Conjunction error rates on a continuous recognition memory test: little evidence for recollection

Area of Science:

• Cognitive Psychology
• Memory Research
• Human Cognition

Background:

• Conjunction memory errors occur when individuals mistakenly recall combined features of previously studied items.
• Dual-process and familiarity-based theories attempt to explain recognition memory.
• Continuous recognition tasks assess memory over time by presenting items repeatedly.

Purpose of the Study:

• To investigate the effect of lag manipulation on conjunction memory errors.
• To test predictions derived from simple familiarity and dual-process models of recognition.
• To examine the role of encoding strategies, such as searching, in memory performance.

Main Methods:

• Two experiments utilized a continuous recognition task.
• Participants studied word pairs and were later tested on their memory for these words and conjunction lures.
• The lag (number of intervening words) between parent words and conjunction lures was systematically varied.

Main Results:

• Experiment 1 showed conjunction errors were highest at the shortest lag (1 word) and decreased with increasing lag.
• Experiment 2 revealed an increase in errors from 0- to 1-word lag, followed by a slight decrease from 1- to 5-word lag.
• Results offered mixed support for existing memory models and suggested active searching is not an optimal encoding strategy.

Conclusions:

• The findings challenge simple familiarity and dual-process accounts of recognition memory.
• Lag effects on conjunction errors are complex and do not consistently support current theoretical frameworks.
• Paradoxically, actively searching memory during encoding may impair subsequent recall accuracy.