The flexible handling of perception-action representations is crucial for cognitive control such as response inhibition, which depends on the catecholaminergic system. However, how cross-frequency interactions support perception-action integration during response inhibition, and how they are modulated by catecholamines, remains unknown. In this placebo-controlled study employing methylphenidate, using electroencephalography (EEG) and a modified Go/Nogo task, we investigate phase-amplitude coupling (PAC) between theta (θ), alpha (α), beta (β), and gamma (γ) oscillations. We demonstrate that these interactions are hierarchically organized, with early α-β PAC supporting perceptual-motor representation, and subsequent β-γ coupling refining downstream processing. Transfer entropy analyses indicate a feed-forward α-β to β-γ influence, suggesting that slower oscillations gate updates in faster bands. Crucially, methylphenidate selectively enhances late β-γ coupling, supporting a functional specialization where α-β rhythms enable access and reconfiguration, while β-γ rhythms mediate local control. These findings suggest a temporally structured mechanism where the catecholaminergic system modulates flexible perception-action integration during response inhibition.

Influence of perforated twisted tapes with vortex generator wings on heat transfer performance and entropy in a heat exchanger tube

Entropy-based methods have gained increasing prominence in analyzing and detecting patterns in complex systems data. Key aspects such as fractality, time reversibility, and nonlinearity are commonly characterized using these methods, often in conjunction with techniques like wavelets and empirical modeling. The goal is to extract complexity indices and identify patterns that traditional methods struggle to capture. Shannon entropy obtains information from data by, e.g., extracting hidden information by removing noisy components. However, its applicability can be limited for data of short length. Approximate, sample and permutation entropies offer more flexible approaches, but their parameter dependence can complicate the interpretation of results. Singular value decomposition (SVD) entropy provides a framework for assessing pattern diversity in terms of an entropy index, reflecting an approximate dimensionality of a given dataset. This review focuses on recent SVD entropy applications across various fields and explores its potential in nonlinearity detection and transfer entropy analysis for time series, illustrated through select cases.

Structure of transfer entropy

EEG dynamics under anesthetics-induced loss and return of responsiveness with A constant rate infusion.

ABSTRACTThe bidirectional non‐linear communication between blood pressure (BP) and heartbeat is critical to cardiovascular homeostasis, which remains poorly understood, especially under hypertensive conditions. We implemented transfer entropy (TE), an information‐theoretic measure of directional coupling, to characterize such bidirectional coupling between BP and heartbeat and its relationships to arterial stiffness and walking performance in older adults. A total of 493 older adults (201 normotensive (NTN), 168 controlled‐hypertensive (controlled‐HTN), and 124 uncontrolled‐HTN) completed simultaneous recordings of resting‐state beat‐to‐beat BP and R‐R interval for ≥ 10 min. The TE from BP to RR (i.e., BP‐RR) and from RR to BP (RR‐BP) was quantified. Participants then completed the assessments of arterial stiffness (i.e., brachial–ankle pulse wave velocity, baPWV) and walking speed in single‐ and dual‐task conditions. The validation using surrogate data confirmed the physiological significance of TE (p < 0.0001). Both BP‐RR and RR‐BP TE were significantly lower in controlled‐ and uncontrolled‐HTN compared to NTN (p < 0.03). In NTN and control‐HTN, higher BP‐RR and/or RR‐BP TEs were associated with slower walking speed (β = −0.25 to −0.16, p < 0.04). Higher BP‐RR TE was associated with lower baPWV (β = −0.17 to −0.16, p < 0.04), while higher RR‐BP TE was associated with greater baPWV (β = 0.17–0.21, p < 0.03). No such significant associations were observed within uncontrolled‐HTN. The observations suggested that TE captures hypertension‐related disruption of bidirectional BP‐heartbeat information flow, reflecting impaired baroreflex feedback, exaggerated feedforward cardiac influence, and dampening with anti‐hypertensive therapy. The distinct associations with vascular stiffness and walking performance suggest TE as a promising marker of cardiovascular integrity and functional reserve in aging.

