Behavioral Influences in the Present Tense

The Expressive Versus Instrumental Distinction

We do not wish to argue the future versus the past because we believe there is another theoretical orphan in treatments about human behavior that has been more ignored. In between the future and the past, this orphan is the classic neglected middle child.

This orphan is the present. More specifically, human behavior may not only be inspired by what the individual believes will happen in the future. It can also be motivated by goals and desires that people wish to achieve here and now. Sociology, and sometimes economics, discusses a distinction that best captures a profitable way to talk about present and future influences on human behavior. This distinction is the one between expressive and instrumental motivation. Actions with expressive motivations have more immediate goals that are fulfilled by executing that action. Such behaviors are ends in themselves. For example, eating ice cream for dessert instantaneously brings a pleasure that a person might seek to gain in the present. In contrast, instrumental motivations are more future based; they are a means to some other end. These motives arise when people execute behaviors because of some future outcome they wish to produce. For example, people choose to invest in the stock market not because owning stock is an end in itself, but to have enough money to retire or to buy their next house. ¹ In psychology, the common terms coming closest to this expressive/instrumental distinction centers on intrinsic versus extrinsic motivation. Those who are intrinsically motivated by their job, for example, derive pleasure directly from the activities associated with the job itself. When extrinsically motivated, people perform their jobs in order to gain some other benefit like social status or a paycheck at the end of the month (Bénabou & Tirole, 2003; Ryan & Deci, 2000).

The expressive versus instrumental distinction has a long history in social science. Behaviors often differ in terms of which types of concerns, expressive or instrumental, motivate them. Some crimes, for example, are purely instrumental in nature. Professional thieves and drug dealers conduct their careers in order to produce an adequate, if illegal, income. They do not engage in these activities because they find them immediately pleasurable. Other crimes are more expressive in that their rewards are more immediate, such as sex offenses and taking drugs. The execution of these behaviors immediately fulfills the goal that individual is concerned with. Such motivations matter, in that the expressive versus instrumental nature of a crime often determines if it is deterrable (Chambliss, 1966).

The expressive/instrumental distinction also serves to explain behaviors that sometimes are a puzzle. It has often been observed that voting in a U.S. Presidential election is difficult to understand in instrumental terms, in that an individual’s vote rarely changes the outcome of the U.S. Presidential race. It would seem rational, instead, to stay home and let other people do the work. Yet millions drive to the polling places to cast their ballot—why do they do so? Theorists suggest that they do so presumably to achieve expressive ends, such as fulfilling one’s civic duty, qualifying for a sense of pride, to express their political values, or to share in a collective experience with others (Brennan & Hamlin, 1998).

Relevance to Interpersonal Trust

The expressive–instrumental distinction matters because determining whether the future matters often means having to disentangle future outcomes from ones fulfilled in the present—that is, distinguishing the instrumental from the expressive. For example, Seligman and colleagues highlight expectation as an important driver of human behavior: People choose based on how they expect a future will unfold. We have found, however, that expectation fails to drive behavior in at least one realm where most theorists think it does.

That realm is interpersonal trust. It is clear that trust is one of the most basic aspects of human life that makes it less nasty, brutish, and short. It is difficult to imagine any social unit remaining stable and productive—whether it is a marriage, a friendship, an organization, or even a democracy—without deep and abiding trust among the individuals in it. Most theorists in economics (McKnight & Chervany, 2001; Rousseau, Sitkin, Burt, & Camerer, 1998) and psychology (Deutsch, 1958; Kramer & Carnevale, 2001; Pruitt & Rubin, 1986; Rotter, 1967; Simpson, 2007) construe this most social of behaviors to be a function of expectation. People will trust others if they believe their trust will be reciprocated and will withhold trust if they think it would be violated. That is, like Seligman and colleagues, they construe trust as a matter of prospection.

However, across several experiments, we have found that such a basic behavior as trust fails to follow the logic of expectation. In these experiments, we presented participants with a one-shot, anonymous choice known in the literature as the trust game (Berg, Dickhaut, & McCabe, 1995). We gave them $5 and asked if they wished to keep it or give it to an anonymous stranger within the experiment. We explained that if they kept the $5, the game would be over. However, if they decided to give the $5 to the stranger, we would inflate the dollar amount involved to $20 and ask the stranger if he or she wished to keep the entire $20 or give $10 back to the participant who originally gave up the $5. In this way, we created a situation that fits an economic definition of trust: making oneself vulnerable to another person with the prospect of profiting from it (Rousseau et al., 1998). Further, a majority of participants freely construed this situation as one of “trust” and “faith in others,” among other descriptors (Dunning, Fetchenhauer, & Schlösser, 2012).

