Predict Performance Changes Research Articles

Exploring the impact of missingness on racial disparities in predictive performance of a machine learning model for emergency department triage.

To investigate how missing data in the patient problem list may impact racial disparities in the predictive performance of a machine learning (ML) model for emergency department (ED) triage. Racial disparities may exist in the missingness of EHR data (eg, systematic differences in access, testing, and/or treatment) that can impact model predictions across racialized patient groups. We use an ML model that predicts patients' risk for adverse events to produce triage-level recommendations, patterned after a clinical decision support tool deployed at multiple EDs. We compared the model's predictive performance on sets of observed (problem list data at the point of triage) versus manipulated (updated to the more complete problem list at the end of the encounter) test data. These differences were compared between Black and non-Hispanic White patient groups using multiple performance measures relevant to health equity. There were modest, but significant, changes in predictive performance comparing the observed to manipulated models across both Black and non-Hispanic White patient groups; c-statistic improvement ranged between 0.027 and 0.058. The manipulation produced no between-group differences in c-statistic by race. However, there were small between-group differences in other performance measures, with greater change for non-Hispanic White patients. Problem list missingness impacted model performance for both patient groups, with marginal differences detected by race. Further exploration is needed to examine how missingness may contribute to racial disparities in clinical model predictions across settings. The novel manipulation method demonstrated may aid future research.

Implications of the Use of Artificial Intelligence Predictive Models in Health Care Settings : A Simulation Study.

Substantial effort has been directed toward demonstrating uses of predictive models in health care. However, implementation of these models into clinical practice may influence patient outcomes, which in turn are captured in electronic health record data. As a result, deployed models may affect the predictive ability of current and future models. To estimate changes in predictive model performance with use through 3 common scenarios: model retraining, sequentially implementing 1 model after another, and intervening in response to a model when 2 are simultaneously implemented. Simulation of model implementation and use in critical care settings at various levels of intervention effectiveness and clinician adherence. Models were either trained or retrained after simulated implementation. Admissions to the intensive care unit (ICU) at Mount Sinai Health System (New York, New York) and Beth Israel Deaconess Medical Center (Boston, Massachusetts). 130 000 critical care admissions across both health systems. Across 3 scenarios, interventions were simulated at varying levels of clinician adherence and effectiveness. Statistical measures of performance, including threshold-independent (area under the curve) and threshold-dependent measures. At fixed 90% sensitivity, in scenario 1 a mortality prediction model lost 9% to 39% specificity after retraining once and in scenario 2 a mortality prediction model lost 8% to 15% specificity when created after the implementation of an acute kidney injury (AKI) prediction model; in scenario 3, models for AKI and mortality prediction implemented simultaneously, each led to reduced effective accuracy of the other by 1% to 28%. In real-world practice, the effectiveness of and adherence to model-based recommendations are rarely known in advance. Only binary classifiers for tabular ICU admissions data were simulated. In simulated ICU settings, a universally effective model-updating approach for maintaining model performance does not seem to exist. Model use may have to be recorded to maintain viability of predictive modeling. National Center for Advancing Translational Sciences.

Fueling the Simulation Machine: Considerations for High-Resolution 3D Microscopy to Support Computational Modeling in Battery Research

