| |
Abstract:
We examined the range of visual priming by investigating
whether priming occurs for word triplets that described a
conceptual script (e.g., DIRECTOR-BRIBE-DISMISSAL) but were not
associatively related. We assumed that priming for these scripts
requires access to general world knowledge not represented at lower
levels (form level and lexical-semantic level). In Experiment 1
participants made lexical decisions. In Experiment 2 participants
indicated whether the three words presented a plausible scenario.
In half of the cases pseudowords (Experiment 1) or plausible
scenarios (Experiment 2) were presented. Event related potentials
(ERPs) were recorded from 13 electrodes and 27 electrodes,
respectively. Reaction times were faster for plausible than
implausible scenarios. This priming effect was larger in Experiment
2 than Experiment 1. In both tasks scripts elicited ERP priming
effects - that is, mean amplitudes in the "N400-window" were more
positive-going for plausible than implausible scenarios. Script
priming was more robust in Experiment 2. In Experiment 1 priming
occurred only over the left hemisphere and in a short time window
(400-500 ms). In Experiment 2 the script priming effect was more
extended in time and more widely distributed across the scalp. In
the left as compared to the right hemisphere the effect was both
larger and earlier. The demonstration of script priming suggests
the operation of a higher-level integration process with a high
flexibility with respect to the kinds of information
exploited.
|
