Ecological Economics 114 (2015) 22–32 Contents lists available at ScienceDirect Ecological Economics journal homepage: www.elsevier.com/locate/ecolecon Surveys From rational actor to efficient complexity manager: Exorcising the ghost of Homo economicus with a unified synthesis of cognition research Jordan Levine ⁎, Kai M.A. Chan, Terre Satterfield Institute for Resources, Environment and Sustainability, University of British Columbia; 2202 Main Mall, Vancouver, BC V6T 1Z4, Canada a r t i c l e i n f o Article history: Received 5 September 2014 Received in revised form 3 March 2015 Accepted 10 March 2015 Available online xxxx Keywords: Unified model Heuristics Biases Analogy Mental models Homo efficens a b s t r a c t It is now commonplace to note that economics' canonical model of humans as rational, self-interested utilitymaximizers (Homo economicus) is both descriptively misleading, and often insufficiently predictive. However, certain outdated assumptions tied to Homo economicus persist, often influencing discourse and research design even in sustainability-oriented fields. We argue this ‘ghost’ of Homo economicus endures because the diversity of findings that confound the canonical model has surfaced across multiple behavioral and cognitive sciences, each with its own terminology and focus area. As such, a unified, accessible synthesis of this new information has yet to emerge. In this paper we review recent insights from across the behavioral and cognitive sciences, and propose an ‘efficient complexity manager’ (ECM) model (Homo efficens) as the best synthesizing option. The crux of this model is that our species works within biological limits to efficiently filter massive environmental complexity. This is achieved largely through analogical—or ‘case-based’—reasoning. We explain this synthesized model using a series of accessible metaphors. Finally, we speculate on how this model may enrich future sustainable development research insofar as it points to fruitful units of analysis, can stimulate methodological innovation, and provide a more explicit theoretical foundation for the field. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Any effort that aims to account for some aspect of collective or individual human behavior must ultimately rely on a theory of human action. The response of twentieth century economic thought to this challenge has been a working model of humans as ‘rational utility maximizers’ (Monroe and Maher, 1995; Sen, 1997; Siebenhuner, 2000; Simon, 1982). This ‘canonical’ model—known popularly as Homo economicus or ‘economic man’ [sic]—assumes, by implication, that people have pre-determined, consistent desires, and that human behavior is therefore the product of a calculated, utilitarian fulfillment of as many of those desires as possible, ostensibly as measured in additive, fungible bits of utility, or ‘utiles.’ Furthermore, in the most radically economistic, but still widely accepted, interpretation of this model, Homo economicus' self-interest is assumed to be primarily, if not entirely, material, concerned chiefly with increasing one's ability to consume, by amassing wealth (Frank, 1987; Sen, 1988). It is now commonplace to note that this canonical model has been resoundingly debunked by a wide array of contemporary findings (Gintis, 2000; Fehr and Gachter, 2000; Henrich et al., 2001; Jager et al., 2000; Kahneman et al., 1982; Roth et al., 1991; Siebenhuner, 2000; Thaler, 2000; Van den Burgh et al., 2000). Yet, as we will outline in ⁎ Corresponding author. E-mail addresses: jlevine@interchange.ubc.ca (J. Levine), kaichan@ires.ubc.ca (K.M.A. Chan), terre.satterfield@ires.ubc.ca (T. Satterfield). http://dx.doi.org/10.1016/j.ecolecon.2015.03.010 0921-8009/© 2015 Elsevier B.V. All rights reserved. more detail below, the figurative ‘ghost’ of Homo economicus appears to stubbornly persist, mainly in the form of lingering inaccurate assumptions that still structure academic and policy discourse in often subtle, undetected ways. We argue this elusive but enduring ‘ghost’ skews researchers' and policymakers' thinking about humans' role in complex social–ecological systems. As such, it is liable to hinder both (a) academic progress on questions of sustainable development, and (b) the pursuit of sustainable development as a normative goal (i.e., as an aim driven by ethical or political concerns). Given the sheer volume of credible evidence debunking Homo economicus, it may at first glance seem perplexing that its legacy, or ‘ghost,’ endures. Yet, in addition to the pivotal importance of the model's core assumptions to the edifice of contemporary welfare economics theory (see Feldman, 2008; Gowdy, 2004), we argue that the unwelcome staying power of these assumptions is further perpetuated by the diversity of partial replacements to Homo economicus proposed by various disciplines, using different lexicons, often with a confusing mix of normative, versus descriptive, flavor (Becker, 2006; Bina and Vaz, 2011; Faber et al., 2002; Ingebrigsten and Jakobsen, 2009; Jager et al., 2000; Siebenhuner, 2000; Van den Burgh et al., 2000). The result is that, without an easily understood, easily remembered and—importantly—unified replacement for Homo economicus as a primary foundational model, researchers and policymakers alike are liable to unwittingly default to earlier debunked theories of purely rational, optimizing, self-interested action. In addition to the political momentum behind associated orthodox theories (Gowdy, 2004), this tendency likely persists because, even J. Levine et al. / Ecological Economics 114 (2015) 22–32 on a purely cognitive level, the canonical set of assumptions is simply more familiar, less complex or merely assumed to be ‘good enough’ for a given job.1 Our aim in this paper is to move toward fully exorcising the ‘ghost’ of Homo economicus from sustainable development research by presenting a synthesis of insights from recent cognitive and behavioral science in consistent, intelligible language. As we argue in more depth below, the conclusion that emerges from such a synthesis is that, rather than assume we are a species of primarily rational, utilitarian optimizers (albeit subject to a number of problematic cognitive ‘blind-spots’; e.g., Kahneman et al., 1982), it is more empirically accurate to model Homo sapiens primarily as cognitively efficient managers of massive complexity. Furthermore, as a mechanism for achieving cognitive efficiency, evidence suggests that people are primarily driven not by a clear-headed, time-and-energy intensive optimization of abstract utiles, but rather by rapid, less cerebrally-taxing, emotionally- and viscerallyfelt responses to a narrow band of environmental complexity (Buck, 1985; Camerer et al., 2005; Murphy and Zajonc, 1993). In fact, emotions appear so inescapably relevant to decision-making that they cannot be neatly separated from higher-level cognition (Haidt, 2001; Pessoa, 2008). The basic drive to achieve desirable emotional states (or ‘affect’) may sometimes be well served or even best served by a clear-headed optimization of an imagined metric of value, such as utiles (i.e., idealized rational utilitarianism). Crucially, however, the former is in no way synonymous with the latter, and it is a mistake to confuse one (pursuit of desirable affect) for the other (abstract utility maximization). In light of this distinction, the remainder of our paper is structured as follows. First, we present some examples of how the ‘ghost’ of Homo economicus persists in contemporary research and practice (Section 2). Next, in Section 3, we contrast the assumptions of the canonical model with a series of descriptive (rather than idealized or normative) principles we have identified over the course of reviewing a range of recent empirical findings on human behavior and cognition. Together, these synthesized principles constitute what we call the efficient complexity manager (ECM) model of the human actor. In Section 4, we present the ECM model in a more formalized fashion, outlining in representational terms the implications of this work for understanding how people proceed from perception, to cognition, to action. The ECM model is, of course, a necessarily drastic simplification of human cognitive processing. As such, we do not present it as a conclusive replacement for other models. Rather, our goal is an accessible synthesis that usefully reflects the points of convergence among the most recent, relevant findings on human cognition from multiple disciplines (primarily within the broader field of behavioral and cognitive science). We argue that this, in turn, can lead to a more empirically-responsible foundation for work on the behavioral aspects of sustainable development challenges. We conclude with some suggestions for future research. 2. Spotting the Ghost Examples of the legacy or ‘ghost’ of Homo economicus are evident in the continued pervasiveness of several trends, both in social science generally and, to some degree, in sustainable development work in particular. At least three trends in this field strike us as particularly telling. These include: • A steady increase in contingent valuation and revealed preference studies, many of which assume preferences to be stable (Adamowicz, 2004; Ariely et al., 2003; Kahneman and Knetch, 1992). Because these studies produce monetary values as outputs, which are therefore commensurate with the language of the global economic system, 1 Ironically, this constitutes a prime example of ‘satisficing’ — an immediate time-andenergy saving, but decidedly sub-optimal, problem solving strategy (Simon, 1982). 