Does letter position coding depend on consonant/vowel status? Evidence with the masked priming technique

Area of Science:

• Cognitive Psychology
• Psycholinguistics
• Computational Linguistics

Background:

• Existing computational models of visual word recognition capture transposed-letter priming.
• These models do not differentiate processing between vowels and consonants.
• Prior research indicates transposed-letter priming effects are absent for vowel transpositions in lexical decision tasks.

Purpose of the Study:

• To investigate whether the consonant/vowel dissociation in transposed-letter priming is task-specific.
• To test the implications of this dissociation for computational models of visual word recognition.

Main Methods:

• Four masked priming experiments were conducted.
• Experiment 1 utilized a lexical decision task.
• Experiments 2-4 employed a same-different task, targeting early perceptual processes.

Main Results:

• A transposed-letter priming effect was observed only for consonant transpositions in the lexical decision task.
• A robust transposed-letter priming effect, irrespective of consonant or vowel status, was found in the same-different task.
• No interaction between consonant/vowel status and transposed-letter priming was detected in the same-different task.

Conclusions:

• The consonant/vowel dissociation in transposed-letter priming is task-specific.
• Findings challenge current input coding schemes that do not account for vowel/consonant processing differences.
• Results provide crucial insights into the front-end processing stages of visual word recognition models.