Electronic communication has become a staple of our world. As rising oil prices restrict travel budgets and a globalized hyper-competitive world demands ever more of its workforce’s schedules, the importance of electronic communication grows with every passing day. Marching to the beat of Moore’s Law, electronic communication has become cheaper, more convenient, and accessible. The world is once again flat (Friedman, 2005).
An explosion of expertise enabled by evaporating barriers to communication pushes knowledge workers to know ever more about less. Specialized researchers facing complex problems increasingly rely on teams ” because no one knows enough to be able to do the experiment by herself” (Hardwig, 1991). Collaboration with others has already become an “inescapable [feature] of much modern knowledge acquisition” (Hardwig, 1991). As industry is transformed by the knowledge economy, academic research foreshadows the collaboration necessary in industry. We must all collaborate.
Never before have we had so many electronic communication options for collaboration. Two decades ago, options available for real time electronic communication were limited to the telephone, an early form of email, or – if you lived in Europe – text messaging. Today, choosing a communication medium invokes the question: “should I text, message, post in a group, tweet, chat, Snapchat, email, audio call, or video call?” In this paper, I discuss a hidden consequence of choosing a communication medium: the potential to develop trust. I will show how choosing a medium to communicate with others also implies choosing a capacity to trust others.
Before we may understand the impact of a communication medium in developing trust between individuals, we must understand the conditions of trust between individuals. I will first discover these conditions by reviewing existing literature on the nature of trust. I will then investigate the properties the conditions, and finally show the impact of communication medium bandwidth on these properties.
The Conditions of Trust
While no standard definition of trust exists, scholars of psychology, sociology, and economics agree risk is essential to conceptualizing trust (Rosseau, 1998). As cited by Crease, Gelvin offers “to trust is to risk; the greater the risk the greater the trust” (2004). Others suggest a more precise characterization of risk as individuals risking vulnerability, and indeed, Rosseau found the most frequently cited definition of trust to be “willingness to be vulnerable” proposed by Mayer, Davis, Schoorman (1995).
Existing literature also identifies confident expectations as a condition for trust. Rosseau claims “confident expectations and a willingness to be vulnerable are critical components of all definitions of trust reflected in the articles” and “trust is a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions of behaviour of another”. However, my investigations into risk (below) show that confident expectations also inform our decision to engage in risk. Confident expectations is an optional component of risk, but a necessary condition of trust. Therefore, I have included my inquiry into confident expectations under the umbrella of risk.
While there is no universal definition of trust in existing literature, scholars agree the two necessary conditions are risk and confident expectations.
If tolerating risk is a pillar of trust, understanding how we evaluate risk will inform how we evaluate when to trust. According to Chiles and McMackin, “Risk is the perceived probability of loss, as interpreted by a decision maker” (Rosseau, 1998). I extend this definition of risk to include willingness to lose.
Unlike probability of loss, willingness to lose is a factor of risk known to the risk decision maker. If I am deciding whether to trust a coworker to look after my lunch without eating it, I consider the value I place on my lunch. While how and why we ascribe value is beyond the scope of this paper, value is often informed by the availability of alternatives. If less alternatives are available should I lose my lunch (there are no food vendors), or the available alternatives are not desirable (food vendors are expensive), I am less willing to lose my lunch and so am less willing to trust because the cost of loss is greater, regardless of probability. Willingness to lose is known to the decision maker, is influenced by what we value, and does not depend on other individuals.
Continuing our lunch experiment: if I believe I will lose my lunch should I leave it in the care of my colleague, then I am unlikely to trust my colleague with my lunch. The defining concept in this statement is belief in my colleague’s actions, and Rosseau suggests probability as a framework for believing which action my colleague will take.
I propose the variables in assessing this probability are: reputation for losing previous lunches and confidence to act as expected.
Perceived confidence to act as expected possesses a unique property not shared by other variables in the risk relation. Willingness to lose, reputation of loss, and possible outcomes may be determined by the risk decision maker without interacting with the individual(s) contending for trust. Perceived confidence to act as expected, however, requires the individual(s) contending for trust to somehow demonstrate their confidence to act as expected in order for the risk decision maker to be capable of perceiving it.
Investigating the nature of confidence is beyond the scope of this paper. For this paper, I treat confidence as a black box and focus on perception of confidence. I assume the individual(s) contending for trust must demonstrate confidence to the risk decision maker, and demonstration requires the individual(s) contending for trust to communicate with the risk decision maker. This communication must be facilitated by a medium, which prior to the 19th century has been either face-to-face or written. In the next section, I contribute that properties of the communication medium influence the risk decision maker’s perception of confidence..
Mediums of Confidence
I argue the perception of confidence is dependent upon the constraints of the communication medium used to convey confidence. Caris-Verhallen et al (1999) show that between 55% and 93% of all in-person communication is non-verbal. For the purposes of this paper, I assume all potential uses (expressing joy, sadness, insecurity, confidence, etc) of a communication medium will make full use of the medium. Therefore, I assume communicating confidence is distributed over both verbal and non-verbal communication when interacting with others in-person. For example, casual observation of verbal communication suggests tone and pitch (verbal) to be relevant to the risk decision maker’s perception of confidence as well as eye movement and facial expression (non-verbal).
Even wIthout considering the myriad of communications mediums available to us within the past two decades, the effects of communication medium on confidence can be observed by comparing telephone communication with in-person communication. The telephone does not permit the risk decision maker to perceive the eye movements and facial expressions of individuals contending for trust. While the relative weight of non-verbal communication compared with verbal communication in establishing trust is beyond the scope of this paper, eye movement and facial expression have a non-zero effect on perception of confidence. Therefore, I propose confidence is a dependent variable of communication medium.
