ERP correlates of masked affective priming with emoticons Montserrat Comesaña a,⇑ , Ana Paula Soares a , Manuel Perea b , Ana P. Piñeiro d , Isabel Fraga c , Ana Pinheiro e a Human Cognition Lab, CIPsi, University of Minho, Braga, Portugal b ERI-Lectura and Departamento de Metodología, Universitat de València, Valencia, Spain c Cognitive Processes & Behaviour Research Group, University of Santiago de Compostela, Spain d Departamento de Psicología, Universidad del Norte, Barranquilla, Colombia e Neuropsychophysiology Lab, CIPsi, University of minho, Braga, Portugal article info Article history: Available online 16 January 2013 Keywords: Emoticons Affective words Masked affective priming ERPs abstract Emoticons seem to enrich computer-mediated communication by improving enjoyment, perceived rich- ness and usefulness of information (Huang, Yen, & Zhang, 2008). Despite their extensive use over the last decades, the way emoticons affect subsequent emotional/cognitive processing is not fully understood. Here we conducted a masked priming experiment that explored the time course of the masked affective priming effect while recording event-related potentials. Type of prime (emoticon vs. word) and prime valence (positive vs. negative) were manipulated to assess their influence in the processing of positive/ negative target words. Results showed a masked affective priming effect in early (N2) and late temporal windows (LPC). This effect was observed in early components for negative target words and in later com- ponents for positive target words. Furthermore, it was restricted to those targets preceded by emoticon primes. Thus, the processing of emoticons seems to be privileged when compared with the words to which they refer. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction In a transcript of Abraham Lincoln’s speech from 1862, we can observe what is possibly the first use of an emoticon [;)]. 1 Initially, emoticons were pictorial representations of facial expressions cre- ated by using punctuation marks and letters. More recently, and in contexts of Messenger-like applications, there has been an evolution to more human-like emoticons (e.g., ). The extensive use of emot- icons starts with the current digital era, in which the communication mediated by computer has (to some extent) replaced face-to-face communication. Emoticons seem to have a positive effect in enjoyment, personal interactions, perceived richness and usefulness of information, thus constituting a valuable addition to more traditional communi- cation methods (see Huang et al., 2008). However, experimental research on how emoticons influence cognitive/affective processes is still scarce. Previous studies have revealed that emoticons activate brain areas related to the processing of affective valence – although not those related to the perception of human faces (Yuasa, Saito, & Mukawa, 2006). This finding suggests that emoti- cons are particularly useful to examine expressions of positive and negative affect since they minimize the influence of feature confounds associated with real faces (see Jolij & Lamme, 2005). In addition, studies of affective blindsight with normal participants have revealed that the affective content of emoticons is automati- cally processed (Jolij & Lamme, 2005; see also White, 1995). In the present paper, we examined the influence of briefly pre- sented emoticons (e.g., the primes vs. ) on the processing of an affective target word [e.g., AZÚCAR (SUGAR)]. For comparison pur- posed, we also employed the words that those emoticons refer to [e.g., the primes alegría (happiness) vs. rabia (rage)] in order to compare the processing of emoticons to that of affective words. This was done by manipulating the affective/evaluative relation- ship between prime and target (positive prime–positive target vs. negative prime–positive target; positive prime–negative target vs. negative prime–negative target) in an affective categorization task (i.e., deciding whether a target word is positive or negative). Affective priming occurs when the response to a target stimulus differs depending on whether both prime and target are either affectively congruent or incongruent. In order to avoid any poten- tial (non-automatic) strategies that may occur when using un- masked visible primes (see Bargh, 1996; Hermans, De Houwer, & Eelen, 2001; Kiefer & Brendel, 2006; Kinoshita & Norris, 2010; Per- ea, Duñabeitia, & Carreiras, 2008) and to track the earliest mo- ments of processing, we used a sandwich-masked priming technique, in which a briefly presented prime is preceded and fol- lowed by a mask (e.g., see Draine & Greenwald, 1998). Importantly, we measured not only behavioral responses, but also Event Related 0747-5632/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chb.2012.10.020 ⇑ Corresponding author. Address: Human Cognition Lab, CIPsi, School of Psychology, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal. Tel.: +351 253 604 270; fax: +351 253 604224. E-mail address: mvila@psi.uminho.pt (M. Comesaña). 1 Retrieved from http://cityroom.blogs.nytimes.com/2009/01/19/hfo-emoticon/. Computers in Human Behavior 29 (2013) 588–595 Contents lists available at SciVerse ScienceDirect Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh