Retrieval practice via corrective feedback: is learning better for targets in an expected or surprising sense?

Area of Science:

• Cognitive Psychology
• Memory Research
• Learning Science

Background:

• Homographic words have multiple meanings, leading to expected (dominant) or surprising (subordinate) associations.
• Previous research indicates recall is better for expected than surprising word associations.

Purpose of the Study:

• To investigate how the act of guessing influences memory for word associations.
• To determine if guess-target congruence impacts learning, independent of word sense (dominant vs. subordinate).

Main Methods:

• Participants learned cue words and associated targets, presented in either dominant or subordinate senses.
• Memory recall was compared between conditions where participants guessed the target versus simply read the cue-target pair.
• A third experiment manipulated word sense activation before guessing or reading to isolate the effect of guess-target congruence.

Main Results:

• Guessing the target significantly improved retention compared to passive reading.
• The benefit of guessing was greater for expected (dominant) targets than surprising (subordinate) ones.
• Experiment 3 showed that guess-target congruence, not just target sense, critically influenced recall accuracy.

Conclusions:

• Active guessing enhances memory learning for word associations.
• The congruence between a guess and the actual target plays a crucial role in memory consolidation.
• Findings clarify that learning benefits are driven by the interaction between guessing and target properties, not solely by inherent word sense characteristics.