Ecological Rationality: A conceptual history

Erwin Dekker & Blaž Remic Erasmus University Rotterdam

Draft, May 2017

Introduction Ecological rationality is a term used by an increasing number of economists to refer to an alternative conception of rationality from behavioral economics. Behavioral economics of the Kahnemann and Tversky type, often referred to as the ‘heuristics and biases’ program, has become widely accepted as the way of integrating psychology and economics (Grüne- Yanoff, Marchionni, and Moscati 2014). The behavioral economics program presents itself as a serious challenge to the neoclassical picture of rational economic man, and argues that a serious reconsideration is necessary, since individuals are only boundedly rational (Tversky and Kahneman 1974; Kahneman 2003; Camerer, Loewenstein, and Rabin 2004; Mullainathan and Thaler 2000). The proponents of ecological rationality do not seek to challenge this claim of bounded rationality, in fact sometimes they go even further in emphasizing the cognitive limitations of individuals, but argue that in their interaction with their (social) environment individuals are nonetheless able to make reasonably good or even rational decisions. This is because they are able to use the environment to their advantage through cues or institutional features of that environment. The two main proponents of ecological rationality are Gerd Gigerenzer and Vernon Smith. On the surface they share the concept of ecological rationality. And both authors refer to each other’s work repeatedly to suggest that they are talking about the same thing (Smith 2003, 469; Gigerenzer 2015, 115–16). What is perhaps even more striking, they both point to Herbert Simon as an important influence on their work. In his most substantive work on rationality Vernon Smith argues that his distinction between constructivist and ecological rationality is in principle related to Herbert Simon’s distinction between objective and subjective rationality. The former meaning rationality from the experimenter’s point of view, and the latter rationality given the perceptual and evaluation premises of the subject. Hence, he argues that ecological rationality simply develops subjective rationality as Simon defined it. Gigerenzer, too, suggests that he is developing this neglected branch of Simon’s work, although he labels it slightly differently. He argues that the second component of Simon’s view on bounded rationality was the structure of the environment, which can help explain why simple heuristics work well, and when they work well. Despite these surface similarities we will demonstrate in this paper that Gigerenzer and Smith have developed distinct notions of ecological rationality, which are partially at odds with each other. Vernon Smith in his Nobel lecture describes ecological rationality as “an un-designed ecological system that emerges out of cultural and biological evolutionary processes: home grown principles of action, norms, traditions, and ‘morality’” (Smith, 2002). Gerd Gigerenzer argues that: “ecological rationality refers to the study of how cognitive strategies exploit the representation and structure of information in the environment to make reasonable judgments and decisions” (Gigerenzer 2000). From these two statements it should be fairly clear that they are talking about quite different research programs, which not only differ in what they study — Gigerenzer mostly heuristics, Smith mostly institutional environments — but also differ in what they seek to explain. Gigerenzer is much closer to psychology and modern decision sciences, which attempt to explain individual choices and decisions, whereas Smith is primarily interested in processes of social interaction and social systems. For Gigerenzer the central puzzle is how individuals manage to achieve their tasks given their limited cognitive abilities. For Vernon Smith the central puzzle is much closer to that of the other Smith, Adam — how can socially beneficial results emerge from actions which are self-centered and based on very limited knowledge. The second goal of our paper is to demonstrate that both types of ecological rationality present alternative combinations of psychology and economics. Especially Vernon Smith’s notion of ecological rationality seems prima facie devoid of psychology, a view recently reinforced by Don Ross (2014). In this paper we show, instead, that Gigerenzer’s and Vernon Smith’s concepts are much better understood as resulting from two rival perspectives in psychology. Gigerenzer’s is a functionalist psychology deeply indebted to the work of Egon Brunswik as he himself acknowledges. The psychology in Vernon Smith’s conception of ecological rationality is that of a situated and embodied cognition. This ‘embodied cognition’ perspective has recently challenged the cognitivist psychology, which has so influenced behavioral economics. This paper will proceed as follows, in the first section we outline what ecological rationality entails and how it is different from the bounded rationality conception in the behavioral economics program. In particular we will highlight how they give different interpretations of what is sometimes called Herbert Simon’s scissors: the cognitive abilities of the individual and the structure of the environment. This will set the stage for the different perspective on psychology we present in section two. In this section we will contrast the cognitive psychology of Simon and later Kahneman and Tversky with both the functionalist psychology of Brunswik and the modern perspective of ‘embodied’ cognition, which underlie the perspectives on ecological rationality. These different psychological perspectives will allow us to differentiate far more clearly between the two types of ecological rationality, and their differences with mainstream views on bounded rationality. In particular, we show that the mainstream perspective on bounded rationality results from a particular type of individualist economics with a particular type of cognitive and individualist psychology. An economics which places more emphasis on institutions and social interaction is in need of a different psychological underpinning. We conclude with implications for thinking about rationality in economics.

1. Economics and Psychology, or Simon’s scissors A recurring reference point for modern bounded rationality theories are the two blades of Herbert Simon’s scissors: “the structure of task environments and the computational capabilities of the actor” (Simon 1990, 7). Both the mainstream bounded rationality program and the alternative perspectives of ecological rationality seek to find a way to incorporate the environment and the capabilities of the actor. And all three research programs seek to trace back their lineage to Herbert Simon. Historically we might wonder what Herbert Simon really meant, and to what extent his research program came to fruition, as Petracca (2017) has recently done. We instead believe it is more helpful to examine how the different research programs have conceptualized the two blades of Simon’s scissors, especially since they all acknowledge the importance of both blades of the scissors. Let us first discuss the behavioural economics program.

