Czech Academy of Sciences, Institute of Computer Science

Previous meta-analyses show that social robots aid learning but were often limited in scope or grouped diverse control conditions together. This meta-analysis examined learning outcomes, focusing on control condition type. We retrieved 146 studies (Google Scholar and reference searches) where a physical social robot was used for training cognitive skills, comprising 183 post-test effect sizes between the robot and the control group, and 372 pre-post effect sizes. Analysis of the 78 studies with control groups indicated that robots generally improved learning, most notably when compared to a no-training control group (d = 0.75). Comparing robots to human teachers yielded an overall positive effect (d = 0.31), although effect sizes varied widely. This variability was explained by the robot’s role: robots in a co-teaching capacity showed a strong positive effect (d = 0.88), while robots replacing the teacher showed no benefit (d = −0.06). LLM-based sentiment analysis indicated that papers from outside Europe received higher positivity scores when describing the robots. We conclude that the effect size is influenced by the robot implementation and the control condition chosen.

The goal of inductive logic programming (ILP) is to search for a hypothesis that generalises training data and background knowledge. The challenge is searching vast hypothesis spaces, which is exacerbated because many logically equivalent hypotheses exist. To address this challenge, we introduce a method to break symmetries in the hypothesis space. We implement our idea in answer set programming. Our experiments on multiple domains, including visual reasoning and game playing, show that our approach can reduce solving times from over an hour to just 17 seconds.

In this work we are expanding upon our previous statistical modeling methodology for reconstruction of daily courses of SO42-, NO3- and NH4+ concentrations in cumulative precipitation samples (Hunova et al 2022, Hunova et al 2024). Here, we investigate relationship between the wet concentrations and concentrations of gaseous pollutants (NOx, SO2, NH3) in a detailed way from several viewpoints using modern and highly flexible statistical approach based on GAM (Generalzied Additive Model) with complexity-penalized spline components. This framework allows for decomposition of influences upon wet concentration into several easily interpretable components as well as for the nonlinearity dictated by basic physical consideration (such as saturation phenomena). The GAM model is formulated in such a way that it deals appropriately with the time-aggregated collections whose length changed historically (from 7 days to 1 day). Since it is quite obvious that the wet concentration in a spatial point sample is related to the gas concentration in a much broader area, we work not only with gaseous concentrations measured at the wet-sample-collection location, but also with numerical air pollution model (CAMx) aerial average estimates in our simultaneous models. Here, it is not clear a priori, how large neighborhood we should take for the most informative CAMx output spatial average and similarly whether it is more natural to take CAMx concentrations at surface or in 50 m. Therefore, we assess both of these features in a formalized way (via AIC model comparison). The modelling approach is illustrated at four professional Czech Hydrometeorological Institute (CHMI) stations (ALIB, JKOS, PPRM, TBKR) with long-term data (2016-2021). There, it turns out that both local measurement and relatively large area CAMx gaseous concentration averages influence water sample ion concentrations significantly. Hight and spatial aggregation differs for different ions. Further analysis using time-varying (TVAR) framework then shows that the influence of the concentration measurements is highly seasonal (local slope changes smoothly but very substantially, favoring spring to summer influences). The work has been done in cooperation with the CHMI and is related to the Technology Agency Czech Republic project ARAMIS, SS02030031).

ABSTRACT A theorem of Frieze from 1985 asserts that the total length of the minimum spanning tree of the complete graph whose edges get independent lengths from the distribution converges to Apéry's constant in probability, as . We generalize this result to sequences of graphs that converge to a graphon . Further, we allow the lengths of the edges to be drawn from different distributions (subject to moderate conditions). The limiting total length of the minimum spanning tree is expressed in terms of a certain branching process defined on , which was studied previously by Bollobás, Janson and Riordan in connection with the giant component in inhomogeneous random graphs.

ABSTRACT Using pikeperch ( Sander lucioperca ) as a model species, we investigated the factors influencing fluctuations in the abundance of piscivorous intra‐annual cohorts and assessed the role of three distinct cohorts: extremely small (ES), ordinary (or mid‐sized) fingerlings, and piscivorous in shaping year‐class strength. Our analysis was based on 13 years of data (2003–2024) collected in the Lipno Reservoir, Czechia. Using LASSO for variable selection among 45 potential covariates and subsequent Generalized Additive Model (GAM) construction, our results revealed that a higher abundance of ES conspecifics and young‐of‐the‐year (YOY) perch, coupled with increased densities of large cladocerans in July, positively impacted the abundance of the piscivorous intra‐annual cohort. The abundance of 1+ pikeperch was positively influenced by the abundance of piscivorous and ordinary cohorts recorded in the preceding year. We concluded that all three intra‐annual cohorts contribute to the year‐class strength of pikeperch, with piscivorous and ordinary cohorts having a more direct effect, while ES intra‐annual cohorts indirectly support the piscivorous cohort by serving as prey.

