Can response congruency effects be obtained in masked priming lexical decision?

Area of Science:

• Cognitive Psychology
• Psycholinguistics
• Computational Linguistics

Background:

• Masked priming lexical decision experiments investigate early stages of lexical access.
• The influence of unrelated wordlike/unwordlike primes on target processing remains under-explored.
• Theoretical models like the Bayesian Reader and interactive-activation models offer differing predictions.

Purpose of the Study:

• To examine whether masked unrelated primes, varying in lexical status and wordlikeness, influence lexical decision tasks.
• To compare experimental findings with predictions from the Bayesian Reader model and interactive-activation models.
• To investigate the role of foil legality in masked priming effects.

Main Methods:

• Conducted three masked priming lexical decision experiments.
• Utilized four types of unrelated primes: high-frequency words, low-frequency words, legal pseudowords, and consonant strings.
• Varied the orthographic legality of nonword foils across experiments.

Main Results:

• When foils were orthographically legal, prime characteristics (lexical status, wordlikeness) did not affect lexical decision responses.
• When foils were illegal nonwords, consonant string primes slowed word target responses and sped up nonword target responses.
• These results align with the Bayesian Reader model under specific assumptions about participant decision-making.

Conclusions:

• Unrelated primes do not influence lexical decisions when foils are orthographically legal, supporting the Bayesian Reader model and selective inhibition.
• The nature of the foil (legal vs. illegal) critically impacts masked priming effects.
• Participants may adapt their decision strategies based on the properties of the stimuli presented, particularly foil legality.