To meet the rapidly evolving energy needs of a growing population, energy storage and conversion devices such as batteries must rapidly evolve as well. Battery materials and other functional energy materials derive their properties not from their static state, but from how they interact with other materials in their service environment. Furthermore, the microstructure of these materials strongly dictates their performance and how they interact with other components (like electrolytes, binders, etc). In order to accelerate this development cycle, researchers are increasingly integrating computational modeling into their materials and device development efforts.Computational analysis is progressing to the point where simulations can incorporate real material microstructures, simulate statistically equivalent structures, and predict performance changes based on changes to these microstructures. At the same time experimental techniques have emerged that can deliver representative volumes of high-quality, high-resolution 3D microstructural images from real devices and materials. The interplay between advanced computational analysis and evolving characterization capabilities means we are rapidly approaching the realization of the “materials by design” paradigm where new materials are designed and tested digitally with input and validation provided by advanced materials characterization techniques. Within this paradigm, however, it is critical that the experimental techniques provide accurate, realistic, and representative input to the computational models. In the case of microstructural information, this means generating 3-dimensional images that accurately represents the materials in question.One key example of this is for materials, like battery materials and electrodes, which contain micro- and nano-scale porosity. Acquiring image data that can reproduce real material properties and feed accurate simulations is challenging and relies on satisfying several conditions. For example, the volume must be representative of the “bulk” material microstructure. The (2D) image resolution must capture the relevant scope of material features, and the 3D image must accurately represent the material in all 3 dimensions and at appropriate length scales. The focused-ion beam scanning electron microscope (FIB-SEM) is capable of imaging these micro- and nano-scale pore networks in 3D but care must be taken when considering these experiments, especially when the intent is to use the images to feed computational models.We present here a review of the factors researchers must consider when tackling these demanding tasks and highlight experimental recommendations for leveraging FIB-SEM tomography to produce the most relevant imaging data in this context. We use a lithium-ion battery cathode as a test case to demonstrate this approach, but the outcome and methods can be leveraged across the materials research space and beyond.

Performance and decision making of a complex skill under monitoring and outcome pressure conditions: Which of them can reinvestment predict?

Despite a huge amount of research conducted on performance under competitive pressure, the different components of a competitive environment such as outcome pressure or monitoring pressure, have less been investigated. The present study aimed to investigate the performance and decision making of a complex skill and their link with reinvestment (a type of self-focus behavior) under different conditions of pressure using a table tennis task. The topspin forehand and backhand shots of 20 expert table tennis athletes were examined under low, monitoring, and outcome pressure conditions in a within-group design. In addition, predictive validities of Decision-Specific Reinvestment Scale (DSRS) and Movement-Specific Reinvestment Scale (MSRS) with the changes of performance and decision making from low- to high-pressure conditions as dependent variables were examined. The repeated measures ANOVA results showed that the effects of pressure conditions on anxiety were significant. Decision accuracy under outcome pressure and decision speed under monitoring and outcome pressure conditions were decreased. The performance of athletes was decreased under monitoring pressure while no changes was observed in their performance under outcome pressure. Decision Reinvestment (a factor of DSRS) could predict decision speed changes under both pressure conditions, while Movement Self-Consciousness (a factor of MSRS) could predict performance changes only under monitoring pressure. Our Findings highlight that the subscales of MSRS and DSRS were activated independently and a competitive pressure situation has different effects on the cognitive and motor aspects of complex skills. It is suggested that expert athletes and coaches consider separate pressure conditions and their relationships with the ongoing task.

Framework of Prediction Model for Mid- to Long-Term Performance Changes of Urban Railway Facilities Based on Performance Evaluation Reports

To maintain railway facilities in an appropriate state, systematic management based on mid- and long-term maintenance plans through future performance prediction must be carried out. To this end, it is necessary to establish and utilize a model that can predict mid- to long-term performance changes of railway facilities by predicting performance changes of individual sub-facilities. However, predicting changes in the performance of all sub-facilities can be difficult as it requires large volumes of data, and railway facilities are a collection of numerous sub-facilities. Therefore, in this study, a framework for a model that can predict mid- to long-term performance changes of railway facilities through analysis of continuously accumulated performance evaluation results is proposed. The model is a system with a series of flows that can classify performance evaluation results by individual sub-facilities, predict performance changes by each sub-facility using statistical methods, and predict mid- to long-term performance changes of the facility. The developed framework was applied to 36,537 sub-facilities comprising 12 lines of two urban railways in South Korea to illustrate the model and verify its applicability and effectiveness. This study contributes in terms of its methodology in establishing a framework for predicting mid- to long-term performance changes, providing the basis for the development of an automated model able to continuously predict performance changes of individual sub-facilities. In practical terms, it is expected that railway facility managers who allow trade-off between reliability and usability can contribute to establishing the mid- to long-term maintenance plans by utilizing the model proposed in this study, instead of subjectively building them.

