ScienceDirect

This article examines objections to the use of artificial intelligence (AI) in decisions affecting humans and argues that most objections rest on anthropocentric assumptions rather than evidence about decision quality. Critics often presume that human decision makers are uniquely capable of moral and contextual judgment, while AI systems are inferior, opaque or hostile. The article challenges this view arguing that modern AI systems are an expression of collective human intelligence and ethics, built on the image of human cognition, trained on curated human decisions and operating on best decision making frameworks. Drawing on empirical studies the article shows that modern AI systems offer unique advantages in decision-making, thus may already in some domains be as good at making good decisions as an average individual human. The article deconstructs main criticisms of AI decision-making and introduces the novel argument of inseparability between human and AI decisions. Human and AI contributions are increasingly intertwined, AI involvement is latent and appropriated by humans, making existing accountability frameworks based on a clear human–AI boundary obsolete. The article advocates for the development of agnostic decision-making frameworks that apply universal accountability to both human and non-human agents and provide a path to better decisions.

Floods are recurrent events that are expected to increase in a climate change context. Nature-based solutions (NBS) are increasingly used in flood risk management, as they offer multiple co-benefits. Different policies and strategies, such as the European Union Biodiversity Strategy for 2030, advocate for implementing NBS. However, they require more land than grey infrastructure. The needed land often belongs to private persons and is thus protected from unlawful, disproportional state interference. Therefore, state authorities should establish a suitable legal framework to enforce the adoption of NBS on privately owned land. These provisions could stem from national or EU regulations concerning disaster risk management, environmental protection, or spatial planning. For those provisions to be successful, it is important to establish a strong partnership with stakeholders.

Objective We examined the prevalence of different types of childhood sexual abuse (CSA) in Lithuania focusing on how these experiences were related to victim's age and relationship to the perpetrator. Methods The participants came from a representative household survey of youth aged 18–29 and the sample consisted of 2000 participants (47.7 % women) with a mean age of 23.9 years (SD = 3.6). The ISPCAN Child Abuse Screening Tool Retrospective version (ICAST-R, Dunne et al., 2009) for young adults was used to investigate childhood exposure to CSA. Both current sociodemographic information and information on circumstances during childhood (e.g., number of siblings, family members with whom the participant lived between 7 and 12 year of age) were collected. Results The prevalence of any CSA experience before the age of 18 was 15.9 %, with higher rates for women (13.5 % and 18.5 % for men and women, respectively). Being spoken to in a sexual way or sexual things being written about the person was the most prevalent form of CSA. In contrast, number of victims that reported sexual intercourse was much lower. The most common type of perpetrator was another young person. Intrafamilial CSA was rare with higher risk in step-relationships. Conclusions Our study provides representative prevalence estimates of the CSA in a region for which such estimates were not previously available. These Lithuanian estimates correspond well with previous literature. Importantly, the findings point to peers being an important perpetrator group in CSA cases. This and other findings can inform the planning of relevant policy measures and actions to both prevent CSA and investigate cases effectively.

Intensive agricultural practices increase agrochemical pollution, particularly nitrogen (N) based fertilizers, which present an environmental risk. This study aims to evaluate long-term (2009–2020) data on soil water regime and nitrate dynamics at an agricultural experimental site on fine-textured soils and to better understand the implications of N management in relation to groundwater pollution. The field site is located in the Biđ field (eastern Croatia), in the proximity of the Sava river. Zero-tension lysimeters were installed at six selected locations. Lysimeters were used to monitor the water regime, i.e., outflows in which nitrate concentration was measured, while additional soil-water samples were collected via 4 and 15-meter-deep monitoring wells. Soil hydraulic parameters were estimated by combining the laboratory measurements, and estimation in RETC software. Water regime and nitrate leaching in lysimeters were simulated using HYDRUS-1D for each year to allow crop rotation and to evaluate their effects individually. The HYDRUS-1D model successfully reproduced lysimeter outflows and nitrate dynamics, which was confirmed with high R2 values (water: 93% above 0.7, and nitrate: 73% above 0.7) indicating the good performance of the model simulating nitrification chain reactions. Principal component analysis (PCA) was performed to identify the relationships among all soil properties and environmental characteristics. The results showed the complex interaction of soil hydraulic properties, precipitation patterns, plant uptake, and N application. All locations have a decreasing trend of nitrate leaching over the investigation period. Most of the lysimeter outflows and elevated nitrate concentrations were connected to the wet period of the year when the soil was saturated, and evapotranspiration was low. The results of this study show that it is important to optimize N fertilizer applications for each particular environmental condition to reduce nitrate loss. The study indicates the importance of long-term field studies, key for agro-hydrological modeling and the improvement of agricultural practices.

Armed conflicts disturb the environment and impair land productivity. Afghanistan has been submerged in conflict for >20 years, affecting the environment dramatically. In this study, we used the Normalised difference vegetation index (NDVI) to investigate vegetation's spatial and temporal changes and the potential underpinned mechanisms. We found a 16.44 % increase in NDVI in Afghanistan from 2000 to 2021. The average NDVI growth rate was 11.33 % (within 5 km distance from the armed conflict), higher in the conflict group than in the non-conflict group. People migration may have reduced the human impacts on the environment. The relative contribution of armed conflict to vegetation growth was 3.17 %. Our results showed that the vegetation in Afghanistan increased, confirming the idea that depopulation increase greenness. Despite the reduced variance explained by the war (R2 values around 0.3), our study provides empirical evidence on the linkages between the war and vegetation change in Afghanistan.

