Neural network model for imprecise regression with interval dependent variables.

This paper presents a computationally feasible method to compute rigorous bounds on the interval-generalization of regression analysis to account for epistemic uncertainty in the output variables. The new iterative method uses machine learning algorithms to fit an imprecise regression model to data that consist of intervals rather than point values. The method is based on a single-layer interval neural network which can be trained to produce an interval prediction. It seeks parameters for the optimal model that minimizes the mean squared error between the actual and predicted interval values of the dependent variable using a first-order gradient-based optimization and interval analysis computations to model the measurement imprecision of the data. An additional extension to a multi-layer neural network is also presented. We consider the explanatory variables to be precise point values, but the measured dependent values are characterized by interval bounds without any probabilistic information. The proposed iterative method estimates the lower and upper bounds of the expectation region, which is an envelope of all possible precise regression lines obtained by ordinary regression analysis based on any configuration of real-valued points from the respective y-intervals and their x-values.

Agent-based models under uncertainty.

Background: Monte Carlo (MC) is often used when trying to assess the consequences of uncertainty in agent-based models (ABMs). However, this approach is not appropriate when the uncertainty is epistemic rather than aleatory, that is, when it represents a lack of knowledge rather than variation. The free-for-all battleship simulation modelled here is inspired by the children's battleship game, where each battleship is an agent. Methods: The models contrast an MC implementation against an interval implementation for epistemic uncertainty. In this case, our epistemic uncertainty is in the form of an imperfect radar. In the interval method, the approach occludes the status of the agents (ships) and precludes an analyst from making decisions about them in real-time. Results: In a highly uncertain environment, after many time steps, there can be many ships remaining whose status is unknown. In contrast, any MC simulation invariably tends to conclude with a small number of the remaining ships after many time steps. Thus, the interval approach misses the quantitative conclusion. However, some quantitative results are generated by the interval implementation, e.g. the identities of the surviving ships, which are revealed to be nearly mutual with the MC implementation, though with fewer identities in total compared to MC. Conclusions: We have demonstrated that it is possible to implement intervals in an ABM, but the results are broad, which may be useful for generating the overall bounds of the system but do not provide insight on the expected outcomes and trends.

Belongingness challenged: Exploring the impact on older adults during the COVID-19 pandemic.

OBJECTIVES: The sense of belonging is a fundamental human need. Enacting it through face-to-face social activities was no longer possible during the COVID-19 pandemic. In this study, we investigate how the sense of belonging, and how it is enacted, changed longitudinally amongst older adults in the UK. In addition, we examine the interplay of the sense of belonging and resilience over time. METHODS: We employed a longitudinal qualitative research design to explore the experiences of older adults during one year of the COVID-19 pandemic (April 2020-April 2021). The analysis was undertaken with constructivist grounded theory. FINDINGS: Before the pandemic older adults were free to engage in social relationships with family and friends, often enacted within social activity groups where they felt valued and gained positive experiences. During the pandemic face to face enactment of belongingness was reduced; adjustments needed to be made to maintain the sense of belonging. The experience of older adults was heterogeneous. We examine three themes. First, how belongingness was enacted prior to the pandemic. Examples include: family holidays, visiting each other, sports activities, eating with friends and family, and visiting cultural events. Second, how participants adapted and maintained their social involvement. Examples include: distanced face-to-face activities; and learning new technology. Third, for some, a belongingness gap emerged and persisted. There was an irretrievable loss of family members or friends, the closure of social groups, or withdrawal from groups as priorities changed. As a consequence, of challenged belongingness, participants expressed increased loneliness, anxiety, social isolation, frustration and, feelings of depression. For many, the disrupted sense of belonging no longer fostered resilience, and some previously resilient participants were no longer resilient.

Is Protecting Older Adults from COVID-19 Ageism? A Comparative Cross-cultural Constructive Grounded Theory from the United Kingdom and Colombia.

The COVID-19 pandemic impacted people's lives all over the world, requiring health and safety measures intended to stop the virus from spreading. This study explores whether an unintended consequence of these measures is a new form of ageism. We explore, using qualitative methods, the experiences of older adults living through the pandemic in the United Kingdom and Colombia. Although there were some small differences between countries, for the most part, the experiences were similar. We found that older adults reported that they were seen as a homogenous group and experienced both benevolent and hostile ageism and a loss of autonomy as a consequence of COVID-19 protection measures. Participants from both countries expressed anger and frustration, and increased anxiety, and felt that their individuality was ignored. We recommend that policy-makers, the media, and wider society consider the impact of such health and safety measures on older adults in preparing for future pandemics and health challenges.

Correction: Is no test better than a bad test: Impact of diagnostic uncertainty on the spread of COVID-19.

[This corrects the article DOI: 10.1371/journal.pone.0240775.].

Is "no test is better than a bad test"? Impact of diagnostic uncertainty in mass testing on the spread of COVID-19.

Testing is viewed as a critical aspect of any strategy to tackle epidemics. Much of the dialogue around testing has concentrated on how countries can scale up capacity, but the uncertainty in testing has not received nearly as much attention beyond asking if a test is accurate enough to be used. Even for highly accurate tests, false positives and false negatives will accumulate as mass testing strategies are employed under pressure, and these misdiagnoses could have major implications on the ability of governments to suppress the virus. The present analysis uses a modified SIR model to understand the implication and magnitude of misdiagnosis in the context of ending lockdown measures. The results indicate that increased testing capacity alone will not provide a solution to lockdown measures. The progression of the epidemic and peak infections is shown to depend heavily on test characteristics, test targeting, and prevalence of the infection. Antibody based immunity passports are rejected as a solution to ending lockdown, as they can put the population at risk if poorly targeted. Similarly, mass screening for active viral infection may only be beneficial if it can be sufficiently well targeted, otherwise reliance on this approach for protection of the population can again put them at risk. A well targeted active viral test combined with a slow release rate is a viable strategy for continuous suppression of the virus.

Comparative, collaborative, and integrative risk governance for emerging technologies.

Various emerging technologies challenge existing governance processes to identify, assess, and manage risk. Though the existing risk-based paradigm has been essential for assessment of many chemical, biological, radiological, and nuclear technologies, a complementary approach may be warranted for the early-stage assessment and management challenges of high uncertainty technologies ranging from nanotechnology to synthetic biology to artificial intelligence, among many others. This paper argues for a risk governance approach that integrates quantitative experimental information alongside qualitative expert insight to characterize and balance the risks, benefits, costs, and societal implications of emerging technologies. Various articles in scholarly literature have highlighted differing points of how to address technological uncertainty, and this article builds upon such knowledge to explain how an emerging technology risk governance process should be driven by a multi-stakeholder effort, incorporate various disparate sources of information, review various endpoints and outcomes, and comparatively assess emerging technology performance against existing conventional products in a given application area. At least in the early stages of development when quantitative data for risk assessment remain incomplete or limited, such an approach can be valuable for policymakers and decision makers to evaluate the impact that such technologies may have upon human and environmental health.

Ecological risk assessment for mink and short-tailed shrew exposed to PCBs, dioxins, and furans in the Housatonic River area.

A probabilistic risk assessment was conducted to characterize risks to a representative piscivorous mammal (mink, Mustela vison) and a representative carnivorous mammal (short-tailed shrew, Blarina brevicauda) exposed to PCBs, dioxins, and furans in the Housatonic River area downstream of the General Electric (GE) facility in Pittsfield, Massachusetts. Contaminant exposure was estimated using a probabilistic total daily intake model and parameterized using life history information of each species and concentrations of PCBs, dioxins, and furans in prey collected in the Housatonic River study area. The effects assessment preferentially relied on dose-response curves but defaulted to benchmarks or other estimates of effect when there were insufficient toxicity data. The risk characterization used a weight of evidence approach. Up to 3 lines of evidence were used to estimate risks to the selected mammal species: 1) probabilistic exposure and effects modeling, 2) field surveys, and 3) species-specific feeding or field studies. The weight of evidence assessment indicated a high risk for mink and an intermediate risk for short-tailed shrew.

Accounting for uncertainty in DNA sequencing data.

