In this study, we applied a widely used insect surveillance metabarcoding workflow, modified through development and implementation of aphid-specific primers, to enhance targeted detection of aphid prey consumed by ladybeetles. We evaluated performance of this prey-targeted metabarcoding approach for dietary analysis and quantified aphid DNA detectability under controlled feeding and starvation regimes. We hypothesised that aphid DNA could be reliably recovered from whole ladybeetle samples using standard surveillance metabarcoding protocols, enabling detection of prey taxa present exclusively as gut contents, which may impact surveillance and diagnostics programs.

Aphid DNA from multiple species was detected up to seventy-two hours post-ingestion and up to one hundred hours following cessation of ad libitum feeding, although detectability varied among aphid species and pre-feeding starvation treatments. These results demonstrate that prey-targeted metabarcoding increases sensitivity for dietary inference in predatory insects and provides a robust framework for detecting cryptic trophic interactions. Our findings further indicate that unintended prey detection may be common in arthropod metabarcoding datasets, with important implications for interpretation in biosurveillance and biosecurity monitoring.

The design of the feeding experiments reflects a deliberate attempt to mimic conditions that ladybeetles encounter in real cropping systems. In the short starvation experiment, adult spotted amber ladybeetles were first allowed to gorge on corn leaf aphids for nine days, simulating a period of high pest abundance in the field, before being starved for seventeen hours and then offered a small number of aphids from a different species. This structure allowed the researchers to distinguish DNA from a recent, quantifiable feeding event from DNA already circulating in the gut from earlier meals. Because the corn leaf aphid was the colony species used for pre-feeding, the alternative prey species chosen for the experimental meal, namely Russian wheat aphids and sow thistle aphids, could be tracked independently through the metabarcoding results. Each ladybeetle was given one and a half hours to feed on five aphids, a window chosen to represent a realistic foraging bout, and the number of aphids consumed was recorded so that detectability could be interpreted in light of actual prey intake.

Sampling at zero, one, two, four, and seventy-two hours after feeding provided a coarse but informative decay curve for aphid DNA inside the predators. The gap between the four-hour and seventy-two-hour timepoints was a logistical compromise, and it means that the precise shape of the detectability decline during that interval remains unresolved. Nevertheless, the recovery of aphid DNA at the seventy-two-hour mark is notable, because it demonstrates that gut contents can remain molecularly detectable for days after a feeding event, long after any physical prey remains would have been processed. For surveillance programs that bulk-collect arthropods into ethanol and sequence them later, this persistence means that a predator caught in a trap may carry molecular evidence of meals consumed well before capture, effectively extending the temporal window of dietary inference far beyond what visual gut dissection could achieve.

The long starvation experiment extended this logic by removing the pre-feeding period altogether. Ladybeetles were starved for ninety-six hours before being offered either corn leaf aphids or cabbage aphids, and sampling continued at zero, one, two, four, twenty-four, forty-eight, and ninety-six hours after feeding. This denser sampling regime captured the early hours of digestion in much finer detail, and the comparison between the two experiments revealed that pre-feeding starvation history materially alters detectability. A hungry predator that has not fed for days may process prey differently than one that has recently gorged, and the resulting differences in DNA decay rates have direct consequences for how field-collected predators should be interpreted. A detection of pest DNA in a beetle starved for several days may indicate a more recent meal than the same detection in a well-fed beetle, complicating any attempt to timestamp predation events from molecular data alone.

Species-level variation in detectability emerged as a consistent theme across both experiments. Aphid DNA from different prey species did not decay at identical rates, which means that the absence of a particular aphid species from a metabarcoding result cannot be taken as evidence that the predator never consumed it. Some prey may simply become undetectable sooner than others, whether because of differences in body size, cuticle thickness, gut chemistry, or the copy number of the target genetic region within the prey genome. This species-specific bias is a well-recognised challenge in dietary metabarcoding generally, but the present study quantifies it in a controlled setting for aphids specifically, providing empirical grounding for interpreting detections and non-detections in agricultural surveillance datasets.

The choice of the spotted amber ladybeetle as the primary predator reflects its status as a known and commercially available biocontrol agent in Australian grain ecosystems. Eggs and adults can be sourced commercially, which makes this species a practical candidate for release programs and, by extension, a logical focal species for post-release monitoring. The inclusion of the large spotted ladybeetle, collected from garden plants near the research campus, added a comparative dimension, allowing the researchers to ask whether detectability patterns generalise across coccinellid species or whether they are idiosyncratic to particular predator-prey pairings. Frass from both adult and larval spotted amber ladybeetles was also sampled, opening the possibility that non-invasive sampling of predator droppings could substitute for, or complement, the destruction of whole specimens during DNA extraction.

