Research

Daily Cognitive Activity

The Daily Cognitive Activity Project is an interdisciplinary research initiative aimed at systematically capturing, modeling, and understanding human cognitive and behavioral dynamics during everyday life. The project addresses a critical gap in cognitive science and mental-health research: the lack of high-resolution, longitudinal, multimodal data linking cognition, physiology, behavior, and context under naturalistic conditions.

The core objective is to construct a large-scale multimodal dataset integrating neural, physiological, behavioral, and self-reported measures to enable early detection and prevention of cognitive and behavioral dysregulation, including stress, burnout, anxiety, and mood instability.

Building on this dataset, the project develops AI models for cognitive time-series analysis using architectures such as LSTM and reinforcement learning with human feedback (RLHF). These models aim to identify subtle risk patterns, support personalized interventions, and stabilize mental states in high-demand environments.

Data acquisition combines portable EEG, smartwatches, wearable video devices, and validated psychometric instruments, allowing continuous monitoring of cognitive control, attention, motivation, mental workload, emotional stability, and physiological health markers.

Astronaut Training - SANA

The project's goal is to develop (1) cognitive-behavioral protocols and (2) a dashboard aimed at studying and improving stress resilience, attention, self-regulation, and decision-making effectiveness in astronaut candidates for the SANA program. The project meets the objectives of the "Advanced Manufacturing, Digital, and Space Technologies" priority, as it primarily focuses on developing a Kazakhstani astronaut training program and creating accompanying software using artificial intelligence (AI). This will create a scientific basis for developing cognitive training and analog mission technology.

Launching SANA-1 project – all-female isolation experiment in the National Space Center, Kazakhstan

Analogue ESV training capsule control room with two participants of the project.

The research problem stems from a limited understanding of the neuropsychological mechanisms of human adaptation to the monotony of space, confined spaces, prolonged cognitive loads, isolation, and sensory deprivation typical of space missions, as well as technologies that could facilitate adaptation. As one of the most advanced countries in space technology development, Kazakhstan actively participates in international space programs. This project addresses the lack of fundamental research in the field of AI, which is used for cognitive training and psychophysiological monitoring of future astronauts.

The results showed a characteristic adaptive arc: enthusiasm at the beginning of the mission, stress in the middle of the mission, which reduced some indicators of cognitive activity, such as memory and attention, and motivation and stability by the end of the mission.
Women represent 11-17% of space program participants/astronauts globally (Hughes-Fulford, 2024; Wolf, 2025; United Nations Office for Outer Space Affairs, 2025), so this program will also support women in STEM.
The project's methodology includes the use of modern AI technologies, cognitive and neuroscience methods, and behavioral experiments. Psychophysiological research is also planned, including joint experiments with colleagues from the Human-Centered Sensing Lab at ETH Zurich.

Dataset and ML for Mental Health

The project “Mental Health in Youth: Cross-Cultural Data-Driven Study” aims to develop and validate predictive models of mental health outcomes among young adults in Kazakhstan and the United Kingdom. The research addresses the growing challenge of mental health disorders—particularly anxiety, depression, and stress—among university students, which represent a major social and economic concern in both countries. Despite increasing awareness, early detection remains limited due to the lack of large-scale, data-driven systems capable of identifying early risk factors and personal trajectories of psychological distress.

The project introduces a cross-cultural, multimodal research framework combining psychological assessment, passive biometric monitoring, and advanced computational modeling. A representative sample of 6,000 young participants will be recruited from universities and communities in Kazakhstan, including Kazakh-British Technical University, Nazarbayev University, and the National Center for Mental Health, with comparative data collected from the University of Edinburgh. Data acquisition will include:

(1) validated psychometric questionnaires (GAD-7, PHQ-9, WEMWBS, MSPSS, PSS),

(2) passive sensing of physiological and behavioral parameters through wearable devices, and

(3) a smartphone-based ecological momentary assessment (EMA) tool collecting self-reported states every few minutes.

The proposed mechanism of mental health disorders in youth on the example of psychosis, anxiety and depression with data collection and ML model rendering. This project aims to investigate psychological mechanisms associating established social risk factors that will be investigated in the collected data. We believe encapsulating social risk factors associated with changes in psychological mechanisms to investigate whether altered psychological mechanisms are associated with the development of (sub)clinical psychopathology in adolescence will guide development of AI models to predict mental health problems.

Page updated

Google Sites

Report abuse