Behind every robust result is a robust method: Perspectives from a case study and publication process in hydrological modelling

Behind every robust result is a robust method: Perspectives from a case study and publication process in hydrological modelling

Development of test protocols for solid oxide electrolysis cells operated under accelerated degradation conditions

In order to be a real competitor to other industrial-scale hydrogen production technologies, solid oxide electrolysis cells (SOEC) must present essential features such as high hydrogen production rates, load flexibility and high system efficiency. However, operating conditions necessary to achieve these objectives can induce undesired performance deterioration and microstructural degradation negatively influencing the long-term stability of SOECs. This work provides the results of experimental investigations on the performance of SOECs operated under different operating conditions. In order to induce and accelerate specific degradation phenomena high steam partial pressures, fast load changes, high voltages, high steam conversion rates and alternating mode were applied. Different kinds and degrees of impact on performance and microstructure were observed and analyzed in detail. Correlations between performance changes and their underlying processes were investigated and useful connections between them were discovered. The major findings are summarized in the form of a suggestion for accelerated stress test (AST) protocols for SOECs. The developed ASTs enable deeper understanding of several degradation phenomena induced by different operating conditions and present methods for identification thereof. Further, the ASTs were estimated to have potential to predict performance changes over long-term operating periods.

Improvement of Software-Defined Network Performance Using Queueing Theory: A Survey

Software-Defined Networking (SDN) is a growing network technology that has brought significant benefits to a wide range of disciplines, from science to technology to various fields. This structure can be used in network-based environments, data centers and various research sites. OpenFlow is one of the most widely used protocols for interaction between a controller and a switch in a Software-Defined Networking. Understanding the performance and limitations of the network defined by open source software, including bottlenecks and security vulnerabilities due to the centralized network structure, are important prerequisites for the efficient deployment of these systems. These points of view led researchers to examine various related mathematical models to address these issues. Queueing theory provides the most important and accurate model for evaluating the performance of the SDN networks affected by these restrictions, which has attracted the attention of researchers recently. Regarding to extensive mathematical modeling, this theory has also been used to improve network efficiency. Researchers have used this theory to investigate operational power and improve time consumption and control of data planes due to the nature of classification and storage of packets in SDN buffers. These methods overcome controller performance bottlenecks and increase SDN control capacity, especially for large distributed networks. In this paper, we examine the queueing models for different applications in different layers of SDN, in which researchers use these methods to monitor network loads, evaluate and predict performance changes due to diversity in network traffic. We introduce and review a collection of articles that explore, different applications of Queueing theory in SDN networks. In addition, in order to increase the efficiency of this research, detailed comparisons are performed in terms of structure, mathematical models and the final simulation results.

Changes in performance and bio-mathematical model performance predictions during 45 days of sleep restriction in a simulated space mission

Lunar habitation and exploration of space beyond low-Earth orbit will require small crews to live in isolation and confinement while maintaining a high level of performance with limited support from mission control. Astronauts only achieve approximately 6 h of sleep per night, but few studies have linked sleep deficiency in space to performance impairment. We studied crewmembers over 45 days during a simulated space mission that included 5 h of sleep opportunity on weekdays and 8 h of sleep on weekends to characterize changes in performance on the psychomotor vigilance task (PVT) and subjective fatigue ratings. We further evaluated how well bio-mathematical models designed to predict performance changes due to sleep loss compared to objective performance. We studied 20 individuals during five missions and found that objective performance, but not subjective fatigue, declined from the beginning to the end of the mission. We found that bio-mathematical models were able to predict average changes across the mission but were less sensitive at predicting individual-level performance. Our findings suggest that sleep should be prioritized in lunar crews to minimize the potential for performance errors. Bio-mathematical models may be useful for aiding crews in schedule design but not for individual-level fitness-for-duty decisions.