Establishing quantitative causal relationships between drought indicators and vegetation degradation in the Chad Basin remained challenging due to statistical limitations of applying traditional Transfer Entropy to finite-length remote sensing time series. This study implemented a Machine Learning Enhanced Transfer Entropy structure to quantify directed information flow from primary drought drivers of precipitation and land surface temperature to vegetation dynamics from 2000 to 2023. A feed-forward neural network trained on 10,000 synthetic samples with known theoretical Transfer Entropies enabled causal inference from 24-year MODIS-derived NDVI, land surface temperature, and precipitation. The trained model was applied over 10 million pixels, producing Transfer Entropy maps. Results showed that precipitation and land surface temperature exerted comparable causal influences on NDVI, with mean Transfer Entropy values of 0.064 and 0.063, ranging from 0.041 to 0.388. Spatial analysis revealed distinct causal hotspots exceeding 75th percentile threshold of 0.069, indicating driver-specific vulnerability zones. The decline in mean annual NDVI from 0.225 in 2019 to 0.194 in 2023, together with spatially divergent hotspots, highlighted the need for geographically targeted land management. The study overcame finite-length time-series limitations and provided a replicable pathway for vulnerability assessment and climate adaptation planning in data-constrained drylands in the Chad Basin in Africa.

Collective decision-making and movement coordination are essential behaviours observed in biological systems, from animal herds to human crowds. Horses are a highly social species with a multilevel society. Herding, where the harem is collected to move in a certain direction, is an often-cited example of agonistic behaviour in horses, yet poorly understood in a granular, quantitative sense. We use transfer entropy to measure herding in a harem group of feral Garrano ponies in Serra D'Arga, Portugal. First, we characterize the harem's leader-follower relationships by quantifying the time lag (average 1.44 s) and duration (average 1.72 s) of influence during herding, establishing variance across social characteristics. Second, we internally validate transfer entropy as a method to detect herding by comparing it with traditional clustering methods. To augment the paucity of existing data, synthetic data is generated from a mathematical model of feral horse harems, demonstrating superior accuracy (0.80) and F1-score (0.76) against traditional clustering and time-series synchrony methods. Third, we provide evidence for herding as an emergent behaviour: leadership influence often propagates indirectly among mares in short bursts of information flow before reaching the entire harem. These results enrich our understanding of horse behaviour and provide a foundation for using transfer entropy to study decision-making in other species.

Financial time-series labeling remains fundamentally limited by three critical deficiencies: temporal rigidity (fixed horizons regardless of market conditions), scale blindness (single-resolution analysis), and correlation-causation conflation. These limitations cause systematic failure during regime shifts. We introduce Adaptive Event-Driven Labeling (AEDL), integrating three core innovations: (1) multi-scale temporal analysis capturing hierarchical market patterns across five time resolutions, (2) causal inference using Granger causality and transfer entropy to filter spurious correlations, and (3) model-agnostic meta-learning (MAML) for adaptive parameter optimization. The framework outputs calibrated probability distributions enabling uncertainty-aware trading strategies. Evaluation on 16 assets spanning 25 years (2000–2025) with rigorous out-of-sample validation demonstrates substantial improvements: AEDL achieves average Sharpe ratio of 0.48 (across all models and assets) while baseline methods average near-zero or negative (Fixed Horizon: −0.29, Triple Barrier: −0.03, Trend Scanning: 0.00). Systematic ablation experiments on a 12-asset subset reveal that selective innovation deployment outperforms both minimal baselines and maximal integration: removing causal inference improves performance to 0.65 Sharpe while maintaining full asset coverage (12/12), whereas adding attention mechanisms reduces applicability to 2/12 assets due to compound filtering effects. These findings demonstrate that judicious component selection outperforms kitchen-sink approaches, with peak individual asset performance exceeding 3.0 Sharpe. Wilcoxon tests confirm statistically significant improvements over Fixed Horizon baseline (p = 0.0024).

