Detecting Error Potentials in Response to Perceived Substitution Spelling Errors

No Thumbnail Available
Issue Date
Journal Title
Journal ISSN
Volume Title
In this thesis we try to determine whether incorrectly spelled words evoke an error potential that is detectable on a single trial level. We do an experiment where the subjects see 400 words in quick succession, and we analyse the EEG data to see if we can detect and classify these error potentials consistently. In this field of research, most of the focus is on P300, P2, N2 and N4 potentials, and almost no attention is on the error potentials. I think these potentials might make for interesting research, because they might give us insights or reinforcements of theories about language and error processing. This research might also be relevant for dyslexia patients. If we have this potential, and dyslectics don’t, we could research whether this is a cause or effect of dyslexia. The results of the experiment showed that, yes, we do seem to have these error potentials. But the classifier I used doesn’t seem to accurately classify erroneous words as such.
Faculteit der Sociale Wetenschappen