There has emerged something of a standard narrative in portraying the history of the relationship between economics and psychology (Hands 2010). A recent historical account by two practitioners of behavioral economics, Angner and Loewenstein (2012), present what can be labeled ‘the standard account’. They divide the history of the relationship in three periods corresponding to an appropriate prevalent approach to economics: classical and early neoclassical economics, post-war neoclassical theory, and the “new” behavioral economics. They then present the attitude of the economists towards psychology through each of those. The most characteristic attitude of the classical and early neoclassical era was one of embracing hedonic psychology, where human behavior is seen as being motivated by the pursuance of pleasure and avoidance of pain (Bruni and Sugden 2007). In other words, there is a direct link between pleasure and utility, and introspection is seen as the proper method to gather the necessary evidence. However, this view gets—according to the standard account—cast out of economics by the developments build on Pareto’s vision of non- psychological economics. The central figure in this development is Paul Samuelson with his “behaviorist” theory of preferences revealed through observable choices, instead of introspection. This apparent rejection of psychology makes the Samuelsonian, or post-war neo-classical economics, the target in this standard account, and the period is often portrayed as a sort of “dark ages” where the theory of revealed preferences plays a role of a grand inquisitor out to excommunicate anybody inclined toward psychological reasoning. According to the standard view, things get finally set straight by the “new” behavioral economics, which emerges as the culmination of the early less successful developments— with old institutionalism, through Keynes’ analysis of the animal spirits, to Herbert Simon’s bounded rationality. As a result of all that has happened before, this time proponents are “cautious not to repeat mistakes committed by early twentieth-century psychologists and identified by the behaviorists” (Angner and Loewenstein 2012, 659). The new synthesis is based on the view of cognitive science that “thinking can be best understood in terms of representational structures in the mind and computational procedures that operate on those structures” (Thagard, quoted in Angner & Loewenstein, 2012, p. 659). Paired with the behavioral decision research program that sought out to expose the limitations of the mind in certain situations of decision-making, this new approach in thinking about economic problems crystallizes in the “heuristics and biases” research program and the “prospect theory” of Daniel Kahneman and Amos Tversky, whose work serves as the main inspiration and has direct lineage to the recent cutting-edge developments. The central concept in the contemporary behavioral economics literature is bounded rationality, which is directly contrasted with the model of full rationality associated with neoclassical economics. The standard view on rationality, most clearly expressed through the axioms of expected utility theory, has been for decades a target of fierce criticisms due to its psychologically unrealistic assumptions about the capabilities of the individual human mind. As a result of these attempts, various accounts challenging the rational choice model have entered the mainstream theoretical discourse under the flag of ‘bounded rationality’ (Conlisk 1996; Sent and Klaes 2005). Since this program emerged in direct opposition to neo-classical economics it sought to prove that in choice, or alternatively decision settings central in the neo-classical program its predictions were incorrect. Especially game- theoretic, and decision-theoretic situations have been a testing ground for the competing theories. This meant that the environment of the bounded-rationality approach has been very much that of game theory: the dictator game and the prisoner’s dilemma stand out as exemplary cases. This environment contains a particular type of institutional context, which is derived from theoretical choice settings. The experiments associated with this program typically seek to ensure that the setting conforms to the theoretical assumptions underlying decision or game theory: common knowledge of the pay-offs, no verbal communication and only a minimal context to remain as close as possible to the theoretical set-up. When more context is introduced into the experiments it was usually to demonstrate some kind of ‘framing’ effect (Tversky and Kahneman 1986). When experimenters in the bounded- rationality program vary the environment the purpose remains to figure out the cognitive response of the individual to different frames. In terms of Simon’s scissors this program is particularly interested in figuring out the cognitive working of the mind, or in Simon’s language the computational capabilities of individuals. It is precisely this clinical environment stripped of all superfluities to isolate the effect of particular factors that is criticized by the proponents of ecological rationality. The first experiment that Vernon Smith performed, was instead a class-room experiment in which subjects were free to talk and walk around, the purpose was not to test a particular response to a choice setting, but to observe exchange behavior. The subjects were given particular behavioral constraints: they could not buy above a certain price or sell below a certain price (Smith 1962). In later experiments the institutional market setting was refined and varied, but the basic idea remained the same: the design of experiments that resembled particular market settings. The outcome Vernon Smith was interested in was not so much individual choices, but instead market prices, suggesting that for him the interesting feature was the way in which institutional settings influence behavior. To return to Simon’s scissors, he therefore had a very different notion of the relevant environment which had to match some stylized market institution, rather than a particular choice situation. The experimental subjects were encouraged to utilize the environment and to make the best choices given that environment. He too was interested in the extent to which market participants were capable of performing these tasks, but more importantly he wanted to test under what institutional conditions the predictions about equilibrium prices held. When he limited the amount of information provided to the experimental subjects, it was not to purify the experiment, but instead to see if with this more limited information we would still see price converge to an equilibrium. When Vernon Smith varied the environment, it was to test how robust the outcomes of price convergence to the theoretically predicted equilibrium price were to changes in the environment. In terms of Simon’s scissors Vernon Smith accepted that individuals had limited capabilities, but was interested when the institutional environment would still lead them to rational outcomes on the market level. Gigerenzer’s experiments take yet a different form. They share with the behavioral economics experiments their study of individual choice in particular decision settings. But they share with Vernon Smith’s experiments their critique of the clinical environment of the behavioral economics experiments. But they take that critique a step further and argue that we can only study choices in real-world settings. Their primary object of study is how well individuals perform real-world tasks. Do they know the city with the most inhabitants in Germany or are they able to estimate the height of a particular building? These real-world questions are essential for Gigerenzer’s idea of ecological rationality, since it is only when individuals are able to use their environment that they can rely on heuristics. Estimating the height of a building in the absence of other structures around is far harder than when such points of reference are present. This anti-clinical design is essential for the approach, since it is argued that individuals make use of particular heuristics which utilize the environment as best as possible. In the case of the city with the most inhabitants, individuals might for example rely on a heuristic of what is the most well-known city in Germany. For this it is essential that individuals can rely on their previous knowledge of the environment. In terms of Simon’s scissors the Gigerenzer approach is primarily interested in how individuals are able to exploit the environment to their advantage relying on heuristics. These heuristics mediate between the limited cognitive abilities of the individual and the structure of the environment. Changing the environment in this approach can only mean studying a different real-world situation, but generally variation in the environment has not been important in this research program. Instead the focus has been on different heuristics and their appropriateness for particular settings. There is another distinguishing feature of the Gigerenzer approach that is essential to highlight. In Simon’s quote above reference was made to the ‘task structure of the environment’. In Vernon Smith’s experiments the task is usually to maximize the personal gains from trade. Within the experiment this is a relatively particular task, but more broadly economists believe that such exchanges enable individuals to pursue their heterogeneous goals, and that is how Vernon Smith interprets his findings. In Gigerenzer’s approach on the other hand the task structure is far more specific, and there typically is some ‘right’ or ‘correct’ answer, that is the task environment is close-ended. Whereas the institutional structure in Vernon Smith’s experiments limits what individuals can do to achieve their open-ended aims, in Gigerenzer the task structure limits what individuals are trying to achieve. In the next section we will trace the origins of these differences. In this section we have demonstrated that the behavioral economics program and the two different ER programs deal differently with the two blades of Simon’s scissors. These differences reflect differences in the psychological foundations of their theories, and differences in the economic foundations of their theories, as we will demonstrate in the next two sections.

2. An alternative psychology? We will at this point take a step back in order to investigate the psychological underpinnings of the two views of ecological rationality. As we can understand by now, cognitive psychology could seem such a natural fit precisely because it matched the way in which the individual had come to be represented in neoclassical economics. But like economics, psychology is also not completely theoretically united, and Edwards (2016) argues that economists, in their attempts to present reconstructions of the history of economics and psychology, tend to be “particularly loose concerning psychology”, and that “their reconstructions give partial accounts” of some terms “while quite overlooking” some others (Edwards 2016, 174). In this section we will argue that, in order to understand the arguments about ecological rationality, we need to delve deeper into different psychological approaches where the environment plays an important role. We begin with cognitivism’s focus on the cognitive processes, and within that the role of Herbert Simon’s scissors for the notion of bounded rationality; continue with functionalism’s focus on action, perception and adaptation, and the somewhat silent influence of Egon Brunswik on Gigerenzer’s thought; as an interlude, we present the ‘extreme’ account of James Gibson, mostly because of a curious coincidence that Vernon Smith borrowed his constructivist-ecological dichotomy from a paper written in this tradition; and round it up with considering the appropriate psychological foundation for Vernon Smith, which is arguably close to the recent developments in the field generally known as embodied cognition that seeks to conceptualize the environment as being part of the cognitive system.

