Cell phone use has become increasingly popular among all generations; especially with younger populations. The instant gratification of connecting to others quickly and easily is very appealing in a fast-paced society. In fact, researchers Diamanduros, Jenkins, and Downs (2007) surveyed 49 college sophomores to inquire about their electronic device habits and found that 98 percent of them used cell phones; with the cell phones being the most used out of all electronic devices included in the survey.  Among the many different amenities that come with cellular phone ownership is text messaging capability. In their survey, Diamanduros et al. found that half of phone users also text message to communicate. Merriam-Webster's online dictionary defines text messaging as "the sending of short text messages electronically especially from one cell phone to another". 

Text messaging has risen in popularity over this last decade. This increase in popularity may be driven by the quick, concise, and convenient nature of text messaging.  Instead of having to take the time to place a phone call to someone, interrupting what both parties are doing, a person can send a text message, forgoing formalities to get right to the point. Another advantage to text messaging is that the receiver is free to answer a text message at their own convenience giving them time to think about the message before they reply. Text messages are also discrete; they can be used to communicate without disrupting other people and they are more private than a phone call. 

One drawback associated with text messaging is the limitations associated with composing the actual message. According to linguist Jonathon Green (2007), text message abbreviations and acronyms are a type of jargon among cell phone users developed out of necessity because many cell phone companies limit the length of each text message to 160 characters.  Also, many phones have only a 12-button key pad rather than a full QWERTY keyboard like those found on most typewriters and computers. With each key representing a number or symbol and multiple letters, it can take up to three key presses to type a single letter.  Green likens text-messaging "jargon" to the lingo used with telegraph and citizen's band (CB) radio users who also abbreviations in their communications to save time and energy. This new text message language includes abbreviations, acronyms, numbers, and symbols to express words, phrases, and emotions. One study examining word abbreviation behavior found that word length was the determining factor for how subjects used abbreviations. Participants abbreviated longer words more frequently than shorter ones as it is almost meaningless to condense shorter words (Hodge & Pennington, 1973). In text messaging for example, users may type lol instead of the phrase, laugh out loud since it is quicker and easier to type the acronym.

With any type of communication, one might wonder how the message is interpreted by the receiver. In a written message the receiving party misses out on body language, facial expressions, and differences in voice pitch and emphasis which are experienced in face-to-face communication. Our question was whether or not the acronyms and abbreviations used in text messaging convey at least as much information as a typical written message. We wanted to know specifically if text messaging acronyms and abbreviations had the same emotional effect as the phrases and words they represent. 

Since acronyms are simply the condensed version of words and phrases, one might assume that they are processed in the same way as words and phrases. Currently, there is limited research on this topic but the existing evidence suggests that acronyms and abbreviations are processed like real words (e.g., Lazlo & Federmeier, 2007; McWilliam, Schepman, & Rodway, 2009). In an ERP study, Laszlo and Federmeier compared neural responses to random letters to known acronyms and abbreviations. Results indicated a difference in neural activity between the two groups such that the results for acronyms and abbreviations resembled neural activity associated with real words. These results suggest that acronyms and abbreviations are processed like words in the brain.

In another study, participants were tested to see if text messaging acronyms and abbreviations are treated like real words (McWilliam et al.). This was done using a modified Stroop task. In the traditional Stroop task, subjects are required to name the color that a word is printed in while ignoring the meaning of the word itself. The word could be the same as the color it is printed in or be an incongruent color word - for example, the word blue displayed in red ink (the correct response would be “red”). Participants are slower to name the color when it is not the same as the meaning of the word (the incongruent condition). This slow down is referred to as the Stroop effect. A similar effect can occur when other stimuli are used. For instance, color naming is slower for meaningful stimuli (e.g., words) compared to non-meaningful stimuli (e.g., random letters). This modified Stroop task is what McWilliam et al. (2009) used to examine the way acronyms and abbreviations are processed. In their study, participants were presented with text message abbreviations, regular words, and non-words. The researchers found that participants were much slower to name the color of the text messaging abbreviations than to name the color of nonwords. This result for acronyms is similar to what is commonly found for word stimuli. From these results, it was suggested that acronyms and abbreviations have been absorbed into the language because participants had difficulty disengaging attention from those stimuli just as they have difficulty disengaging from real words. 