Participants proved unoptimistic about their peers in this game. They believed, on average, that the stranger would keep the money, thinking that only 45% or so would return the $10. Regardless of this, a majority of participants in study after study still gave their $5 away (Fetchenhauer & Dunning, 2009). They did so even though a vast majority would refuse to bet the $5 on a lottery wheel or a coin flip with the same odds and payoffs (Fetchenhauer & Dunning, 2012; Schlösser, Fetchenhauer, & Dunning, 2012). Further, although increasing the odds of winning a lottery prompted participants to bet more, increasing the odds for the same participants that someone would return money in a trust game did not impel them to trust more (Fetchenhauer & Dunning, 2012).

This and other behavior suggests to us that participants were not outcome- or future-oriented in their decisions to trust. They decided to trust even though they expected their trust to be abused rather than rewarded. Instead, they seemed to be pursuing some expressive goal that is achieved simply by choosing to trust. To date, we and (more importantly) our participants do not necessarily seem to know specifically what that expressive goal was (Dunning & Fetchenhauer, 2010a, 2010b; Dunning et al., 2012). It might have been establishing a positive self-image, demonstrating respect for the other individual, or fulfilling some social norm or dictate.

We believe the decisions to trust in our experiments are expressive in nature because we have done something similar to what Seligman and colleagues suggest. We have measured participants’ emotional reactions to the actions they may choose (giving $5 vs. keeping it) as well as the emotions they anticipate they would feel based on the outcomes of those actions. For example, we asked participants how they anticipated they would feel if they gave the $5 and received nothing back.

According to many theories in economics—and the logic laid out by Seligman and colleagues—the decisions people reach should be guided, at least in part, by the emotions people project they will feel when future outcomes arrive. Namely, people will shape their decisions in order to gain pleasure and to avoid any disappointment and regret that outcomes might bring (Bell, 1982; Loewenstein, Weber, Hsee, & Welch, 2001; Loomes & Sugden, 1986; Mellers, Ritov, & Schwartz, 1999). We find, at best, that such anticipated emotions predict trust decisions to only a minor degree.

Instead, the emotions people attach to the decisions themselves prove to be strong predictors of what they choose to do. People who feel more positive about giving $5 are more likely to do so than those who keep it, and these emotions—attached to immediate actions—explained twice as much variance in trust decisions than did emotions attached to possible outcomes (Schlösser et al., 2012). They also explained more variance than did people’s expectations that their trust would be reciprocated. Thus, in terms of emotion, it was not prospections about future feelings that mattered. Rather, it was the emotions attached to the immediate choice faced in the present. More recent work has specified that it is the guilt and anxiety attached to keeping the money that may prompt people to give it to the other person more than economic considerations would suggest (Schlösser et al., 2012).

Decisions in General

We wonder in general how much decisions that look like they are about the future are actually more about the present. We ask this because we have also explored more purely economic decisions and have uncovered the same expressive pattern. In a series of studies, we asked participants to consider purely economic decisions such as whether they would gamble $5 for a possible $10 based on a flip of a coin, or on the role of a die, or on whether a red ball would be pulled out of an urn. In each case, participants reported how they would feel about all possible outcomes. For example, when faced with a coin flip, participants were asked about all four outcomes that come from a 2 (flipping the coin: yes or no) × 2 (outcome: coin flip won or lost) set of possibilities. They also reported how they felt directly about the actions of flipping and not flipping the coin.

Again, in all these studies, it was emotions attached to the action—for example, flipping or not—that predicted participants’ decisions, not the emotions they projected they would feel after arriving at possible future outcomes (Schlösser, Dunning, & Fetchenhauer, 2013). That is, if we take participants’ emotions as valuations of how they felt about actions and outcomes, then it was their valuation of the action occurring in the present that mattered much more than any evaluations based on prospection about future outcomes. As such, it looks like people are focused on the present much more than on the future when they consider what actions to take in purely economic decisions.

Concluding Remarks

In sum, we welcome Seligman and colleagues’ comparison of the strength of past influences versus future prospects in shaping human behavior. However, we view the comparison as incomplete unless one also takes into account the present. In our own research, we have been surprised how often decisions that look prospective in nature turn out, instead, to be more about present concerns. It is not instrumental issues that matter as much as expressive ones, although we have a long way to go to map out the complete geography of expressive concerns that may guide economic and social decision making.

We consider this to be a fruitful area of research, and Seligman and colleagues have provided a broader and potentially richer context for that work. In turn, the question that Seligman and colleagues have laid out may be richer than they themselves conceived. When considering all influences on human behavior, one might have to consider present influences, as well as past and future influences.