2.1 Cognitivism and Herbert Simon The strand of psychology that has had the biggest influence on economics is cognitivism, which has its roots in the so-called cognitive revolution that happened when psychologists started to focus on the actual mental processes within the human mind. That being said, it mainly came about as a stark response to behaviorism that limited its study to the observed behavior. Cognitivists rejected the behaviorist idea of the human mind as a black box and embraced interdisciplinary study of the mind processes that involved psychology, artificial intelligence, and computer science. This resulted in adopting the computer as the central metaphor for the mind, and study focus shifted from the stimulus-response kind of experiments to the study of computer-like features such as memory capacity and computational abilities. In other words, the mind became to be perceived as a general- purpose computer whose underlying principle is to process information. As such, it lent itself nicely to adoption by the economists who sought more realistic foundations for their theories of economic behavior and decision-making, which had come under attack for its unrealistic assumptions about the rational abilities of the economic actors. This eventually resulted in an almost exclusive focus on the workings of an individual mind, while regarding its environment as something external and utterly separated. Thus, what emerged was an image of the mind as a calculating machine whose working gets corrupted by its own ‘flawed’ structure and the environmental influences into a variety of cognitive biases. For example, Kahneman and Tversky, together with the researchers that followed in their footsteps, famously argued that human mind is not adapted to probabilistic thinking, therefore the tasks involving probabilities will inevitably lead to biased and wrong conclusions when compared to the ‘ideal’ solutions suggested by the principles of logic and maximization. Herbert Simon’s initial target, the expected utility theory, was thus saved by being reincorporated as a benchmark for assessing the (ir)rationality of the choices, and cognitivism and standard economic theory have become close allies ever since. As indicated, Simon’s work is in economics most commonly associated with bounded rationality as being shaped by the two blades of the scissors—cognitive abilities and the structure of the environment. However, the legacy of his thought within the discipline is not unambiguous, and not even Simon himself did always pursue his scissors argument with full rigor. Realizing, with Sent (2004), that environmental considerations gradually diminished from Simon's core thought, we adopt Petracca’s (2017) distinction between Simon1 and

Simon2, where the latter refers to the narrow view of bounded rationality being predominately the function of the individual’s cognitive limitations, and the former to a wider understanding that also entails the structure of the environment. This apparent conceptual inconsistency in Simon’s thought brings about a notable consequence for his legacy, since different schools had picked up on different takes on bounded rationality, which have resulted in two traditions that each build on either Simon1 or Simon2: while the latter was picked up by Kahneman and Tversky and further developed within the various offspring of their ‘heuristics and biases’ program; Gigerenzer’s ecological rationality subscribes to the former (Petracca 2017). Apart from that, it has also been suggested that the varied (mis)use of Simon’s insights is arguably a consequence of him “never quite [finishing] what he started (Sent 2005, 230), and thus leaving many gaps in his theory to be filled out by the next generation of scholars1. Therefore, the issue that his theory lends itself to competing interpretations requires a careful consideration of what part of his thought had been appropriated by a given author. Due to its explicit emphasis on the role of the environment, it is the earlier, broader, version of bounded rationality (i.e. Simon1) that is of interest to us. However, this version needs further contextualization as well, since its genesis was spread out over two papers that had subsequently received a rather unequal amount of attention: the 1955 paper, which is concerned only with ”inquiring into the properties of the choosing organism”, followed in 1956 by a “sequel [that deals with] characteristics of the environment and the interrelations of environment and organism” (Simon 1955, 100). Despite the fact that, as Sent (2005) points out, it is the first paper that those economists wanting to refer to bounded rationality predominately choose for citation, to fully understand Simon’s argument about the role of the environment we must draw on the second one. Even though bounded rationality is usually being perceived as the more realistic account of human decision-making and behavior, there is little “reality” in the notion of the environment that Simon sets forth. His environment is stylized, abstracted, described by the patterns that make up its informational structure. It is not the actual world outside that Simon incorporates in the model. Instead, his focus is on what components does environment consist of and what consequences will those have for the purposive behavior of agents. His aim is to model behavior, where the environment has a role to specify, on the one hand, constraints for that behavior, and on the other, environmental situations that will call for specific strategies to be employed by the agent, depending on the particular behavioral goal and the aspirational levels that the organism sets for herself. In this way he

1 “Herbert Simon left us with an unfinished task, a theory of bounded rationality.” (Gigerenzer 2016, 34) can slowly add more complexity to the situation, which enables him to effectively make his case. However, the model remains far from the real world. Similarly, Mirowski (2002) argues that Simon's bounded rationality should not be understood as an attempt to construct a more realistic account of ”how people really think”; rather, it was a necessary tool that Simon had to devise in pursuing his main goal—that of (computer) simulation. Seen from this perspective, it becomes clear why the environment was conceptualized as patterned information, since this is the way the computer ”perceives” it, and computer programs are designed to work with and exploit these patterns. Bounded rationality was conceived in the context of contemplating the design of a general problem- solving program (Simon and Newell completed the first one—Logic Theorist—in 1956)2, which was in Simon's view best approached as a simulation of the way human mind works. This eventually led him to proclaim: “Computers were made in the image of humans” (Simon and Kaplan 1989, 40); albeit, we can add, a particular kind of image. So, as a consequence, the components of the environment got to be defined and conceptualized in a way that was translatable to a computer code. In the same vein, the notion of simulation brings about a crucial consequence also for how the organism is conceptualized. Namely, it is not the ”real” organism that is in focus here, but rather an organism as a collection of characteristics that get discovered by cognitive science. Thus, the organism, just as the environment, is perceived in analytical informational terms as well. Based on these characteristics, Simon would then model the behavior of the organism as it pursues the ‘satisficing’ aspiration levels of the specified behavioral goals, while adapting to the structure of the environment. The focus on simulation (by means of modeling cognitive processes) got coupled in Simon’s theory with his view of the mind as a physical symbol system (i.e. mind as a manipulator of objective representational symbols that get mediated to it from the environment), and that became the bedrock of his version of cognitivism (see Petracca, 2017,

