Social life is constituted of interactions with others – others whom we must rapidly classify as friend or foe, as trustworthy or not. Gauging another's trustworthiness relies on successfully reading nonverbal signals – signals that have been selected through human evolution to serve precisely such a communicative function. These deeply ingrained signals – some part of our phylogenetic heritage and some part of our socially constructed communication system – form the unwritten “order” for cooperative encounters, enabling both individuals and societies to survive and thrive. Yet paradoxically, the same course of evolution has also remunerated, with greater prospects for survival, those who manipulate and falsely manufacture such signals; in short, those who cheat, dissemble and deceive. Put differently, the course of human development has produced a system of presumably reliable signals of veracity, authenticity, trust and trustworthiness, while simultaneously conferring advantages on sham portrayals of those same signals. The use of dishonest signals is not confined to humans; natural selection has also rewarded sophisticated forms of cheating among all manner of living organisms (Greenfield, 2006). Consequently, these nonverbal signals, many of which are universal and have similarities among other species, are among the most important for humans to produce and read and the most useful for computational methods to detect and track.

In what follows, we foreground those aspects of social signaling related to veracity that have widespread use and recognition. These are the kinds of signals that Burgoon and Newton (1991) identified as corresponding to a social meaning model in that they are recurrent expressions that have consensually recognized meanings within a given community. In this chapter, we first provide the reader background on the nature of the aforementioned signals. Next, we discuss automated methods for human nonverbal communication computing, i.e., methods we used for identifying and tracking such signals. Then, we discuss computer vision technologies using sensors operating in different wavelengths of the infrared band and not only the visible band, which is conventionally used. We conclude with recommendations for promising future research directions where the latest technologies can be applied to this elemental aspect of social signaling.

A Biological and Social Imperative

Adaptation is a biological and social imperative – biologically, for the survival of a species; socially, for the survival of a society. Vertebrates and invertebrates alike come equipped with reflexes that produce involuntary survival-related forms of adaptation in the form of fight or flight responses. In the face of a threat, a frightened organism may sound an alarm call, emit an odor, or display a visual signal that is recognized by species mates as fear. The fear triggers behavioral mimicry that leads the entire flock, herd, swarm, or school to take flight en masse. Or, rage by a single individual may fuel a contagion of aggression that turns into mob violence. These reciprocal actions may not be easily suppressed or controlled.

Other forms of adaptation are volitional, intentional, and socially oriented. Humans may copy the speech patterns of their social “superiors” in hopes of being viewed as belonging to the same ingroup. Or one person's antagonistic demeanor toward a target may elicit a docile, calming response by the victim.

Both forms of adaptation – involuntary and voluntary – undergird social organization. As Martin Luther King Jr. observed in his Letter from a Birmingham Jail (1963), “we are caught in an inescapable network of mutuality.” By means of verbal and nonverbal communication, civilized societies negotiate access to scarce resources, work out their interpersonal relationships, and create their social organizations. Thus, communication is fundamentally an adaptive enterprise that reflects and channels these biological and social imperatives. How, when, and why such adaptation takes place is the topic of this chapter.

Forms of Coordination and Adaptation

It is perhaps unsurprising that given its fundamental role in social interaction, terms describing various forms of adaptation have proliferated, leading to conceptual and operational disarray. The same terms have been applied to different phenomena and different terms have been applied to the same phenomenon. Here we introduce the most common usage from scholars of communication, psychology and linguistics who over the course of forty years have largely converged on these definitions. These conceptual and operational definitions are summarized in Table 8.1.

Chapter 9 offered a methodological demonstration of how various statistical analysis techniques spotlight different facets of the adaptation process, with some being more enlightening than others. It also offered empirical evidence for a complex set of adaptation patterns. Because no single investigation is definitive, this chapter presents results from additional dyadic interaction studies to (a) illustrate how statistical methods compare on their informativeness and degree of complementarity and (b) present additional empirical evidence about which adaptation patterns are most prevalent in face–to–face interactions. Before firm conclusions can be drawn, interaction patterns found in one context need to be compared and contrasted with patterns found in others to see whether patterns are stable across contexts or applicable only to specific kinds of contexts such as scripted or structured routines like interviews. These investigations therefore also (c) identify possible moderating factors controlling the different patterns and (d) address the generalizability of our conclusions.

