In recent research testing the non-commutability conjecture of the Generalized Quantum Theory, Maier and Dechamps (2025) reported macroscopic complementary relations empirically documented as non-classical correlations between subjective assessment of likability of colors and variations of objective documentations of these colors. Specifically, they found higher likability mean scores in a condition where objective color parameters were not erased and stored in a result file compared to a condition in which these parameters were erased and inaccessible by experimenters. This effect was robust across four studies. Each study’s design in this series of experiments was intentionally confounded with a “color bias” variable. Only limited sets of colors after random creation were specifically assigned to one of the two conditions, so that the effect of the erasure manipulation could alternatively be explained by the specific features of the colors assigned to conditions (the confound was later statistically ruled out by permutations). The presence of a confound was introduced to avoid the impact of the NT axiom, which forbids a strict causal testing approach. Rather, according to the NT axiom when macroscopic complementary relations are tested in a strictly causal way, initial effect documentations should be followed by a decline within and across studies. In the study presented here, the studies of Maier and Dechamps (2025) were conceptually replicated by a strict, i. e. unconfounded, causal testing strategy, to explicitly test the proposed impact of the NT axiom. In two experimental conditions (within-subjects) objective color parameters were either stored or not stored. Assignment of colors to conditions was randomized on the participant level this time ensuring an unconfounded design. The prediction derived from the NT axiom was that under such circumstances the erasure-dependent likability effect should initially be found and later drastically decline. The results supported this prediction: A higher likability mean score was initially found in the non-erasure compared to the erasure condition with strong Bayesian evidence (BF10 = 39.77), replicating the previous findings but now followed by a decline within the study. These findings are in line with the conjectures of the GQT. Specifically, they indicate that subjective assessments and objective measurements of colors are non-commutable and that the data follow the predictions of the NT axiom exhibiting an “effect and decline” data pattern when strictly causally tested. An additional permutation analysis showed that the probability to find such an effect and decline data pattern by chance was very low (p < .03) indicating that it could hardly be interpreted as “false positive” finding. The impact of this and the previous results on the validity of the GQT in describing psychophysical relations as non-classical correlations between spatio-temporally separated subjective assessments and objective measurements of colors is discussed.
In recent research testing the non-commutability conjecture of the Generalized Quantum Theory, Maier and Dechamps (2025) reported macroscopic complementary relations empirically documented as non-classical correlations between subjective assessment of likability of colors and variations of objective documentations of these colors. Specifically, they found higher likability mean scores in a condition where objective color parameters were not erased and stored in a result file compared to a condition in which these parameters were erased and inaccessible by experimenters. This effect was robust across four studies. Each study’s design in this series of experiments was intentionally confounded with a “color bias” variable. Only limited sets of colors after random creation were specifically assigned to one of the two conditions, so that the effect of the erasure manipulation could alternatively be explained by the specific features of the colors assigned to conditions (the confound was later statistically ruled out by permutations). The presence of a confound was introduced to avoid the impact of the NT axiom, which forbids a strict causal testing approach. Rather, according to the NT axiom when macroscopic complementary relations are tested in a strictly causal way, initial effect documentations should be followed by a decline within and across studies. In the study presented here, the studies of Maier and Dechamps (2025) were conceptually replicated by a strict, i. e. unconfounded, causal testing strategy, to explicitly test the proposed impact of the NT axiom. In two experimental conditions (within-subjects) objective color parameters were either stored or not stored. Assignment of colors to conditions was randomized on the participant level this time ensuring an unconfounded design. The prediction derived from the NT axiom was that under such circumstances the erasure-dependent likability effect should initially be found and later drastically decline. The results supported this prediction: A higher likability mean score was initially found in the non-erasure compared to the erasure condition with strong Bayesian evidence (BF10 = 39.77), replicating the previous findings but now followed by a decline within the study. These findings are in line with the conjectures of the GQT. Specifically, they indicate that subjective assessments and objective measurements of colors are non-commutable and that the data follow the predictions of the NT axiom exhibiting an “effect and decline” data pattern when strictly causally tested. An additional permutation analysis showed that the probability to find such an effect and decline data pattern by chance was very low (p < .03) indicating that it could hardly be interpreted as “false positive” finding. The impact of this and the previous results on the validity of the GQT in describing psychophysical relations as non-classical correlations between spatio-temporally separated subjective assessments and objective measurements of colors is discussed.