Predicting changes in performance due to cognitive fatigue: A multimodal approach based on speech motor coordination and electrodermal activity

Objective Military job and training activities place significant demands on service members’ (SMs’) cognitive resources, increasing risk of injury and degrading performance. Early detection of cognitive fatigue is essential to reduce risk and support optimal function. This paper describes a multimodal approach, based on changes in measures of speech motor coordination and electrodermal activity (EDA), for predicting changes in performance following sustained cognitive effort. Methods Twenty-nine active duty SMs completed computer-based cognitive tasks for 2 h (load period). Measures of speech derived from audio were acquired, along with concurrent measures of EDA, before and after the load period. Cognitive performance was assessed before and during the load period using the Automated Neuropsychological Assessment Metrics Military Battery (ANAM MIL). Subjective assessments of cognitive effort and alertness were obtained intermittently. Results Across the load period, participants’ ratings of cognitive workload increased, while alertness ratings declined. Cognitive performance declined significantly during the first half of the load period. Three speech and arousal features predicted cognitive performance changes during this period with statistically significant accuracy: EDA (r = 0.43, p = 0.01), articulator velocity coordination (r = 0.50, p = 0.00), and vocal creak (r = 0.35, p = 0.03). Fusing predictions from these features predicted performance changes with r = 0.68 (p = 0.00). Conclusions Results suggest that speech and arousal measures may be used to predict changes in performance associated with cognitive fatigue. This work supports ongoing efforts to develop reliable, unobtrusive measures for cognitive state assessment aimed at reducing injury risk, informing return to work decisions, and supporting diverse mobile healthcare applications in civilian and military settings.

Integrating uncertain prior knowledge regarding ecological preferences into multi-species distribution models: Effects of model complexity on predictive performance

Species distribution models (SDMs) are often criticised for lacking explicit linkage to ecological concepts. We aim to improve the ecological basis of SDMs by integrating prior knowledge about ecological preferences of organisms. Additionally, we aim to support a systematic, data-driven review of such prior knowledge by confronting it with independent monitoring data using Bayesian inference. We developed a series of multi-species distribution models (MSDMs) with increasing complexity to predict the probability of occurrence of taxa at sampling sites based on habitat suitability functions that are parameterized with prior ecological knowledge. We subsequently assessed the models` predictive performance with 3-fold cross-validation. So far, if ecological preferences or functional traits have been used in SDMs, they were mainly used as fixed inputs without considering their uncertainty. We take the additional step of considering uncertainty about preference parameters by including them as uncertain prior information that is subsequently updated with Bayesian inference. We apply the series of models in a case study on macroinvertebrates in Swiss streams. We analyse differences in the quality of fit, changes in predictive performance, and the potential to learn about the parameters from the data. We consider ecological preferences for natural and human modified environmental factors including temperature, flow velocity, organic matter concentration, insecticide pollution, and substratum. Results indicate that updating prior knowledge on ecological preferences with Bayesian inference, rather than using it as fixed input, improves model fit and predictive performance. For example, the predictive performance measured by the deviance for validation data improves by 17 % and the explanatory power increases 3.8 times from a model that treats ecological preferences as fixed scores to a model that treats them as uncertain parameters. The spatial distribution of many taxa, including rare taxa with frequencies of occurrence down to about 5 %, which are difficult to model with SDMs that do not consider prior information, can be captured by the new models. Integrating prior knowledge as uncertain parameters in a Bayesian framework establishes ecological interpretable links between taxa and their environment and supports a systematic revision and complementation of databases on ecological preferences, even in case of poor or missing prior knowledge. Model outputs need to be carefully interpreted by modellers and experts on ecological preferences. Increased exchange between these research fields will benefit further integration of ecological preferences into SDMs.

Degree and centrality-based approaches in network-based variable selection: Insights from the Singapore Longitudinal Aging Study.

