-
Uddannelse
Find uddannelse
Studievejledning
Besøg AU
Kvalitet
-
Forskning
Aktuel forskning
Talent
-
Samarbejde
Gymnasier
Myndigheder
Alumner
- Om AU
Dr Sabine Grimm from Cognitive and Biological Psychology (BioCog), Leipzig University is visiting MIB and will give a talk.
18.03.2019 |
| Dato | tor 28 mar |
| Tid | 13:30 — 14:30 |
| Sted | Meeting room 5th floor DNC Building 10G, Nørrebrogade 44, 8000 Aarhuc C |
TITLE:
Memory representations for random pitch patterns: the roles of attention and stimulus variability as measured in the EEG
ABSTRACT:
Extracting recurring patterns from our sound environment is key to the recognition and identification of meaningful auditory events and their segregation from complex auditory scenes. In real life situations, instances of the same sound event are likely to vary considerably regarding their exact acoustic features, e.g., when the same word is uttered by different speakers.
In a series of EEG experiments, we investigated how such acoustic variations in single instances of a recurring (unfamiliar) sound pattern influence the initial stages of forming memory representations for this pattern. In a first set of studies, short random pitch patterns were presented repeatedly before a new pattern occurred. Here we studied event-related potentials (ERP), such as the Mismatch Negativity (MMN; a marker of deviance detection), to determine how the time course of regularity extraction and pattern change detection is affected by introducing variability in the recurring pattern (e.g., by repeating only the pitch interval structure while varying absolute pitch; or by replacing part of the pattern information). In a second set of studies, we looked at repeated patterns that were embedded in a continuous stream of random pitches and studied the Sustained Response of the ERP in order to determine how regularity detection is modulated by attention and variability in the pattern.
Based on the results of those studies, I argue that the strength of established memory representations is diminished by the introduction of acoustic variability in the individual instances of a recurring pattern, while the time course and automaticity of establishing pattern representations is not.