Purpose The hot hand fallacy is generally prevalent in information-driven decisions and calculated risk-taking choices. While gambling fallacious behavior is common in investors who approach the stock markets as gamblers. This paper aims to examine the presence of Gambler’s or hot hand fallacy in the Islamic stock market. Design/methodology/approach This study calculated a relative strength index for 6 and 14 periods and applied it to the daily Jakarta Islamic stock index ranging from 2 January 2010 to 12 September 2024. This study computed dummy variables based on the RSI to track consecutive gains and losses. This study estimated their impacts in both price and trading volumes regressions using ordinary least squares (OLS), Quantiles and wavelet regressions. This study applied Shannon transfer entropy to explore the causality between overbought/oversold scenarios and the Jakarta Islamic index. Findings This study divided the data into oversold and overbought scenarios, the positive (negative) impact of overbought (oversold) conditions is found to be statistically significant in price regression. Only oversold conditions came out to be statistically meaningful in a trading volume regression. Accordingly, results show evidence of the hot hand fallacy in the Jakarta stock market. The effects observed across quantiles exhibit heterogeneity, being insignificant for certain quantiles while corroborating the findings from OLS analysis. Additionally, the wavelet regression reveals a comparable impact, which demonstrates significant effects at high frequencies, specifically for periods not exceeding one week. Research limitations/implications This study results have valuable implications for policymakers, revealing that investors do not have a mere short-term focus and obsessively chase quick outcomes. While not disregarding long-term value beyond mere chance or speculation, they may potentially influence market’ dynamics. Originality/value This study directs the attention to a less-studied domain of behavioral biases in Islamic stock markets, with a particular focus on the Jakarta Islamic Index (JII). While earlier investigations have explored behavioral biases in general stock markets and other biases in Islamic finance, this study addresses a critical gap by examining the gambler’s fallacy and the hot hand fallacy within a Shariah-compliant market.

Circadian rhythms in mammals arise from the spatiotemporal synchronization of ~20,000 neuronal clocks in the suprachiasmatic nucleus (SCN). Although anatomical, molecular, and genetic approaches have revealed diverse SCN cell types, how network-level wiring enables their synchronization remains unclear. To overcome the challenges of inferring functional connectivity from fixed tissue, we developed Mutual Information & Transfer Entropy (MITE), an information-theoretic framework to infer directed cell-cell connections with high fidelity from long-term live-cell imaging. Recording and analyzing 3,290 h of clock gene expression from 8,261 SCN neurons across 17 mice, we uncovered a highly conserved, sparse SCN network organized into two asymmetrically coupled modules: dorsal and ventral. Connectivity analyses revealed five functional SCN cell types independent of neurochemical identity. Notably, only ~30% of vasoactive intestinal peptide neurons exhibited Hub-like connectivity, classifying them as Generators and Broadcasters of synchrony signals. Other spatially stereotyped cell types consistently identified as Bridges, Receivers, or Sinks. Simulations based on MITE-inferred connectomes recapitulated emergent SCN dynamics, including recovery from desynchrony and the daily dorsal-to-ventral phase wave of gene expression. Together, these results demonstrate that MITE enables precise mapping of cellular network topology, revealing the circuit logic and key cell types that mediate circadian synchrony across space and time in the mammalian SCN.

Inference of causal interaction networks of gut microbiota using transfer entropy

The complexity of multi-mode industrial processes poses challenges for traditional fault diagnosis methods in achieving satisfactory results. This challenge arises from the non-stationarity caused by changes in modes, which often overlaps with temporal variations in fault generation processes. To address this issue, we propose a fault diagnosis method based on multi-mode categorization and dynamic transfer entropy graph analysis. This method can uncover the causal information changes resulting from both mode switches and fault occurrence. Firstly, modes are categorized using the criterion of maximum intra-segment similarity and minimum inter-segment similarity, and a static causal graph is constructed for each mode. Secondly, a framework for dynamic transfer entropy graph is established to extract short-term causal information transfer using transfer entropy and compare it with the static causal graph of normal modes. Time points with significant changes in causal relationships are selected as fault time points for further analysis. Finally, an anomaly score is designed to identify critical fault nodes by selecting nodes with significant causal changes before the fault time point, thereby accurately locating the fault root nodes. To validate the performance of the proposed method, a case study using the Tennessee Eastman process is provided. Simulation results demonstrate the effectiveness and feasibility of the proposed method.