23 they hold particular instrumental appeal for conservation and sustainability scientists. However, the underlying premise of many such valuation studies is an often-mistaken assumption that the preferences being ‘measured’ are stable (Kahneman and Knetch, 1992). Empirical findings have firmly established that many preferences are not stable, but rather constructed — i.e., largely the product of contextual factors including framing, scale, scope, reference points, visual cues, information load, the cognitive complexity of the valuation task, and time pressure (Bateman and Mawby, 2004; Gregory et al., 1993; Lichtenstein and Slovic, 2006; Payne et al., 1992). Nonetheless, framing and scoping problems remain under-scrutinized in many published papers (Desvousges et al., 2012; Gomez-Baggathun et al., 2010). Some argue the resulting focus is on generating data that provide monetary valuations, however unstable, rather than understanding realistically presented and empirically defensible choice behavior and tradeoffs (Adamowicz, 2004). ‘Choice experiments’ can help resolve this quandary to an extent, but these methods are still subject to the critique that they do not in fact capture actual, stable, behavior patterns observed across a range of in situ contexts. Rather, they extrapolate preferences from answers obtained in highly controlled situations that do not necessarily reflect or even approximate the diverse range of cognitive and social environments in which most behavior actually occurs (Hudson et al., 2012). This exposes an ongoing assumption in the field that people's decision-making patterns are far more independent from changeable in situ context than contemporary science (Hart et al., 2010) suggests. • A dramatic rise in payment for ecosystem service (PES) schemes, and other market-based mechanisms, stemming from an implicit assumption that monetary incentives are primary motivators and relatively unproblematic (Engel et al., 2008; Kissinger et al., 2013).2 The exuberance with which such schemes have been embraced and promulgated, despite a relative dearth of supporting longitudinal evidence, suggests an underlying assumption that specifically monetary rewards are the best way to incentivize people. This assumption is consistent with a relatively myopic, but popular, interpretation of self-interest, whereby even when supposedly broadly conceived it is conflated with the single-minded pursuit of specifically monetary personal gain (Fong, 2001; Ingebrigsten and Jakobsen, 2009; Sen, 1988). Lamentably this is notwithstanding early advancements to the contrary in rational actor theory itself (Arrow, 1963). This sort of confusion is perpetuated by the continued conflation of normative (i.e., prescriptive or hopeful) ideals, with more empirical-minded, descriptive accounts of actual human behavior (e.g., Becker, 2006; Faber et al., 2002; Ingebrigsten and Jakobsen, 2009; Siebenhuner, 2000). Mixing normative and descriptive theories of human action has little empirical value, but is rather a phenomenon rooted firmly in early modern political philosophy, which tended to combine the two seamlessly and uncritically (Bowles and Gintis, 1993; Thaler, 2000). Unbridled enthusiasm for PES schemes perpetuates this problematic, self-reinforcing conflation, preventing more empirically-informed descriptive models of human motivation from taking hold. • A persistent gap between research output and policy outcomes based in part on prevalent assumptions that people will change their behavior when provided with new information (Ruth, 2006; Shi, 2004; Turner 2 While such schemes can, theoretically, create win-win scenarios for biological conservation and poverty reduction, they can also have a slew of less savory effects, both predictable and unpredictable (Engel et al., 2008; Kissinger et al., 2013). PES schemes can reinforce a commodification of ecosystems, which some argue is a key driver of ecological degradation in and of itself (Gomez-Baggathun et al., 2010) putting long-term sustainability at risk (Siebenhuner, 2000). Others note that primarily financial interventions like PES schemes can lead to both social and macroeconomic dislocations (e.g., the inflation of food prices, or major disruptions in social relations), which can ironically leave poor people worse off in the long run (Bute et al., 2008; Fisher et al., 2010; Frey, 1994). Perhaps most shocking, is that similar problems surfaced in related literature decades earlier, as far back as the 1950s, but have since been overshadowed by newfound contemporary interest In integrating ecosystem function into the increasingly dominant neoclassical paradigm (Baveye et al., 2013). 24 J. Levine et al. / Ecological Economics 114 (2015) 22–32 Fig. 1. Abstracted schematic of the complexity-reduction process. This figure is a synthesized representation, in abstract terms, of the process whereby we use our bodies and minds to filter the vast barrage of new data with which we are perpetually confronted. This iterative process helps us make reasonable inferences that facilitate useful action. (Synthesized from Bar, 2009; Giere and Moffatt, 2003; Lakoff and Johnson, 1999; Fauconnier and Turner, 2002; Slingerland, 2008; Peterson and Flanders, 2002). et al., 1998; Shove, 2010). Although it is certainly not the only culprit, a major driver of the continued gap between research findings and sustainable development policy seems to be an inherited belief that, to help people act in their own or society's best interests, all one must do is provide them with the relevant facts (Bulkeley and Mol, 2003; Macnaghten and Jacobs, 1997). This belief follows logically from the canonical model, which implies people attend equally to all kinds of information and, unaffected by context or emotion, make perfectly consistent, calculated decisions on the basis of said information (Shah and Oppenheimer, 2008). This ‘information-deficit model’ of human behavior, however, has been soundly refuted by empirical data (e.g., Goldstein et al., 2008). Rather, people tend to grasp and attend to some kinds of information (visceral and emotional, or ‘affect laden’) more readily than others (abstract and cerebral) (Borgida and Nisbett, 1977; Miloslavljevic et al., 2012). In addition to content, the source of information, its form, and the way it is framed, are hugely important factors in how or even whether such information is absorbed (Gilovitch et al., 2002; Tversky and Kahneman, 1992). Social norms, along with ingrained habits, can further affect the impact of new information on decision-making and behavior (Goldstein et al., 2008; Klockner et al., 2003). In light of these empirical insights, it seems closing the gap between sustainable development research and policy will require forms of science communication that are more thoroughly disabused of inaccurate assumptions about how people attend to, accept, or reject, new information. 3. A Set of Empirically-derived Principles for a Descriptive Model of the Human Actor Here we focus on a number of convergent trends in contemporary empirical accounts of human cognition and motivation. We present these trends as a set of descriptive principles constituting a synthesis that contrasts with the core assumptions of Homo economicus. Taken as a whole, these principles comprise what we call an ‘efficient complexity manager’ model (ECM) of the human actor or, alternatively, Homo efficens. (See Fig. 3 for a combined visual representation of these principles. See Section 4 for an abstracted formalization.) 3.1. Cognition is Triage: We Are Efficient Within Our Computational, Energetic and Temporal Limits While Homo economicus revolves around the notion of humans' rational maximization of self-interest, contemporary studies of cognition and decision-making3 are converging instead on the importance of humans' cognitive efficiency. From psychology (Gilovitch et al., 2002; Shah and Oppenheimer, 2008; Gigerenzer and Goldstein, 1996; Peterson and Flanders, 2002) to the study of artificial intelligence (Wray et al., 2007), contemporary research suggests a primary task of our mental faculties is to usefully parse the vast barrage of potentially relevant information—i.e., the profound complexity—that continuously confronts us. Importantly, studies of human cognition indicate that people have limited information-gathering and computational power (due the size and nature of our brains and bodies) (Peterson and Flanders, 2002), limited energy (Corcoran and Mussweiler, 2010; Macrae et al., 1994) and, most often, limited time (Weenig and Maarleveld, 2002). To survive and flourish, we thus appear to have evolved intuitive and largely unconscious (Greene and Haidt, 2002) 3 By ‘cognition’ we mean how the human body and mind integrate and make sense of information. It is both a precondition for, and subsumes, the narrower concept of ‘decision making,’ which has connotations of conscious reasoning aimed at a particular actionable outcome. Homo economicus is largely a model of the latter, which takes the former for granted, essentially ignoring the important constraints imposed by the task of selecting and processing relevant information (Doucouliagos, 1994). Here, we argue that an empirically sound understanding of decision-making cannot be usefully separated from broader theories of cognition. Hence, for the purposes of this paper, we tend to conflate the two when presenting our revised set of descriptive principles. J. Levine et al. / Ecological Economics 114 (2015) 22–32 25 Fig. 2. Typology of analogical representations. This figure shows a range of various types of analogical representations (metaphorical ‘maps’) of previous experience that we draw upon when attempting to make sense of how to act in the face of new data. Different disciplines tend to focus on different varieties of representation, and often use different nomenclature. For clarity's sake, here we have ordered some key varieties on two axes: physical structure, and temporal structure. (Synthesized from Slade, 1991; Kolodner, 