The prey assemblage used in the experiments spans a meaningful slice of aphid diversity relevant to Australian grains. Corn leaf aphids and Russian wheat aphids were drawn from laboratory colonies maintained on caged barley plants under controlled temperature and light conditions, ensuring a consistent prey quality and genetic background. Bird cherry-oat aphids, another economically significant grain pest, were also colony-reared. In contrast, cabbage aphids and sow thistle aphids were collected from field mustard and sow thistle plants on the university campus only an hour before the feeding trials began, introducing wild-caught prey with their own microbiomes and genetic variation. California maple aphids, collected from a Japanese maple in a regional botanic garden, served as an additional positive control, broadening the taxonomic coverage of the primer validation. Wild psyllids collected from an Acacia tree on the same campus provided a non-target feeder insect, allowing the researchers to test whether the aphid-specific primers would inadvertently amplify DNA from related hemipterans that a ladybeetle might encounter in the field.

The broader significance of this work lies in what it reveals about the interpretation of existing metabarcoding datasets. Insect surveillance programs that use universal primers on bulk trap samples routinely generate taxonomic lists that may include species for which no physical specimen can be found during visual examination of the sample. Previous work comparing metabarcoding results with morphological identification of trapped arthropods has documented such incongruencies, and the present study offers a plausible mechanistic explanation: at least some of these phantom detections may represent prey DNA inside the guts of predators that were themselves caught in the trap. A ladybeetle full of aphid DNA contributes aphid reads to a bulk sample even though no aphid specimen is physically present, and this can inflate apparent aphid diversity or abundance in a way that misrepresents the actual community composition at the collection site.

For biosecurity monitoring, this distinction matters considerably. If a pest species is detected molecularly in a trap sample, the response may differ depending on whether that detection reflects a living population at the site or merely the gut contents of a predator that fed elsewhere and flew in. The detectability windows quantified here, extending to seventy-two hours after a discrete feeding event and up to one hundred hours after the cessation of sustained feeding, provide a temporal framework for reasoning about this problem. A predator carrying detectable pest DNA consumed that pest within roughly the last few days, which constrains but does not eliminate the geographic uncertainty associated with the detection. Combined with knowledge of predator dispersal distances and local pest phenology, such temporal bounds could help biosecurity officers triage molecular detections and decide which warrant follow-up physical inspection.

For biological control practitioners, the ability to confirm consumption of target pests by released agents addresses a persistent gap in post-release evaluation. Traditional approaches to measuring biocontrol efficacy rely on cage trials, sentinel prey, or population counts before and after release, all of which are labour-intensive and subject to confounding by predator movement and prey dynamics. Molecular dietary analysis offers a complementary line of evidence: if released ladybeetles can be recaptured and shown, through metabarcoding of their gut contents, to have consumed the target aphid species, this provides direct confirmation that the biocontrol agent is engaging with the pest under field conditions. The aphid-specific primers developed in this study increase the sensitivity of such confirmations, particularly for prey that are rare in the diet or consumed in small numbers.

The methodological workflow itself, adapted from a widely adopted Australian insect surveillance metabarcoding protocol, is significant because it demonstrates that existing infrastructure can be repurposed for dietary analysis with relatively modest modification. Rather than requiring a wholly new analytical pipeline, the addition of taxon-specific primers to an established surveillance workflow allows the same laboratory processes that identify community composition to simultaneously probe trophic interactions. This dual capability means that a single sequencing run on a bulk trap sample can yield both a census of the arthropods present and, for predatory specimens within that sample, evidence of what they have been eating. As metabarcoding becomes further embedded in routine agricultural diagnostics, such multiplexed utility increases the value derived from each sample processed.

Finally, the controlled experimental framework established here, with defined feeding events, quantified prey numbers, staged starvation periods, and time-series sampling, provides a template that can be extended to other predator-prey systems. Detectability periods are known to vary with both predator and prey identity, and the factors governing that variation, including prey abundance, predator metabolic rate, and the interplay between feeding rate and starvation history, remain only partially characterised. By quantifying detectability under controlled conditions for a biocontrol-relevant predator and a suite of grain pest aphids, this study contributes empirical parameters that can inform the design of future field studies and the interpretation of molecular surveillance data across agricultural landscapes.