In their study, McWilliam et al. (2009) used only neutral acronyms and abbreviations, but by using this sample they seemingly exclude some of the most common acronyms and abbreviations used in text messaging – emotional items. We wanted to know specifically if text messaging acronyms and abbreviations had the same emotional effect as the phrases and words they represent.  One way to measure the emotionality of words is through an emotional Stroop task.  Just as in the Stroop task, participants are asked to respond to the color a word is printed in, but not to the word itself but in this case the items can have an emotional or neutral valence. Participants take longer to respond to emotional words than to neutral words – an emotional Stroop effect (e.g., Cothran & Larsen, 2008; Dresler, Mériau, Heekeren, & van der Meer, 2009; MacKay et al., 2004). One explanation of the effect is that it takes longer for attention to disengage from emotional words than from neutral words, resulting in slower responses to emotional words.

            Previous research suggests that acronyms and abbreviations are processed like words, but those studies do not reveal the extent of that similarity of processing. The purpose of this study is to examine whether or not the emotional content of acronyms and abbreviations are processed in the same way as the actual words or phrases that they are meant to represent. This is important in today’s society because the use of acronyms and abbreviations are becoming more prevalent as more people are using email, chat rooms, and text messages than ever before. With increased usage of text message acronyms and abbreviations it is vital that we understand how those messages are interpreted. For instance if someone sees LOL do they interpret it with the same emotion they would if they had seen the words laugh out loud?  To investigate this question we used an emotional Stroop task with emotional and neutral words, phrases, acronyms and abbreviations as stimuli. If participants are slower to name the color of emotional acronyms and abbreviations than neutral acronyms and abbreviations it would indicate that emotions are conveyed even in a condensed form of communication. If no emotional Stroop effect is found, it would suggest that, perhaps, some emotional content of the message is lost when acronyms and abbreviations are used.

Methods

Participants

            Thirty students (eight male; average age of 24.8 years) participated in the experiment for extra credit in psychology courses. All but one participant reported sending text messages and of those students 90% report sending text messages daily (52% hourly) and 82% use acronyms and abbreviations in their own text messages. All participants were English speakers having started learning the language before the age of five. No participants reported having colorblindness.

Materials

            All experimental sessions were run and data collected using E-Prime experimental software. Participants were seated approximately 36 inches from the computer monitor. Verbal responses were collected using a microphone. An experimenter seated in the room monitored all responses and recorded participant errors and microphone malfunctions. Trials with microphone malfunctions were excluded from all analyses (3.3% of the total trials).

            One challenge in developing this experiment was finding common text messaging acronyms and abbreviations. Unlike words which have been thoroughly studied and cataloged (e.g., Kuĉera - Francis written frequency count, Kuĉera & Francis, 1967; MRC Psycholinguistic Database; English Lexicon Project, Balota et al., 2007) there are no established norms for text messaging acronyms and abbreviations. Thus, before conducting our experiment we had to find out which acronyms and abbreviations were most familiar to college students and which were perceived as having an emotional valance. To accomplish this, two versions of a survey were created (100 acronyms and abbreviations in each for a total of 200 acronyms and abbreviations) and distributed to students in introductory psychology classes; students received extra credit in their course for participating. In the survey participants were first asked some questions regarding their text-messaging behavior and they were asked for basic demographic information. Then participants were asked to rate each acronym or abbreviation for familiarity and emotionality. Familiarity was indexed on a scale of one to four: 1 = Don’t know what it means; 2 = Seen it before, don’t know what it means; 3 = Seen it before, know what it means; 4 = I use it, know what it means. Emotionality was indexed on a scale of one to five: 1 = Very negative; 2 = Slightly negative; 3 = Not emotional; 4 = Slightly positive; 5 = Very positive. If the participant did not know the meaning of the acronym or abbreviation they were instructed to not rate the emotional content of the item.

