Cargando…

Development of ERN together with an internal model of audio-motor associations

The brain's reactions to error are manifested in several event related potentials (ERP) components, derived from electroencephalographic (EEG) signals. Although these components have been known for decades, their interpretation is still controversial. A current hypothesis (first indicator hypot...

Descripción completa

Detalles Bibliográficos
Autores principales: Lutz, Kai, Puorger, Roman, Cheetham, Marcus, Jancke, Lutz
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2013
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3759802/
https://www.ncbi.nlm.nih.gov/pubmed/24027506
http://dx.doi.org/10.3389/fnhum.2013.00471
Descripción
Sumario:The brain's reactions to error are manifested in several event related potentials (ERP) components, derived from electroencephalographic (EEG) signals. Although these components have been known for decades, their interpretation is still controversial. A current hypothesis (first indicator hypothesis) claims that the first indication of an action being erroneous leads to a negative deflection of the EEG signal over frontal midline areas. In some cases this requires sensory feedback in the form of knowledge of results (KR). If KR is given, then the first negative deflection can be found around 250 ms after feedback presentation (feedback-related negativity, FRN). When KR is not required, a negative deflection is found already around 100 ms after action onset (ERN). This deflection may be evoked when a mismatch between required and actually executed actions is detected. To detect such a mismatch, however, necessitates knowledge about which action is required. To test this assumption, the current study monitored EEG error components during acquisition of an internal model, i.e., acquisition of the knowledge of which actions are needed to reach certain goals. Actions consisted of finger presses on a piano keyboard and goals were tones of a certain pitch to be generated, thus the internal model represented audio-motor mapping. Results show that with increasing proficiency in mapping goals to appropriate actions, the amplitude of the ERN increased, whereas the amplitude of the FRN remained unchanged. Thus, when knowledge is present about which action is required, this supports generation of an ERN around 100 ms, likely by detecting a mismatch between required and performed actions. This is in accordance with the first indicator hypothesis. The present study furthermore lends support to the notion that FRN mainly relies on comparison of sensory targets with sensory feedback.