2 To argue for this point we only need to consider a footnote concerning the title of his (1956) paper, where Simon expresses his indebtedness to Newell for ”numerous enlightening conversations on the subject of this paper” (p. 129). Due to the nature of their work together, it is probably not too audacious to assume that the context of these conversations had little to do with genuine questions of human psychology. Furthermore, his concern with ”how to introduce multiple goals with a minimum complication of the process” (p. 133) reads more like a concern over saving computer power than concern about understanding human nature. p. 24) In other words, Simon developed a cognitivist view that sought to incorporate the informational structure of the environment as a crucial factor for the working of the mind. Reading, processing and adaptation to the perceived information about this structure, based on the particular aspiration levels, resolved the optimization problem, which, in his view, made his theory ‘realistic’. However, by focusing on the cognitive processes and information processing he somewhat neglected the study of the more general traits of observed human behavior. More generally in modern behavioral economics we see the continuation of the strong split between individual and the environment (Simon’s two blades of the scissors), an emphasis on the perfect computer or the perfectly rational individual as the relevant benchmark from which the ‘flawed’ individual deviates, and thirdly artificial environments which are conceptualized as information patterns and more generally derived from theory, rather than stylized after real-world environments.

2.2 Functionalism and Egon Brunswik In general, functionalism is a “doctrine that what makes something a mental state of a particular type does not depend on its internal constitution, but rather on the way it functions, or the role it plays, in the system of which it is a part” (Levin 2016). As such, it can be contrasted to the cognitivist view as described above, which is mainly concerned with defining and describing the processes of the mind. Instead, functionalism’s main explanatory purpose is behavior as expressed in action. In psychology, it is a nowadays a largely defunct tradition that historically developed in opposition to structuralism that focused on the study of mental states through introspection (see e.g. Titchener, 1898), and is ultimately related to behaviorism, whose dominance was eventually overthrown during the cognitive revolution. Initially, it was in its American version defined as “the study of the adaptive role of mental animal life and behavior” (Edwards 2016, 180), while the European strand, as advanced by Karl Bühler, was, similarly, based on the notion that “any given stimulus … will be perceived differently when placed against a different contextual background” (Leary 1987, 118). Therefore, it clearly evokes the arguments about the crucial role of the (physical) environment for the adaptive behavior of organisms. This tradition was in the 1930s picked up by Egon Brunswik, whose work, we observe, has had a significant influence on Gigerenzer’s understanding of ecological rationality. Brunswik’s contributions can be broadly divided in theoretical and methodological part, both of which have been in important ways influential for Gigerenzer’s argument. The former concerns his view of perception as being in its nature probabilistic, while the latter entails his methodological insistence on the necessity of representative sampling of environmental conditions. By describing Brunswik’s theory in more detail, we wish to demonstrate its resemblance to Gigerenzer’s, and thus show that it is as such more relevant than Simon for understanding his notion of ecological rationality. For Brunswik, psychology is ”primarily concerned with the methods of response of the organism to two characteristic features of the environment”. Firstly, the ”environment is a causal texture in which different events are regularly dependent upon each other ... and because of the presence of such causal couplings ... organisms come to accept one event as a local representative for another event”; and secondly, ”such causal couplings are probably always to some degree equivocal.”, (Tolman & Brunswik, 1935, pp. 43–44, italics in original)3. In other words, environment is characterized as a probabilistic texture. By this he means that the environmental cues are entangled (i.e. textured), and that in the process of perception it is never determined which cue will be used for inference (therefore probabilistic). In his terminology, cues function vicariously and he demonstrated this with his lens model of perception (see e.g. Brunswik, 1955). On the diverging side, the model starts with the distal object whose characteristics get mediated to the organism through various environmental cues; on the converging side the cues get utilized by the organism. Since these cues function vicariously, the key to a successful utilization is always a certain level of redundancy in the environmental texture of the cues. It is important to stress that in Brunswik's view utilization is only a part of a wider process of cognitive achievement, which is ”the over-all correspondence between a certain distal and a certain central variable, so that the former could be considered successfully mapped into the latter” (Brunswik 2001, 301–2). In other words, what characterizes the achievement is the level of correspondence between the object and its perception by the

3 See (Brunswik 2001) for a collection of all of the important Brunswik’s writing. organism. This also constitutes Brunswik's answer to “the problem of cognition, which [may be defined] as the problem of the acquisition of knowledge” (ibid., p. 301). Such a view is opposed to the prevalent one in psychology which, in his opinion, “has forgotten that it is a science of organism-environment relationships, and has become a science of the organism” (ibid., p. 300-301)4. As he writes later on in the paper: “Only by detailed analysis of ecological textures can the cognitive problem be restored from mere utilization problems to its full scope of achievement problems and thus again become the key to the core question of psychology, that of the adjustment of the organism to a complex environment” (ibid., p. 312). Achievement, the result of our actions, is thus the most important aspect for Brunswik. Cognition is a means toward achievement, but achievement can also be ‘achieved’ by making use of the environment, or indeed it might result from selection pressures in the environment. Another contribution we want to point out is the notion of ecological validity that plays a crucial role in the process of cognitive achievement. Despite the fact that the term was originated by Brunswik to refer to the degree of correspondence between an object and the organism's perception (2001) and is as such a property of a cue, it has since had a life of its own and is today used to denote the degree to which the experimental setting resembles the real-life. This is rather ironic, since the main methodological innovation that Brunswik proposed was that of representative design, which implies that, in order for the experiment to be generalizable, not only the sample of the participants should be representative (as is usual in the experimental social sciences) but also the conditions themselves. Brunswik was a fierce critic of the clinical design of psychological experiments, because they either removed or radically simplified the environment. In other words it restricted the amount of cues which the individual could choose, and thus failed to study how individuals used their environment.5

4 The quote in its extended form is even pointier: ”If there is anything that still ails psychology in general, and the psychology of cognition specifically, it is the neglect of investigation of environmental or ecological texture in favor of that of the texture of organismic structures and processes. Both historically and systematically psychology has forgotten that it is a science of organism-environment relationships, and has become a science of the organism.” It is important to notice that this passage appears to provide much inspiration for the researchers in the ”ecological rationality” program and has been cited numerously throughout their work (see e.g. (Gigerenzer and Todd 1999). 5 Therefore, ecological validity as used today (as a property of the experimental setting) is somewhat anti- Brunswik distinguishes between two cognitive processes: intuitive perception and analytical reasoning. While the former is characterized by channeled mediation of cues (as described in the lens model), resulting in normal distribution of error (provided perception happens in an environment with a sufficient amount of vicariously functioning cues), and is uncertainty-geared; the latter is characterized by thinking, resulting in precision with scattered error, and is certainty-geared. What more successfully guides behavior is ‘intuitive’ perception, not reasoning and thinking. In this sense, his work also evokes the more recent findings in evolutionary psychology with regard to ‘reasoning instincts’ being “better than rational; that is, they can arrive at successful outcomes that canonical general-purpose rational methods can at best not arrive at as efficiently, and more commonly cannot arrive at all” (Cosmides and Tooby 1994, 329). Brunswik described the relationship between organism and environment as being similar to a married couple, where “as in any marriage, the interrelationship between the two systems has the essential characteristic of a “coming- to-terms” ” (Brunswik 2001, 300). This is a much different vision from Simon’s scissors that do not imply any coming-to-terms between the two, but working together to effectively cut through the problem-space. However, his views lent much worse to the kind of problems that other social sciences, in particular economics, address. The achievement he refers to is a strictly perceptual achievement in the sense of visually ‘reading’ the world, and behavioral achievement in the sense of acting on that reading. Since his theory avoids the problems of symbolic processing, it is not compatible with the usual problems of symbolic cognition. One thing is to infer the size of the rock, and another thing altogether to strategize about the prices of copper, or trust the stranger to deliver what promised. In this regard, the theory is not readily applicable to economic problems. However, it does point to a direction that enables one to imagine a different perspective on the problems of decision making from the one implied by the individualistic cognitive psychology. In particular, the settings characterized by strong uncertainty might be better approached in a more ‘intuitive’ way. It is our claim that

