The research team, led by Lena Arai, Yi Zeng, and Paul J. Zak, designed their experiment to fix what they see as two chronic artifacts in the classic trust-game literature. Since the pioneering experiments of the 1990s, most laboratory trust studies have disallowed pre-decision communication and used trivially small stakes. The rationale was to strip away confounds such as attractiveness, smiles, touch, or ethnicity that might contaminate a clean measure of trust. Yet the behavioral data show that this unenforceable communication, known in game theory as cheap talk, substantially increases cooperation. Outside the laboratory, trust is nearly always preceded by a conversation in which the terms of exchange are discussed. By removing communication, the authors argue, decades of experiments have ignored the most interesting part of the decision to trust.

The second artifact was the small endowments typical of trust experiments, which usually involve college students, a population the literature describes as WEIRD: Western, Educated, Industrialized, Rich, and Democratic. Low-stakes monetary trust experiments typically yield around 90 percent of individuals trusting strangers, a level of cooperation that may say more about the insignificance of the money than about human trust. The new study recruited 76 employed adults between the ages of 25 and 50 from diverse cultural backgrounds, including participants who self-identified as White, Asian, Latino or Hispanic, and African American or Black, and mixed native and non-native English speakers. Participants could earn up to 530 dollars based on their own decisions and those of their dyadic partner, with 50 dollars guaranteed for showing up and all payments made in cash without deception.

The monetary task itself was a modified version of the sequential trust game. Each participant was endowed with 120 dollars and randomly assigned as the first or second decision-maker. The first decision-maker could send 0, 40, 80, or 120 dollars of that endowment to the second decision-maker. Transfers sent were removed from the first player’s account and multiplied in the second player’s account, with the multiplier varying by amount: 40 dollars became 240, 80 dollars became 360, and the full 120 dollars became 480. The second decision-maker then chose how much to return, without multiplication, from the multiplied stake plus the original endowment. Varying the multiplier was a deliberate departure from the standard design, intended to make the decision process more complicated and to ensure participants understood that the amounts involved were meaningful. In the consensus view of experimental economics, the first transfer measures trust and the return transfer measures reciprocity or trustworthiness.

Before making their private decisions at partitioned computer terminals, each dyad sat facing each other in a private room for exactly two minutes. They were advised to discuss their strategies and make a commitment to a particular choice, but were told clearly that these commitments were not enforceable and that all decisions would be made in private. Throughout this conversation, electrodermal activity, the peripheral signature of centrally mediated arousal, was recorded from disposable electrodes on the index and middle fingers of the non-dominant hand. Electrodermal activity captures increased conductivity from sweat in the eccrine glands, and was chosen because it has previously been shown to predict who reneges on promises to reciprocate trust, and because it is straightforward to obtain without interfering with natural conversation. Each participant’s baseline response, measured while sitting alone with headphones for two minutes, was subtracted from the response during communication to control for individual differences.

The central technical question was how to quantify synchrony between two people’s physiological time series. The researchers evaluated two measures. The first, intersubject correlation, is the traditional benchmark used in neuroimaging and electroencephalogram studies: it calculates Pearson correlations for a physiological measure between participant pairs, often within short overlapping windows and with time lags to account for the dynamics of interpersonal interaction, an approach known as windowed cross-lagged correlation. The second, dynamic time warping, is newer to behavioral neuroscience. Developed to align time series that vary in speed, it was famously applied to speech recognition, which made it an apt candidate for electrodermal signals generated during dyadic conversation. Dynamic time warping builds a cost matrix of pairwise distances between two series and finds the warping path that minimizes cumulative distance, effectively stretching or compressing the time axis of one or both signals to achieve better alignment than simple Euclidean distance. Unlike correlation-based approaches applied to nonstationary data, dynamic time warping does not require stationary time series, a property well suited to physiological signals that drift and surge during live interaction.

The behavioral results were themselves unusual for a trust game. Across 222 usable interactions, maximal trust, in which the first decision-maker sent the entire 120-dollar endowment, occurred in 28 percent of cases. Conversely, in 40 percent of interactions, the first decision-maker sent nothing, revealing complete distrust. That nearly half of participants showed complete distrust and one third displayed maximum trust is a distribution not typically seen in other trust games, and the authors attribute it to the meaningful stakes. Notably, the decision to trust did not vary by race, sex, age, education, or income, nor did it depend on whether dyadic partners were the same or different sexes. The amount transferred was uncorrelated with age, education, or income. Even the raw change in skin conductance level from baseline to communication did not differ between those who trusted and those who did not, underscoring that it was not arousal per se but the relationship between two people’s signals that carried the predictive signal.

When the two synchrony measures were compared, the difference was decisive. Intersubject correlations averaged across four window sizes did not vary between trust and no trust, nor by the degree of trust, and nonparametric Kruskal-Wallis tests found none of the individual window values significantly different across the four trust groups. Dynamic time warping told a different story. While average values were identical for all levels of trust compared to no trust, they were statistically different by the degree of trust, and three of the four window widths, 5, 10, and 15, were significantly different from each other. Consistent with the team’s hypothesis, dynamic time warping synchrony during communication was associated with increased first-decision-maker trust for all four windows tested, with p-values below .01. In practical terms, dyads whose electrodermal signals aligned more tightly during the two-minute conversation were more likely to have the first decision-maker sacrifice money to trust the stranger across the table.

To compare the predictive accuracy of the two measures head to head, the researchers estimated binary logistic regressions predicting trust or no trust. Because the four windows of each measure were highly correlated, with correlations ranging from .49 to .99, they generated principal components separately for each synchrony type, using cross-validation to choose the optimal number. The intersubject correlation analysis produced two components explaining 83 percent of the variation in synchrony, but neither was statistically associated with the decision to trust, and the regression achieved a balanced accuracy of only 52 percent. The dynamic time warping analysis produced three components explaining 99 percent of the variation. None alone was statistically significant, but together they predicted the trust decision with a balanced accuracy of 60 percent, with precision and recall both at .68. That is a 15 percent improvement over intersubject correlation, or eight percentage points in the discussion section, and while modest in absolute terms, the authors emphasize that the correlation measures were largely unrelated to the trust decision while the warping windows were statistically associated with it.

The findings open a window onto mechanisms that operate beneath conscious awareness. Cheap talk reveals information that influences choices, including statements, smiles, word choice, gestures, and signs of physiologic stress that indicate lying, but much of this information is processed unconsciously, which is why asking participants why they trusted a stranger yields so little insight. The authors caution against overgeneralizing: the sample was moderate in size and consisted of educated Western adults, the trust game is a scripted social interaction, and future work should replicate the results in field studies with non-scripted tasks, different populations, and phasic skin conductance responses. Still, they argue that future studies should include meaningful stakes, pre-decision communication, and neurophysiologic signatures. Because electrodermal signals emanate from the brainstem and are conserved across individuals, the approach may hold on average across settings. Part of the research program involves mapping these unconscious physiologic tells into observable movements, statements, and language that negotiators could read without sensors. If two minutes of conversation can align the nervous systems of strangers tightly enough to predict who will risk 480 dollars on them, the warping of time between two bodies may be one of the most consequential signals in economic life.

Subject of Research: Dynamic time warping of interpersonal physiologic synchrony predicts high stakes trust

Article Title: Dynamic time warping of interpersonal physiologic synchrony predicts high stakes trust

Article References: Arai, L., Zeng, Y., & Zak, P. J. (2026). Dynamic time warping of interpersonal physiologic synchrony predicts high stakes trust. Frontiers in Behavioral Economics, 5, Article 1826500. https://doi.org/10.3389/frbhe.2026.1826500

Image Credits: AI Generated

DOI: 10.3389/frbhe.2026.1826500

Keywords: neuroeconomics, trust, physiological synchrony, dynamic time warping, electrodermal activity, behavioral economics, Dynamic, time, warping, interpersonal, physiologic, synchrony

The motivation is stark. When the I-35W highway bridge in Minneapolis collapsed in 2007, killing 13 people, the subsequent National Transportation Safety Board investigation traced the failure to gusset plates whose load capacity had been compromised by a design error. Localized weakness had gone unnoticed until catastrophe. The aviation world offers similar warnings: reviews of aircraft fatigue accidents show that cracks frequently initiate at riveted joints and other stress-concentration interfaces, and the January 2024 in-flight separation of a door-plug panel on a Boeing 737-9 MAX underscored how elusive such vulnerabilities can be. In practice, sensors are rarely installed at every location of interest, so the quantities that matter most for fatigue assessment often remain unmeasured.

Existing approaches to this problem fall into two broad camps. Model-based structural health monitoring interprets measured responses through physics-based frameworks such as the eigensystem realization algorithm, subspace identification, augmented Kalman filtering, and Bayesian model updating. These methods are powerful but sensitive to modeling assumptions about material properties, connection stiffness, damping, and boundary conditions, and discrepancies between predicted and measured responses tend to be largest precisely near the local stress concentrations where failure begins. Data-driven methods, from convolutional and recurrent neural networks to sequence-to-sequence models with attention mechanisms, learn relationships directly from signals, but most still assume measurements from multiple distributed sensors or require strain-based inputs. Under truly sparse sensing, estimating internal responses becomes an underdetermined problem, because similar boundary measurements can correspond to different internal deformation states.

The new framework, developed by Jin Hyeok Seok, Ji Wan Seo, Yeong Kwan Jo, Sang Hu Park, Jae Hyuk Lim, and YunHo Kim, takes a different route. Rather than deepening the network to boost its representational capacity, the team split the prediction task into two sequential stages that mirror the physics of structural response. In the first stage, a shallow time-delay neural network converts boundary acceleration into boundary displacement. In the second stage, a second shallow network uses that estimated displacement to infer the internal strain. Each stage therefore learns only a reduced, physically consistent mapping, which keeps the model compact and trainable on limited experimental data while preserving sensitivity to the quantities that matter.

The choice of a shallow time-delay neural network, or TDNN, was deliberate. Each network consists of a single hidden layer with an input delay of eight time steps, allowing it to exploit short-term temporal dependencies in the vibration signal without the data appetite of deep sequence models. The first stage and the single-stage baseline used ten hidden neurons, while the second stage used five. Training employed the Levenberg-Marquardt backpropagation algorithm with full-batch optimization, validation-based early stopping, and a dataset of roughly 9,110 time-series samples split into training, validation, and test subsets at ratios of 70, 15, and 15 percent. A separate prediction dataset of about 121,291 samples, never seen during training, evaluated generalization.

The experimental validation was conducted on a commercial outdoor air-conditioning unit roughly 1.7 meters tall and weighing about 250 kilograms, mounted on an industrial rotary vibration system designed to reproduce transportation-type excitation conditions consistent with MIL-STD-810F guidance. Triaxial accelerometers with a 40 g range were rigidly bolted to the base structure to capture boundary acceleration, while strain gauges were attached to pipe sections inside the unit, locations that are structurally sensitive but sit in narrow spaces poorly suited to long-duration monitoring. Strain signals were initially recorded at 10 kHz and systematically downsampled; 500 Hz preserved the reference waveform with a similarity greater than 0.99 and was selected for training. Boundary displacement was obtained by numerically integrating acceleration with Simpson’s rule, with drift corrected by resetting displacement to zero at the end of each 3.5 Hz excitation cycle.

The results reveal both the promise and the subtlety of the approach. Over a 242-second prediction period, far longer than the training segments, both the one-stage and two-stage models produced stable strain estimates without divergence, adjusting smoothly when excitation intensity changed at around 125 seconds. Yet direct time-domain comparison produced negative coefficients of determination, at -0.66 for the one-stage model and -0.76 for the two-stage model, despite visually similar waveforms. The culprit was progressive phase drift: predicted peaks gradually shifted relative to measured peaks, with offsets growing from roughly 0.134 seconds early in the record to about 0.200 seconds later. After applying dynamic time warping, a post-processing alignment technique, both models achieved an R-squared of approximately 0.93, confirming that the overall waveform structure was faithfully reproduced once the timing offset was compensated.

Where the two architectures genuinely diverged was in peak reconstruction, and this is where the study carries real weight for fatigue engineering. Fatigue damage accumulation, particularly under cycle-counting methods such as Rainflow counting combined with the Palmgren-Miner rule, is dominated by peak amplitudes; if peaks are underestimated, remaining fatigue life can be significantly overestimated. The experimental mean peak strain was about 0.121 microstrain. The one-stage model captured only 68.60 percent of that value, while the two-stage model reached 80.99 percent, an improvement of roughly 12 percentage points. On the independent prediction dataset, peak accuracy fell to 68.94 percent for the one-stage network but rose to 80.67 percent for the two-stage network. The explanation lies in how mean-square-error training treats sparse events: a single-stage network forced to represent both global motion and local deformation in one mapping tends to sacrifice peak fidelity for overall waveform agreement, whereas the staged decomposition lets the second stage focus on local deformation behavior.

Fast Fourier Transform analysis added a mechanistic explanation. Acceleration and strain signals shared a dominant frequency near 6.98 Hz, while displacement was governed by a lower frequency of about 3.48 Hz. Crucially, the high-frequency components associated with rapid strain variations were preserved in the intermediate displacement prediction, meaning the first stage acted not merely as a motion estimator but as a feature-preserving transformation that passed peak-related cues to the second stage. The authors are careful to frame the limits of their work: the experiment covered a single repeated rotary excitation condition with a dominant cycle of about 3.5 Hz, the short 0.016-second input delay window may not capture longer time-varying phase relationships, and progressive loosening of the test fixture may have contributed to the observed phase drift. Real-time, phase-synchronous monitoring would require additional synchronization or phase-correction strategies.

Even within those constraints, the demonstration is striking. Once the relationship between boundary acceleration and internal strain is learned, the trained model can estimate strain at critical internal locations using only a handful of boundary sensors, without a finite element model, state-space formulation, or prescribed loading information. The researchers suggest the framework could support fatigue life evaluation, long-duration response prediction, and digital-twin-based structural state estimation, with future extensions potentially incorporating recurrent, attention-based, or physics-informed architectures while retaining the physically interpretable staged structure. For a field where the most dangerous stresses hide in the least accessible corners of a machine, the message is provocative: sometimes the boundary knows more than it lets on, and a small, well-structured network can be taught to listen.

Subject of Research: Data-driven reconstruction of internal structural strain from sparse boundary acceleration measurements using a two-stage shallow time-delay neural network

Article Title: Data-Driven Estimation of the Critical Structural Strain from Boundary Acceleration Using a 2-Stage Shallow Time-Delay Neural Network

Article References: Seok, J. H., Seo, J. W., Jo, Y. K., Park, S. H., Lim, J. H., & Kim, Y. (2026). Data-Driven Estimation of the Critical Structural Strain from Boundary Acceleration Using a 2-Stage Shallow Time-Delay Neural Network. International Journal of Aeronautical and Space Sciences. https://doi.org/10.1007/s42405-026-01244-1

Image Credits: AI Generated

DOI: 10.1007/s42405-026-01244-1

Keywords: structural health monitoring, time-delay neural network, virtual sensing, strain reconstruction, fatigue assessment, boundary acceleration, machine learning, vibration testing, digital twin, aerospace structures, peak strain prediction, dynamic time warping

The core idea is deceptively simple: treat a hand-drawn signature as a biometric trait, just as banks and document workflows have done for decades with pen and paper. What makes AirSign distinctive is how it captures that signature. Instead of confining users to a predefined two-dimensional plane, or demanding specialized peripherals such as data gloves or a Leap Motion sensor mounted on a desk, the system exploits hardware that modern headsets already carry. Users can sign either with motion controllers or, on devices that support it, with camera-based hand tracking. A single unified gesture, a pinch between the index finger and thumb, signals the intent to write. When the virtual fingertips touch, the system begins sampling the three-dimensional coordinates of the fingertip at the headset’s refresh rate, typically 90 to 120 hertz, and renders the resulting strokes in real time so the user receives immediate visual feedback.