Subject of Research: Metabarcoding and Metagenomics

Article Title: A feast of aphids: assessing the detectability of prey aphid DNA inside ladybeetle predators using taxon-specific metabarcoding

Article References: Willmott, N. J., Scanlan, J. L., Blacket, M. J., Rodoni, B. C., Nancarrow, N., Jones, T. M., & Martoni, F. (2026). A feast of aphids: assessing the detectability of prey aphid DNA inside ladybeetle predators using taxon-specific metabarcoding. Metabarcoding and Metagenomics, 10, Article e196568. https://doi.org/10.3897/mbmg.10.196568

Image Credits: AI Generated

DOI: 10.3897/mbmg.10.196568

Keywords: metabarcoding, ladybeetles, aphids, biological control, integrated pest management, feast, assessing, detectability, prey, aphid, inside, ladybeetle

The research, published in Frontiers in Amphibian and Reptile Science, evaluated a real-world contracted iguana removal effort rather than an artificial experiment, a choice the authors say was essential for understanding what management can actually accomplish. The team, led by Frank J. Mazzotti, Sergio A. Balaguera-Reina, Sidney T. Godfrey, Nathan P. Schwartz, and Melissa A. Miller from the University of Florida’s Fort Lauderdale Research and Education Center, focused on two roughly two-kilometer, unconnected sections of the C-11 Canal in Broward County. One section served as the treatment population, where removals took place; the other served as a control, where iguanas were counted but left alone. The sections were spaced 438 meters apart, separated by a stretch of canal and a street with no vegetation and no iguanas, allowing the researchers to treat them as distinct populations for the purposes of their surveys.

Green iguanas are native to a range stretching from northern Mexico through southern Brazil and portions of the Lesser Antilles, but in Florida they are an established invader with documented adverse effects. They damage native, ornamental, and agricultural plants, disperse non-native plant seeds, foul patios, pools, and canals with feces, and burrow into infrastructure such as runways, seawalls, and levees. Florida hosts more nonnative reptiles and amphibians than anywhere else in the world, with at least 180 introduced species and more than 60 established breeding populations. The green iguana’s expanding numbers, which grew after a historic freeze event in 2010, prompted Florida to designate the species as Prohibited on 29 April 2021, restricting possession and sale with few exceptions.

Conventional removal methods, including hand grabbing, live traps, noosing, tongs, netting, and shooting, are typically applied during the day when iguanas are active. But daytime methods are best suited to small, localized populations. For the large populations found on public lands managed for flood control and water supply, the researchers turned to a nocturnal approach that takes advantage of the iguanas’ tendency to rest in trees at night, especially over water, where sleeping animals are visible and vulnerable. Between mid-February and mid-May 2018, the team conducted 30 nocturnal removal surveys by boat and by truck and foot in the treatment section, using 900-lumen headlamps and a 1,000-lumen tactical flashlight to locate iguanas in the canopy before attempting removal with extendable poles, tongs, nets, or by hand.

The effort removed a total of 808 iguanas, with a mean of 27 per survey and a range of 3 to 106 per night. Of these, 168 were adult males, 299 were adult females, 104 were juvenile males, 233 were juvenile females, and 3 were of unknown sex. No hatchlings were captured because the removal window fell outside the May-to-August hatching season in Florida. Overall, 66 percent of removed iguanas were female and 58 percent were adults. The mean snout-vent length of removed animals was 22.5 centimeters, mean total length was 72.7 centimeters, and mean mass was 0.46 kilograms, with the largest individual reaching 3.40 kilograms. All iguanas were euthanized immediately after removal in accordance with Florida state law on Prohibited species and under an approved University of Florida animal care protocol.

To evaluate whether these removals actually reduced abundance, the researchers conducted 20 diurnal count surveys by boat across four periods: before removals in January and February 2018, immediately after in May 2018, three months after in August 2018, and twelve months after in May 2019. Across all surveys they observed 4,741 green iguanas, with the highest counts during the pre-removal period, when 937 iguanas were seen in the control section and 953 in the treatment section. Immediately after the removal campaign, counts in the treatment section dropped to 328, the lowest observed for that population during the study, while the control section remained comparatively high. Three months later, the treatment section had begun to rebound, and by twelve months the number of iguanas observed in the treatment and control sections was no longer significantly different.

