Good by default. Using heuristic-triggering nudges to promote prosocial behaviour in economic decision-making

Tesis de Máster , 2017 , 74 Páginas , Calificación: 1.3

Economía de las empresas - Negocios - General

Humans are generally perceived as intuitively selfish, particularly in their economic decision-making. Under the Dual-Process Framework (DPF), this view entails that prosocial behaviour requires reflective control over those natural inclinations towards self-interest. However, new lines of research explore the heuristic basis of prosocial actions and their potential use for policy-making. The goal of this thesis is to enquire into intuitive mechanisms which foster prosocial behaviour in economic decision-making and how to use them to promote altruism and cooperation through heuristic-triggering (“pure”) nudges. We found that while in some contexts prosocial choices can be economically rational, heuristic mechanisms also drive them and for some individuals, they do so to a higher extend than deliberative processes. Our literature review identified three main classes of heuristics that significantly drive prosocial behaviour: kin recognition, social and affect heuristics. Based on Intuitive Design principles, we developed a toolkit for heuristic-triggering nudges enabling choice architects to structure and describe choice options, as well as to address implementation issues, with the goal to make prosocial actions the most attractive alternative. One interesting highlight of our analysis is the differentiation of three master frames for evaluating prosocial nudges: increase in prosocial actions, ex-post satisfaction of decision-maker and social welfare outcome. Finally, we derived practical implications of our findings in social and environmental policy interventions.

Table of Content

List of Tables / Figures

1 Setting the Stage
1.1 Introduction
1.2 Relevant Definitions

2 Theoretical Framework
2.1 Decision – making under uncertainty
2.1.1 Rational Choice Theory
2.1.2 Bounded Rationality
2.1.3 Dual-process Theory
2.1.4 Heuristics and Decision Biases
2.2 Prosocial Behaviour
2.2.1 Social preferences
2.2.2 Altruism
2.2.3 Cooperation
2.3 Behavioural Economics and Policy Making
2.3.1 Incentives
2.3.2 Nudge Theory

3 Research Question and Methodology
3.1 Research Question
3.1.1 Formulation of Research Question
3.1.2 Hypotheses
3.1.3 Relevance of the topic
3.2 Methodology

4 Results
4.1 Link Between Prosocial Behaviour and Heuristics
4.1.1 Economic Rationality of Prosocial Behaviour
4.1.2 Heuristics Mechanisms Underlying Prosocial Behaviour
4.2 Use of heuristics to nudge prosocial behaviour
4.2.1 Structuring the choice
4.2.2 Describing the choice options
4.2.3 Addressing implementation issues

5 Conclusion
5.1 Key Findings
5.2 Organizational and practical implications
5.3 Discussion and future research

6 Appendix
6.1 Appendix I: Most common experimental economics games on social preferences
6.2 Appendix II: Summary of nudge tools
6.3 Appendix IV: Classification of relevant incentive theories

7 References

Abstract

Humans are generally perceived as intuitively selfish, particularly in their economic decision-making. Under the Dual-Process Framework (DPF), this view entails that prosocial behaviour requires reflective control over those natural inclinations towards self-interest. However, new lines of research explore the heuristic basis of prosocial actions and their potential use for policy-making. The goal of this thesis is to enquire into intuitive mechanisms which foster prosocial behaviour in economic decision-making and how to use them to promote altruism and cooperation through heuristic-triggering (“pure”) nudges. We found that while in some contexts prosocial choices can be economically rational, heuristic mechanisms also drive them and for some individuals, they do so to a higher extend than deliberative processes. Our literature review identified three main classes of heuristics that significantly drive prosocial behaviour: kin recognition, social and affect heuristics. Based on Intuitive Design principles, we developed a toolkit for heuristic-triggering nudges enabling choice architects to structure and describe choice options, as well as to address implementation issues, with the goal to make prosocial actions the most attractive alternative. One interesting highlight of our analysis is the differentiation of three master frames for evaluating prosocial nudges: increase in prosocial actions, ex-post satisfaction of decision-maker and social welfare outcome. Finally, we derived practical implications of our findings in social and environmental policy interventions.

