Text familiarity, word frequency, and sentential constraints in error detection.

Area of Science:

• Cognitive Psychology
• Human-Computer Interaction
• Linguistics

Background:

• Error detection during proofreading is influenced by word characteristics and sentence predictability.
• Top-down processing, driven by familiarity and constraints, may impact proofreading efficiency.

Purpose of the Study:

• To investigate how word frequency and predictability affect error detection in proofreading.
• To determine if these factors independently or jointly impair the proofreading process.
• To explore the role of pre-task activities (reading, copying, memorizing, paraphrasing) in text familiarity and predictability.

Main Methods:

• Participants engaged in pre-proofreading tasks (reading, copying, memorizing, paraphrasing) with sentences containing errors.
• A proofreading task followed to measure time and accuracy.
• Sentence predictability was manipulated through constraints and familiarity.

Main Results:

• Proofreading times were not affected by prior exposure to sentences.
• Proofreading was slower and less accurate for high-frequency words and highly constrained sentences.
• Prior memorization of highly constrained sentences led to increased error overlooking.

Conclusions:

• Word frequency and sentence predictability significantly impair error detection during proofreading.
• High levels of predictability, particularly from memorization of constrained text, can increase the probability of missing errors.
• Proofreading is susceptible to top-down processing biases influenced by linguistic factors.