Semantic and affective priming of concrete and abstract words

Sánchez-Casas, R. 1 , Ferré, P. 1 , García, T. 2 , Moldovan, C. . 1 , Fraga, I. 3 & Redondo, J. 3

1 CRAMC Universitat Rovira i Virgili (Tarragona, Spain)
2 Universidad de Comillas (Madrid, Spain)
3 Universidad de Santiago de Compostela (Santiago de Compostela, Spain)

It is well known that participants respond faster to a target word when it is preceded by a semantically related prime than when the prime is an unrelated word. This is the semantic priming effect. On the other hand, it has repeatedly been shown that when a target word is preceded by a word of a similar affective valence it is judged or pronounced faster than if it is preceded by a word of an opposite valence. This phenomenon is known as affective priming and, according to some authors, it suggests that people can encode automatically the valence of the prime.
Concerning affective priming, it has to be taken into account that affective congruent stimuli are likely to be also semantically related, therefore, affective priming might also be reflecting semantic priming. However, there are only a small number of studies in the affective priming literature that have controlled for the semantic relationship between primes and targets (Storbeck & Robinson, 2010).
The aim of the present study was to test whether affective priming can be obtained when primes and targets are not semantically related. We conducted a series of experiments in which we manipulated orthogonally the semantic and affective relationship between primes and targets. We also compared the magnitude of semantic and affective priming effects between pairs of concrete and abstract words. Comparisons were made across three types of tasks: a lexical decision, a semantic categorization or an evaluation task. The results showed that semantic priming seems to be a more robust phenomenon than affective priming, as it clearly emerges in all the conditions, whereas affective priming is only observed in more restricted circumstances. These findings are discussed in terms of which is the more obligatory analysis of words at encoding: a semantic analysis or an affective analysis.