Under the hood, every signature becomes a multivariate time series. Each sampled point stores a position vector and a velocity vector, the latter computed from the change in position between consecutive frames divided by the elapsed time. Because the hand-tracking cameras on consumer headsets rely on monocular RGB input without dedicated depth sensors, the raw trajectory is noisy. To tame the jitter, the researchers apply linear interpolation, smoothing each sampled position halfway toward the previous one. This acts as a first-order low-pass filter, balancing responsiveness against stability. Two normalization steps then make signatures comparable regardless of where and how the user was facing when signing: the entire trajectory is translated so its first point sits at the world origin, and it is rotated by a quaternion so that the user’s initial viewing direction aligns with a canonical axis. The result is a signature that can be drawn anywhere, at any orientation, and still be matched against stored templates.

Comparing two signatures captured at different speeds and lengths is the central algorithmic challenge, and it is where Dynamic Time Warping enters the picture. Simple Euclidean distance fails catastrophically when two otherwise identical trajectories are temporally shifted, because points that should correspond are compared with the wrong partners. DTW solves this by warping the time axis of one sequence to find an optimal non-linear alignment with the other, minimizing the cumulative local distance along an admissible warping path that respects boundary, continuity, and monotonicity constraints. The method’s key property for signature verification is its invariance to time shifts: identical but misaligned sequences yield a distance at or near zero. Classic DTW carries a quadratic cost in both time and memory, however, which is prohibitive on a battery-powered headset, so the team turned to FastDTW, a multilevel approximation that coarsens the series, projects a low-resolution warping path upward, and refines it within a constrained radius, achieving linear time and space complexity.

The engineering choices reveal how much practical constraint shapes such systems. An early client-server design was abandoned in favor of fully on-device processing, with signature data saved directly to the headset’s local file system. The DTW computation was rewritten in Rust after benchmarks showed the original Python implementation took 2.37 seconds to compare a 383-point signature with itself, while the Rust version finished in 0.03 seconds, roughly 79 times faster. On a tenfold larger sample of 3,830 points, the gap widened to 480-fold, with Rust completing in 0.13 seconds what Python needed more than a minute to do. Further optimizations, including a Sakoe-Chiba band that restricts the warping path to a diagonal window and a diagonal DTW variant that slashes memory use, cut processing time by more than half on a 70,000-point signature. Classification also runs asynchronously on a background thread, preventing the rendering loop from freezing, a detail that matters enormously in VR, where a stalled frame that no longer matches head movement can induce motion sickness.

Even the choice of which features to feed the matcher mattered. An initial feature set combining position, velocity, and acceleration produced puzzling misclassifications: a handwritten name and an arbitrary wave pattern were judged suspiciously similar to a figure eight, with average DTW distances low enough to slip beneath the acceptance threshold. Ablation analysis showed that acceleration data was injecting noise rather than discriminative power, so the team stripped it out, keeping only position and velocity. The revised feature set produced a much clearer separation between similar and dissimilar signatures, eliminated the false positives, and, as a bonus, shrank the data files and simplified the computation. It is a textbook example of how more sensors and more features do not automatically mean better biometrics.

To evaluate the system, the researchers recruited 28 volunteers, roughly 60 percent male and 40 percent female, aged between 20 and 40, none of whom had meaningful prior headset experience. Each participant produced 25 signatures under two input modalities, controllers and hand tracking, and two experimental conditions. In the fixed-signature scenario, everyone wrote the same word, AIRSIGNX, deliberately stressing the system’s ability to distinguish people who share an identical signature shape. In the custom-signature scenario, each user chose a personal symbol or name, mimicking real-world variability. The team measured false acceptance and false rejection rates across swept thresholds and derived Equal Error Rates, the operating point where the two error rates coincide. Custom signatures proved markedly easier to verify, with EERs as low as 0.030 to 0.061 depending on input method, corresponding to authentication accuracy of roughly 94 to 97 percent. The hostile fixed-signature scenario, which doubles as a simulation of a skilled forgery attack in which an attacker knows the target’s signature shape, yielded EERs between 0.097 and 0.165, still translating to accuracy between roughly 83.5 and 90 percent.

Two further findings stand out. First, when the template gallery was reduced from 20 signatures per user to just 5, error rates rose only marginally, with the custom-signature EER reaching 0.030 with controllers and 0.053 with hand tracking. That stability matters for practical deployment, because it suggests users could enroll quickly without degrading security. Second, controllers consistently outperformed hand tracking on the Area Under the Curve metric, with very large effect sizes in both scenarios, reflecting the spatial jitter, gesture-detection latency, and occasional tracking losses inherent to camera-based hand modeling. Yet for the metric that matters most in practice, the EER on custom signatures, the difference between input methods was statistically negligible. In other words, when inter-class variability is high, the noisier but more convenient hand-tracking mode authenticates just as reliably, especially after smoothing. Users themselves praised the pinch-to-write gesture as natural and ergonomic, with the only recurring complaint being a slight lag in detecting fingertip contact at the start and end of strokes.

