible realizations (John Sutton & Gerard O’Brien, “Distributed traces and the causal theory of constructive memory,” in Current Controversies in Philosophy of Memory, A. Sant’Anna et. al. (eds), London, Routledge, 2024, p. 88). To avoid excessive focus on “traces,” we could directly model memories themselves, assigning occurrence probabilities to certain possible versions of a given memory, or to activation pat-terns capable of triggering memory “recalls” of a particular kind. In any case, pro-ducing the corresponding mnemonic image becomes a kind of “measurement” in the sense that it involves randomly selecting one possible memory state based on its own probability distribution, or on the probability distribution associated with superimposed neural activation patterns likely to correspond to that memory.

Naturally, the term “superposition” must be handled cautiously, as there is no point in invoking quantum probabilities in a macroscopic context (unless we fol-low Roger Penrose and Stuart Hameroff, who hypothesize that consciousness and memory involve quantum processes within neuron microtubules). Nevertheless, the analogy remains valid at a certain level of generality, emphasizing in both cases the inseparability of observer and observed, of measurement and experi-mental context, and consequently a fundamental measurement-related indeter-minacy. Without delving into this matter’s complexities, let’s simply summon Schrödinger’s cat. In his famous thought experiment, the physicist envisioned the animal in “superimposed” states, “entangled” with the entire experimental setup, raising a famous paradox. In reality, we should consider the cat to be nei-ther dead nor alive (double negation)—not dead or alive (disjunctively), nor dead-alive (conjunctively), despite this hypothesis’ speculative interest. Pending evi-dence, before opening the box to verify its condition, the cat remains suspended in quantum superposition: not the combination of two incompatible states, but a fundamentally undetermined state. Essentially, we should acknowledge that the quantum system the cat stands for lacks the substantiality or a permanent property-bearer, even if these properties are contradictory ones. The cat isn’t a stable subject existing, with variable probability degrees, in come mixed state determined in itself before observation or measurement. This roughly summa-rizes the lesson perceptive philosophers, likely including Schrödinger himself, drew from this theoretically entangled situation: ultimately, we should abandon our default ontology—objects with proper properties—to instead discuss fields of potentialities or propensities, the evolution of which is described by the wave equation in terms of detection probability. Perhaps we should avoid ontology altogether, adhering to Schrödinger’s recommendation to view the wave func-tion as a “prediction catalog” linked to particular experimental setups (see Michel Bitbol, L’Aveuglante proximité du réel, Paris, Champs-Flammarion, 1998)? In every case, quantum theory demands serious revision of the spontaneously realistic grammar of predication that attributes well-defined properties to subjects. This approach is aligned with the fundamental lesson the physicist repeatedly empha-sized regarding the supposed individuality of quantum “objects”:

“We must not admit the possibility of continuous observation. Observations are to be regarded as discrete, disconnected events. Between them there are gaps which we cannot fill in. There are cases where we should upset everything if we admitted the