Mood,neuroticism, and the encoding of affective words

A lexical decision task, with words of positive, neutral and negative content, was used to test for affective bias in word encoding. Bias in priming of response was also examined; stimulus onset asynchrony (SOA) between prime and target words was manipulated to distinguish automatic and controlled priming. We tested for bias associated with emotional content of words, and with trait and state subject characteristics of neuroticism and pleasantness of mood. Results showed a variety of biases, but mood-congruence of priming of emotional words was found only in subjects selected for relatively extreme moods. In unprimed lexical decision, the strongest effect was a general processing advantage for negative words, although some variation between subject groups was found. Negative word-pairs also showed greater automatic priming magnitude than positive word pairs. Neutral words were more strongly primed in more neurotic subjects. The magnitude of controlled priming was greater for both types of emotional word than for neutral words. Implications of the results for spreading activation and attentional models of affective bias are discussed. It is concluded that there is a general, automatic processing advantage for negative words, but the tuning of the information-processing system to affective content varies with neuroticism and mood. Results are broadly consistent with hybrid models of affective bias, which propose that mood-congruence is influenced by several distinct mechanisms.