Keywords: economic decision-making, heuristics, social preference, prosocial behaviour, altruism, cooperation, nudge, behavioural economics.

List of Tables / Figures

Figure 1: Models of bounded rationality

Figure 2: Characteristics of cognitive types/systems in the Dual-Process Framework (DPF)

Figure 3: Mathematical representation of altruism and spite

Figure 4: Determinants of Prosocial behaviour on a framing perspective

1 Setting the Stage

1.1 Introduction

Every day, we make roughly 100.000 choices, from trivial ones to those which can have a significant impact on our wellbeing, among which only about 100 are conscious judgements (Eyengar, 2017). When making a decision, we take into account several factors from the environment to choose the course of action that will provide us with the highest level of satisfaction. Because the choices we make, as well our expectations about others’ decisions shape our lives and our environment, researchers, practitioners and policy makers cultivate great interest in understanding the cognitive basis of our decision-making process, especially under uncertainty. In the sphere of economic decision-making, the concept of homo oeconomicus has long been held as the standard canvas to model our choices. This perception relies on the assumptions that the economic agent is rational, and aims at maximizing her utility, namely profit or well-being (Mellers, Schwartz, & Cooke, 1998).

According to the Rational Choice Theory (RCT), individuals’ decision-making is fundamentally self-centred. Under the Dual Process Framework, which differentiates automatic (system/type 1) from reflective processes (system/type 2), this entails that the human intuition would normally favour selfishness. Therefore, acting prosocially, i.e. for the benefit of others, would require a more deliberate and effortful cognitive mechanism. In other terms, the economic agent would need to actively control his natural predisposition towards selfishness in order to incorporate others’ wellbeing in his decision process (Zaki & Mitchell, 2013; Carlson, Aknin, & Liotti, 2016). However, throughout various cultural contexts, researchers have observed that people behave altruistically or spitefully and voluntarily cooperate, even when it reduces their own pay-off, thus introducing the notion of social preferences.

Growing research from diverse disciplines has attempted to understand the motives behind prosocial behaviour, notably altruism and cooperation (Korsgaad & Meglino, 2008). While some have claimed that prosocial behaviour can be economically rational (Declerck & Boone, 2015), others have suggested that prosocial motives operate by mental shortcuts called heuristics (Van Vugt, Schaller, & H. Park, 2005).

In fact, another class of breakthrough in decision theory emanates from the works of Kahneman and Tversky (1979), which demonstrated limitations of the main assumptions of rationality, notably by highlighting the influence of judgment heuristics in our decision-making (Kahneman & Tversky, 2011). On one hand, the susceptibility of economic decisions to emotions and irrational beliefs has been considered as responsible for inefficient and sub-optimal outcomes. On another hand, heuristics have been perceived as ecological rational, i.e. highly efficient and adaptive, often leading to better decisions than theoretically optimal procedures (Gigerenzer & Selten, 2002, p. 47). Van Vugt et al. (2005) suggested that this perspective could be valuable to develop ways to increase prosocial behaviour in the society.

Nevertheless, social scientists and professionals tend to focus more on anti-social behaviour and how to prevent it, than on prosocial behaviour and how to promote it. Along with bans and mandates, extrinsic rewards are part of the policy-making traditional toolkit to foster socially beneficial actions. In recent years, libertarian paternalism - “nudging”, an approach based on behavioural economics insights has gained importance. Thaler and Sunstein (2008, p. 6) define a nudge as “any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives”. A distinction is made between pro-self and pro-social nudges, and between heuristic-triggering, heuristic-blocking and informational nudges. Intuitive Design designates when the decision environment is shaped such as to activate heuristics which contribute to the desirable behaviour (Gigerenzer, 2013).