Statistical modeling using generalized linear mixed models revealed a pattern that complicates straightforward interpretation. Julian date, a proxy for seasonality, was the only variable with a significant negative effect on daytime iguana counts, predicting a reduction of at least one iguana observed per day across the sampled year. Critically, the control population also showed a substantial seasonal decline, falling from 761 to 412 iguanas between the first and final survey periods, a reduction not as extreme as the treatment population’s but still substantial. This suggests that the biology of invasive iguanas, including seasonal patterns in detectability and activity, plays a major role in apparent abundance, independent of any management intervention. The authors caution that catch-per-unit-effort, a common index of abundance in removal studies, can be misleading when detectability varies with season, wariness, or environmental conditions.

The removal data themselves showed clear signs of diminishing returns. Mean catch-per-unit-effort was 5.29 iguanas per hour and 5.49 iguanas per kilometer, but per-survey values decreased from 13.93 iguanas per hour during the first survey to 8.19 during the last. Capture probability, the proportion of detected iguanas actually removed, averaged 0.21 and declined from 0.51 in the first survey to 0.21 in the last, possibly reflecting increasing wariness among surviving animals. The researchers also noted practical constraints: adult iguanas perched beyond the reach of extendable poles or hidden in burrows along the canal banks escaped removal, and the proximity of vehicles, walkways, and buildings ruled out more effective methods such as firearms.

Comparisons with other invasive species management efforts underscore the challenge. Research on green iguanas in the Cayman Islands found that harvest rates below 60 percent can paradoxically cause populations to increase rapidly, as reduced density releases survivors from density-dependent factors regulating survival and fecundity. Studies of invasive tegus in Florida found that a removal rate of 40 percent was needed to decrease the population, with rates of 30 percent allowing growth. The relatively small proportion of iguanas removed in the Broward County study, combined with the capture probability of roughly 21 percent, could explain the rebound in counts observed twelve months after the removal campaign ended.

The authors argue that one-time removal efforts, or repeated efforts conducted over a short duration, are insufficient to control population growth, and they recommend continuous, persistent removal combined with independent monitoring surveys designed to explicitly calculate detection probability, for example through N-mixture models. They also emphasize that removal is a management action, not a management objective in itself; success should be measured against goals such as population reduction, containment, mitigation of impacts on native species, prevention of infrastructure damage, or reduced risks to human health and safety. Targeted removal during the breeding season, when nesting females and their eggs can be taken together, and during the May-to-August hatching season, when young iguanas emerge in large numbers, could improve efficiency, as could supplementing nocturnal removal with burrow targeting, habitat alteration, and exclusion. But the deeper lesson, the researchers conclude, is that fundamental knowledge gaps in the population ecology of green iguanas in Florida, including movements, habitat use, nesting, vital rates, and phenology, currently hinder the development of effective management strategies. Closing those gaps, they argue, is essential to managing this well-established and problematic invader.

Subject of Research: Evaluation of nocturnal bulk removal as a control method for invasive green iguana populations in South Florida

Article Title: Evaluating invasive reptile control efforts: nocturnal bulk removal of green iguanas (Iguana iguana Linnaeus, 1758) in Florida, USA

Article References: Mazzotti, F. J., Balaguera-Reina, S. A., Godfrey, S. T., Schwartz, N. P., & Miller, M. A. (2026). Evaluating invasive reptile control efforts: nocturnal bulk removal of green iguanas (Iguana iguana Linnaeus, 1758) in Florida, USA. Frontiers in Amphibian and Reptile Science, 4, Article 1846355. https://doi.org/10.3389/famrs.2026.1846355

Image Credits: AI Generated

DOI: 10.3389/famrs.2026.1846355

Keywords: green iguana, invasive species, Florida, nocturnal removal, population control, catch-per-unit-effort, wildlife management, Broward County, reptile ecology, detectability, seasonality, Frontiers in Amphibian and Reptile Science

Researchers Ruizhe Huang and Sherrie Wang of the Massachusetts Institute of Technology analyzed more than 1.5 million satellite observations of over 6,000 power plants worldwide, drawing on data from the TROPOspheric Monitoring Instrument, or TROPOMI, aboard the Copernicus Sentinel-5P satellite. Their findings, published in the journal Atmospheric Measurement Techniques, represent the first global, data-driven account of when and where power plant nitrogen dioxide plumes become visible from space. The work reveals that detectability is not simply a function of how much a plant pollutes. Instead, it emerges from an intricate interplay of wind, terrain, surface brightness, and the geometry of the satellite’s view.