Brunswik, since it is applied to a single experiment without a representative sample of conditions where this experiment would be performed. Truth be told, not even Brunswik himself ever performed a fully brunswikian study, and this might probably be one of the main reasons why his methodological thought has had such limited legacy (see Leary, 1987). Gigerenzer’s theory is built on this realization coming from functionalistic psychology. To sum up, we aimed to show that an important alternative to cognitivism, that we need to consider when trying to understand Gigerenzer’s psychological underpinnings, is functionalism. That is, the focus should not be exclusively on the (computational) mental processes, but rather on the functions these processes have when dealing with the environment. While many of them are illusively referred to as ‘intuitive’, it is more important to study what is vital for their success, than exact causal mechanism of their performance. One of the proponents of functionalism, and an important influence on Gigerenzer, is Brunswik, who resolved this dilemma by characterizing the perception of cues as probabilistic. Therefore, while it is not possible to specify precisely how perception works, it is nevertheless crucial that the experimentation happens in the real complex environment where the subject can make use of all the possible cues from that environment in order to achieve greater correspondence between the real world and inferences about it. The rather extensive overview of the Brunswik’s key concepts serves as a demonstration of the extent to which they resemble those of Gigerenzer. The key ones are: the cognitive problem to be resolved is achievement; the focus of analysis is on (intuitive) perception; the main criterion of success is correspondence with the ‘real world; the structure of the environment is probabilistic texture of cues; and the psychological experiments should not strive to isolate a single variable, but be as close to the real environment as possible. As pointed out, Simon’s scissors argument is mostly referred to as the key influence for both bounded rationality and ecological rationality. However, we show that Brunswik is, while mostly being recognized as an important inspiration but not vital part of the theory, far more prominent than acknowledged for Gigerenzer’s ecological rationality perspective. In order to contrast Simon’s theory from Brunswik’s we present the key insights from both in the following table:

Brunswik Simon Cognitive problem achievement computation & choice What guides behavior perception reasoning Success criterion correspondence aspiration levels Structure of the causal texture of cues information pattern environment Organism human being bundle of characteristics What is modeled perception behavior Guiding metaphor married couple scissors ”Action verb” coming-to-terms Satisficing? Intuition yes no

2.3 Gibson and Brunswik’s legacy While Simon’s legacy is well established, Brunswik’s influence is less prominent. Interestingly, Leary even provides Simon’s view on the matter: ”Brunswik was a forerunner of a number of developments, but did not, in fact, have much influence on what developed” (1987, 131). The strand in psychology that did developed along the lines that also Brunswik was pursuing was the so-called ecological psychology by James Gibson. Gibson when pointing out the up-till-then apparent lack of theoretical concepts on which to base environmental thought in psychology, remarks: ”A few psychologists, such as Brunswik (…) have moved in this direction, but none has ended with the sort of theory being put forward here” (Gibson 1979, xv). Gibson thus saw himself as having developed what Brunswik had only indicated. However, this direction was arguably not what Brunswik would have taken, and in the ecological rationality circle Gibson’s theoretical position has been perceived as an ”extreme statement that only the environment need be studied, not the mechanisms of the mind” (Gigerenzer and Todd 1999, 13). This relates to the Gibson’s notion of direct perception, which implies that “the ‘values’ and ‘meanings’ of things in the environment can be directly perceived” (Gibson 1979, 119). As a consequence, the cognitive representations are denied the primary role in the process that Gibson also refers to as information pickup. Cognition does not play a role in this process, and so the focus must shift exclusively to the study of the environment. This legacy has a direct influence on the other subject of this paper Vernon Smith. Smith distinguishes between constructivist and ecological rationality. It is a distinction he learns from Norman6. And it is Norman who develops the distinction in an attempt to differentiate between cognitivism and the psychology of Gibson. Norman explains that

“[the differences between the two approaches] relate to two interrelated topics, the richness of the stimulation reaching our sensory apparatus, and the involvement of “higher” mental processes in the apprehension of our environment (…) [T] he two approaches differ on the aspects of perception they emphasize; the constructivists excel at analyzing the processes and mechanisms underlying perception, while the ecological approach excels at the analysis of the stimulation reaching the observer.” (Norman 2002, 74)

Norman attempts to reconcile the two views by understanding perception as encompassing “both conscious and unconscious effects of sensory stimulation on behavior” (Norman 2002, 73)Reasoning and intuitive perception are thus seen as each contributing its own separate part to the whole process of cognition and acting. In order for such a theory to work there has to be some mediation between the environment and the individual. A point that is also developed by Vernon Smith when he argues that constructivist and ecological rational orders “interact daily in ordinary human interaction” (Smith 2008, 24). This process of interaction is what distinguishes the ecological rationality approaches from the behavioral economics approach in which there is a much stronger separation between the individual (the processor of information), and the environment (the information).