1992; D'Andrade, 1995; Eubanks, 1999; Lakoff and Johnson, 1999; Lakoff, 2007; Atran and Medin, 2010). mechanisms and strategies for making good-enough decisions (particularly in hunter gatherer-type contexts) given these limited resources.4 In this sense, our minds are not unlike doctors and nurses who engage in ‘triage’: a technique used in emergency situations with multiple casualties, whereby medical workers have to rapidly decide who gets treated, how, and in what priority, given limited amounts of time, supplies and personnel. The analogy is apt, in that we are continually inundated with an immense volume of potentially relevant information, and have to use our combined but limited powers of accumulated experience, instinct, and various degrees of conscious deliberation to decide how thoroughly (if at all) new data will be attended to, and in what order. Throughout this process, we must all the while economize for our limited reserves of time, energy and computational power. In other words, our minds are predisposed to sacrifice detail for efficiency. The above vision lies in stark contrast to Homo economicus, who is assumed to make the best possible, fully-conscious, utility-maximizing decisions irrespective of the actual computational, energetic and temporal limits imposed by our physical makeup and surrounding context (Doucouliagos, 1994). In the language of analogy, the canonical model resembles more of an ‘omnipotent, omniscient supercomputer’ than it does the humble but effective ‘triage nurse’ implied by cognitive research. Shifting from an unwitting belief in the former, to an explicit 4 In other words, cognition has and continues to evolve as a system for efficiently sifting through complexity to identify and act upon only what is most relevant for an organism's immediate goals. This system is highly functional, but sometimes succumbs to the inherent tension between efficiency and complexity: it can, occasionally, oversimplify a situation, leading to maladaptive behavior. This is what Kahneman and Frederick (2002) call ‘attribute substitution,’ (see Section 3.7). integration of the latter, seems a pre-requisite for establishing a more empirically-responsible model of the human actor.5 3.2. Different Goals, Different Glasses: What We Perceive is a Function of Our Goals An explicit acknowledgment of our species' computational, energetic and temporal limits begs the question: how, exactly, is it that we humans parse complexity efficiently when making decisions? The canonical model assumes that we hold all relevant information in our minds simultaneously and apply a set of consistent decision rules until we arrive at an objectively optimal outcome (Shah and Oppenheimer, 2008). Yet, from an empirically-informed perspective, how one parses complexity, which includes first distinguishing what is ‘relevant’ from what is not, is largely dictated by one's in situ goals (Peterson and Flanders, 2002), which are themselves shaped by a combination of one's biological needs, one's beliefs and one's immediate context (Eclles and Wigfield, 2002; Hickey and McCaslin, 2001). Furthermore, psychology affirms that the motivations underlying our 5 The central notion of the ‘triage nurse’ metaphor is efficiency under constraints. Since efficiency (defined broadly as a ratio of work output to energy inputs) is also “the bedrock of policy, planning and business approaches to sustainable development” (Jollands, 2006), applying such a lens to human decision making has the additional benefit of aiding disciplinary and scalar consilience within and across ecological economics and other behavioral sciences. A crucial point of the model we present here, however, is that humans tend toward efficiency in a purely cognitive, biological sense (i.e., by economizing their limited time and energy to infer what are usually ‘good enough,’ simplified models of their complex environments). The result is a highly myopic focus on a narrow bandwidth of what is deemed to be potentially relevant information. This is entirely different from the lowerresolution (and hence often erroneous) canonical assumption that humans are perfectly effective optimizers, with unfettered access to all information, equally. 26 J. Levine et al. / Ecological Economics 114 (2015) 22–32 Fig. 3. Metaphorical representation of ‘Homo efficens’, the efficient complexity manager model of the human actor. This representation combines (a) Haidt's (2006) ‘elephant and rider’ metaphor (see Section 3.3) with (b) the other principles described in Section 3. The powerful ‘elephant’ is analogous to our emotions, which lie at the root of our motivational system. The less powerful, but more consciously deliberate ‘rider’ is analogous to the cognitive process that is required in order for us to make useful sense of the vast quantities of data to which we are constantly exposed. In this, the efficient complexity manager (ECM) model, the ‘rider’ is likened to a ‘triage nurse’ (see Section 3.1) in that (s)he is always working under a set of constraints (limited time, energy, and processing power) and must therefore make continual ‘fast-and-frugal’ (i.e., efficient) decisions about what to attend to, and how, so as to guide action. One's particular in situ goals, which are dictated largely by the emotion-based motivation system, determine which ‘glasses’ the rider must wear (see Section 3.2): different goals require attention to different kinds of information, as our limited minds are unable to attend to everything at once (Peterson and Flanders, 2002). To further increase efficiency, the rider always refers to a ‘map’ of some sort to help guide action (see Section 3.7). Figurative ‘maps’ are in fact analogical representations of prior experience (see Fig. 2). Referring to ‘maps’ saves time and energy by precluding the need to reason entirely from scratch. Finally, the other elephant-rider pairs in the distance represent the notion that each of us is in fact part of a much wider network, or ‘herd’, of cognitive processing (see Section 3.5): by interacting, taking cues, and copying from those around us (both past and present), we are able to dramatically reduce each of our own individual cognitive burdens, and hence maximize the efficiency with which we determine appropriate actions. This last phenomenon is known as ‘distributed cognition.’ (Haidt, 2006; Peterson and Flanders, 2002; Peterson, 1999; Giere and Moffatt, 2003; Lakoff and Johnson, 1999; Kahneman, 2003; Shah and Oppenheimer, 2008). goals are experienced emotionally, mediated by affect (Buck, 1985; Camerer et al., 2005; Forgas, 2000; Lowenstein et al., 2001; Murphy and Zajonc, 1993), and that emotions cannot be neatly separated from higher-level cognition (Haidt, 2001; Pessoa, 2008). In this sense, it can be said that even prior to any decision is made, our chosen methods for merely perceiving a situation itself (our metaphorical cognitive ‘glasses,’ of which we have many varieties) depend heavily on our moment-to-moment goals, which are, in turn, less conscious than they are visceral, liable to change readily in reference to context. In other words, our particular momentary goals largely dictate what narrow band of ambient information we attend to, while the rest gets filtered out (see Section 3.4). 3.3. The Emotional Elephant and Rational Rider: Emotions bound Reason, not the Inverse Emotions appear to motivate, contextualize, enable, and arguably precede, more conscious reasoning processes (Damasio, 1994; Lavine, 1998; Zajonc, 2000) — not the other way around (Haidt, 2001). In other words, positive emotional experiences such as receiving peer validation, enjoying short-term gains, or avoiding the experience of shame, ostracism, paralyzing complexity, or loss, are all stronger fundamental drivers of behavior than the more cerebral drive to, e.g., develop a flawless casual understanding of the world around us, or always remain perfectly consistent in speech and action (except insofar as one is able to associate said goals with an expectation of powerful emotional reward). Peterson and Flanders (2002) argue that this is particularly relevant to why people hold tenaciously to rigid theories and belief systems in the face of contradictory or threatening evidence, an issue directly relevant for sustainability science (e.g., the denial of climate change or refusal to accept the reality of non-linear thresholds with respect to the effects of resource depletion). Social psychologist Haidt (2006) introduced a useful metaphor for this tightly linked dynamic between the behemoth of emotional motivation, and the incisive but relatively underpowered faculty of conscious abstract thought: the emotional ‘elephant’ and its rational ‘rider.’ An important point to note in this otherwise apt metaphor, is that the rider has evolved to help steer the elephant, not the other way around. As Haidt (2006) himself explains: [The] rider [is] placed on the elephant's back to help the elephant make better choices. The rider can see farther into the future, and the rider can learn valuable information by talking to other riders or by reading maps, but the rider cannot order the elephant around against its will (17). J. Levine et al. / Ecological Economics 114 (2015) 22–32 In fact, recent evidence from the brain sciences suggests that rational decision-making cannot properly occur in the absence of a seamlessly integrated emotional apparatus: the two are necessarily interlinked in a physiological sense, as well as a metaphorical one (Glimcher et al., 2005). The contrast between the model of cognitive efficiency that begins to emerge from these cumulative analogies, and the notion of a coolly rational Homo economicus, is stark. Whereas the latter implies a near robotic (or ‘Spock’-like6) absence of emotion, paired with the unlimited freedom of an omniscient, unconstrained supercomputer, the former—to begin melding our metaphors—suggests a resourceconstrained metaphorical ‘triage nurse,’ with an array of discrete, goal-specific ‘glasses,’ serving as humble navigator atop a powerful elephant-boss of emotion. We discuss some implications of this more nuanced model for future sustainable development research in the Conclusion. 