Probability effects on event-related potential correlates of recognition memory

Area of Science:

• Cognitive Neuroscience
• Psychology

Background:

• Event-related potential (ERP) studies commonly observe a 'left parietal old/new effect' where recognized items elicit greater parietal positivity than new items.
• This positivity is similar to the P300 or P3b, which is also elicited by low-probability, high-target-value items in non-memory tasks.
• The overlap raises questions about the influence of non-mnemonic factors (probability, target value) on the old/new effect.

Purpose of the Study:

• To investigate the extent to which the left parietal old/new effect is influenced by or interacts with P300 activity.
• To differentiate between mnemonic and non-mnemonic contributions to the observed ERP effects in recognition memory.

Main Methods:

• Compared event-related potentials (ERPs) for correctly identified old and new items.
• Manipulated the ratio of old to new items presented (25:75, 50:50, 75:25).
• Analyzed ERPs at parietal electrode sites, focusing on the latency range of the old/new effect and later components.

Main Results:

• The left parietal old/new effect, in its conventional latency range, remained unaffected by the old/new item ratio manipulation.
• Later parietal ERPs (post-800 ms) to recognized items were influenced by item probability.
• The scalp distribution of this later probability effect differed from the earlier old/new effect.

Conclusions:

• The left parietal old/new effect appears to be a specific indicator of episodic retrieval.
• This effect is largely uncontaminated by non-mnemonic factors such as item probability and target value.
• The findings support the interpretation of the old/new effect as a pure measure of memory retrieval.