2.4 Embodied and Situated Cognition This approach of situating cognition and action between the individual and the environment, or in the interaction between the two is a program that in psychology is known as embodied cognition (Wilson 2002; Clark 1996; Robbins and Aydede 2009). This program emerged as a

6 Understanding this influence we consider crucial for understanding Smith’s theory, bearing in mind his own view on getting inspiration outside of economics: “I importune students to read narrowly within economics, but widely in science. … The economic literature is not the best place to find new inspiration beyond [the] traditional technical methods of modeling.” (Smith 2003, 471) critique of cognitivism and its emphasis on the mind, and seek to reconnect the mind and its environment. Some of its central claims are summed up by Wilson (2002) as (1) cognition is situated, (2) cognition is time pressured, (3) we off-load cognitive work onto the environment, (4) the environment is part of the cognitive system, (5) cognition is for action, (6) off-line cognition is body based. Environment as an active factor explicitly features in Clark and Chamlers’s (1998) notion of active externalism. In their account the environment plays an active role in driving the cognitive processes. As Menary (2006) puts it: ”The organism becomes a cognitive agent by being coupled to the external environment” (Menary 2006, 75). The account coming from this field that is perhaps most known to economists (see e.g. Davis, 2011; Ross, 2014) is the one developed in Clark (Clark 1996). It involves conceptualization of external environment as providing “cognitive scaffolding” that guides reasoning and behavior. This lends itself well to application to the social and institutional environment, where it is rules and norms that take the roles of scaffolds. Thus, to infer from the study of isolated activity as exemplified by a laboratory environment is to miss the point that we are embedded in a social world where our behavior and cognition is guided by numerous scaffolds, from formal to informal ones. This interpretation of the environment as a rich cognitive structure is present in Vernon Smith’s work on ecological rationality. It argues that the social norms, behavior of others, and emergent phenomena such as market prices guide reasoning and behavior. But as we will see below, Vernon Smith sometimes also relies on the strong environmental perspective that we found in Gibson.

3. Two types of ecological rationality Both Gerd Gigerenzer and Vernon Smith use the notion of ‘ecological rationality’ but we are now in a position to disentangle the different origins of their ideas, and to understand the differences between the two authors. Gigerenzer’s idea of ecological rationality, has a relatively direct line, as he himself acknowledges, to the functionalism of Brunswik’s psychology. As a consequence of this intellectual heritage Gigerenzer and those working in his ABC-program focus on the ability of individuals to perform highly specific tasks: “name the capital of Germany, catch a ball, which has more cholesterol cake or pie?” (Gigerenzer, 2008). What Gigerenzer is particularly interested in is how the individual uses heuristics to perform specific tasks, these heuristics are part of what he calls the natural environment. This means that he has less interest in social tasks, although he occasionally touches on them, and has very little to say about institutional environments, or what we might call the cultural, that is non-natural part of the environment. That is different for Vernon Smith, whose intellectual tradition is far less focused on the individual and more on the institutional environment. So when he discusses the challenges of psychology to economics, he discusses how risk-preference is dependent on the institutional context, and how some psychological effects, such as the ‘endowment effect’ tend to disappear when the institutional pressure (of markets) is sufficiently strong. But most importantly Vernon Smith does not try to locate the rationality on the level of the individual. His early class-room experiments tested the convergence of the observed prices in his experiment to the predicted market equilibrium price, rationality was the outcome of a process of interactions in a system guided by certain rules, not a property of the acts of the individual. For him rationality is to be found at the level of system, like it is for the modern theorists of embodied cognition. When we use Norman’s difference between constructivist and ecological approaches, we can see that both Gigerenzer and Vernon Smith end up on the ecological side. Behavioral economics as practiced by Kahnemann and Tversky excels “at analyzing the processes and mechanisms underlying perception”, that is they excel at studying the cognitive processes. Gigerenzer and Vernon Smith, on the other hand, excel at the analysis of the stimuli reaching the observer. They, however, differ radically in how they think of those stimuli. Gigerenzer emphasizes the natural environment in which individual decision making takes place. Vernon Smith emphasizes the cultural context in which social interaction takes place. Following the Brunswik approach to achievement, Gigerenzer mainly studies achievement, and thus is interested whether people make the right choices, but this means that within that environment a clearly ‘correct’ or ‘right’ answer is present. Gigerenzer in that sense embraces the idea that the primary object of study is that of ‘decision-making’. In that decision-making we might rely on the observable behavior of others, which he conceptualizes as social rationality, but what we seek to explain is individual decision- making, in which everything but the individual is in a fundamental sense ‘the environment’. As Gigerenzer and argues, ecological rationality is the ability: “to exploit the structure of the information in natural environments” (Gigerenzer and Todd 1999, 24). Individuals use that environment to make choices, but they do not interact with it. That is different for Vernon Smith whose experiments have always contained interaction, and attempted to resemble institutional market settings, rather than individual decision-making settings. The outcomes of these experiments are typically market prices, rather than individual choices which have to be explained. Individual choice is here not the primary thing to be explained, instead we are interested in the convergence of price to some equilibrium price. The interaction provides information to the individual, is my offer rejected or accepted? Is the price I am asking too high? There is not a clearly ‘correct’ or ‘right’ answer, but instead a range of prices at which trade can take place. The experimental setting is designed so as to replicate essential features of the market economy, but not the complete causal texture that Brunswik and Gigerenzer are interested in. In that sense Vernon Smith’s work is in line with the work of Simon, it is concerned with the informational patterns of the environment which allow the individual to make decisions. This informational pattern, however, is distinctly non-natural in Vernon Smith’s work. And it is also not, as it in Simon and in modern behavioral economics, thought of as a raw data on which computing by the mind has to take place. Instead his work studies the “interactive experience in social and economic institutions” (Smith 1991, 878). His work criticizes the psychologists for having studied cognitive processes in the absence of social contexts, a criticism that is familiar from the work of Brunswik. But where Brunswik’s work treats the environment as given, the social interaction within institutional frameworks gives rise to a richer environment for Vernon Smith. Thus he talks of traditions as ‘encoding the social learning’ which has taken place by experimentation in the past (Smith 1991, 892). Environments are thus cultural products and the outcome of earlier social processes, and experiments which prevent interaction and learning from taking place wrongfully find violations of rationality Smith argues. Once it is recognized that the environment is not simply out there, but is interpreted and changed by individuals, it becomes impossible to think of them as natural environments. Instead Smith argues, many institutional settings have arisen precisely because they are well suited to achieve certain goals. When we drive on the highway street signs are helping us find our destination and indicate other rules which we should follow. And these signs are often designed such that they require minimal cognitive activity from the individual. It is precisely this feature of markets that Hayek has emphasized in classic article on the use of knowledge (Hayek 1945). In that article he compares market prices to a communication system, which helps individuals navigate their environment. Soon after making that argument he quotes Alfred Whitehead: “Civilization advances by extending the number of important operations which we can perform without thinking about them”. His argument, often reduced to the claim that knowledge is dispersed among different individuals, is much more sophisticated, and it is a point in recent years rediscovered in psychology under the banner of ‘distributed cognition’. Hayek argues that prices act like street signs in directing economic action, and are thus guides for behaviour. This point is wholly neglected in Gigerenzer’s account of fast and frugal heuristics found in the natural environment, and related to specific tasks. In Hayek’s account and in Vernon Smith’s account the tasks are not as specific and the environment is not natural: it is social, political, economic and cultural. Markets are used to achieve a wide variety of goals, but it is precisely this flexibility that makes the institutional structure of prices so important. One could compare this to street signs which do not guide us to our destination, but which are designed so that a wide variety of individuals with different goals are able to reach their destination. Or indeed it can be compared to language, as Hayek likes to do, because it can be used for so many different purposes. If one takes this perspective to its extreme the environment contains the rationality, and that is indeed what some economists have suggested. Famous are the experiments by Gode and Sunder (1993) who simulate a market in which the only market rationality required of agents – or achievable for their simulated ‘zero-intelligence traders’ – is that they not bid beyond their budget constraints. They explicitly suggest that the institutional structure of markets is a substitute for individual rationality. That argument is predated by Alchian who argued that “there may have been no motivated individual adapting but, instead, only environmental adopting” (Alchian 1950, 214). And it is the subject of an exchange between Gary Becker and Israel Kirzner on the extent to which conscious adjustment of pricing by entrepreneurs is required (Kirzner 1962; Kirzner 1963; Becker 1963). In fact Vernon Smith was criticized by others that the way his initial experiments were set-up was so institutionally constrained that convergence to the equilibrium price was inevitable. Vernon Smith comes quite close to adopting a view of rationality that places it, like Gibson does, entirely in the environment. His definition of ecological rationality emphasizes: “home grown principles of action, norms, traditions, and ‘morality’”. That is, he argues, that particular patterns of behavior are selected by market institutions, and his view on rules of thumb, norms, traditions and morality is that they are also the outcome of processes of selection. In a particularly strong statement to this effect he states: “The current manager does not know about opportunity cost or even why the policy is what it is; only that he learned it from the last manager. He is an instrument of the ‘law’ of one price in a market” (Smith 1991, 892). This is not the emphasis on intuition and ‘gut feelings and the intelligence of the unconscious’ that we find in Gigerenzer (2007) and that originates in Brunswik, but it emphasizes the intelligence of the institutional context, and the extent to which cognitive work is ‘off-loaded’ to the environment. This leads us to ask not how we could overcome particular cognitive biases on the side of the individual, but instead to look for particular institutional environments which have a high selection pressure, or somewhat more positively for those environments that allow individuals to engage in social learning. Prices are an important guide in this, but as Vernon Smith argues in later studies on norms of fairness, and as the Gigerenzer points out with his wide spectrum of heuristics are certainly not restricted to those. It is relatively straightforward to see that Gigerenzer is influenced deeply by functionalist psychology and in particular by the work of Egon Brunswik, especially since Gigerenzer explicitly acknowledges this. But in the case of Vernon Smith it is tempting to interpret his work as an attempt to keep psychology out of economics. On this view we see some influence of evolutionary thinking, and thus of biology, but not of psychology. It is a suggestion implicitly made by Vernon Smith when he calls his own approach experimental economics, as distinct from experimental psychology. And is the argument put forward by Don Ross, who longs back to a period in which economics was still dismissive of psychology to think that this is all economics and not psychology (Ross 2014). He demonstrates that there is a rich tradition of economists who have argued that economics can and should do without psychology. Psychology was never absent from economics — it was only when behaviorism in psychology was dominant, that it could look that way for a while. But it also seems curious to not seek convergence between fields, which are after all both studying human decision-making. We are now in a position to see that psychology is also not absent from Vernon Smith’s concept of ecological rationality. But, his emphasis on the rationality of market institutions, rather than individual choosers, does fundamentally shift the type of psychology to which we find links. It is the psychology of the modern perspective on embodied cognition, which studies rationality not as a property of individual decisions, but of the ‘system’, that is the group of individuals and the cultural environment as a whole. Furthermore Hayek, on who, Vernon Smith draws so heavily was himself engaged in serious psychological inquiries, that psychology was highly interactionist and sought to distance itself from views of the brain as possessing a structure independent of the environment and the experiences of the individual (Hayek 1952). As Wilson argues in her overview of the ‘embodied’ cognition framework, this leads us to study ‘cognitive systems’ that consists of multiple individuals and an institutional context (Wilson 2002). This emphasis on cognitive systems is an integrative account of the the individual and the environment as opposed to the separationist account that we find in the behavioral economics program. The rationality of markets then does not result from the cognitive capacities of the individuals, but instead from the extent to which learning can take place in a particular institutional environment and the extent to which individuals have been able to off-load cognitive work onto the environment precisely by altering or enriching that environment, as in our example of the street signs. That moreover shows, that economics has lots to teach psychology, and contains a rich tradition of studying particular institutional environments and the social interactions that take place in it (i.e. it already contains lots of knowledge about particular ‘cognitive systems’).