Previous studies reported that acquiring computer programming skills is challenging and might result in high dropout rates. A quasi-experimental design was used to examine the role of different factors in dropping out of an e-based computer programming course. This study applied a knowledge in programming assessment test (20 multiple-choice questions covering the following topics: variables, loops, conditionals, functions, and general knowledge of Python), The Learning Motivating Factors Questionnaire, The Big Five-2, and The Basic Psychological Need Satisfaction & Frustration Scale. Ninety-four participants (38 males and 56 females) completed the course, while 305 participants started it. The mean age of e-learners was 29.96 years (SD 8.27), age range = 18 to 54. The results showed that e-learners who completed the course had higher initial knowledge assessment scores than those who dropped out after the first assessment. Reward and recognition as a motivator were significantly higher in males who completed the course than those who dropped out after the second knowledge assessment. Extraversion was significantly lower in females who completed the course than those who dropped after the first or second knowledge assessment test. Relatedness frustration was significantly higher in those who dropped out after the first knowledge assessment. Due to significant limitations of the sample size, cultural context, measures applied, and research design, the findings would preferably be regarded with caution.

The meteorological conditions may affect COVID-19 transmission. However, the roles of seasonality and macro-climate are still contentious due to the limited time series for early-stage studies. We studied meteorological factors' effects on COVID-19 transmission in Brazil from February 25 to November 15, 2020. We aimed to explore whether this impact showed seasonal characteristics and spatial variations related to the macro-climate. We applied two-way fixed-effect models to identify the effects of meteorological factors on COVID-19 transmission and used spatial analysis to explore their spatial-temporal characteristics with a relatively long-time span. The results showed that cold, dry and windless conditions aggravated COVID-19 transmission. The daily average temperature, humidity, and wind speed negatively affected the daily new cases. Humidity and temperature played a dominant role in this process. For the time series, the influences of meteorological conditions on COVID-19 had a periodic fluctuation of 3–4 months (in line with the seasons in Brazil). The turning points of this fluctuation occurred at the turn of seasons. Spatially, the negative effects of temperature and humidity on COVID-19 transmission clustered in the northeastern and central parts of Brazil. This is consistent with the range of arid climate types. Overall, the seasonality and similar climate types should be considered to estimate the spatial-temporal COVID-19 patterns. Winter is a critical time to be alert for COVID-19, especially in the northern part of Brazil.

Urban green spaces (UGS) are essential for human well-being, contributing towards quality of life and promoting social interactions and inclusion. Accessibility to recreational UGS is key to human well-being in urban areas. However, there is a lack of works focused on recreational UGS accessibility in cities from different countries, using different types of transport. In this work, we studied the accessibility to recreational UGS in Vilnius (Lithuania) and Coimbra (Portugal) through four transport modes (foot, bike, bus and car) within 5- and 10-minute distance. The results showed important differences between cities. Vilnius had higher accessibility to recreational UGS than Coimbra in all transport modes. In both cities, the accessibility was highest by car and lowest by bus. Also, the percentage of population covered by 5- and 10-minute distance UGS accessibility in Vilnius is higher than in Coimbra, despite the lower population density in Vilnius. A large number of recreational UGS, their geographic distribution, the dispersed population pattern, and the location of several urbanizations near recreational UGS increased the inhabitant’s accessibility to recreational UGS in Vilnius. The differences in accessibility between cities were mainly attributed to the large number of recreational UGS and higher road network density, cycleways and footways density, and bus lines in Vilnius, compared to Coimbra. This work identified the areas where recreational UGS accessibility is difficult in using different transport modes and need to be improved in study areas. This is essential to support urban planners and increase cities’ well-being. The findings of this work will be important for the management of the studied urban areas.

Heavy rain events are expected to be more frequent and intense due to climate change, which will likely increase the vulnerability of the urban area to flood episodes. In this context, it is crucial to identify the areas with a high capacity to retain water and the areas where flood protection is more needed. To address this problem, this work aims to map and assess the flood regulation supply and demand ecosystem services in Vilnius, Lithuania. The results showed that Vilnius has a large area with a flood regulation supply, especially in the northeast and southern areas. The areas with high flood regulation demand were located in the old town and several urbanizations scattered inside the city limits. We used the soil-vegetation adjusted index (SAVI) to validate the flood regulation supply index using linear regression. In contrast, the flood regulation demand index was validated by comparing the historical flood points with the model results. SAVI predicted acceptably (r2 = 0.43), and the flooded points identified were located in the areas with high flood regulation demand index. Both models were mismatched (−0.50. P < 0.001), showing that they have different spatial patterns. These results obtained in this work can help planners and decision-makers better plan the areas with a high capacity to retain water and identify the areas where the demand is high.