Unexpected Patterns Research Articles

Incomplete mixed data-driven outlier detection based on local–global neighborhood information

Outlier detection is a crucial task for identifying unexpected patterns, errors, and behaviors; therefore, maximizing the valuable information obtained from ubiquitous, incomplete, redundant, noisy, and mixed data poses a great challenge. To achieve efficient graph-based outlier detection, we enhance the connectivity between similar objects and weaken the connectivity between heterogeneous objects. The network structure proposed in this paper is called “an incomplete local and global neighborhood information (ILGNI) network.” In this network, incomplete mixed data can be exploited considering two aspects; single-attribute local information and multi-attribute global information. Specifically, we initially utilize unsupervised attribute reduction methods to improve data quality. Then, from the perspective of local and global information, we use the level of similarity of objects to design strong-neighborhood and weak-similarity relations to deal with incomplete data. On this basis, the topology reconstruction of a high number of fine-grained neighborhood information networks can be realized. Finally, the outlier scores are calculated considering the stationary distribution of the Markov random walk model over the ILGNI network. The experiments conducted on four real-telecom fraud datasets demonstrate that the proposed algorithm can achieve enhanced outlier detection performance with low time complexity. In addition, the proposed method can effectively mine the information obtained from incomplete data and has high applicability to feature-related and feature-independent datasets.

Repetitive Action and Cognitive Performance

For many High School students, the repetitive action such as chewing gum is the ideal tool to help them achieve a consistent period of clear focus and wakefulness. However, as a chewing gum enthusiast, I have recently noticed a very unexpected pattern in my concentration performance when chewing gum compared to the task performed. Due to this particular reason, a study was designed in order to determine the type of task being performed and its relationship to the effectiveness of repetitive action (this study will only examine the repetitive action of chewing) on concentration when performing different tasks (Complex task & Simple task). The research study includes a review of literature, survey, and data analysis. The review of already existing literature regarding chewing gum and cognitive performance suggests that the action of chewing can serve as a distraction in specific environments (such as during an exam) and participants of the Tucha and Simpson study only showed more sustained attention 30 minutes into the study, with the prior time suffering from less sustained attention than usual. The survey data suggests a pattern between sustained attention in relation to the task performed. While chewing gum, the subject’s performance worsens while conducting a complex task (tasks requiring critical thinking skills), such as completing exams, solving puzzles, and memorizing specific details, but improves while conducting repetitive or hands-on tasks, such as exercising, copying, and coloring. The pattern may also exist between high accuracy tasks and low accuracy tasks, therefore accuracy may also be a distinguishing factor for chewing gum’s cognitive influence.

Authoritative school climate as a context for civic socialization

This study represents a confirmatory attempt to classify schools based on the Authoritative Socialization Theory and test whether schools' classification is invariant across different structural characteristics of schools. Furthermore, it tests whether school typologies are related to school outcomes such as school violence, sense of belonging, and student participation. Through latent class analysis and a multilevel modeling approach, we analyzed data from a representative national survey in 1322 Chilean schools. Consistent with authoritative socialization theory and previous research, a four-class model was retained by considering two school climate dimensions: disciplinary structure and student support. These classes were named authoritative, indifferent, authoritarian, and permissive based on theory. In addition, classes showed to be invariant between schools considering the school administration and the school's religiosity; however, regarding the school size, the model showed to be partially invariant. Finally, the school levels of violence and belonging showed the expected pattern considering the theory and previous evidence; however, the average student participation levels showed an unexpected pattern. We conclude that the proposed methodology to classify schools is helpful and offers some advantages to the current classification methods under this theoretical framework. In substantive terms, our results show that not all students have the same probability of experiencing a nurturing school climate within a school system such as the Chilean one characterized by high levels of school segregation.

Donor selection for adoptive immunotherapy with NK cells in AML patients: Comparison between analysis of lytic NK cell clones and phenotypical identification of alloreactive NK cell repertoire.

