Illustrative Theoretical Findings
Before turning to the specific results, we want to emphasize again that this aspect of our three studies was intended to illustrate the potential of the closeness- generating procedure. Any implications for the substantive theories are clearly highly preliminary and in many cases employed less-than-optimal procedures for opera- tionalizing the major theory-relevant variables.
Studies 1 and 2 focused on adult attachment style (Hazan & Shaver, 1987). There were three key results: (a) Avoidant/dismissive pairs reported less postinterac- tion closeness than any of the other pairings; (b) pairs with a preoccupied partner reported a greater discrepancy between actual and desired postinteraction closeness than any of the other pairings; and (c) for all pairing types, there was an overall change in reported attachment style from before to after the interaction in the direction of greater endorsement of styles consistent with a positive model of other. Each of these findings, if replicated in future research, would bear importantly on theoretical understanding in this area. Note that in each case, the findings bear directly on issues of causal directions that would be difficult to sort out with nonexperi- mental methods.
The illustrative theoretical issue of Study 3 was introversion/extraversion. The main finding was that when no special instructions about getting close were given, extraverted individuals reported achieving more closeness than did introverted individuals; but when closeness was made an explicit task of the procedure, the greater reported closeness for the extraverted individuals essentially disappeared. If this pattern is supported in future research, it would deepen our understanding of the dynamics of the interaction of personality and social behavior. Most important for the present purposes, this finding illustrates the potential of our procedure for yielding theoretically significant results by both systematically controlling pairings of individuals and systematically manipulating the circumstances of their interaction.
How Can This Closeness-Generating Paradigm Help Researchers ?
There are at least four ways that the present procedure might help researchers advance theory in the social psychology of close relationships and related areas. First, the closeness-generating procedure permits individual- difference variables to be measured prior to (and during and after) relationship formation. Second, this procedure permits researchers to control who is in a relationship with whom and separates preexisting individual differences from determinants of pairings such as choice of partner and opportunity constraints. Regarding these first two values of the closeness-generating procedure for research purposes, some other variables that could be examined (in addition to attachment style and introversion/extroversion) are neuroticism, happiness, communication skills, self-esteem, dispositional trust, style of handling conflicts, communal/exchange orientation, and gender. Indeed, some of these variables could themselves be temporarily manipulated, making such a study a true experiment over all variables.
A third way in which this procedure can help researchers is by permitting direct manipulation of various relationship-relevant variables, such as the motivations and expectations of the participants, the kinds of interactions that occur, and the length and intensity of interaction. In the present studies we illustrated these possibilities by comparing the impact of our usual procedures with small talk, by manipulating whether subjects expected their partners to like them, and so forth. These are only a few of a great many possibilities. For example, by creating appropriate interaction tasks, research on relationship awareness (Acitelli, 1988) could manipulate whether subjects were made to use slanguage and focus on the relationship per se; research on relationship investment (e.g., Rusbult, 1983) might actually control the amount of investment subjects made in the process of developing the relationship; research on trust (e.g., Holmes & Rempel, 1989) might actually be able to create or undermine trust; research on particular relationship processes, such as demand-withdraw (e.g., Christensen, 1988), might create those processes in the lab with newly created couples not necessarily predisposed to either process; and so forth. (Also note that by using the closeness- generating procedure to produce a new relationship, there are fewer ethical problems in employing other procedures within the study that might have negative relationship effects.)9
A fourth main way in which this paradigm might be of use is that it puts relationship development into a setting in which it can be readily observed. For example, one could simply give questionnaires after each segment and track change—paralleling procedures used in studies that monitor actual couple development over weeks or months (e.g., Surra, 1987). Or one could observe interaction as the pairs are carrying out the tasks; one might use a procedure like that developed by Ickes, Bissonnette, Garcia, and Stinson (1990), in which inter- actants are videotaped automatically (i.e., without any person actually operating the camera) without their knowledge and then at the end of the experiment given the opportunity to destroy the tape before it is seen by anyone. Or one could assess physiological processes during the interaction in ways that have proven successful in studies of interactions of couples in ongoing relationships (e.g., Levenson 8c Gottman, 1983).
The closeness-generating paradigm described here differs from other experimental procedures for generating interaction used in recent years (e.g., Asendorpf, 1989; Ickes et al., 1990; Thorne, 1987) in that tasks are explicitly structured to create maximum felt closeness in a short period. In addition, unlike many of the procedures used in the self-disclosure research of the 1970s and 1980s, it does not require a confederate and would not ordinarily require deception at all (other than not revealing hypotheses being tested).
Conclusion
The study of close relationships has progressed in the last 15 years by freeing itself from the strictures of controlled interaction and the true experiment (Duck, 1988). But without these tools, we are constantly faced with ambiguities of interpretation and obstacles to identifying the details of hypothesized processes. This article is not a call to abandon the richness of real-world experience, particularly in the area of close relationships. Rather, it is an invitation to alternate field and laboratory, correlational and experimental methods. In this way, work with naturally occurring relationships can be refined and sorted out through experimentally generated relationship experiences. At the same time, precise work with experimentally generated relationships can be inspired by and checked against the reality of relationships as they naturally occur in the world. In short, it is time for researchers of close relationships to find ways to welcome back our wayward friend, the true experiment. We hope that this paradigm we have developed will aid in this process.
APPENDIX
Instructions to Subjects Included With Task Slips Packet
INSTRUCTIONS (Please both read carefully before continuing)
This is a study of interpersonal closeness, and your task, which we think will be quite enjoyable, is simply to get close to your partner. We believe that the best way for you to get close to your partner is for you to share with them and for them to share with you. Of course, when we advise you about getting close to your partner, we are giving advice regarding your behavior in this demonstration only, we are not advising you about your behavior outside of this demonstration.
In order to help you get close we've arranged for the two of you to engage in a kind of sharing game. You're sharing time will be for about one hour, after which time we ask you to fill out a questionnaire concerning your experience of getting close to your partner.