The security analysis is framed with appropriate caution. The researchers consider three threat levels: zero-knowledge attacks by random impostors, captured by the custom-signature EER; shape-aware attacks by forgers who know what the victim writes, captured by the fixed-signature EER; and dynamics-aware attacks, which they defer to dedicated future adversarial testing. Including velocity profiles in the DTW calculation is hypothesized to resist spatial replay attacks, since replicating how fast a signature was drawn is substantially harder than copying its shape, but the team is explicit that this hypothesis still needs rigorous testing. On the performance side, the entire verification cycle, including normalization, feature extraction, and DTW matching, completes within 350 milliseconds and adds less than 2 percent power draw on a standalone Meta Quest 2, confirming that biometric authentication can run continuously without draining the battery or degrading frame rates.

The authors are candid about the study’s limits. Twenty-eight participants cannot represent the full demographic spectrum, all data was collected in a single session, and no head-to-head comparison with classifiers such as support vector machines or deep learning architectures was possible, partly because variable-length 3D time series resist fixed-length feature extraction and the sample size is too small to train neural models fairly. They position AirSign accordingly, not as a novel classification algorithm but as a systems-level contribution: the first demonstration that unconstrained, on-device, three-dimensional signature verification is feasible on ordinary consumer VR hardware. Future work will scale the user cohort, parallelize the DTW computation across CPU cores, refine index-thumb contact detection, mount a proper defense against forged signatures, and benchmark the approach against gaze, motion, and keystroke biometrics. If those steps succeed, the awkward floating keyboard may finally give way to something far more intuitive: simply signing your name into the air.

Subject of Research: VR-based 3D hand-written signature recognition for biometric user authentication in virtual reality

Article Title: AirSign: An innovative VR-based hand-written signature recognition

Article References: Frascarelli, G., Marini, M. R., De Marsico, M., Mecca, A., & Cinque, L. (2026). AirSign: An innovative VR-based hand-written signature recognition. Multimedia Tools and Applications, 85(10), Article 790. https://doi.org/10.1007/s11042-026-21951-x

Image Credits: AI Generated

DOI: 10.1007/s11042-026-21951-x

Keywords: virtual reality, biometrics, signature recognition, Dynamic Time Warping, hand tracking, authentication, Meta Quest, behavioral biometrics, Rust, Unity, human-computer interaction, cybersecurity

The core problem the researchers set out to solve is one that anyone who has followed a paced-breathing app will recognize. Most existing respiratory guidance systems are frequency-driven: they display or play a rhythm, perhaps a rising and falling visual circle, and assume the user’s chest is rising and falling in sympathy. They do not measure whether the shape of the user’s breath, including the crucial ratio of inhalation to exhalation and the relative durations of each phase, actually matches the intended waveform. That matters because decades of psychophysiological research suggest that these waveform details, not just the breaths-per-minute count, influence outcomes such as heart rate variability and perceived stress. Meanwhile, the systems that do measure breathing usually rely on contact sensors, such as chest straps or nasal cannulas, which are uncomfortable over long sessions and add a layer of physical intrusion that is at odds with the goal of relaxation.

The new system replaces both the guesswork and the sensors with a contactless measurement layer. Millimeter-wave radar, operating at wavelengths of a few millimeters, can detect the tiny chest-wall displacements produced by breathing from a short distance away, without any device attached to the body. The technology has already been used to monitor vital signs in premature infants, sleeping adults and drivers, and the study builds on that lineage: FMCW and Doppler radar architectures can resolve respiration with enough fidelity to reconstruct not just breathing rate but the full shape of the breathing cycle. In this work, the radar’s reflected signal is processed to extract the respiratory waveform, which then becomes the raw material for the system’s feedback engine.

Personalization is the second pillar of the strategy. Rather than imposing a generic breathing template, the researchers recorded resting breathing and the well-known 4-7-8 pattern, in which inhalation lasts four counts, breath-holding lasts seven and exhalation lasts eight, from twenty participants. From these recordings, they constructed an individualized progressive guidance curve for each person using a sinusoidal-based segmented motion model. The parameters of this model were fitted with the Grey Wolf Optimizer, a bio-inspired metaheuristic algorithm that searches a parameter space by mimicking the cooperative hunting behavior of wolf packs. The idea is to start each user from a breathing pattern close to their natural resting rhythm and gradually shape it toward the target pattern, reducing the abruptness and discomfort that can come with asking someone to adopt an unfamiliar waveform on the first try.

The centerpiece of the paper is the Breathing Consistency Score, or BCS, a real-time numerical measure of how well the user’s current breathing matches the personalized target. Computing such a score is trickier than it sounds, because two breathing cycles may have the same overall shape but differ in the timing of individual phases, and a naive comparison can miss or misattribute those differences. The team developed a segmented dynamic time warping algorithm, a technique borrowed from signal processing that aligns two time series even when they drift in timing, and added a duration-deviation penalty to it. The penalty makes the score sensitive to errors in the relative lengths of inhalation, breath-hold and exhalation phases, which are precisely the features that respiratory physiology research has linked to autonomic effects. The BCS then serves as the feedback signal that drives the guidance, telling the user, effectively, how close their breath is to the intended curve at any moment.

To test whether this consistency feedback actually helps, the researchers ran a controlled crossover experiment, meaning each participant experienced both the BCS-driven guidance and the conventional frequency-driven guidance, with the order counterbalanced. The team evaluated three outcomes: guidance efficiency, measured as the time it took users to acquire the target breathing pattern; waveform-following consistency, measured by how closely the user’s breath matched the target; and stress reduction, assessed physiologically. Subjective comfort was additionally collected through open-ended feedback from participants after each session. This design allowed each person to serve as their own control, strengthening the comparison between the two guidance strategies.

The results favored the radar-scored approach across the board. BCS-driven guidance significantly improved waveform-following consistency, meaning participants’ actual breaths tracked the personalized target curves more closely. It also reduced the time needed to reach the target pattern, so users spent less of each session struggling to get in sync and more time in the beneficial state. Physiological stress measures fell more under the BCS-driven condition than under frequency-driven guidance. In the open-ended feedback, most participants reported that the consistency-based system felt more comfortable than the conventional alternative, an important practical point for any technology that hopes to be used regularly, especially by people for whom relaxation does not come easily.

The clinical implications reach beyond the laboratory. Slow-paced and structured breathing practices, including the 4-7-8 technique tested here, have accumulated substantial evidence for reducing stress and anxiety, and device-guided slow breathing has been shown in separate work to decrease sympathetic nervous reactivity in people with post-traumatic stress disorder. But adherence and technique are persistent weak points: if a user breathes along only roughly with an audio cue, the intended phase ratios, which drive much of the cardiac effect, may never be achieved. A contactless system that verifies the waveform and adapts the target to the individual could make breathing training more effective for people with anxiety disorders, PTSD, or everyday work-related stress, without requiring chest straps, electrodes or mouthpieces. It could also lower the barrier to long sessions, since comfort complaints about contact sensors are a known limitation of prolonged use.

The technological ingredients assembled here also hint at where the field is heading. Millimeter-wave radar is already inexpensive and compact enough to appear in consumer devices, and it works through clothing and in darkness, making it a natural fit for homes, clinics and vehicles alike. Pairing radar sensing with optimization algorithms like the Grey Wolf Optimizer and alignment techniques like dynamic time warping gives engineers a toolkit for closing the loop between measurement and instruction: the system knows what you are doing, knows what you should be doing, and adjusts the guidance accordingly. The authors suggest this real-time consistency feedback could help individuals with stress-related conditions make better use of respiratory guidance for self-regulation, turning breathing exercises from a matter of faith into a matter of measured practice.

Caveats remain, as they do with any early-stage study. The experiment involved twenty participants, and broader and more diverse cohorts will be needed to confirm how the system performs across ages, body types and clinical populations, and how it copes with the movement artifacts of daily life rather than controlled laboratory conditions. Still, the study addresses a genuine gap at the heart of a booming wellness category. Guided breathing has long asked users to trust an instruction they cannot verify against their own body. By adding a radar eye and a consistency score, the researchers have given that trust something measurable to rest on, and the evidence suggests the payoff comes as faster learning, better technique and deeper calm.

Subject of Research: Contactless millimeter-wave radar-based personalized respiratory guidance using a real-time breathing consistency score

Article Title: Personalized respiratory guidance strategy via millimeter-wave radar-based breathing consistency score

Article References: Yang, C., Wang, R., Zhao, H., Wei, S., Yu, G., Li, J., & Liu, C. (2026). Personalized respiratory guidance strategy via millimeter-wave radar-based breathing consistency score. Medical & Biological Engineering & Computing. https://doi.org/10.1007/s11517-026-03680-3

Image Credits: AI Generated

DOI: 10.1007/s11517-026-03680-3

Keywords: millimeter-wave radar, breathing guidance, Breathing Consistency Score, 4-7-8 breathing, stress reduction, personalized breathing training, dynamic time warping, Grey Wolf Optimizer, respiratory monitoring, anxiety, PTSD, biofeedback

The problem is simple to state. DTW works by constructing a grid in which one time series runs along the horizontal axis and the other along the vertical axis, then searching through that grid for the cheapest path from one corner to the opposite one, where the cost of each step measures how poorly the two signals match at that point. The result of this search is a single number, the total cost of the optimal path, and that number is what nearly every pipeline keeps. The path itself, a winding trace that encodes precisely where the two signals agree perfectly, where they drift apart, and where they must be violently stretched to line up, is discarded as an intermediate artifact. For clinical gait analysis, researchers now argue, this is a costly habit, because the local alignment structure of the path is where the biomechanical story actually lives.

In a paper published in the journal Data Mining and Knowledge Discovery, a team at Université Bourgogne Europe in Dijon, France, introduces DiaSeg, a framework designed to rescue that discarded information. Led by Tresor Y. Koffi, with Amel Hidouri, Corentin Legrand and Aurélie Bertaux as co-contributors, the method takes the optimal warping path produced by DTW and decomposes it into a sequence of diagonal segments, the stretches where the two signals track each other closely and the path climbs steadily along the diagonal of the alignment grid. Breaks between these segments, where the path veers horizontally or vertically to absorb a timing mismatch, are allowed but controlled, so that the segmentation captures meaningful deviations rather than fragmenting into noise.