Natural killer (NK) cell-based adoptive immunotherapy in leukemia patients is an emerging field of interest based on clinical evidence of efficacy and safety. Elderly acute myeloid leukemia (AML) patients have been successfully treated with NK cells from HLA-haploidentical donors, especially when high amounts of alloreactive NK cells were infused. The aim of this study was comparing two approaches to define the size of alloreactive NK cells in haploidentical donors for AML patients recruited in two clinical trials with the acronym "NK-AML" (NCT03955848), and "MRD-NK". The standard methodology was based on the frequency of NK cell clones capable of lysing the related patient-derived cells. The alternative approach consisted of the phenotypic identification of freshly derived NK cells expressing, as inhibitory receptors, only the inhibitory KIR(s) specific for the mismatched KIR-Ligand(s) (HLA-C1, HLA-C2, HLA-Bw4). However, in KIR2DS2+ donors and HLA-C1+ patients, the unavailability of reagents staining only the inhibitory counterpart (KIR2DL2/L3) may lead to an underestimated identification of the alloreactive NK cell subset. Conversely, in the case of HLA-C1 mismatch, the alloreactive NK cell subset could be overestimated due to the ability of KIR2DL2/L3 to recognize with low-affinity also HLA-C2. Especially in this context, the additional exclusion of LIR1-expressing cells might be relevant to refine the size of the alloreactive NK cell subset. We could also associate degranulation assays, using as effector cells IL-2 activated donor peripheral blood mononuclear cells (PBMC) or NK cells upon co-culture with the related patient target cells. The donor alloreactive NK cell subset always displayed the highest functional activity, confirming its identification accuracy by flow cytometry. Despite the phenotypic limitations and considering the proposed corrective actions, a good correlation was shown by the comparison of the two investigated approaches. In addition, the characterization of receptor expression on a fraction of NK cell clones revealed expected but also few unexpected patterns. Thus, in most instances, the quantification of phenotypically defined alloreactive NK cells from PBMC can provide data similar to the analysis of lytic clones, with several advantages, such as a shorter time to achieve the results and, perhaps, higher reproducibility/feasibility in many laboratories.

Using machine learning on clinical data to identify unexpected patterns in groups of COVID-19 patients

As clinicians are faced with a deluge of clinical data, data science can play an important role in highlighting key features driving patient outcomes, aiding in the development of new clinical hypotheses. Insight derived from machine learning can serve as a clinical support tool by connecting care providers with reliable results from big data analysis that identify previously undetected clinical patterns. In this work, we show an example of collaboration between clinicians and data scientists during the COVID-19 pandemic, identifying sub-groups of COVID-19 patients with unanticipated outcomes or who are high-risk for severe disease or death. We apply a random forest classifier model to predict adverse patient outcomes early in the disease course, and we connect our classification results to unsupervised clustering of patient features that may underpin patient risk. The paradigm for using data science for hypothesis generation and clinical decision support, as well as our triaged classification approach and unsupervised clustering methods to determine patient cohorts, are applicable to driving rapid hypothesis generation and iteration in a variety of clinical challenges, including future public health crises.

Better, for worse, or both? Testing environmental sensitivity models with parenting at the level of individual families.

According to environmental sensitivity models, children vary in responsivity to parenting. However, different models propose different patterns, with responsivity to primarily: (1) adverse parenting (adverse sensitive); or (2) supportive parenting (vantage sensitive); or (3) to both (differentially susceptible). This preregistered study tested whether these three responsivity patterns coexist. We used intensive longitudinal data of Dutch adolescents (N = 256, Mage = 14.8, 72% female) who bi-weekly reported on adverse and supportive parenting and their psychological functioning (tmean = 17.7, tmax = 26). Dynamic Structural Equation Models (DSEM) indeed revealed differential parenting effects. As hypothesized, we found that all three responsivity patterns coexisted in our sample: 5% were adverse sensitive, 3% vantage sensitive, and 26% differentially susceptible. No adolescent appeared unsusceptible, however. Instead, we labeled 28% as unperceptive, because they did not perceive any changes in parenting and scored lower on trait environmental sensitivity than others. Furthermore, unexpected patterns emerged, with 37% responding contrary to parenting theories (e.g., decreased psychological functioning after more parental support). Sensitivity analyses with concurrent effects and parent-reported parenting were performed. Overall, findings indicate that theorized responsivity-to-parenting patterns might coexist in the population, and that there are other, previously undetected patterns that go beyond environmental sensitivity models.