3.4. Filter Then Fill-in-the-Blanks: We Mostly Filter Complexity and Complete Patterns Homo economicus implies humans make decisions by applying something akin to the so-called weighted additive rule (Franklin's Rule). This is the idea that one tallies the pros and cons of any given action, assumes the perfect commensurability of said pros and cons, cancels out those of equal weight and arrives at a clear decision outcome (Shah and Oppenheimer, 2008). While some specific contexts may allow for such deliberate and systematic evaluation, empirical findings suggest human cognition, culminating in decision-making, is more akin to (1) the application of a sequence of filters (see Fig. 1), followed by (2) pattern completion. One can think of this heuristically as a “filter and fill-in-the-blanks” process. The act of filtering complexity begins physically, with our body's sensory inputs. Evolutionary theory suggests these have adapted to be able to capture only the small amount of information around us most relevant to our survival (particularly in a hunter–gatherer context), and to our moment-to-moment goals (see Section 3.2) (Peterson and Flanders, 2002).7 Once information has been filtered through our senses, and the metaphorical ‘glasses’ determined by our in situ goals (see Section 3.2), it then progresses to our brain's neural circuitry, the content of which can be thought of, here, as a memory bank of previous experiences (Bar, 2009). In metaphorical terms, it can be useful to imagine this memory bank as an ever-expanding collection of ‘maps,’ compiled and revised continually over the course of our lives, each map representing our understanding of a prototypical object or situation. Importantly, the function of these ‘maps’ (known in the broader literature as ‘mental representations,’ of which there are many varieties — see Fig. 2) is to suggest, based on previous experience, how we should behave in situ to achieve our desired goals (Peterson and Flanders, 2002; D'Andrade, 1995). Importantly, cognitive studies suggest that for this process we rely largely on associative thinking, whereby we compare and contrast current stimuli with what appear to be relevant ‘maps’, i.e., stored memories of previous experience (see Section 3.5)8 (Bar, 2009; Fauconnier and Turner, 2002; Slingerland, 2008). We are then able to reason by analogy to ‘complete the pattern’, as it were. That is to say, if one's brain remembers A + B = C, and, via an intuitive, associative process 6 I.e., Mr. Spock, the famously logical, emotionally unengaged character on the original science-fiction show ‘Star Trek.’ 7 For example, we detect sound waves but not radio waves, and middle-spectrum light but not ultraviolet light, let alone the near-infinite quantity of other potential data of which we are still unaware. 8 On a physiological level, one can think of this process as follows: rather than make sense of each incoming sensory input entirely anew, our mind recruits already existing neural networks to compare and contrast new sensory experiences with memories of previous ones it deems similar or relevant (Bar, 2009; Fauconnier and Turner, 2002). 27 of comparison and contrast, new stimulus ‘D’ is deemed similar in character to ‘A,’ then one's brain ‘completes the pattern’ by inferring that D + B probably also equals C. One then proceeds to act as if that were so (Bar, 2009; Lakoff and Johnson, 1999).9 Phrased differently, once information has been filtered through our senses, as per our in situ goals, we appear pre-disposed to intuitively and pre-consciously scan for similarities and differences with respect to previous experience, and act accordingly (Bar, 2009). This is less effortful (Gilovitch et al., 2002), and hence more efficient, than applying such processes as the idealized weighted additive rule described above (Shah and Oppenheimer, 2008). 3.5. Efficient Cognition is Distributed: The ‘Elephant and Rider’ Belong to a Herd A Homo economicus model assumes people process inputs, and proceed to weigh pros and cons, as autonomous, independent beings. An empirical perspective, however, reminds us that this view is more early modern European idealism than fact (Persky, 1995; Winch, 1992). Rather, our cognitive processes are deeply dependent on our environment. We develop and operate in a context most often replete with socio-cultural and technological means for further reducing one's cognitive load. For instance, we defer to skilled or knowledgeable community members when dealing with a complex issue requiring action, we use modern information technology such as the internet to clarify what to do about virtually any conceivable problem, or we defer to our peer groups to determine contextually appropriate behavior (e.g., Goldstein et al., 2008). This penchant for offloading one's cognitive burden onto a more effective or reliable entity is known as ‘distributed cognition’ (Giere and Moffatt, 2003). By helping to reduce individual mental effort, distributed cognition functions as another key mechanism of efficiency (Zhang and Patel, 2006), continually shaping our cognitive strategies and development. In metaphorical terms, one can think of the ubiquity of distributed modes of cognition by imagining Haidt's (2006) elephant and rider as merely one human–pachyderm pair in a vast herd of other elephants and riders, each of whom is regularly copying, taking cues from, and communicating with, others in the herd. 3.6. Heuristics are the Rule, Not the Exception: Brains are not Lazy Computers Empirical research in sustainable development, and social science broadly, is now replete with examples of flagrant violations of Homo economicus (Gintis, 2000; Gilovitch et al., 2002; Henrich et al., 2001; McDaniels et al., 2003). However, the slew of biases that have been identified to account for these apparent idiosyncrasies of human judgment are often popularly interpreted as just that: idiosyncrasies, flaws, or cracks in an otherwise firmly rationalist foundation. Such an interpretation only makes sense against an empiricallyuninformed backdrop, wherein people are assumed to be fully selfconscious, rational maximizers of abstract utility who nonetheless occasionally and unwisely cut corners by deferring to fast-and-frugal ‘heuristics’ (cognitive shortcuts) for dealing with complexity. That characterization is a radical misunderstanding (Kahneman, 2003), symptomatic of the lingering paradigmatic privilege enjoyed by the canonical model. In contrast, from the unified perspective we are presenting here, heuristics are not ‘shortcuts’ for a system that otherwise reasons everything in a fully self-aware, linear, utilitarian fashion. Rather, it is exactly the inverse: humans possess a cognitive and behavioral system optimized for in situ efficiency, not precise utility maximization. 9 In a sustainable development context, the benefit transfer method of ecosystem service valuation, for instance, is a transparent example of this. 28 J. Levine et al. / Ecological Economics 114 (2015) 22–32 From this view, ‘heuristics’ are especially obvious examples of the fundamental and indispensable process of emotionally-driven, goaloriented, time- and energy-efficient complexity reduction that is continually at work in people's minds. Fully conscious reasoning (which may or may not be explicitly structured on, or well approximated by, a model of abstract utility maximization) can sometimes arise, under certain conditions. However, much of cognition occurs in a semiautomatic, associative manner, shaped profoundly by one's surrounding context in confluence with one's set of underlying motivations, which in turn are experienced emotionally and viscerally (Buck, 1985; Damasio, 1994; Lavine, 1998; Slovic et al., 2007). In other words, cognitive research suggests that Haidt's (2006) allegorical elephant rider is not a hyper-rational supercomputer with unlimited time and energy that sometimes gets lazy. Rather, the rider is, as described in Section 3.1, more like a full-time triage nurse whose default state is to make quick-and-dirty decisions in the face of voluminous complexity, and limited time and energy. Only when conditions are just right does the metaphorical triage nurse experience a lull in pressure, enabling him or her to deliberate at greater length, and with greater precision and foresight, than would normally be affordable. Otherwise, the triage nurse goes with his or her experience-informed, emotionally-mediated ‘gut.’ In the context of sustainable development, looking at things from this perspective can help us de-mystify, and perhaps re-theorize, some of the apparently less ‘rational,’ more bemusing examples of observed behavior alluded to above. With specific respect to sustainable development research, these include the embedding effect (McDaniels et al., 2003), hyperbolic discounting (Gintis, 2000), strong reciprocity (Gintis, 2000), hyper-fair offers in economic games (Henrich et al., 2001), seemingly unrealistic optimism or risk appraisals (Gilovitch et al., 2002), and so on. While an underlying theory of unconstrained rationality would perceive these as anomalies, a vision of the human mind as deeply embedded in shifting social–ecological contexts, engaging in constant information ‘triage’ and motivated to achieve desirable emotional states rather than abstract maximum utility, would in fact predict such kinds of behavior. 