The scale of the analysis is unprecedented. For the United States, the team paired TROPOMI observations from 2019 through 2024 with hourly emissions records from the Environmental Protection Agency’s Continuous Emissions Monitoring Systems, covering the 500 largest power sector emitters of nitrogen oxides. Globally, they used annual emission estimates from the CoCO2 catalog for the 6,000 highest-emitting plants, spanning coal, natural gas, oil, biomass, and waste facilities. After quality filtering, the dataset contained 666,222 high-quality US observations and 875,686 global ones, each capturing a single satellite overpass of a specific plant.

A central challenge was teaching a computer to recognize a plume and correctly attribute it to its source. The researchers developed an automated algorithm that masks potential interference from nearby cities and other power plants, uses wind direction to define a downwind search zone, calculates the background nitrogen dioxide concentration from upwind air, and then applies a dual threshold requiring both statistical significance and an absolute minimum signal strength. Validated against 800 manually labeled observations stratified by continent and emission level, the algorithm achieved 98 percent accuracy in both US and global tests, a level of reliability that made the subsequent analysis possible.

Even with this powerful detection tool, the researchers found that geography matters enormously. Among 171 US plants that survived the study’s strict interference filtering, per-plant detectability ranged from 6 percent to 85 percent of overpasses. Globally, among 1,065 plants, the range stretched from zero to 100 percent, with a median of 27.8 percent. The pattern follows a striking geographic logic: detectability is highest in arid and semi-arid regions such as the American interior West, Iberia, North Africa, Greece and Turkey, and eastern Australia, and lowest in humid, cloudy regions where bright surfaces are rare and cloud cover frequently spoils the retrieval.

To untangle the physical causes, the team trained a machine learning model, a multi-layer perceptron neural network, to predict whether a plume would be detectable from 25 variables describing the sensor, the plant, the meteorology, and the environment. The model performed well, achieving F1 scores above 0.66 and areas under the curve exceeding 0.8. Feature importance analysis using permutation scores and SHAP values identified six dominant factors: the nitrogen oxide emission rate, surface altitude, surface albedo in the nitrogen dioxide spectral window, the satellite’s viewing zenith angle, the plant’s primary fuel type, and wind speed.

Some of these relationships confirm long-standing theory. Higher emissions produce more concentrated plumes and higher detection probability, and brighter surfaces improve the signal-to-noise ratio of the retrieval by increasing the amount of light available for absorption spectroscopy. Wind speed works in the opposite direction: strong winds stretch and dilute a plume, lowering the concentration contrast that the satellite must detect. The dependence on viewing angle proved more subtle, with detectability peaking at intermediate off-nadir angles, near 20 degrees globally and 35 degrees for US plants, where the longer optical path through the lower atmosphere amplifies the absorption signal before oblique geometry begins to degrade the retrieval.

The numbers offer a practical benchmark. For US power plants, an average hourly emission rate of roughly 400 kilograms of nitrogen oxides corresponds to about a 50 percent chance of detection, but under different combinations of wind, albedo, and viewing geometry, the same emission rate can yield detection probabilities anywhere from below 20 percent to above 60 percent. The researchers illustrate this with individual facilities: a low-emitting plant with dark surfaces and modest winds reached only 18 percent detectability, while a plant emitting eight times as much, blessed with bright terrain and calm air, achieved 64 percent. Identical pollution, in short, can be seen or missed depending entirely on local conditions.

The study also exposed a sobering limitation of current satellite monitoring. To ensure that detected plumes could be attributed to a single source, the researchers excluded any plant within 20 kilometers of another major power plant or within 45 to 90 kilometers of a large city, depending on the city’s size. This filtering removed a staggering share of the world’s monitoring targets: 82.3 percent of the 6,000 global plants fell inside interference zones, leaving just 21.1 percent of their combined emissions in the analysis. In the United States, where emission sources are less clustered, 45 percent of emissions survived the filter. The finding underscores how difficult it remains to monitor pollution from the densely industrialized regions where it matters most.

