Event-Related Potential Evidence for Two Functionally Dissociable Sources of Semantic Effects in the Attentional Blink

Francesca Peressotti,Francesca Pesciarelli,Claudio Mulatti,Roberto Dell'Acqua

doi:10.1371/journal.pone.0049099

Francesca Peressotti, Francesca Pesciarelli + Show 2 more

Open Access

https://doi.org/10.1371/journal.pone.0049099

Copy DOI

Abstract

Three target words (T1, T2, and T3) were embedded in a rapid serial visual presentation (RSVP) stream of non-word distractors, and participants were required to report the targets at the end of each RSVP stream. T2 and T3 were semantically related words in half of the RSVP streams, and semantically unrelated words in the other half of the RSVP streams. Using an identical design, a recent study reported distinct reflections of the T2–T3 semantic relationship on the P2 and N400 components of event-related potentials (ERPs) time-locked to T3, suggesting an early, automatic, source of P2 semantic effects and a late, controlled, source of N400 semantic effects. Here, P2 and N400 semantic effects were examined by manipulating list-wide context. Relative to participants performing in a semantically unbiased context, participants over-exposed to filler RSVP streams always including semantically related T2/T3 words reported a dilution of T3-locked P2 semantic effects and a magnification of T3-locked N400 semantic effects. Opposite effects on P2 and N400 ERP components of list-wide semantic context are discussed in relation to recent proposals on the representational status of RSVP targets at processing stages prior to consolidation in visual short-term memory.

Highlights

Unconscious information processing has attracted the interest of researchers from the very early days of scientific psychology and even nowadays the nature of the mechanisms that mediate the influence of unconscious cognition is a highly debated and controversial issue
Data from 12 participants had to be discarded because less than ten trials per cell of the design were left after removal of artifacts and standard rapid serial visual presentation (RSVP) trials associated with incorrect report of T1 or T3
Behavioral analyses were carried out on data from standard and filler RSVP trials, these latter differing in the two groups of tested participants, with one group exposed to semantically related filler RSVP trials, and the other group exposed to semantically unrelated filler RSVP trials

Summary

Introduction

Unconscious information processing has attracted the interest of researchers from the very early days of scientific psychology and even nowadays the nature of the mechanisms that mediate the influence of unconscious cognition is a highly debated and controversial issue. In the vast majority of cases, cognitive studies of unconscious cognition have made use of techniques devised to prevent conscious access to one visual stimulus, traditionally referred to as prime, and to probe its influence on the processing of a different, clearly visible and consciously perceived stimulus, traditionally referred to as target Recent theorization in this field has highlighted a critical subdivision of these techniques according to the way in which conscious access to prime stimuli is limited [1]. With another class of techniques, conscious perception of the prime (or target, depending on variants) is resource-limited by locking out attention mechanisms necessary for conscious perception (e.g., consolidation: [2,3]; tokenization: [4,5]) Prototypical in this class is the rapid serial visual presentation (RSVP) paradigm, in which primes and targets are embedded in streams of spatially overlapping distractors, and displayed at varying stimulus onset asynchronies (SOAs). RSVP stimuli are displayed at frequencies usually varying between 8 to 10 Hz, implying that prime exposure duration is less critical a factor to limit conscious perception using RSVP than the masked priming technique, e.g., [7]

Methods

Results

Discussion

Conclusion