Abstract
Purpose: Repetition effects in face and object processing have been demonstrated in both fMRI and EEG. How are they related to category-specific and item-specific processes? The present study isolates category-specific and item-specific repetition effects in the EEG and assesses their temporal dynamics.
Methods: Nine subjects participated. The EEG was recorded from 61 electrodes while subjects viewed pictures of animals (cats or dogs) or scrambled animals in alternating blocks. Each animal block contained 15 images of one category, and 1–3 images of the other category. In 36 of the 72 blocks, all 15 images of the majority category were identical (“repeated”). In the other 36 blocks, all images were different (“non-repeated”). Each image was displayed for 1 s, followed by a blank period of 400 ms. Subjects were asked to categorize images to “cat” or “dog”. Trials were averaged according to condition (repeated/non-repeated) and whether they had occurred in the first (early) or second part of the block (late). We separately assessed category repetition effects (“early non-repeated” minus “late non-repeated”) and item-specific repetition effects (“late non-repeated” minus “late repeated”).
Results: Item-specific repetition reduced the positivity at around 100 ms in occipital and occipito-temporal electrodes. Starting around 250 ms, a second item-specific effect was evident as a reduced signal in occipital, occipito-temporal and frontal electrodes. A category repetition effect (a smaller negativity for late vs. early non-repeated items from the same category) occurred occipito-temporally around 200 ms.
Conclusion: Item-specific and category-specific repetition effects have different time courses. This suggests that they reflect distinct neural mechanisms acting at different stages of stimulus processing.
Supported by DFG (HE 3504/2-1) and NSF (345920).