The implications reach well beyond this single study. The same variables that govern whether a plume is visible also shape how it appears, meaning that emission-quantification methods could become more accurate by explicitly accounting for wind, albedo, and viewing geometry as auxiliary variables or priors. And as a new generation of satellites comes online, including missions targeting pixel sizes of a few kilometers or even 300 meters, finer spatial resolution should reduce plume dilution and interference, potentially bringing many more of the world’s power plants within reach of space-based monitoring. For now, the study provides the first empirical map of where today’s satellites can and cannot see, a foundation for turning orbital observations into trustworthy global emission accounts.

Subject of Research: Satellite detectability of nitrogen dioxide plumes from power plants

Article Title: Global variability in the detectability of power plant NO2 plumes from space

Article References: Huang, R., & Wang, S. (2026). Global variability in the detectability of power plant NO 2 plumes from space. Atmospheric Measurement Techniques, 19(18), 6099-6124. https://doi.org/10.5194/amt-19-6099-2026

Image Credits: AI Generated

DOI: 10.5194/amt-19-6099-2026

Keywords: nitrogen dioxide, power plants, TROPOMI, satellite remote sensing, air pollution, machine learning, emissions monitoring, plume detection, Sentinel-5P, NOx, surface albedo, detectability

The research builds on a well-known puzzle in clinical psychology. Earlier work by Carol Thompson and colleagues, published in Nursing Research in 2008, showed that when nurses are placed under time pressure, their ability to detect high-risk patient conditions measurably declines. The hit rate for spotting deterioration dropped from 86 percent without time pressure to 68 percent under it, and the key signal detection measure of detectability, known as d-prime, fell regardless of how many years of experience the nurse had. But that landmark study left an important question unanswered: was the decline in accuracy simply the result of a speed-accuracy trade-off, the well-documented tendency of people to sacrifice precision when forced to act quickly, or was time pressure fundamentally altering the perceptual machinery of clinical judgement itself?

To answer that question, the Japanese team designed an ambitious laboratory experiment that treated nursing judgement like a psychophysical signal-detection problem. Twenty-one registered nurses, fourteen of them women with an average of 6.6 years of hospital experience, were recruited through snowball sampling. All had worked on hospital wards or in intensive care units, and all were right-handed. The researchers created eighty twenty-second video stimuli that mimicked the displays of bedside medical monitors, presenting seven vital-sign parameters drawn from the National Early Warning Score 2, or NEWS2, the widely used triage tool that aggregates respiratory rate, blood oxygen, systolic blood pressure, pulse, consciousness level, temperature, and oxygen supplementation into a single risk score.

Half of the eighty datasets represented low-risk patients with aggregate NEWS2 scores between zero and four, while the other half represented high-risk patients scoring seven or eight, the range associated with a substantially elevated probability of cardiac arrest, unplanned intensive care admission, or death within twenty-four hours. Three experienced nurses vetted every stimulus for face validity before testing began. Seated eighty centimeters from a laptop in a quiet, carefully controlled room, with display luminance held at 10.4 candelas per square meter, participants watched each simulated patient and pressed one of two mouse buttons to judge whether urgent intervention was warranted, responding as quickly and accurately as possible and then rating their confidence on a five-point scale.

The headline finding is deceptively simple: nurses assessed the risk of acute deterioration in approximately five seconds, with a mean reaction time of 5,169 milliseconds and an overall accuracy of about 76 percent. That speed is plausible, the authors note, because previous research found clinicians needed roughly six seconds to choose among four diagnostic options after reading a detailed patient case; here the information load was lighter and there were only two response options. Crucially, the measured detectability value of 1.71 landed squarely within the range reported by prior signal detection studies of nursing risk judgement, which have produced d-prime values between roughly 0.98 and 1.75, lending the new paradigm immediate credibility.

But the deeper discovery emerged when the researchers broke performance down by individual NEWS2 score. Repeated-measures analysis of variance revealed that stimuli with a score of zero, meaning every parameter was perfectly normal, produced both faster reactions and higher accuracy than every other score. From this pattern, Hishiya and Ishihara inferred a two-stage architecture for nursing risk perception. The first stage, which they call the normal-abnormal discrimination phase, is a rapid triage of whether anything is wrong at all, completed in under five seconds and apparently mastered by anyone holding a nursing license. The second stage, the risk evaluation phase, kicks in only once an abnormality has been detected, and its duration, estimated as the reaction time difference between score-zero stimuli and everything else, proved to be the real engine of individual performance.