3.7. Analogy as the Unit of Thought: The Elephant Rider Navigates With Maps As an example, Kahneman and Knetch (1992) compellingly argue that respondents to willingness-to-pay (WTP) questions (e.g., for the protection of lakes) often treat the hypothetical transaction not as if they were ‘purchasing an outcome’ (a contrived, unfamiliar task in reference to non-marketed environmental goods like distant lakes) but rather as if they were ‘donating to a cause’ (an only roughly analogous, but much more familiar, and hence less cognitively taxing, exercise). For WTP researchers, who are generally more interested in approximating the total economic value of environmental goods (e.g., lakes) than they are in respondents' donation preferences, this sort of response behavior can be a significant challenge, resulting in the potential under-valuing of public goods (i.e., the embedding effect) (Kahneman and Knetch, 1992; McDaniels et al., 2003). What accounts for this sort of behavior, and why is it surprising? Evidence suggests the answer lies, again, in our propensity toward cognitive efficiency (or ‘information triage’, see Section 3.1), which requires us to use all neurological resources available to make actionable sense out of vast quantities of incoming data as quickly and effortlessly as possible. Above, we described how people do this by drawing on memories of previous experience that appear relevant, and by then ‘completing the pattern’ (see Section 3.4). This can also be helpfully conceptualized as reasoning ‘by analogy’ (Bar, 2009; Lakoff and Johnson, 1999; Slingerland, 2008). Different disciplines focus on different aspects of this process, using different vocabulary. Despite said diversity of nomenclature, each field's conception of this mechanism has in common the notion of analogy: using one, relatively familiar, tangible, or otherwise easily recalled domain of knowledge (source domain) to make sense of, and reason about, another, relatively less familiar, domain (target domain). Key to any notion of analogical reasoning is the idea that one takes a salient attribute of one thing (e.g., the fact that water ‘flows’) and ‘projects’ or ‘maps’ it onto another thing (e.g., electricity), in order to make new and useful inferences (e.g., that electricity can be thought to flow, giving rise to the idea of ‘current’) (Gentner and Gentner, 1983). By engaging in this kind of reasoning, we can make use of lessons already learned from previous experience when navigating new scenarios that we deem roughly analogous (Bar, 2009). This mechanism, which cognitive linguists call the ‘invariance principle’ (Lakoff, 1993), appears to be a core feature of our cognition (Peterson, 1999; Peterson and Flanders, 2002). We are also able to selectively combine aspects of different source domains together to create entirely new, increasingly detailed analogical representations of the world around us (a process called ‘conceptual blending’). This capacity to usefully blend analogies appears in turn to be a fundamental mechanism of human creativity (Fauconnier and Turner, 2002; Slingerland, 2008). Nonetheless, despite the efficiency and creativity that analogical reasoning affords, it does have drawbacks. In the psychological study of heuristics, analogical reasoning gone awry is known as ‘attribute substitution’ (Kahneman, 2003). This occurs when a person erroneously (often unconsciously) assumes similarity between cases that are in fact different in consequential ways, and then goes on to make an error in judgment stemming from that difference (in essence, using the wrong ‘map’). The embedding effect, described above, is merely one such example. Given these findings, we argue it can be helpful to think of people as always relying on some kind of simplified ‘map’ or ‘analogical representation,’ in a broad sense (be it an internal representation such as a ‘mental model,’ a specific, analogous memory, or an external representation, such as a literal, physical image, or text) to frame and guide their interactions with the complex world around them (Peterson and Flanders, 2002; Shore, 1996). (See Fig. 2 for a typology of the various categories of analogical representation described in the cognitive science literature.) As psychologists Peterson and Flanders (2002) argue: “We are doomed to formulate conceptual structures that are much simpler than the complex phenomena they are attempting to account for.” (429). The question is not why we rely on such simplifications; cognitive science suggests this is inevitable given our species' biologicallymediated perceptual, energetic and temporal limitations (Peterson, 1999; Shah and Oppenheimer, 2008). Rather, the more relevant question to answer is what factors influence people to draw on one analogical representation versus another, when, and why. In other words, how does Haidt's (2006) proverbial elephant-rider duo determine which map to use? We consider this question in Section 3.7.1. 3.7.1. Analogies are Chosen by Availability, Associations, and Contextual Cues For those of us studying real human behavior in the context of sustainable development, it is helpful to know which kinds of analogies people will tend to draw on, when, and why. For instance, why would respondents draw on experiences of donation, rather than commerce—or even simple addition—when answering WTP questions about protecting all the lakes in a country, versus just one lake? What factors would help in better designing surveys to accommodate these kinds of tendencies? In cognitive terms, this question could be phrased as: Why do people's brains select one source domain, rather than another, to reason about a given target domain? A diversity of research findings (see Sections 3.7.1.1 and 3.7.1.2) suggest at least two answers to this question, one involving a proximate cause, and the other a more distal cause. J. Levine et al. / Ecological Economics 114 (2015) 22–32 3.7.1.1. Availability. The first, proximate influence on which analogical representation one draws upon is availability (or ‘accessibility’): this is simply the ease with which something comes to mind relative to something else (Tversky and Kahneman, 1973). The more readily available a given domain of thought is, the more likely it is that we will draw on it to reason about any given situation, as doing so requires less effort than drawing on a less available domain. All else held equal, what makes a given domain ‘available,’ or ‘accessible,’ for our minds to reason with involves at least two factors: (1) how frequently the domain is recalled; i.e., one's ‘habits of mind’ (Bang et al., 2007), which, in physical terms, equates to how practiced we are at exercising the particular neuronal pathways associated with that given domain (Fuster, 1999; Read, 1995); and (2) how emotionally charged (or ‘affect’ laden) a given domain is (Keller et al., 2006). The more often one ‘rehearses’ (i.e., accesses or ‘activates’) a particular source domain, and the more intense emotions it evokes, the more available it is. Hence, we are more likely to draw on that domain, or ‘map,’ rather than another, when attempting to interpret a new instance of complexity.10 3.7.1.2. Context and Associative Networks. The second, more distal cause determining which source domain an individual will be disposed to draw on has to do with what creates relative differences in availability to begin with (i.e., when all else is not held equal as above). At least two other factors in addition to the basic effects of rehearsal and emotional content further determine availability. These are (1) context, broadly conceived; and, (2) the particular network of associations among various emotions, ideas, memories and concepts that each of us has (what Bower (1981) coined ‘associative networks’). There are a plethora of findings in psychology detailing the effects of context (often called ‘priming’ effects) on how we make sense of complexity and, ultimately, how we take decisions (Kusev et al., 2012; Tversky and Kahneman, 1981; Smith and Semin, 2004; Storbeck and Clore, 2008). Context, here, should be thought of in its broadest possible sense. Through a largely unconscious process of association (Claxton, 1999; Dijksterhuis and Nordgren, 2006), virtually anything a person perceives or experiences at a given time can have an influence on his or her thinking, on the analogical representation (or ‘map’) of the situation the person's brain (re)constructs and, ultimately, on his or her choice of action. This includes external context, such as how presenting bicultural people with images associated with one or the other of their two learned cultures can produce different responses to the same question (Benet-Martinez et al., 2002). It also includes ‘internal’ context: one's real-time emotional state, object of mental attention, prior train of thought, or preexisting underlying goals (e.g., Cohen et al. (1996) demonstration of how the emotions evoked by a perceived insult can directly alter how an individual behaves in an experiment minutes later11). All of these contextual factors interact continuously with our networks of mental and emotional associations to help us draw on apparently useful ‘maps’—i.e., appropriate memories from which to reason analogically—when dealing with a new instance of complexity.12 10 See, e.g., Lichtenstein and Slovic (2006) and Roeser (2012) on the role of affect, associative imagery, and availability in risk perception, particularly as concerns climate change. 11 The notion that different emotional states can activate different associative memory networks, and thus affect decision-making, is known in the cognitive science literature as the ‘somatic marker hypothesis’ (Damasio, 1996). 