Long-run exposure to low-dose radiation reduces cognitive performance

This paper examines the effect of long-run exposure to low-dose radiation on cognitive performance. We focus on the fallout from the Chernobyl accident, which increased the level of ground radiation in large parts of Europe. To identify a causal effect, we exploit unexpected rainfall patterns in a critical time window after the disaster as well as the trajectory of the radioactive plume, which determine local fallout but have no plausible direct effect on test scores. Based on geo-coded survey data from Germany, we show that people exposed to higher radiation perform significantly worse in standardized cognitive tests 25 years later. An increase in initial exposure by one standard deviation reduces cognitive test scores by around 5% of a standard deviation.

Learning from Our Study Organisms about Sexual Selection: Lessons from the Ocellated Wrasse.

AbstractSexual selection is a powerful force shaping not only the details but also the breadth of what we see in nature. Yet so much unexplained variation remains. Organisms often solve the "problem" of how to pass on their genes in ways that do not fit our current expectations. I argue here that integrating empirical surprises will push our understanding of sexual selection forward. Such "nonmodel" organisms (i.e., species that do not do what we think they should do) challenge us to think deeply, integrate puzzling results, question our assumptions, and consider the new (and arguably better) questions these unexpected patterns pose. In this article, I share how puzzling observations from my long-term research on the ocellated wrasse (Symphodus ocellatus) have shaped my understanding of sexual selection and suggested new questions about the interplay among sexual selection, plasticity, and social interactions. My general premise, however, is not that others should study these questions. Instead, I argue for a change in the culture of our field-to consider unexpected results a welcome opportunity to generate new questions and learn new things about sexual selection. Those of us in positions of power (e.g., as editors, reviewers, and authors) need to lead the way.

A lean method for selecting determinants when developing behavior change interventions

ABSTRACT When developing behavior change interventions in a systematic way, it is important to select determinants relevant to the target behavior. Data is needed to gain insight into the determinant structures (the relative strengths of associations between determinants and behavior) and their univariate distributions. This insight is crucial to select the most relevant determinants, but at the same time institutions tasked with behavior change (e.g. prevention organizations, municipal health services) often operate under prohibitive resource constraints, which also extend to how easily they can collect data from a sample. This paper introduces CIBERlite – an approach that furnishes the intervention developer with an idea of the relevance of a limited number of determinants using short measurements informed by theory. The first study (N = 401) in a series of three explores the convergent validity of short and full measurements of determinants derived from the Reasoned Action Approach. The short measurements are used in the main study (N = 415) that serves as a proof-of-concept for the CIBERlite plot, an efficient visualization combining data of determinant structures and their univariate distributions for eight behaviors. The unexpected patterns detected in the main study led to an expert estimation study (N = 45), which shows that individual experts have difficulty in predicting how people score on determinants. This stresses the importance of conducting determinant studies and CIBERlite is a valuable alternative to do so if resources are limited.

Diagnostic modeling of the shoreline variation along the Jiangsu Coast, China

Intertidal flats are of great socio-economic and ecological importance in defending the coastal cities from flooding, providing resources for land reclamations and habits for wildlife. On the intertidal flats, milder profiles are usually featured with finer sediment. However, we find the opposite relationship between the alongshore variation in intertidal slope and sediment grain size on the intertidal flat along the Jiangsu Coast. With a conceptual figure of the hydrodynamics and shoreline evolution on this coast, we hypothesize that the unexpected pattern is caused by the alongshore gradient in hydrodynamic forcing. In order to test our hypothesis, we carry out a series of numerical model simulations in a highly schematized manner to investigate the real mechanism behind this unexpected pattern. Through the analysis, we find that only the southwards coarsening pattern is inconsistent with the shoreline evolution pattern. This inconsistency is not induced by alongshore hydrodynamic gradient, and can only be explained by different sediment provenances. We also find that the alongshore shoreline evolution pattern is not only determined by the alongshore gradient in hydrodynamic forcing, but also influenced by the alongshore variation in bed composition. In the erosion/sedimentation transition zone, the bed composition factor plays the major role.

