Making sense of emotion effects in lexical processing
1 : Cognitive, Perceptual and Brain Sciences Research Department University College London
(UCL)
* : Corresponding author
26 Bedford Way, London WC1H 0AP -
United Kingdom
Although emotional connotations of words have long been considered irrelevant in most language processing studies, a growing body of evidence now suggests that words' emotional content affects their processing. Words also offer the opportunity to investigate emotion effects in a highly controlled manner, as we now understand much more about specific lexical, sublexical and semantic characteristics of words which affect their processing. Nonetheless, current lexical processing studies have highly incompatible results despite using exactly the same tasks (e.g. Briesemeister, Kuchinke & Jacobs, 2011; Estes & Adelman, 2009 & ESCAN 2012 talk; Kousta, Vinson & Vigliocco, 2009). First, studies report differences in the direction of effects (overall disadvantage for negative words, vs. advantage for emotional words whether positive or negative). Second there is disagreement about whether emotional effects are continous or discrete. To better understand how these conflicting findings may arise from such similar experiments, we conducted various analyses on a single data set: using different regression models in which valence was categorically vs. continuously defined, and crucially in which we controlled other lexical and semantic variables that were not taken into account in some studies. These analyses clearly show how, once a greater degree of control on lexical and semantic variables is applied, it appears that emotional connotations, whether positive or negative, confer a benefit for lexical processing.
| Subject : | : | Poster |
| Topics | : | PO: Emotions 2 |
| PDF version | : |  PDF version |
