Senior Research Associate in Machine Learning for Spontaneous Inner Speech Detection

The Department of Psychology at Lancaster University is seeking to appoint a Senior Research Associate in Machine Learning (ML), at 0.8 FTE (4 days per week) to develop novel computational methods for objectively detecting inner speech – the voice in our heads – from EEG signals.

This post is indefinite, subject to funding end date. Funding is currently only available for 7 months, from February 2026 to September 2026.

THE PROJECT

Inner speech – talking to yourself in your mind – appears fundamental to human consciousness, thinking, and self-reflection. Yet we have no reliable way to objectively detect or measure it as it happens spontaneously in everyday life. This project tackles one of cognitive neuroscience's most challenging problems: detecting fleeting, spontaneous inner speech from the "haystack" of ongoing brain activity.

The Challenge

Traditional classification approaches have shown limited success, likely because spontaneous inner speech is sparse and co-occurs with other brain activities. Unlike controlled laboratory tasks, we do not have precise temporal labels for when inner speech occurs – instead, we use experimental designs where inner speech is more likely in some conditions than others.

Can alternative ML paradigms - such as methods suited to weakly-supervised/unsupervised settings, transfer learning from controlled speech tasks to naturalistic cognition, or contrastive learning to identify state changes - better capture these spontaneous inner speech events?

Your Role

You will work alongside Dr Bo Yao (Lancaster University) and Professor Xin Yao (Lingnan University, Hong Kong) to develop a novel ML approach for inner speech detection.

By the end of this project, we aim to have:

• A working implementation of one novel approach for detecting inner speech from EEG
• Systematic validation demonstrating performance relative to neuroscience-informed baseline methods
• Documented comparison of approaches with insights into what works, what doesn't, and why
• Lead manuscript preparation for publication

You'll work with high-density EEG data that capture inner speech during silent speech tasks as well as naturalistic cognition, with access to Lancaster's high-performance computing facilities. This is a fast-paced, iterative project requiring rapid prototyping and adaptation to emerging results.

ABOUT YOU

Essential Requirements

• PhD Awarded in Machine Learning, Computer Science, Computational Neuroscience, or related field with substantial ML expertise
• Demonstrable experience developing and implementing deep learning models for time-series or sequential data (e.g., EEG, biosignals, audio, sensor data, or similar)
• Strong Python skills and proficiency with PyTorch or TensorFlow
• Proven ability to work independently on complex technical problems while meeting milestones
• Excellent communication skills: clear technical writing and ability to explain ML concepts to interdisciplinary audiences
• Excitement about tackling unsolved problems where standard approaches haven't worked

For this short-term project, candidates must have the right to work in the UK for the duration of the project.

Desirable Qualifications

• Experience with ML paradigms suited to unlabelled or sparsely-labelled sequential data (self-supervised learning, anomaly detection, domain adaptation, semi-supervised methods, or similar approaches)
• Familiarity with model interpretability techniques (attention mechanisms, saliency mapping, feature visualisation)
• Prior work with neuroimaging data (EEG, MEG, fMRI) or biosignals
• Track record of first-author publications in ML or computational neuroscience

WHY THIS ROLE?

What we offer:

• Intellectual freedom to explore novel approaches – you'll have real ownership of your methods
• Rare cross-disciplinary experience – master neuroscience methods while applying your ML expertise
• Fast-track impact – see your methods in use within months
• Flexibility – 0.8 FTE gives you space for other projects or life commitments
• Strong mentorship – collaboration with experts in neuroscience and AI
• Portfolio diversification – valuable for ML researchers seeking academic experience or transitioning between sectors
• Potential first-author publication(s) in an emerging field at the intersection of AI and consciousness science

THE DEPARTMENT

The Department of Psychology at Lancaster is a vibrant research and teaching community with around 40 academic staff and strong technical support. You'll work within the Language, Inner Speech, and Neuroscience (LISN) Lab, which advances inner speech research through theoretical innovation, methodological development, and computational approaches.

Benefits include:

• 25 days annual leave (pro-rata) plus University closure days and bank holidays
• Employee pension scheme and flexible benefits
• A Department committed to Equality, Diversity, and Inclusion, currently holding an Athena Swan Silver Award
• Support for flexible working arrangements
• For more information: Jobs - Lancaster University

With your online application, please submit:

• CV (standard academic CV)

• Cover letter (max 2 pages) addressing:

• Your most relevant ML project involving time-series, sequential, or noisy/weakly-supervised data – describe your specific role, methods used, and outcomes

• One potential approach you would explore for detecting sparse, unlabelled events in noisy multivariate time-series (no neuroscience background required – we want to see how you think about challenging signal detection problems)
• Why this project appeals to you at this career stage

3.Code sample (optional): Link to GitHub repo or notebook demonstrating your ML implementation style

Informal enquiries are strongly encouraged. Please contact Dr Bo Yao at

Lancaster University is committed to equality, diversity, and inclusion. We welcome applications from all backgrounds and underrepresented groups.