Severe processing capacity limits for sub-lexical features of letter strings

Area of Science:

• Cognitive Psychology
• Visual Perception
• Reading Science

Background:

• The human visual system processes multiple words during reading.
• Previous research indicates parallel processing of low-level visual features but serial processing of lexical attributes.
• An internal bottleneck is hypothesized between early visual and lexical analysis stages.

Purpose of the Study:

• To investigate the location of the bottleneck in word recognition.
• To determine if the bottleneck occurs during letter recognition or phonological decoding.
• To understand the limits of simultaneous information processing during reading.

Main Methods:

• A dual-task behavioral paradigm was employed.
• Participants performed letter identification and phonological decoding tasks on two simultaneously presented letter strings.
• Attention was manipulated between focused (one string) and divided (both strings) conditions.

Main Results:

• High costs of divided attention supported a serial processing model, indicating only one letter string could be processed per trial.
• A stimulus processing trade-off was observed, characteristic of serial processing.
• Performance on one string negatively impacted performance on the other.

Conclusions:

• The bottleneck constraining word recognition is located at a sub-lexical level.
• This limitation may stem from constraints in the efficiency of letter recognition.
• Reading multiple words simultaneously is constrained by early-stage visual processing limits.