Postdoctoral Researcher in Multimodal Human Sensing and Advanced Behavioral Data Analysis

Description

EPFL is seeking a postdoctoral researcher to join a Swiss National Science Foundation (SNSF)-funded project investigating human motivation and stress responsiveness. The project will develop and validate individually calibrated behavioral tasks in immersive virtual reality (VR) to quantify effort-based motivation, vigor, persistence and goal-directed versus habitual control. It will then test how acute stress alters these processes, combining behavioral performance, physiological monitoring, movement-based phenotyping, and advanced statistical and computational analyses.

The successful candidate will play a key role in ensuring that these complex data streams are acquired, synchronized, quality-controlled, modeled and interpreted with the highest level of technical and analytical rigor.

Responsibilities

The postdoctoral researcher will lead the technical and quantitative core of the project. Responsibilities will include:

• Developing, implementing, optimizing and troubleshooting immersive behavioral tasks in close interaction with the PI and lab members.
• Integrating behavioral task events with physiological acquisition systems, movement tracking, positioning data and experimental logs.
• Establishing robust acquisition, synchronization, calibration and quality-control procedures across multimodal data streams.
• Troubleshooting software, hardware, sensors, timing, synchronization, data acquisition and experimental workflow issues.
• Extracting and analyzing multimodal behavioral features from head, hand, body and positional tracking data.
• Processing and analyzing physiological signals, including ECG / HRV, electrodermal activity, respiration and autonomic stress indices.
• Developing reproducible pipelines for data preprocessing, feature extraction, statistical modeling, visualization and documentation.
• Implementing advanced statistical and computational analyses, including trial-level models, mixed-effects models, clustering or latent-profile analyses, dimensionality reduction, predictive modeling, cross-validation and interpretable feature analysis.
• Integrating behavioral, kinematic, physiological, endocrine and questionnaire-based measures to characterize individual differences in motivation and stress responsiveness.
• Contributing to experimental design, task calibration, pilot testing, participant testing, manuscript preparation, conference presentations and open-science deliverables.
• Supervising students and contributing to the technical and quantitative training of junior lab members.

Qualifications

Applicants should have a PhD in biomedical engineering, electrical engineering, computer science, data science, computational neuroscience, human movement science, psychophysiology, cognitive neuroscience, psychology with strong quantitative expertise, or a related discipline.

Essential qualifications include:

• Strong programming skills, preferably in Python and/or R, with experience building reproducible data-analysis workflows.
• Excellent quantitative and statistical reasoning, including a clear understanding of model assumptions, uncertainty, validation, data structure, and the limitations of different analytical approaches.
• Experience with multimodal human behavioral data, time-series data, sensor-based data, physiological signals, movement tracking, or related complex datasets.
• Some knowledge of Unity game engine development and experience with network programming with C# or equivalent, at sufficient level to maintain existing data acquisition setup.
• Experience with physiological data acquisition and/or signal processing, ideally including ECG, heart-rate variability, electrodermal activity, respiration, wearable sensors, or related measures.
• Ability to troubleshoot complex experimental setups involving software, hardware, sensors, timing, synchronization, and data acquisition.
• Experience with advanced statistical or computational methods, such as mixed-effects models, hierarchical models, Bayesian models, trial-level analyses, dimensionality reduction, clustering, latent profiles, predictive modeling, or model comparison.
• Ability to build robust pipelines rather than simply apply standard analysis packages.
• Strong interest in human behavior, motivation, stress, individual differences, and quantitative approaches to behavioral neuroscience.
• Excellent organizational, communication, and documentation skills.

Prior experience with virtual reality is welcome but not required. Candidates with strong backgrounds in experimental systems, human sensing, physiological acquisition, signal processing, movement analysis, computational modeling or advanced behavioral data analysis are strongly encouraged to apply.

Additional assets include experience with Biopac or comparable physiological acquisition systems; real-time data acquisition; motion capture or positional tracking; Bayesian or hierarchical modeling; reinforcement-learning models; effort-discounting or decision-making models; survival or hazard models; gradient-boosted decision trees or related predictive approaches; interpretable feature-importance methods; Git/GitHub; Linux or command-line workflows; preregistered analyses; and open-science practices.

Scientific environment

The successful candidate will join the Laboratory of Behavioral Genetics at EPFL, an interdisciplinary environment focused on the biological, behavioral and individual-difference mechanisms of stress, motivation, anxiety and resilience.