Safety of co-administration of mRNA COVID-19 and seasonal inactivated influenza vaccines in the vaccine adverse event reporting system (VAERS) during July 1, 2021–June 30, 2022

BackgroundCOVID-19 vaccines may be co-administered with other recommended vaccines, including seasonal influenza vaccines. However, few studies have evaluated the safety of co-administration of mRNA COVID-19 and seasonal influenza vaccines. ObjectiveTo describe reports to the Vaccine Adverse Event Reporting System (VAERS) after co-administration of mRNA COVID-19 and seasonal influenza vaccines. MethodsWe searched the VAERS database for reports of adverse events (AEs) following co-administration of mRNA COVID-19 and seasonal influenza vaccines and following a first booster dose mRNA COVID-19 vaccine alone, during July 1, 2021–June 30, 2022. We assessed the characteristics of these reports and described the most frequently reported MedDRA preferred terms (PTs). Clinicians reviewed available medical records for serious reports and reports of adverse events of special interest (AESI) and categorized the main diagnosis by system organ class. ResultsFrom July 1, 2021 through June 30, 2022, VAERS received 2,449 reports of adverse events following co-administration of mRNA COVID-19 and seasonal influenza vaccines. Median age of vaccinees was 48 years (IQR: 31, 66); 387 (15.8%) were classified as serious. Most reports (1,713; 69.3%) described co-administration of a first booster dose of an mRNA COVID-19 vaccine with seasonal influenza vaccine. The most common AEs among non-serious reports were injection site reactions (193; 14.5%), headache (181; 13.6%), and pain (171; 12.8%). The most common AEs among reports classified as serious were dyspnea (38; 14.9%), COVID-19 infection (32; 12.6%), and chest pain (27; 10.6%). DiscussionThis review of reports to VAERS following co-administration of mRNA COVID-19 and seasonal influenza vaccines did not reveal any unusual or unexpected patterns of AEs. Increased reporting of certain events (e.g., COVID-19 disease) was expected. CDC will continue to monitor the safety of co-administration of mRNA COVID-19 and seasonal influenza vaccines, including co-administration involving bivalent mRNA COVID-19 booster vaccines that have been recommended for people ages ≥ 6 months in the United States.

State and Fate of Glaciers in the Val Veny (Mont-Blanc Range, Italy): Contribution of Optical Satellite Products

The glaciers of the Val Veny (Italian side of the Mont-Blanc Massif) have been the site of numerous field observations during the last decades, in particular for the study of glacial fluctuations or surface processes related to the debris cover. Here, we propose to examine how satellite observations can complement field measurements on the state and fate of the Val Veny glaciers. Indeed, satellite products obtained in a quasi-systematic way allow to account not only for the loss of surface and volume, but also for the changes in their flow velocities. The overall pattern we document is a glacier thinning and slowdown of the ice flow, with an estimated shrinkage of 25% by 2050 and a volume loss ranging between 30 and 43% depending on the data source used for the estimation of the initial volume. In such a context, a portion of the upper reaches of Brenva Glacier shows an unexpected pattern of thickening and increase in ice flow that rises questions on its origin. Finally, the uncertainties in the estimation of ice thicknesses remain important and have repercussions on the future evolution of the glaciers and their contribution from a hydrological point of view. By 2050, we estimate that the water contribution due to the volume loss of Val Veny glaciers could decrease by 40%.

Mask-Wearing and Emotional Intensity Perceptions

The widespread use of sanitary face masks due to the COVID-19 pandemic renewed interest in facial emotion perception while wearing masks. Usage of face masks during this time were considered a nonpharmaceutical intervention to mitigate the spread of the virus. We examined whether wearing face masks affects the intensity of emotion perception and judgments of approachability while also considering the sex and age of the rater. Emerging adults (ages 18–25, M = 19.5, 16 men, 44 women, n = 60) and adults (ages 26–65, M = 44.8, 10 men, 31 women, n = 41) viewed photos of faces of a young adult man, middle-aged man, young adult woman, and a middle-aged woman, masked and unmasked, with happy, sad, neutral, and angry expressions. ANOVAs repeated onthe face mask and no face mask conditions showed significant reductions in emotion intensity for happy (p < .001, η2 = .22) and sad faces (p < .001, η2 = .35), no differences for angry faces (p = .16, η2 = .02), and the opposite (increased intensity) with neutral faces (p = .03, η2 = .02). Unmasked happy faces were rated as more approachable than masked happy faces. Unmasked angry faces were rated as less approachable than masked angry faces but only by emerging adults. No differences appeared for sad emotions. Neutral faces again showed an unexpected pattern, with masks increasing approachability.