We begin with a detailed description of another dyadic interaction experiment, followed by a briefer summary of findings related to reciprocity and compensation taken from two experiments on deception and suspicion.

A SECOND MULTIMETHOD DYADIC INTERACTION EXPERIMENT

Whereas the investigation reported in Chapter 9 employed an interview task and setting, the current one employed discussions of social and moral dilemmas. An interview context is highly constrained by the prevalence of role–bound behavior. A discussion format, however, allows much more participant freedom; participants are constrained mainly by social norms and “laboratory expectations.”

Many of the theories discussed so far implicate a strong biological influence. Several theories also add psychological and sociological components in the form of affect, needs, expectations, and norms. What sets these theories apart from the preceding ones is the increasing emphasis on behavior as communication and on the functions and meanings that such behaviors entail. Functional and meaning–centered perspectives permeate many contemporary approaches to the study of interpersonal and nonverbal communication generally. We therefore review the more salient critical attributes of these perspectives before examining particular theories.

A FUNCTIONAL PERSPECTIVE

A functional approach to interaction adaptation acknowledges that interactions are multifunctional: Interactants bring multiple goals, objectives, needs, and wishes to be accomplished through communication. Among the “functions” that communication serves are message production and processing (also labeled information exchange or message transmission), identification and identity management, impression formation and management, relational communication and relationship management, emotion expression and management, conversation structuring and management, social influence and control, and personal resource management (see, e.g., Argyle, 1972; Burgoon, 1994; Burgoon & Saine, 1978; Dillard, Segrin, & Harden, 1989; Ekman & Friesen, 1969b; Goffman, 1959; Graham, Argyle, & Furnham, 1980; Patterson, 1983, 1990, 1991). Interactants must encode comprehensible messages (the message production function) and attend to, interpret, and evaluate the partner's messages (the message processing function) in the process of conducting a conversation.

Sometimes a great notion springs up at you. Many of the conclusions in this book arose that way. Other times, an idea sidles up, nudging you time and time again from the periphery. That seems to better fit how this book came about.

The first seed was probably planted at Michigan State University, where the climate, if inhospitable to the body, was highly conducive to interpersonal explorations. There, Judee taught a graduate seminar called Theories of Interpersonal Intimacy. The seed germinated at the University of Arizona, with the aid of the warm desert sun and further graduate seminars, titled Dyadic Interaction Processes and Theories of Intimacy Exchange. But it truly began to take root when the three of us, now working as a research team, realized that no single volume or article could offer us a comprehensive analysis to guide our own theorizing and research in this at once exciting and frustrating area.

And so the book grew, with much nurturing, weeding, and pruning from a host of student “gardeners” – people like Jerry Hale, Dave Buller, Milt Shatzer, Mary Diez, Rodney Reynolds, Lynn Aho, Beth Le Poire, Aileen Buslig, Renee Klingle, Patricia Rockwell, Michael Payne, Jamie Comstock, Cindy White, Pamela Koch, Eusebio Alvaro, Leu Strope, Kristen Burge, Mike Voloudakis, Megan Sheehan, and Carol Hensley – whose penetrating questions, summaries, editings, and critiques provided essential nutrients and to whom we owe a debt of gratitude.

In Chapters 2–5, we saw that the extensive research on interaction adaptation patterns is replete with seemingly conflicting findings and that few unequivocal generalizations have emerged so far. In Chapters 6 and 7 we saw that part of the difficulty has been the mix of conceptual and operational definitions that have been employed. In this and the next two chapters, we consider the role that statistical analysis plays in confounding the picture. As will be apparent, conclusions about interaction adaptation patterns are tied to the type of statistical analysis procedure used. The problem is not that researchers have been using inappropriate methods but rather, they have often failed to recognize how their choice of analysis influenced conclusions about adaptation. Thus, studies using one method might find “reciprocity” while studies using another might find “compensation.” Instead of this being a case of contradictory findings, it may merely reflect two different methods – for example, use of a between–subjects versus within–subjects design.

Our intent in reviewing statistical analysis alternatives, then, is not to identify which methods are inherently “better” but rather to distinguish different methods according to their objectives, the forms of measurement for which they are applicable, the restrictiveness of their assumptions, the kinds of estimates they supply, and so forth, so that results emanating from different analyses can be interpreted properly.

That conclusions are linked to analysis choices is not a new idea. Several past investigations have compared methods in an effort to determine empirically how they differ.