            Forty-four students responded to this initial survey. Of those students, 95.4% reported text-messaging on a daily basis; 66% of those students text-message hourly. In addition, 91% of students responding to the survey reported using acronyms and abbreviations in their own text messages. Participant ratings were used to select stimuli for the emotional Stroop experiment. In order to be considered for the experiment, stimuli had to be familiar to at least 70% of survey respondents. From those, emotional stimuli were selected if at least 60% of respondents rated the item as being negative of positive while neutral stimuli had to be rated as “not emotional” by at least 60% of respondents. From the list of familiar acronyms and abbreviations that were rated as being emotional or neutral, 52 (26 neutral, 26 emotional) were selected for use in the emotional Stroop experiment. In addition to the list of acronyms and abbreviations, the corresponding words and phrases were also gathered for use in the experiment (e.g., talk to you later for TTYL). Thus there were four types of stimuli used in the experiment: emotional acronyms and abbreviations, neutral acronyms and abbreviations, emotional words and phrases, and neutral words and phrases. Each participant was exposed to each stimulus type. The presentation (acronym or words) of the stimulus was varied between participants, for example, while Participant 1 would see TTYL, Participant 2 would see talk to you later. No participant saw two stimuli with the same meaning (both TTYL and talk to you later). Because this design was used, two experimental lists were created; the same number of participants completed each list.

Procedure

            Participants first read an informed consent document identifying risks and benefits associated with the experiment. Essentially, participants were informed at that point that some words presented in the experiment may be emotion evoking and may even be considered to be taboo. After reading and signing the informed consent document participants completed a short questionnaire used to gather demographic and text-messaging behavior information. These questions were the same as those given to survey respondents. One additional question asked if the participant was colorblind.

            Before beginning the experiment participants were told they would see words, phrases, acronyms, and abbreviations presented in different colors. Their task was to say aloud the color the item was presented in. Participants were also instructed to respond as quickly as possible without making too many mistakes. After the instructions, participants completed a series of practice trials in which they simply had to name the color that XXXXX’s were presented in. Upon completion of the practice, participants were given another chance to ask questions before starting the experiment.

            During all experimental trials the participant was presented with a white fixation cross for 500 ms, followed by a 100 ms blank inter-stimulus interval, followed by a stimulus presented in one of four colors (red, green, yellow, blue). The stimulus (emotional acronym, neutral acronym, emotional words, neutral words) remained on the screen until the participant triggered the voice key via the microphone. The next trial began after a 2 sec blank inter-trial interval. A black background was used throughout the experiment. Participants were given three breaks during the experiment during which they could take as much time as needed before continuing. At the conclusion of the experiment participants were debriefed.

Results

            Responses faster than 200 ms were excluded from all analyses as were response slower than three times the standard deviation for each participant. This outlier procedure resulted in the exclusion of 1.7% of the data.

            A 2 (emotionality – emotional, neutral) X 2 (stimulus type – acronym or abbreviation, word or phrase) repeated-measures ANOVA of participant response times revealed a significant main effect of emotionality, F (1, 29) = 8.47, p = .007, indicating that, overall, participants were slower to name the color of emotional items (M = 678.5) than neutral items (M = 652.5). An interaction between emotionality and stimulus type was also found, F (1, 29) = 5.20, p = .030.

            Follow-up comparisons revealed a significant difference between emotional (M = 684.5) and neutral words and phrases (M = 639.2), t (29) = 3.06, p = .005, but no difference between emotional (M = 672.4) and neutral acronyms and abbreviations (M = 665.8), t (29) = 0.73, p = .473.

            A repeated-measures ANOVA with the same variables was also conducted for error rates. It revealed an interaction between emotionality and stimulus type, F (1, 29) = 7.65, p = .010. Follow-up comparisons revealed a significant difference between emotional (M = 2.6%) and neutral words and phrases (M = 0.5%), t (29) = 2.50, p = .018, but no difference between emotional (M = 0.5%) and neutral acronyms and abbreviations (M = 1.8%), t (29) = -1.54, p = .134. 

            In sum, an emotional Stroop effect was found for emotional words and phrases in both participant response times and error rates, but no emotional Stroop effect was found for emotional acronyms and abbreviations.

Discussion

As our results show, we found a delay in response time and an increase in error rates when participants viewed words and phrases with emotional context, but no effect of emotion was found for acronyms and abbreviations. Thus, our results indicate that an emotional Stroop effect occurred for words and phrases, but not for acronyms and abbreviations. Even though research suggests that acronyms are processed similarly to words and phrases, our study demonstrates that the emotional content of acronyms are not processed as the words and phrases they represent.  Therefore, it appears that emotion is lost when a word or phrase is minimized to an abbreviation or acronym.

            There are several possible explanations for the lack of emotional Stroop effect associated with the abbreviated emotional stimuli. One possibility is that participants were not familiar with the text messaging acronyms and abbreviations used in the study. Although the stimuli were selected based on rating from other students, the participants in the study were not explicitly asked whether or not they knew the meanings of the acronyms and abbreviations. This is potentially problematic as participants should not have emotional responses to emotional acronyms and abbreviations they do not know. In future studies it would be helpful to have participants identify the acronyms and abbreviations after the experiment in order to rule out this possibility.