FS3change: A Scalable Method for Change Pattern Mining

Mining change patterns can give unique understanding on the evolution of dynamically changing systems like social relation graphs, weblinks, hardware descriptions and models. A more recent focus is source code change pattern mining that may qualitatively justify expected or uncover unexpected patterns. These patterns then offer a basis, e.g., for program language evolution or auto-completion support. We present a change pattern mining method that greatly expands the limits of input data and pattern complexity, over existing methods. We propose scalability solutions on conceptual and algorithmic level, thereby evolving the state-of-the-art sampling-based frequent subgraph mining method FS <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> , resulting in 75% reduction in memory consumption and a speedup of 6500 for a large scale dataset. Patterns can have 100,000s of occurrences for which manual review is impossible and may lead to misinterpretation. We propose the novel content track approach for interactively exploring pattern contents in context, based on marginal distributions. We evaluate our approach by mining 1,000 open source projects contributing a total of 558 million changes and 2 billion contextual connections among them, thereby, demonstrating its scalability. A manual interpretation of 19 patterns shows sensible mined patterns allowing to deduct implications for language design and demonstrating the soundness of the approach.

Using Functional Clustering to Diagnose Person Misfit

The purpose of this study is to demonstrate clustering methods within a functional data analysis (FDA) framework for identifying subgroups of individuals that may be exhibiting categories of misfit. Person response functions (PRFs) estimated within a FDA framework (FDA-PRFs) provide graphical displays that can aid in the identification of persons that have responded unexpectedly to items comprising an achievement test. Typical person fit statistics are also useful for detecting unexpected response patterns, but they do not provide insight into the underlying behaviors responsible for those responses. However, different responding behaviors tend to produce FDA-PRFs of different shapes, and may provide additional information regarding the reasons for misfit. Functional clustering methods are useful for categorizing respondents into subgroups based on the shapes of their FDA-PRFs. In this study, a small simulation illustrates the potential of clustering FDA-PRFs for identifying persons displaying common types of responding behaviors. The methodology is also applied to data from a high school biology assessment and a mathematics achievement test. Clustering FDA-PRFs offers a promising methodology for operationalizing person fit evaluations in large-scale assessments, and may be a valuable step in person fit assessment when used in conjunction with traditional indices of psychometric quality.

Heuristic interpretation as rational inference: A computational model of the N400 and P600 in language processing

Much inquiry in psycholinguistics has focused on evidence from the N400 and P600 components of the event-related potential (ERP) signal—and a central theoretical challenge in this area is accounting for the so-called “semantic P600”, which involves unexpected patterns in these components relative to traditional theories of the underlying mechanisms. In this paper we present a computational model of the language processing mechanisms underlying these ERP components, which builds on existing psycholinguistic theories in positing a heuristic interpretation stage of processing, but which deviates from existing theories in formulating this heuristic interpretation process as probabilistic selection via a noisy channel model, and in quantifying and accounting for fine-grained variation in statistical and representational properties of individual stimuli. Our model successfully simulates N400 and P600 patterns from eight psycholinguistic experiments, reflecting the full range of N400-only, P600-only, and biphasic N400-P600 effects, and its behaviors shed light on a number of key patterns that have presented challenges for existing theories. The model’s success indicates that a strong account for the processing mechanisms underlying these effects is one in which language comprehension involves a probabilistic heuristic interpretation stage resembling a noisy channel process, feeding into subsequent processes that assess target word fit and reconcile between heuristic and literal interpretations. The model’s success also indicates that these mechanisms are critically sensitive to statistical variation in individual stimuli, and that modeling the effects of this variation is essential to account for the full range of observed effects in language processing.