By analysing findings from experiments in behavioural economics, cognitive science and psychology, as well as empirical evidence, this thesis examines how heuristics can contribute to prosocial behaviour and whether they can be used to foster cooperation and altruism in economic decision-making through nudges. Thus, our research question is: How can prosocial behaviour be nudged using heuristics in economic decision making? We articulated this inquiry around two hypotheses:

H1: Judgement heuristics can incite prosocial behaviour.

H2: Prosocial behaviour can be aroused through heuristic-triggering nudges

To test these hypotheses, we carry out a literature review, mainly on the topics of decision making under uncertainty, prosocial behaviour and nudge theory. Our focus is on individual choice and does not extend to collective/social decision making.

Structure of the thesis

After defining relevant concepts that are used throughout our exploratory research, we outline a theoretical framework summarizing the current research findings on those concepts. In the third part of this thesis, we explain our methodology, our research question and its hypotheses, as well as the relevance of this enquiry. The fourth part presents our results, structured around our two hypotheses. The final section of this document summarizes our findings, highlighting their potential implications and opening the debate for further research.

1.2 Relevant Definitions

Economic decision making

The cognitive process through which people make choices among a set of alternative possibilities is defined as decision making (Wang & Ruhe, 2010). Economic decision making involves monetary costs and benefits, and entails specific approaches and optimization strategies (Hoggett, Edwards, & Medlin, 2002).

Rationality

In traditional economic models, individuals are assumed to be rational, which means that they can state complete and transitive preferences between alternatives and do so in order to maximize something (Edwards, 1954).

Dual-Process Framework (DPF)

One important psychology framework to study learning, thinking, memory and action is the DPF (Lizardo, 2016). Its core principle is that our behaviour is determined by the interplay between autonomic or intuitive and controlled or higher order reasoning processing (Barett & Tugade, 2006; Evans & Stanovich, 2013). In behavioural economics, the DPF is used to analyse the cognitive process of judgement and decision making, with the two distinct systems/types 1 and 2 (Alós-Ferrer C. &., 2014; Evans & Stanovich, 2013) governing respectively rapid and reflective processes.

Intuition

Type 1 processes (system 1) designate our fast, autonomous and automatic responses to stimuli, that do not require controlled attention. They tend to be associative, to develop and actualize themselves on the basis of experience, repetition, emotions and evolutionary features, such as fight or flight responses (Kahneman & Tversky, 2011; Evans & Stanovich, 2013). These attributes of the system 1 are commonly termed intuition.

Heuristics & Heuristic Cues

To be faster, system 1 tends to proceed though heuristics (Kahneman & Tversky, 2011). Judgement heuristics are procedures in decision making that reduce the number of possible alternatives (Lewis, 2008) through hard-wired cognitive shortcuts or rules of thumb (Herbert, 2011; Nevid, 2012; Kahneman & Tversky, 2011). A heuristic cue is thereby any stimuli or signal leading to a mental shortcut.

Cognitive Bias

Most of the time, heuristics enhance efficiency in decision making, especially for routine choices (del Campo, Pauser, Steiner, & Vetschera, 2016). However, these mental shortcuts may lead to systematic errors, called cognitive biases (Tversky & Kahneman, 1975) which distort individuals´ judgement, perception of reality and logic, making the decision-maker prone to irrational choices (Ariely, 2008).

Prosocial Behaviour

When people voluntarily decide to adopt a behaviour intended to benefit others (Eisenberg, Fabes, & Spinrad, 2007), that is helping, sharing, cooperating or donating (Brief & Motowidlo, 1986), it is qualified as prosocial behaviour.

Social Preferences

While making certain decisions, people can be motivated, not solely by self-interest but also positive or negative pay-offs for others, named reference agents (Fehr & Fischbacher, 2002). These preferences, which are other-oriented and not necessarily outcome oriented for the decision-maker himself (Fong, 2000) are defined as social preferences.

Social Dilemma

Situations involving conflicts between intermediate self-interest and long-term collective interests can arise in decision-making (Van Lange, 2013; Parks, 2000). Those are social dilemma, particularly present in societal, business and environmental challenges.

