High lexical construal levels decrease negativity bias in online reading: evidence from recognition and eye-tracking experiments

Negativity bias in online reading widely occurs and leads to harmful results at the individual and social levels, but few studies have examined how to reduce this bias from the perspective of lexical processing. This study examined whether and how high lexical construal levels decrease negativity bias through a recognition memory experiment (Study 1) and an eye-tracking experiment (Study 2). In Study 1 ( N = 70), results showed a negativity bias in recognition memory, which was attenuated under high construal levels, primarily through a reduction in the discrimination sensitivity ( A ’) advantage for negative words. In Study 2 ( N = 53), eye-tracking revealed that at low construal levels, negative words were processed more quickly than positive words, indicating automated processing. Critically, at high construal levels, this pattern reversed, with positive words being processed more efficiently. These findings demonstrate that high lexical construal levels reduce negativity bias not by equalizing processing, but by attenuating the automatic advantage for negative words and shifting efficiency toward positive information.