Objective: We investigated early brain activity (under 200 ms after the stimulus onset) related to the encoding and the retrieval of verbal information.
Methods: First, we compared ERPs produced by words which were encoded to ERPs produced by words from following test phases (correctly identified repetitions and correctly classified new words) in two different experiments. Experiment 1 consisted of an intentional learning paradigm and experiment 2 consisted of an incidental learning paradigm. In addition, we conducted a control experiment (experiment 3), which was a continuous recognition task with two different repetition intervals. Secondly, we conducted a magnetoencephalographic (MEG) study to further investigate early brain activity (experiment 4). The same intentional learning paradigm as in experiment 1 was used.
Results: We found that ERPs elicited by correctly classified test words (repeated words and new words) of both experiment 1 and experiment 2 were significantly more negative going than the ERPs elicited by the study words. This effect was apparent between 100 ms and 200 ms after the stimulus onset and was distributed over occipital and parietal scalp locations. In the control task (experiment 3), these early potential differences were missing (for both repetition intervals). Early event-related fields (ERFs) were also found to depend on the situation of the study phase and the test phase. This activity difference peaked at 120 ms after the stimulus onset. The distributions of the difference magnetic fields were occipito-parietal and thus consistent with the findings of experiment 1 (EEG-experiment).
Conclusion: Whether the effect we defined in the present study is due to an increase of activity during the test situation or due to a decrease of activity during the study situation remains unclear. However, it might reflect attentional processes within a word recognition task depending on whether a word is encoded or an effort of word retrieval has to be made.