Search

The effect of anthropogenic impact on groundwater resources in Southeast Asian countries is subject to scientific debate, since it is increasingly considered a contributor to the observed contamination of groundwater with arsenic. It can be diverse, and range from direct pollution, through alteration of hydraulic conditions to indirect effects on geochemical conditions and reactions. While direct pollution can be observed easily, the hydraulic and indirect impact on groundwater quality are more difficult to determine. Nevertheless, considering a generally increasing anthropogenic impact, its complex effects on groundwater chemistry and quality need to be kept in mind, especially if the causes are less obvious. In order to detect this impact, different approaches have been applied over the past few decades. This paper reviews the multiple indicators and approaches used in various studies to trace anthropogenic influences on geochemical conditions leading to high arsenic groundwater content in Bangladesh and South/Southeast Asia. Future research has to address increasingly the spatial complexity and subsurface heterogeneity through clay layers and hydrogeological differences, as well as the temporal development of local contaminant plumes., Asien, Nr. 156/157 (2020): ASIEN (Juli/Oktober)

Visually opaque information affects people’s inferences about another person’s emotional state. For instance, knowledge about prior moral conduct biases the perception of faces, so that faces with neutral expressions may appear as if they were showing negative emotions. Interactions between humans and social robots are predicted to increase in the near future. Here we investigate whether affective knowledge influences the perception and evaluation of robot faces. In a fully counterbalanced design, participants hear positive, neutral and negative stories (e.g., “has shown aggressive behaviour”, “teaches social skills to children with autism”, “works as a hotel receptionist”) about 36 real human-like robots and subsequently rate each robot's trustworthiness and the valence of their facial expressions. We expect affective information to modulate people’s perception and evaluation of the robots’ faces. If so, the findings would suggest that humans see robots as intentional agents, whose past conduct can help predict their future behaviour. This study may therefore provide valuable insights for the design of social robots and inform the ethical discourse surrounding the responsibility of artificial agents.

Can the language we speak influence what we consciously perceive? While evidence accumulates that language modulates visual discrimination, little is known about the relation between language structure and access to visual consciousness. In two previous experiments we employed EEG and the attentional blink paradigm in which targets are often unnoticed. In Experiment 1, native Greek speakers, who distinguish categorically between light and dark shades of blue, showed higher hit rates for this contrast compared to a verbally unmarked green contrast. Electrophysiological signatures of early visual processing (P1 component) predicted this behavioral advantage. Experiment 2 replicated the study with German speakers who place the “Greek” blues into one category, yielding no differences between the blue and the green contrast in behavior and electrophysiological signatures. The present study is a pre-registered behavioral replication in which we will test Russian native speakers, who also distinguish categorically between light and dark shades of blue. We expect to replicate the behavioral effects found in Experiment 1.

How does affective person knowledge influence mental representations of emotional faces? We investigate this by having participants perform a reverse correlation task in which they create classification images of faces that pertain to one negatively valenced character and one positively valenced character. In a subsequent rating task, an independent sampe of participants provides a number of ratings for the classification images. We expect clasification images to be influenced by affective person knowledge, possibly even making a smile disappear.

Visually opaque information affects people’s inferences about another person’s emotional state. For instance, knowledge about prior moral conduct biases the perception of faces, so that faces with neutral expressions may appear as if they were showing negative emotions. Interactions between humans and social robots are predicted to increase in the near future. Here we investigate whether affective knowledge influences the perception and evaluation of robot faces. In a fully counterbalanced design, participants hear positive, neutral and negative stories (e.g., “has shown aggressive behavior”, “teaches social skills to children with autism”, “works as a hotel receptionist”) about 18 real human-like robots and subsequently rate each robot's trustworthiness and the valence of their facial expressions. We expect affective information to modulate people’s perception and evaluation of the robots’ faces. We expect to find evidence in favor of this idea in ERP components associated with different stages of visual processing of robot faces, as well as ratings of trustworthiness and facial expressions. If confirmed, the findings would show that humans can read good or bad intentions into robot faces. This would suggest that we can see robots as intentional agents, whose past conduct can help predict their future behavior. This study may therefore provide valuable insights for the design of social robots and inform the ethical discourse surrounding the responsibility of artificial agents.