12 Interestingly, with reference to public understanding of ecological science, it is worth noting that the greater the number of cues to which an individual feels he or she must consciously attend, the more likely it appears that individual will default to a very simplistic analogical representation (Ableson and Levi, 1985), imaginably as a response to the emotional discomfort of cognitive overload (Peterson and Flanders, 2002). 29 4. A Formalized Synthesis: The Efficient Complexity Manager (ECM) Model Given the attributes described above, we refer to our synthesized characterization of the human actor as the ‘efficient complexity manager’ (ECM) model or, simply, Homo efficens. To reiterate, this model is not intended to be a conclusive or comprehensive account of human cognition and motivation. Rather, it is meant as a simplified expression of how recent findings can be unified in a consistent fashion to help better integrate empirical insights from various disciplines into a coherent, contemporary theory. To clarify, immediately below we demonstrate in eight condensed, chronological steps how the ECM model suggests people proceed from perception, to cognition, and ultimately to action. While in reality the process is somewhat non-linear, here we present it linearly, purely for clarity's sake: 1) Perceive Stimulus. We encounter a new instance of complexity — the ‘target domain’ (Lakoff and Johnson, 1999; Fauconnier and Turner, 2002; Slingerland, 2008) — interpreted via our body's sensory inputs. (See Section 3.4 and Fig. 1.) 2) Draw on Context. We intuitively draw on contextual information to further reduce the time and effort required to usefully interpret the stimulus. Context, here, includes framing, and all other ‘external’ social and environmental conditions (what we see, hear or sense at the time), as well as ‘internal’ mental, physiological and emotional conditions (what is ‘on our mind,’ our affective state, and pre-existing goals and motivations, also experienced largely in terms of emotion) (Damasio, 1994, 2000; Lavine, 1998; Slovic et al., 2007). This contextual information helps our mind narrow in on which source domain(s) (maps) of previous experience are most relevant in making further sense of the stimulus (physiologically speaking, which clusters of previously instantiated neural connections to use to determine appropriate action) (Kokinov and Petrov, 2001; Peterson, 1999). (See Sections 3.7 and 3.7.1.) 3) Draw on Associative Networks. Given the perceived attributes of the stimulus, and all contextual cues, we then use association to search our figurative ‘databank’ of memories for analogous cases (in metaphorical terms, useful ‘maps’; in physiological terms, relevant clusters of neural connections). This process is not unlike using detailed search terms (analogous, here, to contextual cues in Step 2, directly above) to query a tagged database such as an internet search engine (analogous to the mind's associative networks of memory) for relevant domains of knowledge from which to draw in making sense of the stimulus (Bar, 2009). Domains (maps) that we use relatively often (i.e., more frequently recalled or ‘rehearsed’) will be easier for our mind to access, and we will thus be more likely to draw on those domains relative to others (Fuster, 1999). (See Section 3.7.1.2.) 4) (Re)construct an Analogical Representation. Our mind then draws from the source domains of knowledge we deem relevant in light of Steps 3 and 4 to (re)construct an analogical representation (a contextually refined ‘map’) of the real-world stimulus that our mind is attempting to interpret (Fauconnier and Turner, 2002). Depending on context, this process can range from being highly conscious, to being entirely unconscious and nearly instantaneous (Kahneman, 2003). At this point, we may also deem it appropriate to defer to other sources (e.g., more knowledgeable individuals, social norms, technology) when constructing a simplified representation of the situation (i.e., reliance on distributed cognition) (Giere and Moffatt, 2003). (See Sections 3.5 and 3.7.) 4a) Identify Goals. Parallel to Steps 1–4 above, we are continually reshaping our goals dependent on our surrounding context and emotional state. New stimuli, new contexts, new associations and new analogical representations, feed back into this process to create goals specific to time and place. These goals may include, but are not limited to, maximizing our personal material gain. 30 5) 6) 7) 8) J. Levine et al. / Ecological Economics 114 (2015) 22–32 They may also include whatever else produces a sense of emotional reward given the particular context (e.g., be that averting a negative feeling such as shame or fear, or increasing a positive feeling such as a sense of security, inclusion or social worth) (Buck, 1985; Eclles and Wigfield, 2002; Markus and Kitayama, 1991; Peterson and Flanders, 2002). Because this process is so intimately linked with emotion and affect (Haidt's (2006) figurative ‘elephant’), it is largely pre-rational (Greene and Haidt, 2002). (See Section 3.3.) Reason. In light of the goals determined in Step 4a, we then use the representation (map) constructed in Step 4 to reason about how we should act toward the stimulus. As with Step 4, this can range from an intuitive, time- and energy-efficient, largely unconscious process (i.e., what Kahneman (2003), Stanovich and West (2000) call ‘System 1’) to a more deliberate, energy-intensive, fully conscious process, when specific conditions allow (what Kahneman (2003), Stanovich and West (2000) call ‘System 2’). Act. We act accordingly. Observe and Remember the Outcome. We then observe the outcome of our chosen action, and said outcome is stored in our databank of memories (i.e., our ‘collection of maps’) to be drawn upon in future iterations of Step 3 (Bar, 2009). Iterate. The process is iterative: each outcome serves as a new stimulus, and the cycle repeats ad infinitum. Our mind thus continually accumulates lived experience, perpetually deepening its databank of associative memory. Linearly, Steps 1 to 4 of this process can be represented as follows: Stimulus þ Contextðexternal& internalÞ þ Associative Networks or S þ C þ AN→AR: produces → Analogical Representation Steps 5, 6 and 7 can be represented as: Reason→Act→Observe þ Remember or R→A→O þ R: In total, the iterative process can be represented visually as in Fig. 4. The numbers in parentheses (1, 2, etc.) highlight the approximate (if simplified) sequential ordering of steps (e.g., Step 1 and Step 2). 5. Conclusion A core challenge to the field of ecological economics is crafting a clear and unique identity that distinguishes it from the neoclassical economics paradigm, while avoiding the pitfall of becoming a catch-all for any and all transdisciplinary or heterodox work (Ropke, 2005; Spash, 2012). We believe that a strongly evidence-based, explicitly descriptive (as opposed to normative or wishful) model of the human actor can add to the growing theoretical foundation that is key to consolidating such an identity for the field. Thus, here we have presented a descriptive model of the human actor (the efficient complexity manager model, or ‘Homo efficens’) that synthesizes diverse empirical findings and theory on cognition, motivation and decision-making into what we hope is a manageable and useful summation. While the canonical model is premised on the notion of rational self-interest, often narrowly construed—the product of a blurred mix of both normative thought and questionable evidence—the ECM model revolves instead around the notion of humans' cognitive efficiency, achieved through mechanisms that are perpetually guided and contextualized by emotion. Fig. 4. Sequential flow diagram of the efficient complexity manager (ECM) model. This diagram represents how the efficient complexity manager (ECM) model, or Homo efficens, accounts for how people reduce complexity into actionable behavior while economizing for time and energy. The numbers in brackets (1-8) indicate the sequence of steps in an approximate linear order. See Section 4 for a full description. (Synthesized from Bar, 2009; Damasio, 1994, 2000; Fauconnier and Turner, 2002; Haidt, 2001; Kahneman, 2003; Lakoff and Johnson, 1999; Lavine, 1998; Peterson and Flanders, 2002; Slovic et al., 2007). J. Levine et al. / Ecological Economics 114 (2015) 22–32 Importantly, this model is intended as a purely descriptive synthesis of recent findings from relevant disciplines, as distinct from any normative vision of an idealized human actor. While there is certainly a role to be played both by normative visions, and by descriptive accounts, respectively, we believe clearly making such a distinction is crucial for future progress on this topic. With further respect to the issue of progress, we reiterate that the ECM model is by no means comprehensive or final and should not be interpreted as such. Rather, it is intended as a first-cut synthesis of descriptive human actor research, which, ideally, will help ecological economists—and other sustainable development specialists—to ‘exorcise the ghost’ of Homo economicus from their respective fields. Finally, the synthesis we have provided here also implies some fruitful avenues for future research, both in ecological economics, and in the realm of sustainable development more broadly. These include, in particular: studies and methods that further examine the nuances of (1) in situ choice behavior (e.g., Schlapfer and Fischhoff, 2012; Welsh and Kuhling, 2009); (2) the role of mental models, analogy and metaphor in both participants' and researchers' thinking, (e.g., Anderies et al., 2011; Norton and Noonan, 2007; Raymond et al., 2013; Schulter, 2009) and, (3) the role of affect and emotion in the selection of said mental models, and in economic and environmental decision-making more broadly (e.g., Arana and Leon, 2008, 2009; Boyer, 2008; Finucane et al., 2000). Many researchers are already employing methods aimed at analyzing precisely these