Conclusion In this paper we have argued the following points  Although Gigerenzer and those associated with his research program and Vernon Smith and his allies both use the term ‘ecological rationality’ we have demonstrated that they use it in fundamentally different ways  Gigernzer’s perspective emphasizes the importance of heuristics which help the individual guide himself through natural environment which involve specific tasks. Vernon Smith’s perspective emphasizes the importance of the (emergent) cognitive properties of the environment which guide the individual and operates through selection, resulting in rational or reasonably good outcomes at the ‘system’ level.  Their concepts of ecological rationality originate from different psychological traditions and influences. Gigerenzer’s ecological rationality is rooted in the functionalist psychology of Egon Brunswik, Vernon Smith’s ecological rationality on the other hand is closer to modern views in psychology on situated cognition and Smith’s account originates in part from a more institutional economic tradition.  In particular Vernon Smith’s account of ecological rationality has serious implications for how we think about and conceptualize rationality in economics, because an implication of his thinking is that rationality is to be found at the ‘system’-level rather than at the individual level.  These alternatives to the behavioral economics program are rival combinations of psychology and economics, and not attempts to do economics without psychology, as has sometimes been argued.  In order to study the relative usefulness of different types of psychological perspectives we have argued it is important to understand the rival schools in psychology. This is in stark contrast to the way in which the behavioral economics program has presented its own history, which suggests that they have combined psychology with economics.  We have also hinted at the fact that in the social sciences are in need of a less cognitive psychology, which recognizes the importance of the social and institutional environment for human behavior.