Kin recognition and selection

Individuals’ ability to recognize their genetic relations (Hepper, 2005) has an important influence on human social behaviour and preferences (Holmes & Sherman, 1983). This recognition process is a basis for kin selection theory or Hamilton´s genetical theory, which explains how prosocial behaviour occurs towards relatives to indirectly increase one´s reproductive value (Hamilton, 1963; Eberhard, 1975).

Nudging

Nudging can be defined as “altering people´s behaviour in a predictable way” (Thaler & Sunstein, 2008) by using instruments meant to change their choice architecture without limiting their freedom.

2 Theoretical Framework

In this section, we summarize the most relevant research findings on each concept that are used to address our research question.

2.1 Decision – making under uncertainty

The analysis of people´s decision behaviour has sparked interest in several disciplines, from psychology to mathematics, and became a fundamental topic in economics. Decision theory arose as set of models and approaches to study the reasoning underlying an agent´s choice (Steele & Stefánsson, 2016), with a normative branch, which outlines how decisions should be done and a descriptive approach aiming at analysing how decisions are actually made (Suhonen, 2007; Bradley, 2014). In Richard Bradley´s words (2014), decision theory is “the study of how choices are and should be made in a variety of different contexts”. Also, a prescriptive scope involves helping people make optimal decisions (Suhonen, 2007).

When deciding, an agent takes into consideration her goal, resources, information, standards for evaluating outcomes as well as environmental features, called “state of the world”. If the true state of the world and the outcome for each alternative are known, decision theorists talk about “certainty”, i.e. deterministic knowledge. Non-certainty is divided into three categories: uncertainty, risk and ignorance. Contrary to uncertainty, risk can be quantified, with one action leading to a set of possible outcomes, and the decision-maker knows the probability for each outcome; he has complete probabilistic knowledge. When these probabilities are unknown or not meaningful, the agent faces decision-making under uncertainty. In most cases, the agent has a certain level of knowledge on these probabilities, as most decision problems fall between risk and uncertainty, that is partial probabilistic knowledge. Ignorance is when nothing is known about the states (Bradley, 2014; Hansson, 2005).

2.1.1 Rational Choice Theory

At the centre of the normative decision theory is the concept of rationality. In that sense, a decision is considered “good” if it is deemed rational (Hansson, 2005). Traditional economic theories introduced the concept of homo economicus, portraying the human being as a self-interest seeking rational agent, aiming at maximizing her own utility in an optimal way (Rittenberg & Timothy, 2009). Thus, the rational economic agent also searches for efficiency and whether she has a lot or no information, she should select the option that gives her the higher level of satisfaction. Besides, the object of satisfaction of the agent can vary from monetary wealth even to self-destructive purposes; the defining aspect of rational choice theory is not what it assumes about people´s objectives, but rather their expected determination to pursue them. Nevertheless, the expectation of selfishness remains dominant in microeconomics (Green & Fox, 2007). Moreover, norms other than rationality tend to be regarded as external to decision theory (Hansson, 2005).

Although the homo economicus was formalized the first time in the works of John Stuart Mill (1836; 1874; Persky, 1995) in the late nineteenth century, the perception of human beings as mainly driven by the pursue of their self-interest dates back to the eighteenth century, with Smith and Ricardo (1986). Some researchers trace the rational choice theory´s origins as early as the age of reason, securing an intellectual position in Thomas Hobbes´ Leviathan (Oppenheimer, 2008), where Hobbes attempts to explain political institutions via individuals´ decisions.

In the 19th century, the rational choice theory became a dominant model for economic behaviour, notably leading to mathematical representations of its assumptions. Some models of rational choice indeed assume that people have sufficient cognitive abilities to make intuitive decisions “as if” resulting from mathematical computations (Green & Fox, 2007). Becker (2013) later even proposed applying this modelling of human behaviour to various other aspects of life and social disciplines, including marriage, crime and punishment, family and other social interactions.