We describe a network-based method to obtain a subset of representative variables from clinical data of subjects of the second Singapore Longitudinal Aging Study (SLAS-2), while preserving to a good extent the predictive performance of the full set with regards to a multi-faceted index of successful aging, SAGE. To examine differences in predictive performance of high-degree nodes (“hubs”) and high-centrality ones (“cores”), we implement four subsetting strategies (two degree-based, two centrality-based) and obtain four surrogate sets of variables, which we use as input features for machine learning models to predict the SAGE index of subjects. All four models have variables belonging to the physical, cardiovascular, cognitive and immunological domains among their fifteen most important predictors. A fifth domain (leisure-time activities, LTA) is also present in some form. From a comparison of the surrogate sets’ size and predictive performance, a centrality-based approach (selection of the most central variable-nodes within each cluster) yielded the smallest-sized surrogate set, while having high prediction accuracy (measured by its model’s area-under-curve, AUC) in comparison to its analogous degree-based strategy (selection of the highest-degree nodes per cluster). Inclusion of the next most-central variables yielded negligible changes in predictive performance while more than doubling the surrogate set size. The centrality-based approach thus yields a surrogate set which offers a good balance between number of variables and prediction performance, and can act as a representative subset of the SLAS-2 clinical dataset.

Understanding contribution and independence of multiple biomarkers for predicting response to atezolizumab.

2567 Background: No biomarker satisfactorily predict response to anti-PD-L1 therapies. Biomarker studies suffer from small sample size, presence of disease subtypes, and lack of simultaneous measurement of multiple biomarkers. The IMvigor210 dataset (Mariathasan et al., Nature 2018) provides baseline measurements for multiple biomarkers of response to atezolizumab (n range: 105-298) coupled with genomewide RNAseq profiles. We examined predictive performance of individual biomarkers and combined information from multiple biomarkers to measure changes in predictive performance. Methods: We built classification models (PR/CR vs. PD/SD) using genes and gene sets that provide information on pathways (mSigDB), immune components (xCell, Cibersort), and predictors of response (IMPRES, Immunophenoscore, and TIDE). Prognostic features were removed based on survival association in TCGA. All experiments were done with repeated five-fold double cross validation. Predictions from the gene sets model were used as a single biomarker. PD-L1 expression by IHC in tumor core and immune cells, tumor mutation burden(TMB), neo-antigen burden (NB), location of metastatic disease, immune phenotype and genomic subtypes were then systematically merged with the gene set based model. Results: NB was the best predictor of response (AUC 0.77), while a model combining NB, TMB, ECOG and expression signatures was marginally better (AUC 0.81) with a chance of over fitting. Chi-square tests for independence suggested that examined biomarkers do not provide independent information explaining lack of increase in AUC. Signatures for TP53 mutations, M1 macrophages, CD8+ T effector cell and DNA repair, among others, were present frequently in classification using gene expression information (AUC 0.71), suggesting their independent contributions to response. Adding gene expression information to NB didn’t improve AUC for response but provided better survival stratification. Conclusions: Integration of examined biomarkers with machine learning did not improve response prediction significantly. We are now examining sizes of subgroups defined by combination of low NB/TMB with these biomarkers.

Performance Degradation Predictions Based on Microstructural Evolution Due to Grain Coarsening Effects in Solid Oxide Fuel Cell Electrodes

Due to the time required to test fuel cell degradation experimentally, a physics-based model which can predict degradation can be a valuable tool. Grain coarsening and the resulting microstructure evolution is a primary mode of degradation. In this study, a multi-physics model of a fuel cell is presented which can predict performance loss as a function of time caused by coarsening in the electrodes of an LSM-YSZ/YSZ/Ni-YSZ SOFC. Microstructural properties are updated as a function of time from their initial values using functional relations derived from experimental data. Specie and charge transport equations are solved to predict performance changes with time from the microstructural changes. The model is first calibrated such that it correctly captures the performance of a button cell. Performance change over time predicted by the model is compared to experimental data for a cell operated for 775 hours. It is found that the model predicts a slower degradation rate than the rate which occurs experimentally. This is reasonable as other forms of degradation are not being accounted for. The model is then used to make long term predictions of cell performance loss due to grain coarsening out to 40,000 hours for various operating temperatures and initial microstructures.