Mediums may emphasize or deemphasizes indicators of confidence. Particularly, as a medium’s bandwidth approaches zero, the medium enforces a disproportionate emphasis on increasingly poor indicators of confidence; as a medium’s bandwidth decreases, it becomes easier to emit false confidence and falsely interpret confidence. I define a poor indicator of confidence by the ease of which the indicator is falsified, and bandwidth by its technical definition from the discipline of electrical engineering. Table 1.1 lists common communications mediums and the typical bandwidth required to support the medium.
10kb – 1+Mb
9 Kbps – 100Kbps
300Kbps – 1.5Mbps
Table 1.1 – Communication Medium Bandwidth
For example, a teletype machine permitted the most powerful leaders of America and the USSR to quiver behind their boardrooms at the possibility of global nuclear war while presenting each other with confident teletypes demanding the other stand down. It was, in part, a rambling discoherent message from the USSR during the Cuban Missile Crisis which permitted Robert Kennedy to perceive the confidence of his counterpart may not be what they believed. The arrival of this letter marked a turning point in the standoff, giving him the confidence to begin deescalation.
The teletype medium afforded more emphasis to the words and their properties as indicators of confidence in place of other indicators, such as non-verbal body language cues and immediacy of the response. While the USSR’s true confidence (which in fairness, was not dissimilar from the US) was ultimately discovered through analysis afforded by the teletype medium, the medium also afforded their confidence to be successfully falsified for most of the crisis. If the negotiation between the superpowers had been conducted via a high bandwidth communication medium such as Google Hangouts, harder to falsify indicators of confidence, such as immediacy of response, may have resulted in faster deescalation.
However, an accurate indicator of confidence in one medium may not translate as an accurate indicator of confidence in another medium. Consider the immediacy indicator which may have aided deescalation of the Cuban Missile Crisis in the Google Hangouts medium, and apply it to text messaging. At first, it may appear a long-delayed response to a sensitive question might imply the respondent lacks confidence. However, another property of this medium discredits immediacy as an indicator of confidence: text messaging is not a real-time medium. Unlike the Google Hangouts medium, text messaging does not assume the sole focus of the respondent. The respondent may be driving, cooking, or in an airplane, and any of these activities would be reasonable alternatives explanations for delayed response other than lack of confidence. However, the individual asking a sensitive question is not necessarily aware of this preoccupation, and so immediacy in the low-bandwidth text messaging medium also fits the definition of lower bandwidth mediums encouraging false interpretations of confidence.
I offer high bandwidth communications mediums afford higher quality indicators of confidence, which then may be used in conjunction with reputation of loss, possible outcomes and willingness to lose to evaluate risk. I will now show how we come to identify certain behaviors as indicators of confidence.
A Library of Correlations
Up to this point I have assumed a risk decision maker evaluating the probability of loss using only information pertaining to either herself or the individual(s) contending for trust. In practice, we are not so rational. We do not need to be. There is an easier method of evaluating risk than deeply analyzing the factors influencing the specific case at hand.
I argue we create a library of successful correlations – correlation of successful outcomes to the behavior that preceded the successful outcome, and use these correlations to inform future risk decisions. This is consistent with Rosseau’s claim that “risk taking buttresses a sense of trust when the expected behavior materializes.” These successful correlations are the benchmark for evaluating of what we call confidence; these successful correlations are why we can say we perceive confidence. When we say we are confident in an outcome, we are actually perceiving behavior previously associated with successful outcomes, and on this basis claiming to expect more successful outcomes if we perceive similar behavior.
However, because correlations depend on properties of the communication medium, the library of successful correlations may only be used to accurately inform future risk decisions of the same communication medium type. As a new communication medium’s properties deviate from the properties of the communication medium responsible for generating our personal library of successful outcome correlations, our ability to predict probability of loss decreases as well.
Within the past twenty years, choosing a communication medium has become a decision facing most of the developed and developing world. As Moore’s law continues to offer us higher bandwidth communication mediums at an accelerating pace, evaluating communication mediums will become a staple of modern existence. Different communication mediums will permit relationships to develop to different strengths. In this paper, I have analyzed the role of the communication medium in establishing trust.
Specifically, in this paper I have shown not all communications mediums permit the same quantity or quality of confidence indicators. These confidence indicators form an vital component of the risk relationship for evaluating the probability of loss. Since probability of loss is one of two conditions necessary for trust, I have shown that by choosing a communication medium, we also choose our potential to trust others.
I have also shown that we create libraries correlating behavior to successful outcomes, and use these correlations to create and add credibility to confidence indicators. I have shown these libraries to be medium-specific.
I believe the primary consequence is simple: choosing a communication medium should be thought of as an extension of the investment in the relationship.
Caris-Verhallen, Wilma M.c.m., Ada Kerkstra, and Jozien M. Bensing. “Non-verbal Behaviour in Nurse-elderly Patient Communication.” Journal of Advanced Nursing 29.4 (1999): 808-18. Web.
Friedman, Thomas L. The World Is Flat: A Brief History of the Twenty-first Century. New York: Farrar, Straus and Giroux, 2005. Print.
Hardwig, John. “The Role of Trust in Knowledge.” THE JOURNAL OF PHILOSOPHY 88.12 (1991): 693-708. Web.
Rousseau, Denise M., Sim B. Sitkin, Ronald S. Burt, and Colin Camerer. “Not So Different After All: A Cross-Discipline View Of Trust.” Academy of Management Review 23.3 (1998): 393-404. Web.