BALDEY: A database of auditory lexical decisions

Area of Science:

• Psycholinguistics
• Computational Linguistics
• Phonetics

Background:

• The auditory lexical decision task is crucial for understanding spoken word recognition.
• Existing datasets often lack comprehensive phonological, morphological, and acoustic information for spoken words.
• The BALDEY database addresses this gap by providing a rich resource for Dutch spoken word research.

Purpose of the Study:

• To introduce and describe the BALDEY (Biggest Auditory Lexical Decision Experiment Yet) database.
• To demonstrate the utility of the BALDEY database through initial analyses of word neighborhood measures and frequency effects.
• To provide a publicly available resource for further research into auditory lexical decision and spoken word processing.

Main Methods:

• An auditory lexical decision experiment was conducted with 5541 Dutch spoken content words and pseudowords.
• Stimuli varied in phonological features, syllable count, stress patterns, and morphological complexity.
• Response times and accuracy rates were collected, alongside detailed phonological, acoustic, and morphological information for each item.

Main Results:

• Initial analyses explored the predictive power of word neighborhood measures on lexical decision outcomes.
• The study investigated the influence of various frequency measures (written, spoken, subtitles, ratings) on response times and accuracy.
• The BALDEY data file contains response times, accuracy rates, and rich linguistic information for each stimulus.

Conclusions:

• The BALDEY database is a valuable and comprehensive resource for studying auditory lexical decision and spoken word recognition in Dutch.
• The initial analyses highlight the importance of phonological and morphological factors, as well as frequency, in spoken word processing.
• The publicly available BALDEY database facilitates numerous future research avenues in psycholinguistics and related fields.