factors. Underlying assumptions about how and why humans behave the way they do can having far-reaching, cascading effects on multiple levels of a research agenda. As such, we hope that the unified, empirically-informed model of the human actor we have presented here will help provide an even clearer, more solid theoretical foundation for the continuation and refinement of innovative sustainable development work. Acknowledgments We wish to thank the University of British Columbia's Faculty of Graduate and Postdoctoral Studies for providing a ‘Four Year Fellowship’(#6456) stipend to the primary author during the research and writing stages of this paper. We also with to acknowledge the Social Sciences and Humanities Research Council (SSHRC) of Canada (#8612009-1106; #820-2006-0040; #820-2008-3026; #435-2013-2017) and the Natural Sciences and Engineering Research Council (NSERC) of Canada (#365144-08) for their support. Special thanks go to Dr. Edward Slingerland of the University of British Columbia's Centre for Human, Evolution, Cognition and Culture for his helpful insights, and Dr. Jordan Peterson of the University of Toronto for his thoughtprovoking body of work. References Ableson, R.P., Levi, A., 1985. Decision making and decision theory. In: Lindzey, G., Aronson, E. (Eds.), Vol. 1, in The Handbook of Social Psychology. Random House, New York, pp. 231–309. Adamowicz, W.L., 2004. What's it worth? An examination of historical trends and future directions in environmental valuation. Aust. J. Agric. Resour. Econ. 48 (3), 419–443. Anderies, J.M., et al., 2011. The challenge of understanding decisions in experimental studies of common pool resource governance. Ecol. Econ. 70 (9), 1571–1579. Arana, J.E., Leon, C.J., 2008. Do emotions matter? Coherent preferences under anchoring and emotional effects. Ecol. Econ. 66 (4), 700–711. Arana, J.E., Leon, C.J., 2009. Understanding the use of non-compensatory decision rules in discrete choice experiments. Ecol. Econ. 68 (8–9), 2316–2326. Ariely, D., Loewenstein, G., Prelec, D., 2003. “Coherent arbitrariness”: stable demand curves without stable preferences. Q. J. Econ. 118 (1), 73–106. Arrow, K.J., 1963. Social Choice and Individual Values. John Wiley & Sons, New York. Atran, S., Medin, D., 2010. The Native Mind and the Cultural Construction of Nature. The MIT Press, Cambridge, Massachusetts. Bang, M., Medin, D.L., Atran, S., 2007. Cultural mosaics and mental models of nature. PNAS 104 (35), 13868–13874. 31 Bar, M., 2009. The proactive brain: memory for predictions. Philos. Trans. R. Soc. B 364 (1521), 1235–1243. Bateman, I., Mawby, J., 2004. First impressions count: interviewer appearance and information effects in state preference studies. Ecol. Econ. 49 (1), 47–55. Baveye, P., Baveye, J., Gowdy, J., 2013. Monetary evaluation of ecosystem services: it matters to get the timeline right. Ecol. Econ. 95, 231–235. Becker, C., 2006. The human actor in ecological economics: philosophical approach and research perspectives. Ecol. Econ. 60 (1), 17–23. Benet-Martinez, V., Leu, J., Lee, F., Morris, M., 2002. Negotiating biculturalism: cultural frame switching in biculturals with oppositional versus compatible cultural identities. J. Cross-Cult. Psychol. 33 (5), 492–516. Bina, O., Vaz, S.G., 2011. Humans, environment and economies: from vicious relationships to virtuous responsibility. Ecol. Econ. 72, 170–178. Borgida, E., Nisbett, R.E., 1977. The differential impact of abstract vs. concrete information on decision. J. Appl. Soc. Psychol. 7 (3), 258–271. Bower, G.H., 1981. Mood and memory. Am. Psychol. 36 (2), 129–148. Bowles, S., Gintis, H., 1993. The revenge of homo economicus: contested exchange and the revival of political economy. J. Econ. Perspect. 7 (1), 83–102. Boyer, P., 2008. Evolutionary economics of mental time travel? Trends Cogn. Sci. 12 (6), 219–224. Buck, R., 1985. Prime theory: an integrated view of motivation and emotion. Psychol. Rev. 92 (3), 389–413. Bulkeley, H., Mol, A.P.J., 2003. Participation and environmental governance: consensus, ambivalence and debate. Environ. Values 12 (2), 143–154 (May). Bute, E.H., Lipper, L., Stringer, R., ZIlberman, D., 2008. Payments for ecosystem services and poverty reduction: concepts, issues and empirical perspectives. Environ. Dev. Econ. 13 (3), 245–254. Camerer, C., Loewenstein, G., Prelec, D., 2005. Neuroeconomics: how neuroscience can inform economics. J. Econ. Lit. 53, 9–64. Claxton, G., 1999. Hare Brain, Tortoise Mind: How Intelligence Increases When you Think Less. HarperCollins, New York. Cohen, D., Nisbett, R.E., Bowdle, B.F., Schwarz, N., 1996. Insult, aggression and the southern culture of honor: an experimental ethnography. J. Pers. Soc. Psychol. 70 (5), 945–960. Corcoran, K., Mussweiler, T., 2010. The cognitive miser's perspective: social comparison as a heuristic in self-judgements. Eur. Rev. Soc. Psychol. 21 (1), 78–113. Damasio, A.R., 1994. Descartes' Error: Emotion, Reason and the Human Brain. HarperCollins, New York. Damasio, A.R., 1996. The somatic marker hypothesis and the possible functions of the prefrontal cortex. Philos. Trans. R. Soc. Lond. B 351, 1413–1420. Damasio, A.R., 2000. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. Houghton Mifflin Harcourt, Boston. D'Andrade, R.G., 1995. Development of Cognitive Anthropology. Cambridge University Press, Cambridge. Desvousges, W., Mathews, K., Train, K., 2012. Adequate responsiveness to scope in contingent valuation. Ecol. Econ. 84, 121–128. Dijksterhuis, A., Nordgren, L.F., 2006. A theory of unconscious thought. Perspect. Psychol. Sci. 1 (2), 95–109. Doucouliagos, C., 1994. A note on the evolution of homo economicus. J. Econ. Issues 28 (3), 877–883. Eclles, J.S., Wigfield, A., 2002. Motivational beliefs, values and goals. Annu. Rev. Psychol. 53, 109–132. Engel, S., Pagiola, S., Wunder, S., 2008. Designing payments for environmental services in theory and practice: an overview of the issues. Ecol. Econ. 65 (4), 663–674. Eubanks, P., 1999. The story of conceptual metaphor: what motivates metaphoric mappings? Poetics Today 20 (3), 419–442. Faber, M., Petersen, T., Schiller, J., 2002. Homo oeconomicus and homo politicus in ecological economics. Ecol. Econ. 40 (3), 323–333. Fauconnier, G., Turner, M., 2002. The Way We Think: Conceptual Blending and the Mind's Hidden Complexities. Basic Books, New York. Fehr, E., Gachter, S., 2000. Fairness and retaliation: the economics of reciprocity. J. Econ. Perspect. 14 (3), 159–181. Feldman, A., 2008. Welfare economics. In: Durlauf, S., Blume, L. (Eds.), The New Palgrave Dictionary of Economics, Second edition (online [accessed 9 Jan 2014]). Finucane, M.L., Alhakami, A., Slovic, P., Johnson, S.M., 2000. The affect heuristic in judgments of risks and benefits. J. Behav. Decis. Mak. 13, 1–17. Fisher, B., Kulindwa, K., Mwanyoka, I., Kerry Turner, R., Burgess, N.D., 2010. Common pool resource management and PES: Lessons and constraints for water PES in Tanzania. Ecol. Econ. 69 (6), 1253–1261. Fong, C., 2001. Social preferences, self-interest, and the demand for redistribution. J. Public Econ. 82 (2), 225–246. Forgas, J. (Ed.), 2000. Feeling and Thinking: The Role of Affect in Social Cognition. Cambridge University Press, Cambridge. Frank, R.H., 1987. If Homo economicus could choose his own utility function, would he want one with a conscience? Am. Econ. Rev. 77 (4), 593–604. Frey, B., 1994. How intrinsic motivation is crowded out and in. Ration. Soc. 6 (3), 334–352. Fuster, J.M., 1999. Memory in the Cerebral Cortex: An Empirical Approach to Neural Networks in the Human and Nonhuman Primate. MIT Press, Cambridge, MA. Gentner, D., Gentner, D.R., 1983. Flowing waters or teeming clouds: mental models of electricity. In: Gentner, D., Steven, A.L. (Eds.), Mental Models. Lawrence Erlbaum Associates, New Jersey, pp. 99–129. Giere, R.N., Moffatt, B., 2003. Distributed cognition: where the cognitive and the social merge. Soc. Stud. Sci. 33 (2), 301–310. Gigerenzer, G., Goldstein, D., 1996. Reasoning the fast and frugal way: models of bounded rationality. Psychol. Rev. 103 (4), 650–669. 32 J. Levine et al. / Ecological Economics 114 (2015) 22–32 Gilovitch, T., Griffin, D., Kahneman, D., 2002. Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge University Press, New York. Gintis, H., 2000. Beyond Homo economicus: evidence from experimental economics. Ecol. Econ. 35 (3), 311–322. Glimcher, P., Dorris, M., Bayer, H., 2005. Physiological utility theory and the neuroeconomics of choice. Games Econ. Behav. 52 (2), 213–256. Goldstein, N.J., Cialdini, R.B., Criskevicius, V., 2008. A room with a viewpoint: using social norms to motivate environmental conservation in hotels. J. Consum. Res. 35 (3), 472–482. Gomez-Baggathun, E., de Groot, R., Lomas, P., Montes, C., 2010. The history of ecosystem services in economic theory and practice: from early notions to markets and payment schemes. Ecol. Econ. 69 (6), 1209–1218. Gowdy, J., 2004. The revolution in welfare economics and its implications for environmental valuation and policy. Land Econ. 80 (2), 239–257. Greene, J., Haidt, J., 2002. How (and where) does moral judgment work? Trends Cogn. Sci. 6 (12), 517–523. Gregory, R., Lichtenstein, S., Slovic, P., 1993. Valuing environmental resources: a constructive approach. J. Risk Uncertain. 