The Impact of Visual Segmentation on Lexical Word Recognition

Area of Science:

• Cognitive Psychology
• Linguistics
• Human-Computer Interaction

Background:

• English word recognition involves segmenting text into orthographic units corresponding to sounds.
• Irregular English orthography complicates this segmentation process compared to regular languages.
• Visual interventions aim to simplify segmentation and reduce reading's computational cost.

Purpose of the Study:

• To investigate if visual segmentation techniques can aid reading and word recognition.
• To quantify the impact of different visual segmentation methods on reading speed.
• To explore the interaction between typographic encoding and visual variables in text.

Main Methods:

• Conducted pre-registered lexical decision experiments with 192 participants.
• Tested five visual segmentation types: outlines, spacing, connections, underlines, and color.
• Measured word identification times for each visual intervention.

Main Results:

• All tested visual segmentation techniques slowed down word identification.
• The slowing effect ranged from 32.7ms (color) to 70.7ms (connections).
• Moderate training with these interventions did not accelerate the lexical decision task for typical adult readers.

Conclusions:

• Visual segmentation interventions, despite intentions, do not accelerate word recognition in typical adult readers.
• These findings provide empirical data on the reading cost of common text visualizations.
• The interaction of visual variables with typography suggests novel theoretical directions for reading research.