Even though the definition of rationality itself is still heavily discussed among economists, overall there tend to be two views of this concept: one as consistency of choice and another as a theory of intentional behaviour (Blume & Easley, 2008). The consistency view is centred around the concept of revealed preferences, which based on radical behaviourism and according to the classical preference theory must respect the following conditions: pairwise comparable, complete and transitive. Hence, the individual should be able to order her preferences, checking options against one another. Three additional assumptions relate these preference properties to human choice:

- Maximization: individuals are assumed to always choose their most favourite option, when available.
- Stability: the preference ordering should not vary across time.
- Uniqueness: all people only have one preference ordering.

The intentional view of rationality is based on “folk” psychology, with choice driven by beliefs and values. This approach rests on an instrumental vision of rationality, in which the efficient pursuit of an end is the concern, not its nature (Blume & Easley, 2008). While reasoning about means to achieve a goal is the realm of instrumental rationality, focus on ends is labelled value rationality (Weber M. , 1978).

Bentham´s early formulations on utilitarianism, based on an evaluation of pleasure and pain potentially derived from a choice (Bentham, 1789; Edwards, 1954, p. 382) engendered the utility maximization aspect of economic rationality. Utility, as viewed by Jevons et al. is a shortcut to a theory of value. Only between from the late 19th century, with the revisions of Mill of his own work and Fisher and Pareto´s research, utility adopted a purely ordinal interpretation (Blume & Easley, 2007).

A rational economic agent hence seeks to maximize her utility; for companies, it may be profits and they are expected to always choose the alternative that yields the highest (Hansson, 2005, p. 21). When the choices are risky, they are assumed to maximize her expected utility (Edwards, 1954, p. 381), in which the utility value of each alternative (usually expressed in monetary value) is weighed with the probability, as objective frequencies of each state of nature (Hansson, 2005). The von Neumann-Morgenstern expected utility theorem (1953) introduced the independence and continuity axioms under which rational preferences can be expressed in an interval scale (Dubra, Maccheroni, & Ok, 2004).

Researchers in economics sometimes view rationality in its consistency approach as a form of discipline, minimizing the influence of their own beliefs and values over the data analysis (Blume & Easley, 2008). Hence, the rational choice theory tends to be used as a baseline model, from which certain assumptions can be relaxed or expanded to allow for more descriptive power for actual decision making. Nevertheless, this theory is subject to heavy criticism, sparkling from various disciplines. Modern research in economics outlined numerous paradoxes and experiments, which weakened and even discredited its assumptions. Notably, empirical and experimental findings would indicate systematic deviations from traditional economic models´ predictions (Pollitt & Shaorshadze, 2011; Allais, 1990). Sociologists as well heavily criticized rational choice theory (J. Goode, 1997). When the behaviour of other actors, as well as their expectations about others´ responses are considered, game theory models have relevance (Green & Fox, 2007).

2.1.2 Bounded Rationality

Despite its intellectual elegance, the concept of a perfectly rational economic agent as portrayed by Savage has been defied by empirical and experimental evidence, which suggest that human beings do not conform to that ideal (Gigerenzer & Selten, 2002). Herbert Simon (1959) introduced the term “satisficing” to describe the search process through which the decision maker aims to attain a certain aspiration level. He later argued that when risk and rationality, as well as imperfect knowledge come to modify its basic assumptions, rationality is bounded, consequently making it harder to the find the optimum. Furthermore, rationality can be bounded by complexity and by altering the nature of the agent´s goals (Simon, 1972). Bounded rationality does not only address uncertain future consequences, but also uncertainty of future preferences (March, 1978). Further research on bounded rationality hasn´t generated a comprehensive round view on this concept, but rather focused on specific bounds of rationality, most notably on the limitations in humans´ computational capacities and resources (Gigerenzer & Selten, 2002).

