Be aware of the rifle but do not forget the stench: differential effects of fear and disgust on lexical processing and memory

ABSTRACT

The aim of this study was to investigate the role of discrete emotions in lexical processing and memory, focusing on disgust and fear. We compared neutral words to disgust-related words and fear-related words in three experiments. In Experiments 1 and 2, participants performed a lexical decision task (LDT), and in Experiment 3 an affective categorisation task. These tasks were followed by an unexpected memory task. The results of the LDT experiments showed slower reaction times for both types of negative words with respect to neutral words, plus a higher percentage of errors, this being more consistent for fear-related words (Experiments 1 and 2) than for disgust-related words (Experiment 2). Furthermore, only disgusting words exhibited a higher recall accuracy than neutral words in the memory task. Moreover, the advantage in memory for disgusting words disappeared when participants carried out an affective categorisation task during encoding (Experiment 3), suggesting that the superiority in memory for disgusting words observed in Experiments 1 and 2 could be due to greater elaborative processing. Taken together, these findings point to the relevance of discrete emotions in explaining the effects of the emotional content on lexical processing and memory.

KEYWORDS:

• Discrete emotions
• word processing
• memory
• lexical decision task
• emotional dimensions

Acknowledgments

We would like to thank Drs Briesemeister, Kuchinke and Jacobs for kindly providing us with their stimulus list.

Notes

1. It might be argued that the differences in memory between disgusting words and fearful words can be explained by dimensions other than valence and arousal. If this is the case, the present results would not constitute evidence in favour of the discrete emotions approach. In particular, an anonymous reviewer pointed out the relevance of the dominance/potency/control dimension. In order to explore whether this dimension has contributed to our results, we collected the control values of our stimuli (we obtained the ratings for 45 words from Redondo et al., 2007, and for the remaining 63 words we conducted a questionnaire of dominance/control that was filled in by 36 participants). Then, we did an ANOVA with these ratings by including “type of word” (disgusting, fearful, neutral) as a factor. The results revealed a significant effect of type of word, F (2, 105) = 42.29, p < .001. Pairwise comparisons showed that there were differences between the three sets of words in perceived control/potency (M = 4.49, 3.85 and 5.38 for disgusting, fearful and neutral words, respectively).

In order to know the extent to which this difference in control may have affected our results, we looked for studies that tested the effects of this variable on lexical processing and memory. Concerning the LDT, we were able to find only a few studies evaluating the effects of dominance/control (apart from valence and activation) in this task (Wurm & Vakoch, 1996; Wurm, Vakoch, & Seaman, 2004). The results here are mixed, in that high potency scores were associated to either faster (Wurm & Vakoch, 1996) or slower (Wurm et al., 2004) RTs, depending on the study.

It should be noted, however, that the definition of “potency” in these studies does not correspond to that commonly used in the field (i.e. the definition used by Bradley & Lang, 1999). Thus, in the studies of Warm et al., the dimension of potency refers to “how strong or weak” a thing is, whereas Bradley and Lang defined potency as the capacity (power, control) that people feel when faced with that thing. For this reason, it is difficult to compare these results with ours. Regarding memory, we were able to find only a single study investigating the effects of potency/dominance (Mneimne, Wellington, Walton, & Powers, 2015). The authors there investigated the hemispheric lateralisation of memory for the affective dimensions of valence and dominance. Thus, they compared memory for positive words (with high dominance) and two types of negative words (negative words having high or low dominance) that were presented to each hemisphere. When words were presented to the left hemisphere, there was no effect of dominance. In contrast, when they were presented to the right hemisphere, low dominance words were remembered better than high-dominance words. Again, this study cannot be directly compared to our work due to methodological differences. However, considering the results, if low dominance is associated with better recall, fearful words should have been those remembered better in our study, which was clearly not the case.

For the above-mentioned reasons, it is unlikely that our results can be explained by the dimension of control/potency. They seem to be better accounted for by a discrete emotions approach.

Additional information

Funding

This work was supported by grants PSI2015-63525-P; PSI2015-68368-P (MINECO/FEDER), Ref 2016PFR-URV-B2-37 from the Universitat Rovira i Virgili and Ref. H2015/HUM-3327 from the Comunidad de Madrid. The second author has a grant from the Universitat Rovira i Virgili (2015PMF-PIPF-16).