In the Generalized Quantum Theory (GQT), a theory designed to describe psychophysical phenomena, the subjective and objective aspects of reality are considered to be complementary. The research presented here focused on investigating such a macroscopic complementarity between subjective and objective reality descriptions within the context of color assessment. In particular, the non-commutability conjecture of the GQT posits that two observables derived from the subjective and objective subsystems do not simultaneously provide specific eigenvalues. Rather, the act of measurement within one subsystem, e. g. the performance of objective measurements on a stimulus, modifies the state of the entire system, including the eigenvalues of the other subsystem, e. g. the subjective experience of this stimulus. This conjecture was tested empirically in four studies and in three overall analyses of the complete data set combining all four studies. The experimental manipulation involved a measurement variation of one aspect of the supposed complementary pair, namely the storage (non-erasure condition) or deletion (erasure condition) of objective color parameters (hue and lightness). It was anticipated that the erasure manipulation would influence the other part of the pair, namely the subjective evaluations of color with respect to brightness and likability. The primary hypothesis, which tested erasure-dependent effects on variations in subjective brightness scores was confirmed in Studies 1 and 2 but could not be replicated in Studies 3 and 4. The initial findings obtained with this dependent variable can thus be considered false positives and the primary hypothesis could not be confirmed. However, the second hypothesis, which tested erasure-effects on subjective likability mean scores revealed the following results: The exploratory findings from Studies 1 and 2, as well as the confirmatory findings from Studies 3 and the preregistered Study 4, indicated an erasure effect on subjective likability, whereby higher likability scores were observed in the non-erasure condition compared to the erasure condition in each study. Since the experimental design was intentionally contaminated by a “color bias” alternative explanation to stabilize the effect documentation, a clear causal erasure interpretation cannot be provided at the single study level. To overcome this contamination an overall analysis strategy was pre-planned and performed on the combined data set. Several overall analyses ruled out the “color bias” alternative explanation and another confound. In conclusion, the overall results provided evidence for an erasure effect on subjective likability supporting the model of GQT. This can be interpreted as macroscopic complementary relations between objective and subjective reality descriptions as proposed by the GQT. The implications of these findings for our understanding of the nature of reality and for the validity of the GQT are discussed.
The Generalized Quantum Theory (GQT) proposes that subjective experiences and objective physical measurements constitute macroscopic complementary subsystems related through acausal, non-local entanglement correlations. The non-commutability conjecture of the GQT predicts that measurement within one subsystem modifies the state of the entire system, including the eigenvalues of the complementary subsystem. Maier and Dechamps (2025) provided preliminary evidence for this conjecture by demonstrating that the storage (non-erasure) versus erasure of objective color parameters (hue and lightness) systematically affected subjective likability ratings across four studies (higher subjective likability mean score in the non-erasure compared to the erasure condition). The present research conceptually replicated these color-erasure findings in a different stimulus domain. In four studies (total N = 647), participants rated photographs of food dishes on subjective tastiness while the objective food parameters (calorie values and stimulus IDs) were either stored (non-erasure condition) or permanently deleted (erasure condition). As in the color studies, limited stimulus sets were fixedly assigned to conditions within each study, deliberately introducing a stimulus-bias confound to protect the acausal complementary relation from the destructive impact of the non-transmission (NT) axiom. Following a predefined three-step analysis plan, preregistered Study 1 (Step 1) documented a strong erasure effect on tastiness ratings. Participants evaluated food dishes as more appetizing in the non-erasure compared to the erasure condition (BF10 = 1309.22, d = .66), and Studies 2a–2c (Step 2) replicated this effect with three independently drawn stimulus sets (BF10s ranging from 10.40 to 26 × 10¹⁵, ds ranging from .26 to .50). In Step 3, preregistered overall analyses including a stratified permutation test (p = .002), a multilevel model (β = 4.72, p = .011), and sensitivity analyses ruled out the biased stimulus assignment alternative. These findings provide cross-domain evidence for the non-commutability conjecture: the documentation of objective stimulus parameters appears to acausally co-relate with subjective evaluations, consistent with macroscopic complementary relations as proposed by the GQT. The implications for substance dualism, physicalism, and dual-aspect monism are discussed.
This response addresses commentaries by Thomas Rabeyron, Ian Tierney, and Harald Walach on two articles in this issue concerning macroscopic complementary relations between subjective and objective aspects of reality, framed within the Generalized Quantum Theory (GQT). The central concern shared by the commentators is whether the erasure-confound paradigm can robustly avoid the non-transmission (NT) axiom-induced decline that has affected earlier experimental paradigms in this research area. We argue that the NT axiom expresses an epistemic restriction arising whenever causal-deterministic documentation is applied to phenomena that contain an autonomous-subjective component. Under this reading, the erasure-confound paradigm achieves only a reduced-objective documentation of the effect at the single-study level, because the confound prevents an unambiguous causal interpretation, while the effect is established indirectly across studies through the exclusion of the alternative explanation (the “color selection bias” confound). This design decouples evidence accumulation from the arrow of time and is proposed to circumvent the decline. We then put the decline prediction to an empirical test by reporting three previously unpublished color-erasure-with-confound studies (Studies 5–7), which together with the four studies of Maier and Dechamps (2025) form the complete record of seven studies in this paradigm, with no file drawer. The erasure effect appeared in the predicted direction in every study (sign test p = .008), and combined stratified permutation analyses across three independent 36-color sets ruled out the color-selection-bias confound as a joint explanation (Fisher p = .010; Edgington p = .018). An inverse-variance weighted meta-regression on the seven effect sizes returned a non-significant downward slope (b = −0.018, 95% CI [−0.043, 0.006], one-tailed p = .07), and the seventh study yielded an effect still clearly present (BF₁₀ = 3,679.73, d = .30). The data are consistent with a stable, reduced-objectively documented acausal psychophysical correlation rather than the attrition predicted by the NT axiom under direct replication. Further specific points raised by individual commentators, such as the experimenter-psi interpretation and the choice of subjective and objective measurements, are addressed in shorter form.