The citizen science approach enables the collection of extensive ecological data while reducing time and cost, offering a valuable tool to address complex conservation challenges. However, its precision in monitoring morphologically similar freshwater fish species remains insufficiently explored. This study evaluates the effectiveness of citizen science in distinguishing between the critically endangered crucian carp ( Carassius carassius ) and the invasive gibel carp ( Carassius gibelio ), two closely related species often confused due to their similar appearance. The native crucian carp has experienced a sharp population decline in Central Europe and is currently classified as critically endangered in the Czech Republic. In response, the citizen science initiative, save the crucian carp was launched to map the species’ distribution and support its conservation and restoration. We assessed the project’s effectiveness using data submitted by 953 participants, which included current and historical occurrence records for both crucian and gibel carp. The analysis focused on species identification accuracy, public engagement, and the influence of media outreach on participation. Field verification of reported sightings revealed a 35% accuracy rate for crucian carp identification. A positive correlation was observed between respondents’ species identification quiz scores and the number of valid occurrence tips submitted. Participants aged 31–50 showed the highest engagement in conservation efforts. Notably, media campaigns significantly boosted public involvement, highlighting the importance of awareness-building in fostering biodiversity stewardship. Data collected through citizen science exhibited broader coverage and greater reporting regularity compared to records from the Nature Conservation Agency of the Czech Republic, underscoring the value of citizen contributions in shaping conservation strategies. Overall, this study illustrates the promise of citizen science as a scalable and impactful approach for monitoring and conserving threatened freshwater fish species.

Individual variability in psilocybin response is a major challenge for psychedelic-assisted therapy, with structural brain features potentially serving as predictive biomarkers. (Lewis et al. Biomedicines 8(2):34 2020) reported that rostral anterior cingulate cortex thickness predicted emotional experiences under psilocybin, suggesting cortical morphometry as a marker of psychedelic responsivity. This study sought to replicate and extend these findings by examining associations between cingulate thickness and psilocybin-induced altered states of consciousness using comprehensive assessment and rigorous statistical control. Twenty-five healthy participants underwent a double-blind, placebo-controlled crossover design with psilocybin (0.26 mg/kg) and placebo. High-resolution T1-weighted magnetic resonance imaging (MRI) measured cortical thickness across cingulate subregions. Subjective effects were assessed with the Altered States of Consciousness (ASC) questionnaire. Analyses applied false discovery rate (FDR) correction for multiple comparisons. The primary Lewis et al. finding—that rostral anterior cingulate cortex thickness predicts emotional psilocybin responses—showed a comparable effect size (β = 0.523 vs. their range 0.324–0.572) that did not achieve statistical significance (p = 0.297), likely reflecting limited statistical power given our smaller sample (N = 25 vs. N = 55). We identified an anterior–posterior gradient in cingulate thickness that significantly predicted psychedelic experience intensity (r = 0.676, FDR p = 0.0004). Findings indicate that spatial organization within the cingulate cortex provides a neuroanatomical marker of variability in psychedelic response. Results highlight the importance of organizational patterns within the cingulate cortex, rather than focal regional measures, when predicting psychedelic effects.

The electroencephalogram (EEG) reflecting brain activity may be characterised through brief periods of stable neural activity patterns that recur over time and are referred to as microstates. Microstates are related to a range of cognitive processes, and their analysis has become an increasingly popular tool for studying human brain function. While microstates have been extensively studied in humans, their presence and characteristics in animal models have yet to be as thoroughly investigated. This study aims to address this gap by detecting and characterising microstates in EEGs of rats collected using a superficial electrode system corresponding to homological areas of the human 10–20 system. Specifically, we demonstrate the presence of microstates in rats’ EEGs, i.e., those that may be captured by the same metrics as in humans. We identified five microstate EEG maps in rats, explaining 71% of the variance in our dataset (N = 30). The explained variance, mean temporal coverage values (0.2), and average duration (0.26 s) are comparable to the human-derived EEG microstates. Via a source localisation technique, the cingulate cortex, precuneus, and insula were found to be associated with the microstates’ temporal dynamics. Among the microstates that showed a broadband character, we also found those that showed an association with the theta and alpha bands. These findings have important implications for the use of microstates as a preclinical tool for investigating brain functions, detecting new biomarkers of brain diseases, and translating this knowledge to humans.