Selective attention in cross-situational statistical learning: evidence from eye tracking

Area of Science:

• Cognitive Science
• Developmental Psychology
• Linguistics

Background:

• Adults and children effectively learn statistical evidence from ambiguous contexts with multiple words and objects.
• Cross-situational learning involves mapping novel words to referents across various situations.

Purpose of the Study:

• To demonstrate the critical role of selective attention in cross-situational learning mechanisms.
• To investigate if an associative mechanism incorporating selective attention explains eye-tracking data during learning.

Main Methods:

• Collected eye movement data from participants during a cross-situational statistical learning task.
• Analyzed gaze patterns to differentiate between strong, average, and weak learners.
• Examined how attention is controlled and statistical evidence is accumulated in real-time.

Main Results:

• Gaze patterns revealed how learners direct attention after hearing a word.
• Selective attention to objects was observed when multiple items competed for focus.
• Evidence accumulation and integration across words, objects, and mappings were detailed.

Conclusions:

• Selective attention is a key mechanism enabling successful cross-situational statistical learning.
• Eye movement data provide insights into the real-time cognitive processes underlying word learning.
• Findings support associative learning models that incorporate attentional control.