Bibliography Alchian, Armen A. 1950. “Uncertainty, Evolution, and Economic Theory.” Journal of Political Economy 58: 211–21. Angner, Erik, and George Loewenstein. 2012. “Behavioral Economics.” In Philosophy of Economics, edited by Uskali Maki, 641–90. Amsterdam: North Holland. Becker, Gary S. 1963. “Rational Action and Economic Theory: A Reply to I . Kirzner.” Journal of Political Economy 71: 82–83. Bruni, Luigino, and Robert Sugden. 2007. “The Road Not Taken: How Psychology Was Removed from Economics , and How It Might Be Brought Back.” The Economic Journal 117: 146–73. Brunswik, Egon. 1955. “Representative Design and Probability Theory in a Functional Psychology.” Psychological Review 62 (3): 193–217. ———. 2001. The Essential Brunswik : Beginnings, Explications, Applications. Edited by Kenneth R Hammond and Thomas R Stewart. Oxford, UK: Oxford University Press. Camerer, Colin F, George Loewenstein, and Matthew Rabin. 2004. Advances in Behavioral Economics. Princeton, NJ: Princeton University Press. Clark, Andy. 1996. Being There: Putting Brain, Mind, and World Together Again. MIT Press. Clark, Andy, and David Chalmers. 1998. “The Extended Mind.” Analysis 58 (1): 7–19. Conlisk, John. 1996. “Why Bounded Rationality ?” Journal of Economic Literature 34: 669– 700. Cosmides, Leda, and John Tooby. 1994. “Better than Rational: Evolutionary Psychology and the Invisible Hand.” The American Economic Review: Papers and Proceedings 84: 327– 32. Davis, John B. 2011. Individuals and Identity in Economics. Cambridge: Cambridge University Press. Edwards, José. 2016. “Behaviorism and Control in the History of Economics and Psychology.” History of Political Economy 48 (annual supplement): 170–97. doi:10.1215/00182702-3619262. Gibson, James J. 1979. The Ecological Approach to Visual Perception. New York: Psychology Press. Gigerenzer, Gerd. 2000. Adaptive Thinking: Rationality in the Real World. Oxford, UK: Oxford University Press. ———. 2007. Gut Feelings: The Intelligence of the Unconscious. New York: Viking. ———. 2015. Simply Rational: Decision Making in the Real World. Oxford, UK: Oxford University Press. ———. 2016. “Towards a Rational Theory of Heuristics.” In Mind, Models, and Milieux, edited by Roger Frantz and Leslie Marsh, 34–59. Palgrave Macmillan. Gigerenzer, Gerd, and Peter M Todd. 1999. Simple Heuristics That Make Us Smart. Oxford, UK: Oxford University Press. Gode, Dhananjay K, and Shyam Sunder. 1993. “Allocative Efficiency of Markets with Zero- Intelligence Traders: Market as a Partial Substitute for Individual Rationality.” Journal of Political Economy 101: 119–37. Grüne-Yanoff, Till, Caterina Marchionni, and Ivan Moscati. 2014. “Introduction: Methodologies of Bounded Rationality.” Journal of Economic Methodology 21. Taylor & Francis: 325–42. doi:10.1080/1350178X.2014.972140. Hands, D Wade. 2010. “Economics, Psychology and the History of Consumer Choice Theory.” Cambridge Journal of Economics 34 (4): 633–48. doi:10.1093/cje/bep045. Hayek, Friedrich A. 1945. “The Use of Knowledge in Society.” The American Economic Review 35 (4). American Economic Association: 519–30. ———. 1952. The Sensory Order: An Inquire into the Foundations of Theoretical Psychology. Chicago: Chicago University Press. doi:10.1017/CBO9781107415324.004. Kahneman, Daniel. 2003. “Maps of Bounded Rationality: Psychology for Behavioral Economics.” American Economic Review 93: 1449–75. Kirzner, Israel M. 1962. “Rational Action and Economic Theory.” Journal of Political Economy 70: 380–85. ———. 1963. “Rational Action and Economic Theory: Rejoinder.” Journal of Political Economy 71: 84–85. Leary, David E. 1987. “From Act Psychology to Problematic Functionalism : The Place of Egon Brunswik in the History of Psychology.” In Psychology in Twentieth-Century Thought and Society, edited by Mitchell G Ash and William R Woodard, 115–42. Cambridge, MA: Cambridge University Press. Levin, Janet. 2016. “Functionalism.” The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/win2016/entries/functionalism/. Menary, Richard. 2006. “Attacking the Bounds of Cognition.” Philosophical Psychology 19 (3): 329–44. doi:10.1080/09515080600690557. Mirowski, Philip. 2002. Machine Dreams: Economics Becomes a Cyborg Science. Cambridge, UK: Cambridge University Press. Mullainathan, Sendhil, and Richard H Thaler. 2000. “Behavioral Economics.” 7948. NBER Working Paper Series. doi:JEL No. D0, H0, JO. Norman, Joel. 2002. “Two Visual Systems and Two Theories of Perception: An Attempt to Reconcile the Constructivist and Ecological Approaches.” Behavioral and Brain Sciences 25 (1): 73–144. doi:10.1017/S0140525X0200002X. Petracca, Enrico. 2017. “A Cognition Paradigm Clash: Simon, Situated Cognition and the Interpretation of Bounded Rationality.” Journal of Economic Methodology 24. Routledge: 20–40. doi:10.1080/1350178X.2017.1279742. Robbins, Philip, and Murat Aydede. 2009. The Cambridge Handbook of Situated Cognition. Cambridge, UK: Cambridge University Press. Ross, Don. 2014. “Psychological versus Economic Models of Bounded Rationality.” Journal of Economic Methodology 21: 411–27. doi:10.1080/1350178X.2014.965910. Sent, Esther-Mirjam. 2004. “The Legacy of Herbert Simon in Game Theory.” Journal of Economic Behavior and Organization 53 (3): 303–17. doi:10.1016/j.jebo.2002.10.006. ———. 2005. “Simplifying Herbert Simon.” History of Political Economy 37: 227–32. doi:10.1215/00182702-37-2-227. Sent, Esther-Mirjam, and Matthias Klaes. 2005. “A Conceptual History of the Emergence of Bounded Rationality.” History of Political Economy 37: 27–59. doi:10.1215/00182702- 37-1-27. Simon, Herbert A. 1955. “A Behavioral Model of Rational Choice.” The Quarterly Journal of Economics 69 (1): 99–118. ———. 1956. “Rational Choice and the Structure of the Environment.” Psychological Review 63 (2): 129–38. doi:10.1037/h0042769. ———. 1990. “Invariants of Human Behavior.” Annual Review of Psychology 41: 1–19. Simon, Herbert A, and Craig A Kaplan. 1989. “Foundations of Cognitive Science.” In Foundations of Cognitive Science, edited by Michael I Posner. Cambridge, MA: MIT Press. Smith, Vernon L. 1962. “An Experimental Study of Competitive Market Behavior.” The Journal of Political Economy 70: 111–37. doi:10.1086/258609. ———. 1991. “Rational Choice: The Contrast between Economics and Psychology.” Journal of Political Economy 99: 877–97. ———. 2003. “Constructivist and Ecological Rationality in Economics.” The American Economic Review 93: 465–508. ———. 2008. Rationality in Economics. Cambridge, UK: Cambridge University Press. Tolman, Edward C, and Egon Brunswik. 1935. “The Organism and the Causal Texture of the Environment.” Psychological Review 42 (1): 43–77. doi:10.1037/h0062156. Tversky, Amos, and Daniel Kahneman. 1974. “Judgment under Uncertainty: Heuristics and Biases.” Science 185 (4157): 1124–31. ———. 1986. “Rational Choice and the Framing of Decisions.” The Journal of Business 59: s251–78. Wilson, Margaret. 2002. “Six Views of Embodied Cognition.” Psychonomic Bulletin & Review 9: 625–36. doi:10.3758/BF03196322.