Temporal Closeness Promotes Imitation of Meaningful Gestures in Face-to-Face Communication

Abstract

The present research investigated how temporal distance affects imitation of gestures during a conversation. According to construal level theory, psychological closeness (vs. distance) leads to preference for more pictorial and embodied (vs. symbolic and amodal) forms of communication as well as to greater context sensitivity. We thus hypothesized that people are more sensitive to the contextual meaning of gestures and show more imitation of meaningful (vs. meaningless) gestures when primed with temporal closeness (compared to distance). Participants (N = 91) were assigned to the role of a job interviewer for an internship that would begin in 1 week (vs. in 1 year) and interacted with an ostensible applicant who displayed various meaningful and meaningless gestures. When the internship was to start in 1 week, participants imitated meaningful gestures more often than meaningless gestures. When the internship was to start in 1 year, participants imitated both gesture types equally frequently. This finding indicates that people are more likely to embody verbal statements of an interaction partner when feeling psychologically close to the subject of the conversation.

Appendices

Appendix 1: Pretest of Gestures

We developed a fictional job interview with six critical questions and answers. The pretest was designed to validate the meaningfulness of the scripted gestures accompanying each answer. We video-recorded a confederate answering the six interview questions together with corresponding scripted gestures. In each case, the confederate used a total of four gestures, two meaningful and two meaningless ones (see Table 1), which were arranged a priori in a fixed quasi-randomized order. For the first three answers, the confederate used 12 different gestures: six meaningful and six meaningless. For the remaining three answers, she used the same 12 gestures but in their meaningless version (when the gesture was meaningful before) and meaningful version (when the gesture was meaningless before), respectively. To keep the visual field of gestures constant, all gestures occurred within the physical space of the confederate’s body center (Gullberg and Holmqvist 1999; McNeill 1992). The confederate was trained to perform both versions of a gesture equally often, intense, and long, independent of her accompanying verbal statement.

For the pretest, 31 undergraduate psychology students were randomly assigned to one of two distance conditions (close: n = 15 in 1 week vs. distant: n = 16 in 1 year). As shown in Table 1, distance was manipulated by the phrase “in 1 week” (“in 1 year”) which was included in each interview question. Participants were asked to read each of the six interview questions on a computer screen and watch the applicant’s respective answer. For each of the 24 gestures that participants watched during the whole interview, they answered five items on a paper–pencil questionnaire in front of them. Four items asked participants to rate how important, meaningful, appropriate, and incidental the gesture was in the given situation, according to a 9-point Likert-scale (1 = not at all to 9 = very much). One item asked participants how likely they would spontaneously imitate the gesture in a real interview if they were the job interviewer. In (re)viewing the video to answer the items, participants could stop, fast forward, or reverse the track as often as they liked.

We found satisfactory internal reliability between the three items measuring the meaningfulness of the gestures (important, meaningful, appropriate; Cronbach’s alphas from .63 to .97), and thus calculated the mean of these three items for the meaningful and meaningless gestures as a measure of meaningfulness. One item (“How incidental is the gesture in the situation?”) substantially reduced the reliability of the meaningfulness scale and was excluded. There was no difference between meaningful and meaningless gestures on this item, nor an interaction with condition, Fs < 1.36.

For the meaningfulness measure, a 2 (Distance: in 1 week vs. in 1 year) × 2 (Meaning of gesture: meaningful vs. meaningless) mixed analysis of variance (ANOVA) revealed a main effect of meaning, F(1, 29) = 116.32, p < .001, η ²_p  = .80, f = 2.00. Participants rated gestures that emphasized the verbal statement of the applicant as significantly more meaningful (M = 6.33, SD = 1.08) than gestures that did not (M = 4.19, SD = 0.94), M _D = 2.14, SE _D = 0.20, 95% CI [1.74, 2.55]. There was neither a main effect of distance nor a significant interaction between meaning of gesture and distance, Fs < 1.

For the self-reported likelihood of imitating the gesture in a real interview, a 2 (Distance: in 1 week vs. 1 year) × 2 (Meaning of gesture: meaningful vs. meaningless) mixed ANOVA also revealed a main effect of meaningfulness, F(1, 29) = 61.10, p < .001, η ²_p  = .68, f = 1.46. Participants reported imitating meaningful gestures with a significantly higher likelihood (M = 47.10%, SD = 19.82) than meaningless gestures (M = 26.16%, SD = 12.03). There was neither a main effect of distance nor a significant interaction between meaningfulness and distance, Fs < 1. Ratings for the likelihood of gestural imitation were not influenced by the distance manipulation.

In summary, by varying the accompanying verbal statement in our study design we could successfully create a meaningless versus a meaningful version of each gesture. In the main study, participants and the confederate engaged in a simulation of a real job interview to investigate actual imitation behavior.

Appendix 2: Post Interview Questionnaire

The final (exploratory) questionnaire contained two items about the evaluation of the applicant (i.e., “How likely is it that you as human resources manager would employ the person?”, “How good do you think the applicant’s answers were?”, r = .76, p < .001), one item about the extent to which they thought the applicant’s gestures fit what she was saying, five items about how much they liked the applicant (e.g., “How much do you like the person?”, “How close did you feel to the person?”, Cronbach’s α = .85), five items about their motivation during the interview (“How interesting was the interview?”, “How motivated were you to conduct the interview well?”, Cronbach’s α = .78), three items about their mood during the interview (e.g., “Did you feel comfortable during the interview?”, Cronbach’s α = .90), and seven items from the perspective-taking sub-scale of the Interpersonal Reactivity Index (IRI; Davis 1980; e.g., “Before criticizing somebody, I try to imagine how I would feel if I were in their place”, Cronbach’s α = .73). Participants rated all of these items on a 9-point Likert scale (1 = not at all to 9 = very much). In addition, participants indicated the extent to which they felt socially close to the applicant via the Inclusion of the Other in the Self Scale (IOS; Aron et al. 1992). One item asked which kind of tasks they would assign to the applicant if she started the internship (from making copies to tasks with more responsible such as customer care). For correlations between the study variables with the temporal distance manipulation and the main dependent variable frequency of imitation, see Table 2 in Appendix 2.

Table 2 Pearson’s Product Moment Correlation Matrix Between the Distance Manipulation and all Measured Variables (Behavioral Identification Form as a measure of level of construal, mean number of imitated meaningless and meaningful gestures, overall post-interview evaluation of the applicant, liking of the applicant (the confederate), self-reported perspective taking measured with IRI subscale (Davis 1980), interviewers’ self- reported motivation during the interview, self-reported mood during the interview, overall interview duration in seconds)

Afterwards, participants filled out a modified version of the Behavioral Identification Form (BIF; Vallacher and Wegner 1989) as explained in the main text body. Finally, an open-ended question probed if participants correctly recollected the time point for the start of the internship. Six participants were not able to freely recall when the applicant would have started her internship (i.e., in 1 week or in 1 year), and were therefore excluded from further analyses (n = 2 in 1 week; n = 4 in 1 year), which was correctly remembered by 93.4% of the participants (close condition: 95.6%; distance condition: 91.3%). At the end, they were asked to report their age, sex, education, and any suspicion they had regarding the experimental procedure and purpose of the study. No participant correctly guessed the purpose of the study.