A Litmus Test for Plant Consciousness: Pattern–Temporal Synergy in a Relation-First Ontology

Abstract

Plant cognition has progressed from anecdote to rigor, yet the field still lacks a quantitative test for when distributed plant activity crosses into unified—perhaps conscious—processing. I introduce Pattern–Temporal Synergy (PTS), a substrate-agnostic metric rooted in Dynergeia, a relation-first ontology that defines consciousness as reflexive coherence among five universal patterns—self-reference, division-creation, information integration, responsiveness and flux—phase-locked inside a system’s binding window (τ). Each pattern is operationalized with established signal-processing measures; their mean strength is multiplied by their mean synergy and released only if a τ-specific coherence gate is met.Three preregistered hypotheses anchor the study: H1 baseline PTS > 0 in intact plants; H2 4 % diethyl-ether collapses PTS below threshold φ; H3 PTS rebounds on wash-out. A multispecies protocol—Mimosa pudica, Arabidopsis thaliana, Picea abies—combines 64-channel surface electrodes, glutamate-sensitive Ca²⁺ imaging and micro-optode O₂/heat-flux probes. Sliding 3 × τ windows with phase-shuffled surrogates yield z-scored PTS trajectories, adjudicated by preregistered effect-size criteria.By turning decades of qualitative insight into falsifiable numbers, PTS offers plant biology a litmus test for conscious-level processing, directly challenges Integrated Information Theory and supplies a road-map for cross-kingdom comparisons—including neuromorphic silicon. Confirmatory results would shift debates on plant sentience from speculation to data; null results would equally refine what consciousness requires.

Plant phenotyping relevance

植物の電気・Ca²⁺・酸素・熱流束を統合して植物状態を定量化する新規指標PTSを開発し、事前登録仮説とサロゲート比較で検証する研究であり、表現型取得・解析法が中心である。

Code and data availability

The article is a theoretical/proposal manuscript describing a planned plant-phenotyping experiment (PTS metric on Mimosa, Arabidopsis, Picea). No public phenotype datasets, plant images/sensor data, author analysis code, or trained models are reported; only generic third-party libraries (MNE-Python, IDTxl, dit, Network

No evidence-backed public reproduction asset is currently recorded.