The effect of daytime napping and full-night sleep on the consolidation of declarative and procedural information.

SummaryMany studies investigating sleep and memory consolidation have evaluated full‐night sleep rather than alternative sleep periods such as daytime naps. This multi‐centre study followed up on, and was compared with, an earlier full‐night study (Schabus et al., 2004) investigating the relevance of daytime naps for the consolidation of declarative and procedural memory. Seventy‐six participants were randomly assigned to a nap or wake group, and performed a declarative word‐pair association or procedural mirror‐tracing task. Performance changes from before to after a 90‐min retention interval filled with sleep or quiet wakefulness were evaluated between groups. Associations between performance changes, sleep architecture, spindles, and slow oscillations were investigated. For the declarative task we observed a trend towards stronger forgetting across a wake period compared with a nap period, and a trend towards memory increase over the full‐night. For the procedural task, accuracy was significantly decreased following daytime wakefulness, showed a trend to increase with a daytime nap, and significantly increased across full‐night sleep. For the nap protocol, neither sleep stages, spindles, nor slow oscillations predicted performance changes. A direct comparison of day and nighttime sleep revealed that daytime naps are characterized by significantly lower spindle density, but higher spindle activity and amplitude compared with full‐night sleep. In summary, data indicate that daytime naps protect procedural memories from deterioration, whereas full‐night sleep improves performance. Given behavioural and physiological differences between day and nighttime sleep, future studies should try to characterize potential differential effects of full‐night and daytime sleep with regard to sleep‐dependent memory consolidation.

Improved species‐occurrence predictions in data‐poor regions: using large‐scale data and bias correction with down‐weighted Poisson regression and Maxent

Species distribution modelling (SDM) has become an essential method in ecology and conservation. In the absence of survey data, the majority of SDMs are calibrated with opportunistic presence‐only data, incurring substantial sampling bias. We address the challenge of correcting for sampling bias in the data‐sparse situations. We modelled the relative intensity of bat records in their entire range using three modelling algorithms under the point‐process modelling framework (GLMs with subset selection, GLMs fitted with an elastic‐net penalty, and Maxent). To correct for sampling bias, we applied model‐based bias correction by incorporating spatial information on site accessibility or sampling efforts. We evaluated the effect of bias correction on the models’ predictive performance (AUC and TSS), calculated on spatial‐block cross‐validation and a holdout data set. When evaluated with independent, but also sampling‐biased test data, correction for sampling bias led to improved predictions. The predictive performance of the three modelling algorithms was very similar. Elastic‐net models have intermediate performance, with slight advantage for GLMs on cross‐validation and Maxent on hold‐out evaluation. Model‐based bias correction is very useful in data‐sparse situations, where detailed data are not available to apply other bias correction methods. However, bias correction success depends on how well the selected bias variables describe the sources of bias. In this study, accessibility covariates described bias in our data better than the effort covariate, and their use led to larger changes in predictive performance. Objectively evaluating bias correction requires bias‐free presence–absence test data, and without them the real improvement for describing a species’ environmental niche cannot be assessed.

Prediction of Performance Degradation Due to Grain Coarsening Effects in Solid Oxide Fuel Cells

Long term degradation of solid oxide fuel cells is one of the biggest impediments to commercialization. Due to the inordinate amount of time required to properly test degradation experimentally, physics based models which can predict long term degradation can be crucial time and money saving tools. At the cell level, one of the primary modes of degradation comes from grain coarsening and the resulting changes in microstructure properties. In this study a multi-physics model of a single fuel cell is presented which aims to predict performance loss as a function of time caused by coarsening in the electrodes of an LSM/YSZ/Ni SOFC. Microstructural properties are changed as a function of time from their initial values using functional relations derived from data from experiments and phase field models. Specie and charge transport equations are solved using these parameters to predict performance changes with time which result from the aforementioned microstructural changes. The model is first calibrated such that it correctly captures the performance and impedance response of a particular button cell at a given time without any degradation. This same model is run for a duration of 750 hours with performance data being measured for the same cell including impedance measurements being taken intermittently. Performance change over time predicted by the model are compared to experimental data. It is found that the model predicts a slower degradation rate than the rate which occurs experimentally. This is deemed reasonable as other forms of degradation are not being accounted for. The model is then used to make long term predictions of cell performance loss due to grain coarsening. The resulting degradation rate is considered to be the theoretical minimum which occurs when all other modes of degradation are not present. Work is underway to include other forms of degradation in the model so that a complete and robust degradation prediction tool is available for design analysis without the need for costly long term degradation measurements.

