Search Abstracts | Symposia | Slide Sessions | Poster Sessions
The influence of task type on ERP effects in the PWI Paradigm
Poster Session E, Sunday, September 14, 11:00 am - 12:30 pm, Field House
Jacklyn Jardel1, Sofia E. Ortega1, Lauren E. Magliocco1, Katherine J. Midgley1, Phillip J. Holcomb1; 1San Diego State University
Many studies have utilized the Picture-Word Interference (PWI) paradigm to study lexical selection and production. These studies often report typical N400 priming effects (reduced N400s) in identity conditions (e.g., the word “DOG” with a picture of a dog) and faster reaction times (RTs) for related versus unrelated pairs. However, semantic conditions (e.g., the word “CAT” with a picture of a dog) often report behavioral interference (slower picture naming RTs) for related versus unrelated pairs but still typically show priming on the N400 (i.e., larger N400s for unrelated than related word-picture pairs). A recent study by Holcomb et al. (2025) examined this incongruity between behavioral and event-related potential (ERP) data in the semantic condition using a RT based analysis. Each item in the semantic condition was grouped based on their average RT with items that showed faster RTs with related versus unrelated primes in the facilitation bin and RTs showing the reverse were placed in the interference bin. Using this novel scheme, we found a polarity reversal in the N400 for the interference RT trials, potentially indicating traditional ERP analysis may obscure effects reflective of interference behavior. However, it is still unclear why interference occurs. Some propose interference occurs during lexical selection, others during production. To better understand when and why this interference behavior occurs, this study explores how task intention modulates these effects. In both experiments, participants engaged in a typical PWI paradigm with 120 word-picture pairs shown twice, once with a related prime word and once with an unrelated prime. Sixty of the pairs were in the identity condition and 60 in the semantic condition. In Experiment one, participants pressed one button every time they saw a vegetable picture (about 16% of the stimuli) and another for any non-vegetable picture. Experiment two used the same stimuli, but the task was to press one button every time they saw a related prime picture pair and another if the pair was unrelated. Experiment one revealed significant priming effects only in the identity condition, and not the semantic. When examining bins formed based on the interference RT items from Holcomb et al. (2025), ERPs did not reveal any priming effects, while the comparable facilitation RT bins revealed a strong positivity in the late N400 window with significantly more negativity for related versus unrelated pairs in the 700-1200ms window. In Experiment two, priming effects appeared for both the identity and semantic conditions, as well as in the facilitation and interference RT groups. Together, these results indicate that task affects early stimuli processing. Further, without the requirement of production, interference effects in the interference RT bins disappear, indicating that the intent to produce a word plays a key role in the resulting interference.
Topic Areas: Meaning: Lexical Semantics, Language Production