The effect of interactivity on initial interactions: the influence of information valence and modality and information richness on computer‐mediated interaction

Abstract

Research investigating relationship development through computer‐mediated channels has failed to acknowledge the importance of initial interactions. Increasingly, multimodal forms of communication, such as audio‐ and videoconferencing, in addition to text‐only formats have emerged on socially oriented websites designed for relationship initiation. Utilizing the principle of interactivity as a conceptual framework, the present study investigates whether increased structural interactivity provided by the additional aural and visual modalities influences initial interaction processes and outcomes online. The results indicate that increased availability of nonverbal modalities, combined with the valence of the information acquired, significantly affected interaction involvement and mutuality, which are markers of processual interactivity, as well as the interaction outcomes of uncertainty, predicted outcome value, and information seeking. Additional analysis revealed that mutuality mediated the effect of richness on postinteraction uncertainty level and the evaluations of future relationship potential.

Keywords:

• Initial Interactions
• Computer‐Mediated Communication
• Interactivity; Relationship Development
• Modality and Information Richness

Notes

Artemio Ramirez, Jr. (Ph.D., University of Arizona) is Assistant Professor in the School of Communication at the Ohio State University, Columbus, Ohio. Judee K. Burgoon (Ed.D., West Virginia University) is Professor of Communication, Professor of Family Studies and Human Development, and Director for Human Communication Research, Center for the Management of Information at the University of Arizona, Tucson, Arizona. Portions of this research were supported by a grant from the U.S. Army Research Institute (Contract #DASW01‐98‐009). The views, opinions, and/or findings in this report are those of the author and should not be construed as an official Department of the Army position, policy, or decision. An earlier version of this paper was presented at the 2001 annual meeting of the National Communication Association, Atlanta, GA. Correspondence should be addressed to Artemio Ramirez, Jr. at the School of Communication, the Ohio State University, 3143 Derby Hall, 154 North Oval Mall, Columbus, OH 43210‐1138, USA. Email: [email protected]

As referenced in previous studies utilizing the principle of interactivity (e.g., Burgoon, Buller, & Floyd, 2001; Burgoon et al., 2002), contingency encompasses the level of dependence and responsiveness in the message exchange between individuals; transformation refers to the dynamic and evolutionary nature of the interaction as indicated by behavioral interdependence and feedback between individuals; participation refers to the extent to which individuals may send and receive verbal and/or nonverbal messages, ranging from full interactive participation to passive observation; and synchronicity encompasses the degree that messages are exchanged in “real time,” such as in chats, audioconferencing, and videoconferencing, versus delayed transmission and reception, as in e‐mail or distribution lists.

Other potential moderators include propinquity, defined as the physical location of individuals in relation to each other; identification, such as the extent to which individuals are known to each other, anonymous (physically or socially), and/or maintain conjured identities within and/or across interactions; concurrency, or whether messages are exchanged sequentially or simultaneously; anthropomorphism, or the degree to which senders/receivers embody “human‐like” qualities, as in the case of Microsoft computer agent helpers, or are actually human; and retrievability, or whether messages are recorded such that they may be recovered for their examination in the future.

FtF interaction is included in the present study for the following reasons. First, the principle of interactivity recognizes FtF as the richest mode of communication and thus serves as an endpoint of a richness continuum. Second, FtF interaction is commonly treated as a benchmark against which computer‐mediated formats are assessed. Its inclusion allows for comparisons with such studies.

The questions were: (a) What do like or dislike about your classes? (b) Discuss something from your past that you feel guilty about. (c) What type of job would you like to have? (d) What are the best and worst things that have ever happened to you? (e) How would your male and female friends describe you? Participants were allowed to determine the order in which they were discussed. Subsequent analyses did not produce any significant effects due to ordering.

Four text chat, two audioconferencing, three videoconferencing, and one FtF interactions required the extra time. Follow‐up analyses failed to reveal any significant effect due to additional time.

A similar procedure was employed in previous studies. Sample disclosures were based upon a small‐scale pilot test and can be acquired from the first author upon request.

The four items used to measure interaction involvement were, “My partner was highly involved in the interaction,” “My partner was interested in our interaction,” “My partner was attentive to our discussion,” and “My partner was attuned to our discussion.” The three items used to measure perceived receptivity were “My partner was open to my ideas,” “My partner was receptive to conversing with me,” and “My partner was responsive to my comments.” The seven items used to measure perceived understanding included “I feel the interaction was a good experience,” “I feel accepted by my partner,” and “I left the interaction feeling I was important.” A similar version of the perceived connectedness measure was also used in Burgoon et al. (2002).

An additional manipulation check was conducted to insure the success of the information valence manipulation. Two trained graduate assistants coded one‐half of the interactions, distributed equally across information valence and interaction format conditions. After observing each interaction independently, each assistant globally assessed the extent to which the confederate disclosed positive or negative information on a 7‐point semantic differential scale (positive information–negative information). An interrater reliability of .94 was achieved. The results paralleled those reported in the text. The disclosures of the confederates in the positive condition were perceived more positively (M=6.21, SD=1.03) than were those in the negative condition (M=2.72, SD=1.41), F(1, 39)=79.90, p<.001, η²=.68.

According to statisticians (e.g., Keppel, 1991; Winer, 1971), nonorthogonal planned contrasts are acceptable when deemed theoretically relevant. Keppel (1991) further asserts that the risk of Type I is minimal as long as the overall number of contrasts conducted is reasonable and does not overwhelmingly exceed n−1.

Although underpowered, an additional analysis involving only the mediated conditions produced the following significant results. Interaction involvement: main effect for information valence, F(1, 59)=12.20, p=.001, η²=.18; and an information valence by richness interaction, F(2, 59)=3.81, p<.05, η²=.12. Mutuality: a main effect for information valence, F(1, 59)=17.71, p<.001, η²=.25; additionally, the main effect for richness approached significance, F(2, 59)=3.01, p=.06. POV: main effects for information valence, F(1, 59)=59.90, p<.001, η²=.53; and richness, F(2, 59)=6.05, p<.01, η²=.18. General uncertainty level: a main effect for information valence, F(1, 59)=4.92, p<.05, η²=.08. Information seeking: a main effect for information valence, F(1, 59)=6.39, p<.05, η²=.11.

Information valence was not examined, as it is not a structural affordance according to the principle of interactivity. Mediation analysis was not conducted for information seeking, since, according to both URT and predicted outcome value theory, any such effects should be themselves moderated by either uncertainty level or POV (see Sunnafrank, 1986, for clarification). Results of the initial analyses are available upon request.

Additional information

Notes on contributors

Artemio Ramirez, Jr

Artemio Ramirez, Jr. (Ph.D., University of Arizona) is Assistant Professor in the School of Communication at the Ohio State University, Columbus, Ohio. Judee K. Burgoon (Ed.D., West Virginia University) is Professor of Communication, Professor of Family Studies and Human Development, and Director for Human Communication Research, Center for the Management of Information at the University of Arizona, Tucson, Arizona. Portions of this research were supported by a grant from the U.S. Army Research Institute (Contract #DASW01‐98‐009). The views, opinions, and/or findings in this report are those of the author and should not be construed as an official Department of the Army position, policy, or decision. An earlier version of this paper was presented at the 2001 annual meeting of the National Communication Association, Atlanta, GA. Correspondence should be addressed to Artemio Ramirez, Jr. at the School of Communication, the Ohio State University, 3143 Derby Hall, 154 North Oval Mall, Columbus, OH 43210‐1138, USA. Email: [email protected]