Test-retest reliability and stability of N400 effects in a word-pair semantic priming paradigm

Objective: Elicited by any meaningful stimulus, the N400 event-related potential (ERP) component is reduced when the stimulus is related to a preceding one. This N400 semantic priming effect has been used to probe abnormal semantic relationship processing in clinical disorders, and suggested as a possible biomarker for treatment studies. Validating N400 semantic priming effects as a clinical biomarker requires characterizing their test-retest reliability. Methods: We assessed test-retest reliability of N400 semantic priming in 16 healthy adults who viewed the same related and unrelated prime-target word pairs in two sessions one week apart. Results: As expected, N400 amplitudes were smaller for related versus unrelated targets across sessions. N400 priming effects (amplitude differences between unrelated and related targets) were highly correlated across sessions (r=0.85, P< 0.0001), but smaller in the second session due to larger N400s to related targets. Conclusions: N400 priming effects have high reliability over a one-week interval. They may decrease with repeat testing, possibly because of motivational changes. Significance: Use of N400 priming effects in treatment studies should account for possible magnitude decreases with repeat testing. Further research is needed to delineate N400 priming effects' test-retest reliability and stability in different age and clinical groups, and with different stimulus types.