Each extracted diagonal segment is then characterized by five geometric features, capturing properties such as its length and its position within the alignment. The crucial design choice is that these features require no domain-specific engineering: no biomechanics expert needs to define what a heel strike or a swing phase looks like in advance. The framework turns a black-box distance measure into a vocabulary of temporal primitives that can be fed directly into unsupervised learning algorithms, letting the data reveal its own structure. The source code implementing the entire pipeline has been made publicly available by the authors, lowering the barrier for other groups to adopt and extend the approach.

To test whether this recovered structure actually means anything clinically, the team validated DiaSeg on data from 91 subjects spanning six clinical conditions: healthy aging, Parkinson’s disease, Huntington’s disease, amyotrophic lateral sclerosis, brain tumor, and stroke. This is a demanding test bed, because these conditions damage the nervous system in very different ways and would be expected to disrupt gait coordination at different points in the walking cycle. If the diagonal segments carried no real information, one would expect their patterns to look like noise. Instead, the opposite happened.

The first finding is that diagonal segments form consistent unsupervised patterns even when the algorithm is given no labels at all. Clustering the segment features produced groupings with a silhouette score of 0.33, a moderate but meaningful level of structure, and these data-driven groupings aligned with biomechanical phase annotations supplied independently by domain knowledge. More striking still, external validation confirmed a near-perfect separation of healthy and pathological gait, with an adjusted Rand index reaching 0.986, a measure of clustering agreement where values close to one indicate almost identical partitions. In other words, without being told who was sick and who was well, the method recovered a structure that essentially matched the clinical labels.

The second finding concerns how pathology actually expresses itself in these features. When used in a supervised setting, the segments discriminated between conditions with an accuracy of 69 percent, and unsupervised clustering at the patient level reached 75 percent. But the researchers observed something subtle: disease did not typically manifest through the properties of any single segment. Instead, pathology appeared as a distributional shift in segment length, an alteration in the overall statistics of how coordination is parcelled across the gait cycle rather than a single dramatic defect. This suggests that neurological disorders degrade movement in a diffuse, statistical way, reshaping the whole texture of the walk rather than breaking one identifiable component. When segment-level features were combined with conventional cycle-level features, classification accuracy climbed to 91.7 percent.

The third finding addresses an honest trade-off. Purely cycle-based methods, which summarize an entire stride into a single representation, achieved a slightly higher accuracy of 91 percent in this study. So why bother with segments at all? The answer is interpretability. A global representation can tell a clinician that two walks differ, but it cannot say where they differ. Diagonal segments, by contrast, provide phase-specific localization: they can pinpoint where within the gait cycle coordination breaks down, which is precisely the kind of information a neurologist or physical therapist needs when assessing a patient with Parkinson’s disease or recovering from a stroke. The two approaches are therefore complementary rather than competing, one optimizing raw discrimination and the other optimizing explanation.

The implications extend beyond gait. DTW is arguably the dominant similarity measure for time series across domains, from speech and handwriting to sensor streams and financial data, and the standard practice of discarding the warping path is equally ubiquitous in all of them. DiaSeg demonstrates a general recipe: extract the diagonal structure of the alignment, characterize it geometrically, and use those primitives for pattern discovery. The work also connects to a broader tradition in signal analysis, including recurrence plots and their quantification, which likewise seek to mine the internal geometry of comparison operations rather than collapsing everything into a single scalar. There is a growing recognition in the machine learning community that interpretability and accuracy need not be enemies, and this study offers a concrete, tested example of having both.

For the clinic, the near-term promise is a diagnostic and monitoring tool that can explain itself. Neurodegenerative diseases such as Parkinson’s and Huntington’s are progressive, and clinicians increasingly rely on quantitative movement measures to track their course and evaluate interventions, including emerging tele-rehabilitation approaches. A framework that localizes where coordination degrades within each stride, and that does so without hand-crafted features, could make such monitoring more informative and more transparent. The work emerged in part from the ENABLE project, a European collaborative effort on evaluating motor capacities and tele-rehabilitation in children with neuromotor disorders, underscoring the applied motivation behind what might otherwise look like an abstract data mining contribution. The method has limitations that future work must address, including its dependence on the quality of the underlying DTW alignment and the moderate size of the validation cohort, but the central message stands. The warping path that scientists have been discarding for nearly half a century turns out to be not a by-product, but a map, and DiaSeg provides the first systematic guide to reading it.

Subject of Research: Interpretable gait analysis using diagonal segment extraction from Dynamic Time Warping paths for neurological disease assessment

Article Title: DiaSeg: diagonal segment extraction from DTW paths for interpretable gait analysis

Article References: Koffi, T. Y., Hidouri, A., Legrand, C., & Bertaux, A. (2026). DiaSeg: diagonal segment extraction from DTW paths for interpretable gait analysis. Data Mining and Knowledge Discovery, 40(6), Article 107. https://doi.org/10.1007/s10618-026-01276-x

Image Credits: AI Generated

DOI: 10.1007/s10618-026-01276-x

Keywords: Dynamic Time Warping, gait analysis, DiaSeg, time series clustering, unsupervised pattern discovery, neurodegenerative disease, interpretable machine learning, Parkinson's disease, data mining, biomechanics, diagonal segments, pattern recognition

The research, led by Thomas Seacrist, Elizabeth E. Walshe, David Grethlein, Megan S. Ryerson, Flaura K. Winston and colleagues at the Children’s Hospital of Philadelphia and partner institutions, demonstrates that time-series clustering, an unsupervised machine learning technique, can automatically identify distinct visual search strategies from eye-tracking recordings collected during simulated driving. Instead of requiring human coders to watch hours of video and label where each driver looked, the approach compares entire streams of gaze data directly and groups drivers whose scanning patterns resemble one another.

The significance of this shift is hard to overstate for the field. Traditionally, characterizing a visual search strategy has meant video coding: trained analysts review synchronized footage of the road scene and the driver’s eyes, mark fixations on areas of interest, and translate those marks into summary measures. The process is labor-intensive, subjective at the margins, and effectively caps the size of datasets a lab can analyze. That cap matters because the consequences of insufficient visual search are severe. A driver who fails to scan far enough ahead may miss a pedestrian stepping out from behind an obstructed view, or may enter a blind curve without the anticipatory glances that experienced drivers deploy.

To test their method, the team collected eye-tracking data from 36 young drivers aged 16 to 24 as they completed a virtual driving assessment, a validated simulated driving test used to gauge performance behind the wheel. The cohort is deliberately focused on a high-risk population. Crash rates among newly licensed and teenage drivers remain dramatically elevated compared with older, experienced motorists, and errors in scanning and hazard anticipation figure prominently among the mistakes that precede serious crashes involving young novices.

At the heart of the method is a measure called dynamic localized coordinate aligned warping, abbreviated DCLAW, an extension of the well-known dynamic time warping algorithm. Dynamic time warping, originally developed for aligning spoken word recordings, allows two time series that unfold at different speeds to be compared by stretching and compressing them along the time axis. DCLAW adapts this idea to eye-tracking data, which arrive as rapidly sampled coordinates of gaze position and rarely align neatly between one driver and another. Two drivers might sweep their eyes across the same sequence of road regions but at slightly different moments or paces; a naive comparison would call them dissimilar, while DCLAW can recognize the underlying strategy as essentially the same.

Once pairwise similarities between all drivers’ raw gaze streams had been computed, the researchers applied k-medoids clustering, an unsupervised algorithm that partitions data into groups organized around actual representative examples called medoids. Because the method is unsupervised, it requires no preconceived categories of good or bad scanning behavior. The number and structure of the clusters emerge from the data itself, and the researchers then examined each cluster’s medoid, the driver whose time series sits at the center of the group, to characterize what defined that particular visual search strategy.

The results showed that time-series clustering successfully identified generalizable visual search strategies during a curved roadway scenario, one of the more demanding situations in driving. Curves demand anticipatory glances toward the tangent of the bend and disciplined checking of the lane ahead, and differences in how young drivers allocate attention there are linked to crash risk. The fact that clustering raw data recovered meaningful, generalizable strategy groups in this setting suggests the technique can capture behaviorally important variation without any manual preprocessing of the eye-tracking signal.

The practical implications extend well beyond the driving simulator. Because the method removes the need for labor-intensive manual coding, it opens the door to analyzing far larger and more diverse datasets, the kind of scale needed for findings to generalize across populations, driving environments and research questions. Larger samples could reveal how visual search strategies differ by age, experience, fatigue, distraction or neurological condition, and could support the evaluation of training interventions designed to teach novice drivers to scan more like experts. Similar approaches could prove valuable in aviation, air traffic control, construction safety, medicine and any domain where situational awareness depends on where people look and when.

The study also reflects a broader trend in behavioral science, in which methods developed in the data mining community, including time-series clustering, shapelet-based classification and related techniques, are being redeployed to make sense of rich, high-frequency behavioral recordings. Eye-trackers have become cheaper and more ubiquitous, generating torrents of gaze data that conventional analysis pipelines were never designed to handle. Techniques that operate directly on raw time series, rather than on heavily processed summaries, promise to preserve the temporal structure of behavior that those summaries often discard.

The authors have made their data and code publicly available through the Children’s Hospital of Philadelphia’s GitHub repository, lowering the barrier for other teams to adopt and extend the approach. For a research area long constrained by the slow economics of video coding, the demonstration that raw eye-tracking streams can be clustered into interpretable visual search strategies marks a genuine methodological milestone, one that could accelerate the science of how humans take in the visual world during complex, safety-critical tasks.

Subject of Research: A scalable time-series clustering method for characterizing visual search strategies from raw eye-tracking data in young drivers

Article Title: A scalable method for characterizing visual search strategies: A novel application of time-series clustering to raw eye-tracking data

Article References: Seacrist, T., Walshe, E. E., Grethlein, D., Ryerson, M. S., & Winston, F. K. (2026). A scalable method for characterizing visual search strategies: A novel application of time-series clustering to raw eye-tracking data. Behavior Research Methods, 58(10), Article 290. https://doi.org/10.3758/s13428-026-03154-2

Image Credits: AI Generated

DOI: 10.3758/s13428-026-03154-2

Keywords: eye-tracking, visual search strategy, time-series clustering, machine learning, situational awareness, young drivers, driving simulation, k-medoids, dynamic time warping, hazard perception, Behavior Research Methods, driver safety