The correlations here are striking. Nurses who took longer in this risk evaluation phase showed significantly lower detectability, with a correlation coefficient of minus 0.52, and a significantly more conservative response bias, with a correlation of plus 0.65. In signal detection terms, d-prime measures how well a perceiver separates true signals from noise, while the criterion C captures the liberal or conservative leaning of their decisions. A liberal nurse over-flags danger, triggering unnecessary interventions that can divert scarce staff from patients who genuinely need them; a conservative nurse under-flags, and a deteriorating patient may slip through unnoticed. The finding that the time spent evaluating abnormal stimuli predicts both accuracy and bias suggests that this evaluative stage, not basic abnormality detection, is where clinical judgement quality is forged, and where two of the twenty-one participants even showed the phases running in parallel rather than in sequence.

Notably, the study found no evidence of a speed-accuracy trade-off across individuals. The correlation between d-prime and overall reaction time was weakly negative and statistically non-significant, which the authors interpret as evidence that the accuracy losses previously observed under time pressure reflect genuine changes in perceptual-behavioral information processing caused by the situation, rather than a deliberate slowing-down trade. This distinction matters enormously for patient safety: if time pressure corrupts the processing pipeline itself rather than merely shifting a strategic dial, then interventions must target the processing environment, not just train nurses to be more careful when rushed.

Equally intriguing were the physiological results. The researchers attached silver-silver chloride electrodes to participants’ palms and forearms and recorded skin potential level, a measure of sympathetic nervous system arousal, sampled continuously at 1,000 hertz throughout the task. While arousal levels did not correlate with accuracy or speed, a striking pattern emerged around experience: nurses who reported frequent subjective involvement with acute deterioration showed significantly lower normalized skin potential than their colleagues with rare exposure. In other words, nurses who regularly face patient crises appear to run their judgement machinery at a calmer physiological register, echoing earlier findings that expert nurses maintain stable heart rates during clinical decisions while novices do not, and hinting at a low-arousal, low-cognitive-load strategy honed by repeated real-world exposure to emergencies.

Perhaps the most socially resonant discovery involves personality. Among the five basic traits measured with the Ten Item Personality Inventory, only agreeableness correlated with response bias, and it did so positively and significantly, with a coefficient of 0.57. Agreeable nurses leaned conservative, more inclined to judge ambiguous patients as low-risk. The authors speculate a gently uncomfortable possibility: highly agreeable nurses may weigh the busyness of their colleagues, consciously or not, and hesitate to escalate cases that would demand scarce human resources. Previous work has already linked organizational culture to risk-detection bias, and together these findings point toward a practical intervention already envisaged by the researchers: giving individual nurses structured feedback on their own decision-making tendencies, quantified through the same signal detection lens used in this study.

The research is not without limits, which its authors candidly acknowledge. The sample of twenty-one nurses is small, the stimuli omitted patient background details such as illness history and medication, and high-risk cases appeared in half the trials, far more often than the five to ten percent prevalence seen in real hospitals, a mismatch known in vision science to shift detection accuracy and decision criteria. Yet the study stands as a proof of concept for something genuinely new in nursing science: a way to probe the unconscious, intuitive machinery of clinical judgement without relying on what nurses can articulate about their own thinking. Since nursing judgement is frequently intuitive, built on rapid pattern recognition that practitioners themselves cannot fully explain, behavioral and psychophysical tools like reaction time analysis and signal detection theory may expose what subjective self-report never can. If follow-up studies manipulate time pressure, multitasking, and realistic case prevalence, the five-second risk radar documented here could become the foundation for a new generation of nursing education, staffing policy, and patient safety science, one calibrated not to what nurses say they do, but to what their minds demonstrably do in the critical first seconds of a crisis.

Subject of Research: The information processing mechanisms underlying nurses' clinical risk judgement, examined with signal detection theory and reaction time analysis

Article Title: Exploring information processing underlying risk judgement: Implications from signal detection theory and reaction time analysis

Article References: Hishiya, R., & Ishihara, M. (2026). Exploring information processing underlying risk judgement: Implications from signal detection theory and reaction time analysis. Heliyon, 12(14), Article e45422. https://doi.org/10.1016/j.heliyon.2026.e45422

Image Credits: AI Generated

DOI: 10.1016/j.heliyon.2026.e45422

Keywords: nursing, risk judgement, signal detection theory, reaction time, clinical decision-making, patient deterioration, NEWS2, skin potential level, response bias, detectability, time pressure, cognitive processing