7, 177–197. Haidt, J., 2001. The emotional dog and its rational tail: a social intuitionist approach to moral judgment. Psychol. Rev. 108, 814–834. Haidt, J., 2006. The Happiness Hypothesis. Basic Books, New York. Hart, S.J., Green, S.R., Casp, M., Belger, A., 2010. Emotional priming effects during Stroop task performance. NeuroImage 49 (3), 2662–2670. Henrich, J., et al., 2001. In search of homo economicus: behavioural experiments in 15 small-scale societies. Am. Econ. Rev. 91 (2), 73–78. Hickey, D.T., McCaslin, M.A., 2001. A comparative, sociocultural analysis of context and motivation. In: Volet, S., Jarvela, S. (Eds.), Motivation in Learning Contexts: Theoretical Advances and Methodological Implications. Elsevier Science Ltd., Oxford, pp. 33–56. Hudson, D., Gallardo, R.K., Hanson, T.R., 2012. A comparison of choice experiments and actual grocery store behaviour: an empirical application to seafood products. J. Agric. Appl. Econ. 44 (1), 49–62. Ingebrigsten, S., Jakobsen, O., 2009. Moral development of the economic actor. Ecol. Econ. 68 (11), 2777–2784. Jager, W., Janssen, M.A., De Vries, H.J.M., De Greef, J., Vlek, C.A.J., 2000. Behaviour in commons dilemmas: homo economicus and homo psychologicus in an ecological– economic model. Ecol. Econ. 35 (3), 357–379. Jollands, N., 2006. Concepts of efficiency in ecologial economics: Sisyphus and the decision maker. Ecol. Econ. 56 (3), 359–372. Kahneman, D., 2003. A perspective on judgment and choice: mapping bounded rationality. Am. Psychol. 58 (9), 697–720. Kahneman, D., Frederick, S., 2002. Representativeness revisited: attribute substitution in intuitive judgment. In: Gilovich, T., Griffin, D., Kahneman, D. (Eds.), Heuristics and Biases. Cambridge University Press, New York, pp. 49–81. Kahneman, D., Knetch, J., 1992. Valuing public goods: the purchase of moral satisfaction. J. Environ. Econ. Manag. 22 (1), 57–70. Kahneman, D., Slovic, P., Tversky, A., 1982. Judgment Under Uncertainty: Heuristics and Biases. Cambridge University Press, Cambridge. Keller, C., Siegrist, M., Gutscher, H., 2006. The role of the affect and availability heuristics in risk communication. Risk Anal. 26 (3), 631–639. Kissinger, G., Patterson, C., Neufeldt, G., 2013. Payments for ecosystem services schemes: project-level insights on benefits for ecosystem and the rural poor. Working Paper. World Agroforestry Centre, Nairobi. Klockner, C.A., Matthies, E., Hunecke, M., 2003. Problems of operationalizing habits and integrating habits in normative decision-making models. J. Appl. Soc. Psychol. 33 (2), 396–417. Kokinov, B.N., Petrov, A.A., 2001. Integrating memory and reasoning in analogymaking: the AMBR model. In: Genter, D., Holyoak, K.J., Kokinov, B.N. (Eds.), The Analogical Mind: Perspectives from Cognitive Science. MIT Press, Cambridge, MA, pp. 59–124. Kolodner, J., 1992. An introduction to case-based reasoning. Artif. Intell. Rev. 6, 3–34. Kusev, P., van Schaik, P., Aldrovandi, S., 2012. Preferences induced by availability: evidence from priming. J. Neurosci. Psychol. Econ. 5 (4), 250–258. Lakoff, G., 1993. The contemporary theory of metaphor. In: Ortony, A. (Ed.), Metaphor and Thought. Cambridge University Press, Cambridge, pp. 202–252. Lakoff, G., 2007. Cognitive models and prototype theory. In: Evans, V., Bergen, B., Zinken, J. (Eds.), The Cognitive Linguistics Reader. Equinox, London, pp. 130–167. Lakoff, G., Johnson, M., 1999. Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought. Basic Books, New York. Lavine, H., 1998. On the primacy of affect in the determination of attitudes and behaviour: the moderating role of affective–cognitive ambivalence. J. Exp. Soc. Psychol. 34 (4), 398–421. Lichtenstein, S., Slovic, P., 2006. The Construction of Preference. Cambridge University Press, New York. Lowenstein, G.F., Hsee, C.K., Weber, E.U., Welch, N., 2001. Risk as feelings. Psychol. Bull. 127 (2), 267–286. Macnaghten, P., Jacobs, M., 1997. Public identification with sustainable development: investigating cultural barriers to participation. Glob. Environ. Chang. 7 (1), 5–24. Macrae, C.N., Milne, A.B., Bodenhausen, G.V., 1994. Stereotypes as energy-saving devices: a peek inside the cognitive toolbox. J. Pers. Soc. Psychol. 66 (1), 37–47. Markus, H., Kitayama, S., 1991. Culture and the self: implications for cognition, emotion and motivation. Psychol. Rev. 98 (2), 224–253. McDaniels, T., Gregory, R., Arvai, J., Chuenpagdee, R., 2003. Decision structuring to alleviate embedding in environmental evaluation. Ecol. Econ. 46 (1), 33–46. Miloslavljevic, M., Navalpakkam, V., Koch, C., Rangel, A., 2012. Relative visual saliency differences induce sizable bias in consumer choice. J. Consum. Psychol. 22, 64–74. Monroe, K.R., Maher, K.H., 1995. Psychology and rational actor theory. Polit. Psychol. 16 (1), 1–21 (March). Murphy, S., Zajonc, R.B., 1993. Affect, cognition and awareness: affective priming with optimal and suboptimal stimulus exposures. J. Pers. Soc. Psychol. 64 (5), 723–739. Norton, B., Noonan, D.S., 2007. Ecology and valuation: big changes needed. Ecol. Econ. 64 (3), 644–675. Payne, J.W., Bettman, J.R., Johnson, E.J., 1992. Behavioral decision research: a constructive processing perspective. Annu. Rev. Psychol. 43, 87–132. Persky, J., 1995. Retrospectives: the ethology of homo economicus. J. Econ. Perspect. 9 (2), 221–223. Pessoa, L., 2008. On the relationship between emotion and cognition. Nat. Rev. Neurosci. 9 (148–158). Peterson, J.B., 1999. Maps of Meaning. Routledge, New York. Peterson, J., Flanders, J.L., 2002. Complexity management theory: motivation for ideological rigidity and social conflict. Cortex 38, 429–458. Raymond, C.M., Singh, G.G., Benessaiah, K., Bernhardt, J.R., Levine, J., Nelson, H., Turner, N.J., Norton, B., Tam, J., Chan, K.M.A., 2013. Ecosystem services and beyond: using multiple metaphors to understand human–environment relationships. Bioscience 63 (7), 536–546. Read, J.D., 1995. The availability heuristic in person identification: the sometimes misleading consequences of enhanced contextual information. Appl. Cogn. Psychol. 9, 91–121. Roeser, S., 2012. Risk communication, public engagement, and climate change: a role for emotions. Risk Anal. 32 (6), 1033–1040. Ropke, I., 2005. Trends in the development of ecological economics from the late 1980s to the early 2000s. Ecol. Econ. 55 (2), 262–290. Roth, A.E., Vesna, P., Okuno-Fujiwara, M., Zamir, S., 1991. Bargaining and market behavior in Jerusalem, Ljubljana, Pittsburgh, and Tokyo: an experimental study. Am. Econ. Rev. 81 (5), 1068–1095 (December). Ruth, M., 2006. A quest for the economics of sustainability and the sustainability of economics. Ecol. Econ. 53 (3), 332–342. Schlapfer, F., Fischhoff, B., 2012. Task familiarity and contextual cues predict hypothetical bias in a meta-analysis of stated preference studies. Ecol. Econ. 81, 44–47. Schulter, A., 2009. Institutional change and ecological economics: the role of mental models and sufficient reason. Inst. Sustain. 4, 315–340. Sen, A.K., 1988. On Ethics and Economics. Blackwell, Oxford. Sen, A.K., 1997. Rational fools: a critique of the behavioral foundations of economic theory. Philos. Public Aff. 6 (4), 317–344. Shah, A.K., Oppenheimer, D.M., 2008. Heuristics made easy: an effort-reduction framework. Psychol. Bull. 134 (2), 207–222 (March). Shi, T., 2004. Ecological economics as a policy science: rhetoric or commitment towards an improved decision-making process on sustainability. Ecol. Econ. 48 (1), 23–36. Shore, B., 1996. Culture in Mind: Cognition, Culture and the Problem of Meaning. Oxford University Press, Oxford. Shove, E., 2010. Beyond the ABC: climate change policy and theories of social change. Environ. Plan. A 42, 1273–1285. Siebenhuner, B., 2000. Homo sustinens: towards a new conception of humans for the science of sustainability. Ecol. Econ. 32 (1), 15–25. Simon, H.A., 1982. Models of Bounded Rationality. MIT Press, Cambridge, MA. Slade, S., 1991. Case-based reasoning: a research paradigm. AI Mag. 21 (42), 55. Slingerland, E., 2008. What Science Offers the Humanities: Integrating Body and Culture. Cambridge University Press, Cambridge. Slovic, P., Fincucane, M.L., Peters, E., MacGregor, D.G., 2007. The affect heuristic. Eur. J. Oper. Res. 177 (3), 1333–1352. Smith, E.T., Semin, G.R., 2004. Socially situated cognition: cognition in its social context. Adv. Exp. Soc. Psychol. 36, 53–117. Spash, C., 2012. New foundations for ecological economics. Ecol. Econ. 77, 36–47. Stanovich, K.E., West, R.F., 2000. Individual differences in reasoning: implications for the rationality debate? Behav. Brain Sci. 23 (5), 645–665. Storbeck, J., Clore, G.L., 2008. The affective regulation of cognitive priming. Emotion 8 (2), 208–215. Thaler, R., 2000. From homo economicus to homo sapiens. J. Econ. Perspect. 14 (1), 133–141. Turner, R.K., Adger, W.N., Brouwer, R., 1998. Ecosystem service value, research needs, and policy relevance: a commentary. Ecol. Econ. 25 (1), 61–65. Tversky, A., Kahneman, D., 1973. Availability: a heuristic for judging frequency and probability. Cogn. Psychol. 5, 207–232. Tversky, A., Kahneman, D., 1981. The framing of decisions and the psychology of choice. Science 211 (4481), 453–458. Tversky, A., Kahneman, D., 1992. Advances in prospect theory: cumulative representation of uncertainty. J. Risk Uncertain. 5 (4), 297–323. Van den Burgh, J., Ferrer-i-Carbonell, A., Munda, G., 2000. Alternative models of individual behaviour and implications for environmental policy. Ecol. Econ. 32 (1), 43–61. Weenig, M.W.H., Maarleveld, M., 2002. The impact of time constraint on information search strategies in complex choice tasks. J. Econ. Psychol. 23 (6), 689–702. Welsh, H., Kuhling, J., 2009. Determinants of pro-environmental consumption: the role of reference groups and routine behaviour. Ecol. Econ. 69 (1), 166–176. Winch, D., 1992. Adam Smith: Scottish moral philosopher as political economist. Hist. J. 35 (1), 91–113. Wray, R., et al., 2007. Towards a complete, multi-level cognitive architecture. Proceedings of ICCM 2007 Eighth International Conference on Cognitive Modelling, pp. 325–330. Zajonc, R.B., 2000. Closing the debate over the independence of affect. In: Forgas, J. (Ed.), Feeling and Thinking: The Role of Affect in Social Cognition. Cambridge University Press, Cambridge, pp. 31–58. Zhang, J., Patel, V.L., 2006. Distributed cognition, representation and affordance. Pragmat. Cogn. 14 (2), 31–58.