Title :
Single-trial identification of failed memory retrieval
Author :
Eunho Noh ; Mollison, Matthew V. ; Curran, Tim ; de Sa, Virginia R.
Author_Institution :
Dept. of Electr. & Comput. Eng., UCSD, La Jolla, CA, USA
Abstract :
We show that it is possible to distinguish unsuccessful from successful retrieval of study items based on single-trial EEG recorded during the test phase of 3 separate recognition memory experiments. The overall classification accuracy across all 34 classification problems was 58.4 %. The classification accuracy monotonically increased to 68.03 % by only classifying trials with high classifier confidence levels. The likelihood of remembering a study item for trials with the 10% highest and lowest classifier outputs were 0.80 and 0.45 respectively. This suggests that the classifier outputs are reflecting the level of retrieval during the test phase. These findings combined with previous single-trial results predicting subsequent memory from EEG recorded during and prior to memory encoding will provide a basis for a passive brain-computer interface (BCI) system for improving memory.
Keywords :
brain-computer interfaces; electroencephalography; information retrieval; signal classification; BCI system; classification problem; classifier confidence levels; classifier outputs; failed memory retrieval; overall classification accuracy; passive brain-computer interface system; recognition memory experiments; single-trial EEG recording; single-trial identification; study item retrieval; Accuracy; Brain; Electroencephalography; Encoding; Image coding; Image color analysis; Memory management;
Conference_Titel :
Signals, Systems and Computers, 2014 48th Asilomar Conference on
Print_ISBN :
978-1-4799-8295-0
DOI :
10.1109/ACSSC.2014.7094388