In their book Bounded rationality: The adaptive Toolbox, Gigerenzer and Selten (2002) define bounded rationality as “the rational principles that underlie non-optimizing adaptive behaviour of real people”. They distinguish bounded rationality from irrationality, stating that altering some assumptions of rationality does not compare to complete ignorance. Their adaptive toolbox is articulated around three premises:

- Psychological Plausibility: a vision of descriptive decision making that is realistic about humans´ actual emotional, cognitive and behavioural repertoire, which are unrealistic ideals deprived of time, knowledge and computational limits). It consists of a search for alternatives and search for cues (fast and frugal heuristics serving as reasons to prefer one alternative to the other).

Abbildung in dieser Leseprobe nicht enthalten

Figure 1: Models of bounded rationality

Source: (Gigerenzer & Selten, 2002, p. 39).

- Domain Specificity: heuristics allow for faster and more frugal operations, as they are specialized and defined for a specific goal, in contrast to general purpose algorithms such as subjective expected utility (SEU). We cover heuristics and cognitive biases more in detail in the section 2.1.4 of this thesis.
- Ecological Rationality: while psychological plausibility suggests looking inside the mind, ecological rationality urges for an adaptation to the environmental structures and their matching with domain specific heuristics. It concerns the match between a strategy and an environment, where simple heuristics tend to be more robust and accurate then regression.

As the driving force for the emergence of Behavioural Decision Theory (BDT), bounded rationality is a core concept in descriptive decision studies, as it allows for a more realistic and contingent understanding of the human mind in decision-making (Dhar & Novemsky, 2008). For economic decision making, it allowed to incorporate variables that cannot necessarily be monetarized and factors previously considered irrelevant because not quantifiable. Furthermore, extended models of bounded rationality contributed to outline theories of alternative rationalities such as limited, contextual, process, game or adaptive rationality (March, 1978).

2.1.3 Dual-process Theory

The general dual-process theory is a framework used to model cognitive processes around two types of cognition: one that is evolutionary old, unconscious and fast , i.e. type/system 1 and another that is more recent, controlled and slow, i.e. type/system 2 (Kahneman & Tversky, 2011; Lizardo, 2016; Evans & Stanovich, 2013; Evans J. S., 2008; Evans & Frankish, 2009; Kahneman, 2003). While the dissociation of these processes can produce contradictory results for the same task (Gawronski & Bodenhausen, 2006), they may also operate in tandem and used at all phases of learning, storing and using culture in thinking or action (Lizardo, 2016).

Abbildung in dieser Leseprobe nicht enthalten

Figure 2: Characteristics of cognitive types/systems in the Dual-Process Framework (DPF)

Source: Own illustration, following Kahneman (2003), Lizardo et al. (2016), Kahneman & Tversky (2011) and Evans (2008)

The DPF emerged in various social sciences as an answer to the ongoing debate on the predominance of an explicit logical processing of symbols or of implicit and automatic construction and recognition in human cognitive processes (Lizardo, 2016; Kahneman, 2003). The key premise of duality in human cognition was first advanced by William James in the 19th century (Barrett, Tugade, & Engle, 2004) but only in the late 70s was it brought to prominence by Jonathan Evans (1984) with the distinction of two processes: heuristic and analytical.

System 1/Type 1 is the realm of intuitive, fast and automatic processes, which do not require heavy cognitive strain. During experimental studies, it is activated under time pressure, when the subject doesn’t have to think through. Moreover, system/type 1 is constantly actualized by system 2/type 2 processes. For instance, the first time one carries out an operation, it can be slow and effortful, but the more she practises, the less reflective doing that activity become. A move from full cognition to intuition allow repeated experiences to generate associations and cognitive ease. Likewise, system 2 mostly adopts suggestions from system 1, building a set of beliefs and culture that then affect even our most analytic processes. Conflict between the two systems is also common in our daily lives, whenever we are struggling between our first intention and a deep thought about a situation (Kahneman & Tversky, 2011; Barrett, Tugade, & Engle, 2004; Lizardo, 2016).

Alos-Ferrer and Strack (2014) outlined the applications of the dual-process framework to economic decision-making, notably its contribution to the incorporation of bounded rationality into economic theory and through the multiple-selves models. Arguing that the neoclassical models of decision making has been focusing on modelling system 2, they further suggest that the dual-process theories provided a useful language to structure debates in economics on dualities such as rational/cognitive and affect grounds for action, especially deviations from rationality assumptions (p. 8). The multiple-selves models add another layer to the contingency of cognitive processes in economic decision making, showing that depending on the situation at hand, one system can be more useful in some situations rather than others, for instance in social dilemmas such as in the Public Goods game. The automaticity of one motive (social or selfish) hence depends on the stimuli from the environment and one the decision-maker’s previous experiences.

Several other economists suggest that the duality of human cognitive processes should be taken into account in order to understand economic decision making. They demonstrated its relevance during games and experiments on trust, mindsets and gender, risk aversion, time preference, social motives and on issues related to policy making (Posten, Ockenfels, & Mussweiler, 2014; Hügelschäfer & Achtziger, 2014; Brocas & Carrillo, 2008; Fudenberg & Levine, 2006; Fudenberg & Levine, 2011; Findley & Caliendo, 2014).

A difference must be made between the general dual process framework (DPF) and Dual process models (DPMs) that can be applied to a variety of issues related to culture such as learning, remembering, thinking and acting (Lizardo, 2016). Lizardo et al. further warn against reducing specific DPMs to a distinction between “emotion” and “cognition” (p. 29). Alos-Ferrer and Strack also make clear that the dual-process theories are a mere theoretical scaffolding of a psychology framework that may not completely fit into the economic analysis of decision-making (2014, p. 9).

2.1.4 Heuristics and Decision Biases

With his fast and automatic properties, system 1 is the realm of judgement heuristics. Judgement heuristics are procedures in decision making that reduce the number of possible alternatives (Lewis, 2008) through hard-wired cognitive shortcuts or rules of thumb (Herbert, 2011; Nevid, 2012; Kahneman & Tversky, 2011). Drawing from Simon´s research on bounded rationality, the psychologists Tversky and Kahneman provided remarkable insights on violations of rationality’s basic assumptions (Bazerman, 2017). Using experimental methods and demonstrations, they drew attention on heuristics and cognitive biases at play in decision-making with the paper “Judgment Under Uncertainty: Heuristics and Biases" in 1974. Their research, consecrated by Kahneman´s Nobel Memorial Prize in Economic Sciences in 2002 upon Tversky’s death, continues to be a guiding influence on modern microeconomics and decision science (Fiedler & von Sydow, 2015).

Even though they are mostly recognized for their deviation from rational reasoning in decision making, judgement heuristics often lead to better decisions than theoretically optimal procedures, because they can exploit environmental structure and can be robust or adaptive to new situations (Gigerenzer & Selten, 2002, p. 47). As part of the “adaptive toolbox” of our cognitive process when facing uncertainty, they enable fast and computationally cheap decisions, allowing agents to be ecologically rational, that is considering and adapting to their decision-making environment (Gigerenzer & Selten, 2002; Kahneman & Tversky, 2011). Hence, heuristics allow us to use our cognitive abilities efficiently, notably by learning from experience.

Examining the cognitive efforts of senior decision makers in a hazardous and highly uncertain environment, Elizabeth Maitland and André Sammartino demonstrated that in high-stake strategy decisions, heuristics may serve as a powerful decision-making tool (Maitland & Sammartino, 2015). In their Adaptive Toolbox, Gigerenzer and Selten (2002) present heuristics as domain-specific components of bounded rationality, with building blocks serving 3 functions:

- Search rules: to find alternatives or cues, which are designed for situations in which the alternatives are already known. Random search, ordered search and search by imitation of conspecifics are building blocks for these rules. Search heuristics are particularly successful in noisy environments.
- Stopping rules: involve easily ascertained criteria and allow the search to stop when a first alternative as high or better than the aspiration level or as soon as the first cue favouring one alternative is found (Gigerenzer & Goldstein, Reasoning the fast and frugal way: models of bounded rationality, 1996). For some adaptive questions such as mating or parenting, emotions can function as more effective and stable stopping rules than cognitive tools.
- Decision rules: while rationality models rely on weighing and summing, heuristics relying on only one cue can make more accurate predictions than complex computations, while ignoring a lot of information. Indeed, some situations present no trade-off between simplicity and accuracy.

