A cross-linguistic speech error investigation of functional complexity

Area of Science:

• Psycholinguistics
• Linguistic Typology
• Speech Production

Background:

• Cross-linguistic studies are crucial for understanding universal and language-specific aspects of speech production.
• Previous research often focused on English, potentially limiting generalizability.
• The relationship between language structure and speech error patterns requires systematic investigation.

Purpose of the Study:

• To systematically examine speech errors across diverse languages (English, Hindi, Japanese, Spanish, Turkish).
• To develop a methodology for creating parallel corpora of speech error data.
• To test hypotheses regarding language complexity, processing, and structural influences on errors.

Main Methods:

• Developed a methodology for generating parallel corpora of speech error data.
• Collected and analyzed speech error data from native speakers of five languages.
• Compared error frequencies and patterns across languages to test specific hypotheses.

Main Results:

• No significant differences in overall error rates were found across the five languages, supporting the hypothesis that languages are equally complex.
• English-based generalizations about language production largely held true across languages, indicating similar processing mechanisms.
• Speech errors disproportionately occurred in morphologically complex areas (e.g., inflectional morphology in Turkish/Spanish, closed-class forms in Japanese), demonstrating language-specific patterns.

Conclusions:

• Speech production systems exhibit universal characteristics but are also shaped by language-specific structural complexities.
• Speech error analysis provides valuable insights into the interplay between linguistic structure and cognitive processes in real-time language use.
• The findings underscore the importance of cross-linguistic research in psycholinguistics and speech production studies.