Adapted wavelet transform improves time-frequency representations: a study of auditory elicited P300-like event-related potentials in rats.

Nelly Richard, Bettina Laursen, Morten Grupe, Asbjørn M  Drewes, Carina Graversen, Helge Bjarup Dissing Sørensen, Jesper F. Bastlund

    Research output: Contribution to journalJournal articleResearchpeer-review

    1 Downloads (Pure)

    Abstract

    Objective. Active auditory oddball paradigms are simple tone discrimination tasks used to study the P300 deflection of event-related potentials (ERPs). These ERPs may be quantified by time-frequency analysis. As auditory stimuli cause early high frequency and late low frequency ERP oscillations, the continuous wavelet transform (CWT) is often chosen for decomposition due to its multi-resolution properties. However, as the conventional CWT traditionally applies only one mother wavelet to represent the entire spectrum, the time-frequency resolution is not optimal across all scales. To account for this, we developed and validated a novel method specifically refined to analyse P300-like ERPs in rats. Approach. An adapted CWT (aCWT) was implemented to preserve high time-frequency resolution across all scales by commissioning of multiple wavelets operating at different scales. First, decomposition of simulated ERPs was illustrated using the classical CWT and the aCWT. Next, the two methods were applied to EEG recordings obtained from prefrontal cortex in rats performing a two-tone auditory discrimination task. Main results. While only early ERP frequency changes between responses to target and non-target tones were detected by the CWT, both early and late changes were successfully described with strong accuracy by the aCWT in rat ERPs. Increased frontal gamma power and phase synchrony was observed particularly within theta and gamma frequency bands during deviant tones. Significance. The study suggests superior performance of the aCWT over the CWT in terms of detailed quantification of time-frequency properties of ERPs. Our methodological investigation indicates that accurate and complete assessment of time-frequency components of short-time neural signals is feasible with the novel analysis approach which may be advantageous for characterisation of several types of evoked potentials in particularly rodents.
    Original languageEnglish
    Article number026012
    JournalJournal of Neural Engineering
    Volume14
    Issue number2
    Number of pages10
    ISSN1741-2560
    DOIs
    Publication statusPublished - 2017

    Keywords

    • Electroencephalography
    • P300 event-related potentials
    • Phase-locking factor
    • Spectral power
    • Wavelet
    • Rat

    Fingerprint

    Dive into the research topics of 'Adapted wavelet transform improves time-frequency representations: a study of auditory elicited P300-like event-related potentials in rats.'. Together they form a unique fingerprint.

    Cite this