Attending is not enough: Responding to targets is needed for across-trial statistical learning

Area of Science:

• Cognitive Psychology
• Perception and Attention
• Human Learning

Background:

• Observers can learn across-trial regularities, demonstrated by faster responses to predicted target locations.
• Previous research indicated learning occurred when responses were made to both targets in a predictive pair.

Purpose of the Study:

• To investigate whether active response execution to both targets in a pair is necessary for across-trial regularity learning.
• To determine if merely attending to target locations is sufficient for statistical learning.

Main Methods:

• Participants performed a visual search task for a shape singleton target, responding to an internal feature.
• Two across-trial regularities concerning target locations were implemented.
• Experiments manipulated response requirements, with some pairs requiring responses to both targets and others to only one.

Main Results:

• Across-trial learning occurred only when responses were required for both targets in a predictive pair (Experiment 1).
• Withholding a response to one target in a pair prevented learning.
• This learning deficit was not due to carry-over inhibition, as learning persisted when response requirements were later altered (Experiment 2).

Conclusions:

• The execution of a simple motor response to both targets in a pair is essential for across-trial statistical learning.
• Attentional capture by target locations alone does not facilitate this form of learning.
• Active response engagement is a critical factor in learning sequential environmental regularities.