This study investigates the risk spillover network among major stock market indices of G20 countries from 2003 to 2024. Transfer entropy is employed to measure the asymmetric and nonlinear information flow between stock markets. Based on this key metric, a directed and weighted risk spillover network among stock markets is constructed using the threshold method during periods of extreme events. Utilizing complex network theories, such as PageRank and betweenness centrality, the study analyzes the macro-topological characteristics of the risk spillover network and identifies key nodes. The findings not only demonstrate strong information interaction among G20 stock markets but also show that European and North American markets exhibit regional clustering characteristics, while emerging markets serve as bridging nodes in the risk spillover network. These results offer theoretical and practical insights for portfolio management, risk monitoring, and cross-border financial regulation and crisis management.

Numerical investigation of laminar flow heat transfer characteristics and entropy generation in microchannel with shark-skin bionic structure

Transfer entropy analysis reveals interaction dynamics between termite castes

Addressing Bias in machine learning feature importance for food quality assessment.

Purpose This study explores the evolving interactions between scientific research and technological innovation in the support vector machines (SVMs) field, aiming to reveal their coupling mechanisms and innovation-driven dynamics. Design/methodology/approach We analysed 84,369 publications and 14,271 patents. Using rapid automatic keyword 123,514 keywords were extracted (109,996 scientific, 13,518 technical and 2,199 shared) to construct knowledge networks. Coupling strength, structural entropy and effective transfer entropy (ETE) were applied to assess network integration and directional influence between science and technology (S&amp;T). Findings The scientific network is consistently larger yet sparser than the technical one (average density 1.08E−03 vs 1.19E−02), though both show small-world effects (clustering coefficients: 0.796 vs 0.778). Coupling has intensified over time, reaching 0.5116 between the 2023 scientific and 2022 technical networks. ETE results indicate two phases: from 2004 to 2017, science strongly drove technology (ETE = 0.2888), while from 2018 to 2023, technology exerted a weaker reverse influence on science (ETE = 0.0153). Structural entropy analysis highlights stable scientific communities but volatile technical ones, especially after 2017. Originality/value By integrating content and structural perspectives, this study quantitatively demonstrates the shift from science-driven to technology-driven innovation in SVMs, offering new insights for research management and policy formulation.

Transfer entropy is a widely used measure for quantifying directed information flows in complex systems. While the challenges of estimating transfer entropy for continuous data are well known, it has two major shortcomings for data of finite cardinality: it exhibits a substantial positive bias for sparse bin counts, and it has no clear means to assess statistical significance. By computing information content in finite data streams without explicitly considering symbols as instances of random variables, we derive a transfer entropy measure which is asymptotically equivalent to the standard plug-in estimator but remedies these issues for time series of small size and/or high cardinality, permitting a fully nonparametric assessment of statistical significance without simulation.

The dynamics of physiological systems are impacted by both exercise and hypoxia. Network models can be used to map the interactions between various physiological components in environmental physiology and exercise using the concepts of information theory. This cross-over study compared three normobaric conditions: control, simulated altitude of 2500m (fraction of inspired oxygen: ≈ 15.1%) and 3500m ( ≈ 13.5%), and rest vs. isometric exercise through the lens of network physiology. The 12 participants (6M and 6 F; 22.25±2.42 years; 23.01±3.24kg/m2) spent ∼30min in a tent coupled to an altitude simulator, whose last 3min consisted of a series of nine unilateral isometric maximal contractions of quadriceps. A metabolic system in breath-by-breath mode was used to register cardiorespiratory variables. In-degree, out-degree, and transfer entropy (TE) were computed to capture the information flow between variables. A weighted Jaccard Similarity Index was used to assess network similarities. The increase of in exercise over rest was slightly more prominent during hypoxia (P=0.054, η2 p=0.232). Normoxia-hypoxia networks were more similar during resting than exercise. Rest-exercise networks were less similar to each other during simulated altitude of ∼2500m (P=0.008, η2 p=0.353). Neither TE during rest nor during exercise nor the / ratio significantly predicted the occurrence of symptoms. Unexpectedly, compared to mild-grade hypoxia, low-grade hypoxia induced more changes in physiological connectivity, with the majority of the connections converging on putative hidden nodes that we suggest are oxygen delivery-dependent. Network approaches could offer new developments in exercise and environmental physiology.