Manual aiming in healthy aging: does proprioceptive acuity make the difference?

The present study examines whether non-active older adults are more dependent on visual information when executing aiming movements and whether age-related declines in proprioception play a mediating role herein. Young (N=40) and older adults (N=38) were divided into physically active and non-active subgroups based on self-reported sports participation levels. In experiment 1, participants executed wrist-aiming movements with and without visual feedback. In experiment 2, passive proprioceptive acuity was assessed using wrist motion detection and position matching tests. Results showed similar aiming accuracy across age groups both with and without visual feedback, but older adults exhibited longer movement times, prolonged homing-in phase, and made more corrective submovements. Passive proprioceptive acuity was significantly affected by physical activity level and age, with participants in the active group scoring better than their non-active peers. However, these declines did not predict performance changes on the aiming task. Taken together, our observations suggest that decline in proprioceptive acuity did not predict performance changes on the aiming task and older adults were able to compensate for their decreased motion and position sense when allowed sufficient time. In line with these observations, we proposed that older adults are able to compensate for their decline in proprioception by increasing their reliance on predictive models.

Tracing the value of data for flood loss modelling

Flood loss modelling is associated with considerable uncertainty. If prediction uncertainty of flood loss models is large, the reliability of model outcomes is questionable, and thus challenges the practical usefulness. A key problem in flood loss estimation is the transfer of models to geographical regions and to flood events that may differ from the ones used for model development. Variations in local characteristics and continuous system changes require regional adjustments and continuous updating with current evidence. However, acquiring data on damage influencing factors is usually very costly. Therefore, it is of relevance to assess the value of additional data in terms of model performance improvement. We use empirical flood loss data on direct damage to residential buildings available from computer aided telephone interviews that were compiled after major floods in Germany. This unique data base allows us to trace the changes in predictive model performance by incrementally extending the data base used to derive flood loss models. Two models are considered: a uni-variable stage damage function and RF-FLEMO, a multi-variable probabilistic model approach using Random Forests. Additional data are useful to improve model predictive performance and increase model reliability, however the gains also seem to depend on the model approach.

Parietal plasticity after training with a complex video game is associated with individual differences in improvements in an untrained working memory task

Researchers have devoted considerable attention and resources to cognitive training, yet there have been few examinations of the relationship between individual differences in patterns of brain activity during the training task and training benefits on untrained tasks (i.e., transfer). While a predominant hypothesis suggests that training will transfer if there is training-induced plasticity in brain regions important for the untrained task, this theory lacks sufficient empirical support. To address this issue we investigated the relationship between individual differences in training-induced changes in brain activity during a cognitive training videogame, and whether those changes explained individual differences in the resulting changes in performance in untrained tasks. Forty-five young adults trained with a videogame that challenges working memory, attention, and motor control for 15 2-h sessions. Before and after training, all subjects received neuropsychological assessments targeting working memory, attention, and procedural learning to assess transfer. Subjects also underwent pre- and post-functional magnetic resonance imaging (fMRI) scans while they played the training videogame to assess how these patterns of brain activity change in response to training. For regions implicated in working memory, such as the superior parietal lobe (SPL), individual differences in the post-minus-pre changes in activation predicted performance changes in an untrained working memory task. These findings suggest that training-induced plasticity in the functional representation of a training task may play a role in individual differences in transfer. Our data support and extend previous literature that has examined the association between training related cognitive changes and associated changes in underlying neural networks. We discuss the role of individual differences in brain function in training generalizability and make suggestions for future cognitive training research.