            A second explanation for the lack of an emotional Stroop effect for acronyms and abbreviations is that those stimuli may be so new to the participant that they do not have the same semantic richness as regular words. A parallel case exists for those who learn a second language later in life. For instance, it has been shown that those individuals do not have the same emotional response to taboo words in their second language even though they have strong emotional responses to taboo words in their first language (e.g., Harris, Ayçiçeði, Berko Gleason, 2003). Thus it is possible that acronyms and abbreviations operate like a second language. If the “language” were learned at an earlier age, e.g., before the critical period, perhaps acronyms and abbreviations would convey the intended emotion. This hypothesis could be tested by recruiting individuals who grew up communicating via text messages for the experiment. With the increasingly pervasive use of cell phones and text messaging, even young children are texting. Within a few years, these children will be entering college; therefore it is possible that very different results could be obtained for the same study within a few years.

            A third explanation for the absence of an emotional Stroop effect for acronyms and abbreviations could simply be that the effect is too small to be statistically significant. We argued that the emotional Stroop can be used as a tool to index emotional responses, but this measure might lack the sensitivity to detect emotion elicited by acronyms and abbreviations. In future research participants’ physiological responses such as the galvanic skin response, heart rate, and blood pressure could be examined in response to emotional acronyms and abbreviations.  

            Another possibility is that the acronyms and abbreviations used in text messages are not meant to carry the same emotional content as their constituent words. For instance, do people feel emotion when they write an acronym or abbreviation? If not, then there is little reason to expect the receiver to experience emotion when they see an emotional acronym or abbreviation. Further research could be used to asses where emotion is “lost” in this communication process.

            Whatever the reason for the absence of the emotional Stroop, further research is warranted to investigate the way people interpret text messaging acronyms and abbreviations. We found no evidence to suggest that people feel emotion when reading a text message abbreviation or acronym. This finding could have considerable implications. For instance, some companies now advertise their products and services through mass text messages. Perhaps those ads would be more effective if they were spelled out, avoiding acronyms and abbreviations. There may also be implications for the use of acronyms and abbreviations in the workplace. For example, to increase the sense of urgency one might ask that a project be completed as soon as possible instead of ASAP.

References

Balota, D. A., Yap, M. J., Cortese, M. J., Hutchison, K. A., Kessler, B., Loftis, B., . . . Treiman, R. (2007). The English lexicon project. Behavior Research Methods, 39, 445-459.

Cothran, D. L. & Larsen, R. (2008). Comparison of inhibition in two timed reaction tasks: The color and emotion Stroop tasks. The Journal of Psychology, 142, 373-385.

Diamanduros, T., Jenkins, S. J., and Downs, E. (2007). Analysis of technology ownership and selective use among undergraduates. College Student Journal, 41, 970-976.

Dresler, T., Mériau, K., Heekeren, H. R., & van der Meer, E. (2009). Emotional Stroop task: effect of word arousal and subject anxiety of emotional interference. Psychological           Research, 73, 364-371.

Harris, C. L., Ayçiçeði, A., Berko Gleason, J. (2003). Taboo words and reprimands elicit greater autonomic reactivity in a first language than in a second language. Applied       Psycholinguistics, 24, 561-579.

Hodge, M. & Pennington, F. (1973). Some studies of word abbreviation behavior. Journal of Experimental Psychology, 98(2), 350-361.

Kuĉera,  H. & Francis, W. H. (1967). Computational analysis of present-day American English. Providence, RI: Brown University.

Laszlo, S. & Federmeier, K, D. (2007). Better the DVL you know: Acronyms reveal the contribution of familiarity to single-word reading. Psychological Science, 18, 122-126.

MacKay, D. G., Shafto, M., Taylor, J. K., Marian, D. E., Abrams, L., & Dyer, J. R. (2004). Relations between emotion, memory, and attention: Evidence from taboo Stroop, lexical         decision, and immediate memory tasks. Memory & Cognition, 32, 474-488.

McWilliam, L., Schepman, A., & Rodway, P. (2009). The linguistic status of text message abbreviations: An exploration using a Stroop task. Computers in Human Behavior, 25, 970-974.