Science is defined in part by an honest exposition of the uncertainties that arise in measurements and propagate through calculations and inferences, so that the reliabilities of its conclusions are made apparent. The recent rapid development of high-throughput DNA sequencing technologies has dramatically increased the number of measurements made at the biochemical and molecular level. These data come from many different DNA-sequencing technologies, each with their own platform-specific errors and biases, which vary widely. Several statistical studies have tried to measure error rates for basic determinations, but there are no general schemes to project these uncertainties so as to assess the surety of the conclusions drawn about genetic, epigenetic, and more general biological questions. We review here the state of uncertainty quantification in DNA sequencing applications, describe sources of error, and propose methods that can be used for accounting and propagating these errors and their uncertainties through subsequent calculations.

Strategies for risk communication: evolution, evidence, experience.

This volume presents the proceedings of the symposium entitled Strategies for Risk Communication: Evolution, Evidence, Experience. The symposium was held in Montauk, Long Island, New York on May 15-17, 2006. It explored practical methods and robust theories of risk communication informed by recent research in risk perception, neuroscience, and the evolutionary social sciences. The symposium focused on what experimental, survey, and brain imaging research has uncovered about how humans process and perceive uncertainty and risks and what the evolutionary history of humans suggests about how we understand and respond to risks. The purpose of the symposium and of this collection of papers is to begin to synthesize the findings from these diverse fields and inform the development of practical strategies for risk communication.

A frequency/consequence-based technique for visualizing and communicating uncertainty and perception of risk.

This chapter presents an approach under development for communicating uncertainty regarding risk. The approach relies on a risk imaging technology that decomposes risk into two basic elements: (i) the frequency of each kind of harm associated with a hazard and (ii) the adversity of each of those harms. Because different kinds of harm are often measured along incompatible dimensions, adversity is quantified on an ordinal scale. Frequency is quantified on a ratio scale. Sampling error, measurement error, and bias all contribute to uncertainty about frequency. Differences in opinion, measurement error, and choice of dimensions lead to uncertainty about adversity. In this chapter, risk is imaged as an area circumscribed by uncertainty bounds around all of the harms. This area is called the risk profile of a hazard. Different individuals and groups respond to uncertainty and risk differently, and the risk profile can be further focused to visualize particular risk perceptions. These alternate risk visualizations may be contrasted and compared across management choices or across different risk perceivers to facilitate communication and decision making. To illustrate the method, we image published clinical trial data.

Evolved altruism, strong reciprocity, and perception of risk.

Humans have a long history of coping with particular recurring risks. We expect natural selection to have resulted in specific physiological and psychological adaptations that respond well to these risks. Why, then, does it seem so difficult to communicate risk? We suggest that the human mind has been structured by natural selection to use a mental calculus for reckoning uncertainty and making decisions in the face of risk that can be substantially different from probability theory, propositional calculus (logic), or economic rationality (utility maximization). We argue that this is because of the unique armamentarium of strategies humans have evolved to cope with the risks faced during our long history living as hunter-gatherers. In particular, we believe the risk of social contract violation (not contributing a fair share to cooperative endeavors) was an important selective factor because reciprocity, reciprocal altruism, and cooperation are primary adaptations to the most important risks our ancestors faced.

Comparison of deterministic and probabilistic calculation of ecological soil screening levels.

The U.S. Environmental Protection Agency (U.S. EPA) is sponsoring development of ecological soil screening levels (Eco-SSLs) for terrestrial wildlife. These are intended to be used to identify chemicals of potential ecological concern at Superfund sites. Ecological soil screening levels represent concentrations of contaminants in soils that are believed to be protective of ecological receptors. An exposure model, based on soil- and food-ingestion rates and the relationship between the concentrations of contaminants in soil and food, has been developed for estimation of wildlife Eco-SSLs. It is important to understand how conservative and protective these values are, how parameterization of the model influences the resulting Eco-SSL, and how the treatment of uncertainty impacts results. The Eco-SSLs were calculated for meadow voles (Microtus pennsylvanicus) and northern short-tailed shrews (Blarina brevicauda) for lead and DDT using deterministic and probabilistic methods. Conclusions obtained include that use of central-tendency point estimates may result in hazard quotients much larger than one; that a Monte Carlo approach also leads to hazard quotients that can be substantially larger than one; that, if no hazard quotients larger than one are allowed, any probabilistic approach is identical to a worst-case approach; and that an improvement in the quality and amount of data is necessary to increase confidence that Eco-SSLs are protective at their intended levels of conservatism.