Selection for high growth displaces costs of digestion from SDA to SMR in divergent ecotypes of rainbow trout

AbstractTo understand the drivers of variation in digestive performance and their effects on growth, we examined relationships among food consumption, digestive metabolism, food processing efficiency, and growth rate in juveniles of fast‐growing piscivore versus slow‐growing insectivore ecotypes of rainbow trout reared at satiation rations on the same diet (i.e., commercial food pellets). Relative to slow‐growing insectivores with lower basal metabolism, and despite a much larger maximum food ration, faster‐growing piscivores presented an unexpected pattern of higher digestive efficiency through a reduction in the absolute costs of postprandial metabolism coupled with shorter or similar gut residence time and higher assimilation efficiency. These results suggest that the increase in digestive metabolism following the ingestion of larger meals can be mitigated by displacing the costs of digestion from SDA to SMR. Reducing total digestion costs (SDA + SMR) while maintaining higher assimilation efficiency may be possible through potential adaptations including (1) increased intestinal absorption capacity; (2) economies of scale that shorten gut transit time with increasing ration level; or (3) a permanently larger digestive tract that increases maintenance costs but reduces the need for cyclic upregulation and associated overhead costs.

Designing a resilient supply chain through a robust adaptive model predictive control policy under perishable goods and uncertain forecast information

We consider the inventory control problem for Supply Chains (SC) with deteriorating items and an uncertain future customer demand freely varying inside a given compact set. The problem is to define a resilient Replenishment Policy (RP) keeping the actual inventory level as close as possible to a desired reference trajectory. This requirement should be satisfied despite uncertainties on the decay factor of stocked goods and unexpected customer demand patterns outside the bounds of the compact set. We propose a method based on a Resilient Robust Model Predictive Control (RRMPC) approach. This requires dealing with a Min-Max Constrained Optimization Problem (MMCOP). To reduce the numerical complexity of the algorithm, the control signal is parametrized using B-spline functions.

Immune gene expression in the mosquito vector Culex quinquefasciatus during an avian malaria infection.

Plasmodium relictum is the most widespread avian malaria parasite in the world. It is listed as one of the 100 most dangerous invasive species, having been responsible for the extinction of several endemic bird species, and the near-demise of several others. Here we present the first transcriptomic study focused on the effect of P. relictum on the immune system of its vector (the mosquito Culex quinquefasciatus) at different times post-infection. We show that over 50% of immune genes identified as being part of the Toll pathway and 30%-40% of the immune genes identified within the Imd pathway are overexpressed during the critical period spanning the parasite's oocyst and sporozoite formation (8-12 days), revealing the crucial role played by both these pathways in this natural mosquito-Plasmodium combination. Comparison of infected mosquitoes with their uninfected counterparts also revealed some unexpected immune RNA expression patterns earlier and later in the infection: significant differences in expression of several immune effectors were observed as early as 30 min after ingestion of the infected blood meal. In addition, in the later stages of the infection (towards the end of the mosquito lifespan), we observed an unexpected increase in immune investment in uninfected, but not in infected, mosquitoes. In conclusion, our work extends the comparative transcriptomic analyses of malaria-infected mosquitoes beyond human and rodent parasites and provides insights into the degree of conservation of immune pathways and into the selective pressures exerted by Plasmodium parasites on their vectors.

Effects of Attention and Congruence on Judging True and False COVID-19 News Headlines

ake news has become increasingly prevalent in the United States, especially during the outbreak of the COVID-19 pandemic in the U.S. since March 2020. This has caused a significant number of Americans to believe that COVID-19 is a hoax rather than a real problem. Fake news spreads easily and quickly through social media, where people are often inundated with advertisements and thumbnails of false headlines along with pseudo-reliable media outlets. The purpose of the current study is to analyze how divided attention and the congruence of news headlines and news sources affect the identification of true and false news headlines about the COVID-19 pandemic. Participants (N = 27) saw 48 news headlines paired with 48 sources and judged the validity of the headline. In the congruent condition, true and false headlines were paired with reliable and unreliable sources. In the incongruent condition, these pairings were mixed up: true headlines were accompanied by unreliable sources, and false headlines reliable sources. In 24 out of 48 trials, participants also did mental math problems while evaluating the news (divided attention). The results showed that when dividing attention, participants performed better in trials either with congruently matched headlines and sources or with incongruent matchings of false headlines and reliable sources. There were also unexpected patterns of results, which are further discussed along with some methodological revisions for future research. Keywords: attention, congruence, news media, social media, coronavirus