Automatic Affective Appraisal of Words

Abstract

Slower colour-naming reaction times (RTs) for negatively valenced Stroop words (e.g. SELFISH printed in green ink) than for positively valenced words (e.g. TOLERANT printed in blue ink) have been taken as evidence of an automatic vigilance mechanism that directs attention towards undesirable events and objects. These findings were replicated when words were shown at fixation but not when they were shown away from the locus of fixation. The failure of negative word valence to capture spatial attention is compared with the capture that has been found with angry and threatening facial expressions. From an ecological view of social perception both findings are unsurprising. Unlike words, facial expressions have evolved to provide the species with adaptively important information.