Research Areas

My research spans multiple domains in neuroscience and biometrics, with a focus on understanding how physiological signals reflect cognitive processes and human behavior.

Current Research Directions

EEG and ECG markers of Tinnitus (Tinnitus and Hyperacusis Research Lab)

Tinnitus affects approximately 15% of the population, creating phantom auditory sensations that can severely impact quality of life. Our research investigates the neural and physiological mechanisms underlying tinnitus perception, with particular focus on sleep disturbances and autonomic nervous system dysfunction in chronic tinnitus patients.

Current Research Focus:

• Sleep-Tinnitus Interactions: Investigating how chronic tinnitus disrupts sleep architecture using polysomnography and examining EEG sleep spindles, slow waves, and REM characteristics in tinnitus patients versus controls

• Residual Inhibition Mechanisms: Using high-density EEG to understand the neural oscillatory changes that occur during residual inhibition - the temporary suppression of tinnitus following sound exposure

• Heart Rate Variability Analysis: Developing and comparing advanced time-frequency methods (STFT, Wavelet, EMD) to assess autonomic nervous system function in tinnitus patients during stress and rest conditions

• EEG Biomarkers: Identifying specific neural oscillatory patterns that correlate with tinnitus severity and treatment response

Methodological Approach:

Our work combines clinical EEG recordings with advanced signal processing techniques including independent component analysis (ICA) for artifact removal, time-frequency decomposition for oscillatory analysis, and machine learning approaches for pattern classification. We coordinate data collection protocols, manage large-scale EEG datasets, and perform statistical analyses using mixed-effects models to account for individual differences in tinnitus presentation.

Continuous Discrete Wavelet Transform algorithm visualization showing time-frequency decomposition of neural signals

Auditory Processing (Cognition, Audition and Time Lab)

The ability to detect brief gaps in continuous sounds is fundamental to speech perception. Our research examines the neural mechanisms underlying gap detection using the mismatch negativity (MMN) - an event-related potential that reflects the brain’s automatic detection of acoustic changes.

Research Questions:

• How does the brain’s automatic change detection system respond to increasingly brief temporal gaps in sound?
• What developmental changes occur in gap detection abilities across the lifespan, from childhood through aging?
• How do spatial auditory factors (earphones vs. speakers) influence temporal processing mechanisms?
• What are the source locations and connectivity patterns underlying mismatch negativity responses?

Current Investigations:

• Gap Detection MMN Studies: Systematically examining neural responses to gaps ranging from 2-40 milliseconds using EEG, extending previous work to understand how changes in auditory stimulation parameters affect neural and behavioural responses.

• Developmental Auditory Processing: Comparing gap detection abilities across five age groups (children, adolescents, young adults, middle-aged, older adults) to map lifespan changes in temporal auditory processing

• Spatial Audio Processing: Investigating whether the delivery method of auditory stimuli (earphones vs. free-field speakers) influences the neural mechanisms of temporal gap detection

• EEG Source Analysis: Using advanced source localization techniques to identify the cortical and subcortical generators of mismatch negativity responses

Technical Contributions:

I coordinate complex multi-session EEG data collection protocols, develop standardized preprocessing pipelines for developmental datasets, and implement advanced time-frequency analysis methods to extract oscillatory neural responses. Our work employs rigorous statistical approaches including permutation testing and cluster-based analyses to handle the multiple comparisons inherent in high-dimensional EEG data.

Standard ERP components showing neural responses to auditory stimuli

Biosignal Analysis

EEG Signal Processing

Modern neuroscience research requires sophisticated signal processing approaches to extract meaningful information from the complex, noisy signals recorded from the human brain. My work focuses on developing and implementing robust preprocessing and analysis pipelines that can handle the challenges of clinical and research EEG data.

Time-Frequency Analysis:

• Multi-Resolution Decomposition: Applying empirical mode decomposition (EMD), discrete wavelet transforms, and short-time Fourier transforms to analyze neural oscillations across multiple temporal and spectral scales
• Instantaneous Frequency Analysis: Using Hilbert transforms and complex Morlet wavelets to track moment-to-moment changes in oscillatory power and frequency
• Cross-Frequency Coupling: Investigating phase-amplitude coupling and other cross-frequency interactions that reflect neural communication mechanisms

Connectivity and Network Analysis:

• Phase-Based Connectivity: Computing phase-locking values and phase lag indices to assess functional connectivity between brain regions
• Source-Level Analysis: Implementing beamforming and minimum norm estimation techniques to project sensor-level connectivity measures to anatomically defined brain regions

ECG and Physiological Monitoring

Cardiovascular signals provide a window into autonomic nervous system function and emotional processing. My work develops sophisticated approaches to extract physiological markers that complement neuroimaging data and provide insights into mind-body interactions.

Heart Rate Variability Analysis:

• Time-Domain Methods: Computing standard measures including RMSSD, pNN50, and triangular indices to assess parasympathetic nervous system activity
• Frequency-Domain Analysis: Implementing advanced spectral analysis techniques including autoregressive modeling, multitaper methods, and wavelet-based approaches to decompose HRV into sympathetic and parasympathetic components

Multi-Modal Integration:

• Dual-Measure Cardiac Rhythm Analysis: Developing methods to reconstruct electrocardiography (ECG) data from simultaneously recorded Photoplethysmography (PPG) data.

Methodological Expertise

Experimental Design and Research Coordination

• Complex Paradigm Design: Developing multi-session experimental protocols that balance statistical power with participant burden, incorporating counterbalancing schemes and randomization procedures for robust causal inference
• Clinical Research Coordination: Managing IRB protocols, informed consent procedures, and data collection workflows for sensitive populations including tinnitus patients and pediatric participants
• Cross-Cultural Research: Adapting experimental paradigms for bilingual populations and ensuring cultural sensitivity in research design

Knowledge Translation and Collaboration

• Interdisciplinary Integration: Bridging neuroscience, audiology, and biomedical engineering to develop comprehensive research approaches that address complex clinical questions
• Mentorship and Training: Developing training protocols for research assistants and graduate students, ensuring consistent data quality across team members with varying experience levels

Research Innovation and Problem-Solving

• Multi-Site Study Coordination: Establishing standardized protocols and quality control procedures across multiple research sites to ensure data consistency and comparability

Previous Research (Graduate Work)

Visual Attention and Brain Oscillations

Investigating how alpha oscillations (8-12 Hz) in the brain coordinate spatial and temporal attention. My work has shown that these neural rhythms play a crucial role in filtering relevant information while suppressing distractions.

Key Methods:

• Electroencephalography (EEG)
• Time-frequency analysis
• Event-related potentials (ERPs)

Moral Perception and Processing

Exploring how the brain rapidly processes moral information and distinguishes moral from non-moral content. This research reveals the temporal dynamics of moral cognition using electrophysiology.

Key Findings:

• Moral words are distinguished from non-moral words within 300ms
• Early frontal processing followed by left-posterior differentiation
• Implications for understanding moral decision-making