Effects of negative content on the processing of gender information: An event-related potential study

Abstract

Previous research on emotion in language has mainly concerned the impact of emotional information on several aspects of lexico-semantic analyses of single words. However, affective influences on morphosyntactic processing are less understood. In the present study, we focused on the impact of negative valence in the processing of gender agreement relations. Event-related potentials (ERPs) were recorded while participants read three-word phrases and performed a syntactic judgment task. Negative and neutral adjectives could agree or disagree in gender with the preceding noun. At an electrophysiological level, the amplitude of a left anterior negativity (LAN) to gender agreement mismatches decreased in negative words, relative to neutral words. The behavioral data suggested that LAN amplitudes might be indexing the processing costs associated with the detection of gender agreement errors, since the detection of gender mismatches resulted in faster and more accurate responses than did the detection of correct gender agreement relations. According to this view, it seems that negative content facilitated the processes implicated in the early detection of gender agreement mismatches. However, gender agreement violations in negative words triggered processes involved in the reanalysis and repair of the syntactic structure, as reflected in larger P600 amplitudes to incorrect than to correct phrases, irrespective of their emotional valence.

Subliminal Emotional Words Impact Syntactic Processing: Evidence from Performance and Event-Related Brain Potentials

“…They observed that the amplitude of the AN increased for negative adjectives containing morphosyntactic violations, while it decreased for positive adjectives, in comparison with neutral ones. The results of the study by Hinojosa et al (2014) follow a similar line. On the other hand, P600 modulations have been reported when presenting happy and sad film clips preceding subject-verb agreement violations (Vissers et al, 2010; Verhees et al, 2015).…”

“…In view of Jiménez-Ortega et al (2014) and supraliminal studies with similar procedures (Martín-Loeches et al, 2012; Hinojosa et al, 2014), we expect that the syntactic processing of the sentence will be impacted by subliminal emotional words. The expected results are an increase in anterior negativity in response to morphosyntactic anomalies when the subliminal adjective is emotionally negative, as well as the absence of this component when the adjective is positive, as in Martín-Loeches et al (2012) using supraliminal presentations.…”

“…Although the modulation of language processing by emotions has often been investigated by using isolated words as targets, or focusing on semantic sentence processing (Kiefer et al, 2007), recent studies have also explored emotional effects on the syntactic processing of a sentence (Vissers et al, 2010; Jiménez-Ortega et al, 2012; Martín-Loeches et al, 2012; Hinojosa et al, 2014; Verhees et al, 2015). The effects of emotional words on syntactic processing as reflected in ERP data seem to have been proved, in spite of the traditional view that syntax is an encapsulated process (Fodor, 1983; Hauser et al, 2002).…”

Time course of gender agreement violations containing emotional words

“…More recently, Hinojosa et al (2014) have studied the processing of negative words with gender violations. They reported a decrease in the amplitude of the LAN component when the mismatching word was negative.…”

Understanding approach and avoidance in verbal descriptions of everyday actions: An ERP study

“…Martín-Loeches et al (2012) found that negative in contrast to positive words caused greater amplitudes of a left anterior negativity (LAN) ERP when participants were detecting syntactic anomalies, which involved a deeper processing of negativity, and a negative bias. Moreover, Hinojosa et al (2014) have shown that negative valenced words caused shorter LAN amplitude than neutral words in detecting errors in a syntactic judgment task (gender agreement), which was interpreted in terms of processing priority for emotional words in contrast to neutral words. In addition, with a gender agreement task, Díaz-Lago, Fraga, and Acuña-Fariña (2015) showed a late positive (P600) ERP associated with word emotionality and grammaticality.…”