P 35 The success of learning from errors – the role of high-gamma EEG and peripheral physiological responses
| Autor: | Tonio Ball, Sofie Berberich, Wolfram Burgard, Martin Völker, Andreas Schulze-Bonhage, Lukas D. J. Fiederer, E. Andreev, S. Contzen |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
medicine.medical_specialty
Communication Visual perception medicine.diagnostic_test business.industry Electroencephalography Audiology Sensory Systems Physiological responses Learning from errors Peripheral Neurology Physiology (medical) medicine Heart rate variability Eye tracking Neurology (clinical) Psychology business Eriksen flanker task |
| Zdroj: | Clinical Neurophysiology. 128:e345-e346 |
| ISSN: | 1388-2457 |
| Popis: | How humans learn from errors is a matter of great interest in neuroscience, yet many aspects of the underlying brain mechanisms and the role of peripheral physiological processes in error- based learning still remain unclear. Here, we investigated error-related learning in 30 healthy subjects during a classical Eriksen flanker task. Subjects had to use either the left or right hand to respond to the visual stimuli and were instructed to avoid errors in the form of ”slow” trials exceeding a pre-defined time limit. EEG data were acquired with a 128-channel system optimized for detection of high-frequency EEG components. To ensure an optimal signal-to-noise ratio, recordings took place in an electromagnetically shielded cabin and with full optical decoupling of all devices. We acquired peripheral physiological data using a high resolution binocular eye tracker and a two-channel ECG and computed the individual heart rate variability as a parameter previously implicated in learning processes. Consistent with the instruction stressing the importance of speed over accuracy, the number of slow trials decreased during the experiment ( Fig. 1 A). Further, we could show a significant post-error heart-rate deceleration, while pupil size was significantly bigger following error trials compared to correct ones. Both parameters showed an opposite yet similar temporal profile (Fig. 1B). In the EEG data, we found high-gamma band (HGB) spectral power increases in the frequency range up to 120 Hz approx. 100 ms post-error and over the frontal midline region. Across subjects, the amplitude of the individual frontal-midline HGB response was significantly correlated with the individual learning performance, as well as the heart rate variability and error-related pupil size response. Our findings reveal a novel response pattern consisting of central (frontal midline HGB) and peripheral (pupil size, heart-rate variability) components predictive of the individual success in learning from errors and thus open up a new window on the mechanisms contributing to learning via the error-monitoring network. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect