Institute of Education

Publications

• Book

  Kirill Ditkovsky, Fridlyanova S., Gokhberg L. et al.

  Science. Technology. Innovation: 2024 : Pocket Data Book

  This pocket data book contains main S&T and innovation indicators for the Russian Federation. The publication includes the most recent statistical data on R&D input and output, as well as international comparisons. The data book includes information of the Russian Federal State Statistics Service (Rosstat), Ministry of Science and Higher Education of the Russian Federation, Russian Federal Service for Intellectual Property (Rospatent), Organisation for Economic Co-operation and Development (OECD), European Statistical Office (Eurostat), UNESCO, World Intellectual Property Organisation (WIPO), national statistical offices of other countries, and results of methodological and analytical studies of the HSE Institute for Statistical Studies and Economics of Knowledge. In some cases, 2022 data are preliminary.

  National Research University Higher School of Economics, 2024.

• Article

  Eremenko J. A., Martinez-Saito M., Naumova K. et al.

  Neurophysiological markers of hedonic taste assessment

  In recent years, there has been significant interest in studying hedonistic food experiences using neurophysiological methods within the field of neuroscience. Understanding the neural mechanisms behind the perception of pleasure and aversion to food stimuli has implications for both the fundamental processes of human sensory and emotional taste processing and for addressing health-related issues such as obesity and eating disorders.

  In our study, we used functional near-infrared spectroscopy (fNIRS), heart rate (HR), electrodermal activity (EDA), and electromyography (EMG) to measure brain and peripheral nervous system reactions to pleasant and neutral foods.

  Using fNIRS, we found that while the insula was activated in response to subjectively pleasant food, the right precentral gyrus was activated in response to neutral taste. This finding suggests interhemispheric asymmetry in relation to neutral taste. The EMG results demonstrated that the corrugator facial muscle was sensitive to neutral food, whereas the zygomaticus facial muscle was sensitive to pleasant food. EDA responses were larger for pleasant food, and HR increased in response to all food stimuli.

  The convergence of neuroimaging and peripheral physiological results provides a powerful framework for advancing our understanding of hedonistic responses to food. By elucidating the neural and physiological bases of food hedonics, researchers can gain insights into the complex nature of human appetite.

  Food Quality and Preference. 2024.

• Book chapter

  Lysova-Golomzina E., Kapuza A.

  VALIDATION OF THRESHOLD CONCEPTS IN CASE OF SOCIAL SCIENCE AT MIDDLE SCHOOL USING CONCEPT MAPS

  In bk.: 16th International Conference on Education and New Learning Technologies, Palma, Spain. 1-3 July, 2024. EDULEARN24 Proceedings. IATED, 2024. P. 2290-2295.

• Working paper

  Kardanova E., Ivanova A., Tarasova K. et al.

  A Novel Psychometrics-Based Approach to Developing Professional Competency Benchmark for Large Language Models

  The era of large language models (LLM) raises questions not only about how to train models, but also about how to evaluate them. Despite numerous existing benchmarks, insufficient attention is often given to creating assessments that test LLMs in a valid and reliable manner. To address this challenge, we accommodate the Evidence-centered design (ECD) methodology and propose a comprehensive approach to benchmark development based on rigorous psychometric principles. In this paper, we have made the first attempt to illustrate this approach by creating a new benchmark in the field of pedagogy and education, highlighting the limitations of existing benchmark development approach and taking into account the development of LLMs. We conclude that a new approach to benchmarking is required to match the growing complexity of AI applications in the educational context. We construct a novel benchmark guided by the Bloom's taxonomy and rigorously designed by a consortium of education experts trained in test development. Thus the current benchmark provides an academically robust and practical assessment tool tailored for LLMs, rather than human participants. Tested empirically on the GPT model in the Russian language, it evaluates model performance across varied task complexities, revealing critical gaps in current LLM capabilities. Our results indicate that while generative AI tools hold significant promise for education - potentially supporting tasks such as personalized tutoring, real-time feedback, and multilingual learning - their reliability as autonomous teachers' assistants right now remain rather limited, particularly in tasks requiring deeper cognitive engagement.

  Computation and Language (cs.CL); Artificial Intelligence (cs.AI). cs.CL. arXiv, 2024