Comparison of Modified Yield-Line and Punching Shear Capacities for Concrete Traffic Barriers and Bridge Rails

The traditional, triangular yield-line method used by most departments of transportation for analyzing concrete traffic barriers and bridge rails has been largely unchanged since 1978. Testing of concrete barriers since this time has indicated that the triangular yield-line method is not qualitatively representative of observed damage patterns and is overconservative. Further, the conversion from NCHRP Report 350 to the crash test criteria from the Manual for Assessing Safety Hardware (MASH) will result in increases to lateral impact loads; therefore, overconservative analysis practices may result in many concrete barriers being unnecessarily deemed inadequate. In this research, alternative analysis methods for concrete barriers were extracted from an extensive literature review of concrete barrier investigations. These methods were applied to a sample of eight concrete barriers to demonstrate and compare their effects on capacity estimates. Alternative methods included trapezoidal yield-line mechanisms, effects of impact heights lower than the top of the barrier, punching shear evaluation, and consideration of expected material strengths. Capacity estimates of the selected barriers were increased by an average of 47 percent when alternative methods were cumulatively applied. Although the traditional method does not consider punching shear, the capacity of one of the eight barriers was controlled by punching shear rather than by yield-line flexure. With the alternative methods applied, seven of the eight barriers were deemed adequate relative to the increased lateral loads corresponding to MASH criteria for Test Levels 2 through 5. By contrast, if analyzed according to the traditional method, three of the eight barriers would have been deemed insufficient considering MASH loads.

Unsupervised Online Assessment of Visual Working Memory in 4- to 10-Year-Old Children: Array Size Influences Capacity Estimates and Task Performance

The events of the COVID-19 Pandemic forced many psychologists to abandon lab-based approaches and embrace online experimental techniques. Although lab-based testing will always be the gold standard of experimental precision, several protocols have evolved to enable supervised online testing for paradigms that require direct observation and/or interaction with participants. However, many tasks can be completed online in an unsupervised way, reducing reliance on lab-based resources (e.g., personnel and equipment), increasing flexibility for families, and reducing participant anxiety and/or demand characteristics. The current project demonstrates the feasibility and utility of unsupervised online testing by incorporating a classic change-detection task that has been well-validated in previous lab-based research. In addition to serving as proof-of-concept, our results demonstrate that large online samples are quick and easy to acquire, facilitating novel research questions and speeding the dissemination of results. To accomplish this, we assessed visual working memory (VWM) in 4- to 10-year-old children in an unsupervised online change-detection task using arrays of 1–4 colored circles. Maximum capacity (max K) was calculated across the four array sizes for each child, and estimates were found to be on-par with previously published lab-based findings. Importantly, capacity estimates varied markedly across array size, with estimates derived from larger arrays systematically underestimating VWM capacity for our youngest participants. A linear mixed effect analysis (LME) confirmed this observation, revealing significant quadratic trends for 4- through 7-year-old children, with capacity estimates that initially increased with increasing array size and subsequently decreased, often resulting in estimates that were lower than those obtained from smaller arrays. Follow-up analyses demonstrated that these regressions may have been based on explicit guessing strategies for array sizes perceived too difficult to attempt for our youngest children. This suggests important interactions between VWM performance, age, and array size, and further suggests estimates such as optimal array size might capture both quantitative aspects of VWM performance and qualitative effects of attentional engagement/disengagement. Overall, findings suggest that unsupervised online testing of VWM produces reasonably good estimates and may afford many benefits over traditional lab-based testing, though efforts must be made to ensure task comprehension and compliance.

Linking habitat and population dynamics to inform conservation benchmarks for data-limited salmon stocks

Management of data-limited populations is a key challenge to the sustainability of fisheries around the world. For example, sockeye salmon (Oncorhynchus nerka) spawn and rear in many remote coastal watersheds of British Columbia (BC), Canada, making population assessment a challenge. Estimating conservation and management targets for these populations is particularly relevant given their importance to First Nations and commercial fisheries. Most sockeye salmon have obligate lake-rearing as juveniles, and total abundance is typically limited by production in rearing lakes. Although methods have been developed to estimate population capacity based on nursery lake photosynthetic rate (PR) and lake area or volume, they have not yet been widely incorporated into stock-recruit analyses. We tested the value of combining lake-based capacity estimates with traditional stock-recruit based approaches to assess population status using a hierarchical-Bayesian stock-recruit model for 70 populations across coastal BC. This analysis revealed regional variation in sockeye population productivity (Ricker α), with coastal stocks exhibiting lower mean productivity than those in interior watersheds. Using moderately-informative PR estimates of capacity as priors reduced model uncertainty, with a more than five-fold reduction in credible interval width for estimates of conservation benchmarks (e.g. SMAX - spawner abundance at carrying capacity). We estimated that almost half of these remote sockeye stocks are below one commonly applied conservation benchmarks (SMSY), despite substantial reductions in fishing pressure in recent decades. Thus, habitat-based capacity estimates can dramatically reduce scientific uncertainty in model estimates of management targets that underpin sustainable sockeye fisheries. More generally, our analysis reveals opportunities to integrate spatial analyses of habitat characteristics with population models to inform conservation and management of exploited species where population data are limited.

Reliability of researcher capacity estimates and count data dispersion: a comparison of Poisson, negative binomial, and Conway-Maxwell-Poisson models

Item-response models from the psychometric literature have been proposed for the estimation of researcher capacity. Canonical items that can be incorporated in such models to reflect researcher performance are count data (e.g., number of publications, number of citations). Count data can be modeled by Rasch’s Poisson counts model that assumes equidispersion (i.e., mean and variance must coincide). However, the mean can be larger as compared to the variance (i.e., underdispersion), or b) smaller as compared to the variance (i.e., overdispersion). Ignoring the presence of overdispersion (underdispersion) can cause standard errors to be liberal (conservative), when the Poisson model is used. Indeed, number of publications or number of citations are known to display overdispersion. Underdispersion, however, is far less acknowledged in the literature. In the current investigation the flexible Conway-Maxwell-Poisson count model is used to examine reliability estimates of capacity in relation to various dispersion patterns. It is shown, that reliability of capacity estimates of inventors drops from .84 (Poisson) to .68 (Conway-Maxwell-Poisson) or .69 (negative binomial). Moreover, with some items displaying overdispersion and some items displaying underdispersion, the dispersion pattern in a reanalysis of Mutz and Daniel’s (2018b) researcher data was found to be more complex as compared to previous results. To conclude, a careful examination of competing models including the Conway-Maxwell-Poisson count model should be undertaken prior to any evaluation and interpretation of capacity reliability. Moreover, this work shows that count data psychometric models are well suited for decisions with a focus on top researchers, because conditional reliability estimates (i.e., reliability depending on the level of capacity) were highest for the best researchers.