This study aims to conceptually replicate a previously observed "color values erasure effect" on subjective color preferences. The research investigates whether the storage or deletion of objective color data influences participants' subjective likability ratings of colors, as predicted by Generalized Quantum Theory's macroscopic complementarity principle. Purpose The primary purpose is to test whether participants rate colors as more likable when the objective color values are stored compared to when they are deleted immediately after data collection. This builds on four previous studies that demonstrated this erasure effect using limited color sets. The current study employs a within-subjects design where each participant evaluates 36 colors (18 in each condition) on a 101-point likability scale. Methodology Participants will view colors randomly selected from a 360-degree color circle (30-70% lightness range, 100% saturation) and rate how much they like each color. Crucially, participants remain unaware of the erasure manipulation. The same 18 colors will be assigned to the "stored" condition and the same 18 colors to the "deleted" condition across all participants - an intentional confound designed to avoid "strict causality tests" that theoretically might destroy the complementarity effect. Expected Outcomes We expect two main findings: (1) Higher mean likability ratings in the stored condition compared to the deleted condition (tested via sequential Bayesian analysis with BF > 10 criterion), and (2) The observed effect size should occur in less than 5% of random color-to-condition permutations, suggesting the effect is not simply due to color bias. The study will use sequential data collection up to N=5,000 participants, with smartphone use prohibited based on previous findings of reduced effect sizes on mobile devices.
The Complementarity Principle (CP), introduced by Nils Bohr, described the wave-particle duality of quantum phenomena and its dependence on the measurement setup exploring the quantum states. In general, this principle summarized the fact that two mutually exclusive and contradictory states of quantum events can be reconciled as a measurement-dependent occurrence of these states. Originally, the CP was postulated only for the realm of quantum mechanics, but recently the Generalized Quantum Theory (GQT) extended its area of applicability to the macroscopic domain. According to GQT complementarity relations are also hypothesized to exist between macroscopic objective and subjective measurements of the same concept. This implies objective measurement-dependent variations of subjective experience. To test this hypothesis, seven studies were conducted in which individual variations in subjective intelligence ratings were tested in relation to the availability or non-availability of corresponding objective individual intelligence data. Only one pre-registered study (Study 2) showed strong Bayesian evidence for H1 (BF10 > 10), indicating, as predicted, higher subjective intelligence ratings when objective data were erased compared to a condition in which objective data were available within a male sample. This effect could not be replicated in direct replication attempts, nor did a moderator search in subsequent studies find any robust systematic variation in the data. The results seem to question the validity of the macroscopic complementarity conjecture derived from the GQT. On the other hand, they could also be interpreted as the “effect and decline” data pattern that the Model of Pragmatic Information would predict when conducting empirical confirmations of macroscopic complementarity relations. Possible future research strategies to clarify these different interpretations are discussed.
A vivid discussion revolves around the role of the human mind in the quantum measurement process. While some authors argue that conscious observation is a necessary element to achieve the transition from quantum to classical states during measurement (Wigner, 1963), some go even further and propose a more active influence of the human mind on the probabilities of quantum measurement outcomes (e.g. Atmanspacher, Römer, & Walach, 2002; Penrose & Hameroff, 2011). This proposition was tested in micro-psychokinesis (micro-Pk) research were intentional observer effects on quantum random number generators (RNGs) were investigated. In the studies presented here we extended this line of research and tested the impact unconscious goals on micro-Pk. Our focus was cigarette addiction as an unconscious drive and hypothesized that regular cigarette smokers would influence the outcome of a quantum RNG that determined whether the participant was going to see a smoking-related or a neutral picture. Study 1 revealed strong evidence for micro-Pk (BF = 66.06) supporting the H1. As expected, no deviation from chance was found with non-smokers. In Study 2, a pre-registered highly powered replication attempt, failed to reproduce this result and showed strong evidence for H0 (BF = 11.07). When the data from both studies are combined a remarkable change in effect across time (resembling a combination of appearance followed by decline) can be seen only in the smoker’s subsample. Appearance-and-decline effects were absent in the non-smokers sample and in a simulation. Based on von Lucadou’s Model of Pragmatic Information we suggest that (micro-)Pk effects follow a systematic pattern comparable to a dampened harmonic oscillation. This concept may shed new light on past and future Pk research.