Meaningless, but not emotionless: Evidence from a megastudy of pseudoword lexical decisions

We examined the influence of morphological structure and emotional content on the recognition of pseudowords in a megastudy using a lexical decision task. To this end, 1,025 native Spanish speakers responded to 1,200 experimental words and 4,800 experimental pseudowords with different morphological structures derived from emotional and neutral base words, as well as to 1,200 filler words and 1,200 filler pseudowords. Reaction times (RTs) and accuracy were analyzed using linear and generalized linear mixed-effects models, revealing an effect of morphological composition. Pseudowords containing both a real root and a real suffix elicited the slowest response times and the highest error rates, followed by those containing one real morphological component (e.g., root + non-suffix or non-root + suffix). Pseudowords lacking identifiable morphological components (e.g., non-root + non-suffix) were rejected the fastest and with the greatest accuracy. With regard to affective features, higher valence and arousal elicited higher error rates and slowed rejection times. This suggests that emotionally charged pseudowords are more likely to be perceived as real words. Furthermore, these effects interacted with morphological structure, suggesting that affective and sublexical cues contribute jointly to pseudoword processing. Overall, we show that emotional information can be accessed even in the absence of lexical representations, possibly via the activation of representations from existing words and/or by exploiting systematic links between certain sublexical features and emotional meaning. Additionally, the current study provides a database ( https://osf.io/mh3xe/ ) of reaction times and accuracy data for 2,400 Spanish words and 6,000 Spanish pseudowords that vary in emotional and morphological composition.