Given the relative “laziness” of system 2, heuristics are used to save some cognitive power and effort (Shah & Oppenheimer, 2008). Furthermore, when they derive from repeated experiences, they support the learning process. The theory of attribute substitution places the origin of heuristics in a natural human inclination to unconsciously answer an easier question rather than dealing with a more complex problem (Kahneman & Frederick, 2002; Newell, Lagnado, & Shanks, 2007). This substitution is only expected to happen when the target attribute (the actual attribute of the judgement or answer to the problem) is relatively inaccessible, when the associated or heuristic attribute is highly accessible and when the substitution is undetected and uncorrected by the system 2 (Kahneman & Frederick, 2002).

Besides, the evolutionary origins of heuristics are demonstrated by several researchers to explain the adaptive fit of such cognitive processes and their pervasiveness in human nature (Haselton, Nettle, & Andrews, 2005, p. 726; Gigerenzer, 2008, p. 23; Haselton & Nettle, 2006). Indeed, to navigate through highly uncertain environments where data is scarce, humans have evolved to make inferences, to “fill the blanks” and/or use rules of thumb in order to make decisions for survival or mating (Gigerenzer, 2013, p. 9). The accuracy-effort trade-off provided by heuristics justifies why people would use them instead of going through a strenuous computation process, especially when they do not possess the necessary computational abilities. As a matter of fact, heuristics play a central function in ecological rationality, to the degree that they are adapted to the structure of the environment. But a heuristic does not necessarily evolve because of that environment but is functional, not a veridical view of the world (Gigerenzer & Gaissmaier, 2011). Most commonly cited types of heuristics are availability, representativeness, anchoring and judgement and affect heuristics.

Under most circumstances, heuristics work well but sometimes they may lead to detrimental decision outcomes, causing “cognitive biases”. Ecological rationality entails that in some environment heuristics will succeed, while in others, they may be less adapted and lead to systematic errors (Gigerenzer & Gaissmaier, 2011, p. 457).

Because cognitive biases can be responsible for substantial damages in economic decision making, such as the overconfidence bias, the endowment effect, discriminatory practices and loss aversion, an effort to be aware of one’s biases and to reduce them is encouraged. The process of debiasing can occur either within the individual who may adopt better decision-making strategies, or by a change in external factors such as decision environment settings or incentive structures. Debiasing can generally proceed in 3 ways: Incentives, nudging and training (Morewedge, 2015; Mullainathan, 2015; Thaler & Sunstein, 2008; Larrick, 2004, p. 316).The sections 2.3.1 and 2.3.2 of this thesis will explore the incentives and nudging approaches of debiasing.

Contrary to common perception, both systems 1 and 2 can be responsible for cognitive biases (Gigerenzer & Gaissmaier, 2011). Gigerenzer criticized Kahneman and Tversky’s classification of certain cognitive biases as “errors”. Criticism to Kahneman and Tversky famous research program is on the empirical, methodological and normative levels and most remarkably questions to very necessity to “debias” oneself from certain heuristics (Vranas, 2000; Fiedler & von Sydow, 2015, p. 150).

2.2 Prosocial Behaviour

"We are not ready to suspect any person of being defective in selfishness" (Smith A. , 1969, p. 446) "It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest" (Smith A. , 1986, p. 14)

Prosocial behaviour is defined as any voluntary action intended to benefit others (Eisenberg & Mussen, 1989; Eisenberg, Fabes, & Spinrad, 2007), such as helping, sharing, cooperating or donating (Brief & Motowidlo, 1986). Some researchers also consider obeying rules and conforming to socially accepted norms as prosocial (Bushman & Roy Baumeister, 2007). Moreover, prosocial behaviour is not just the mere opposite of antisocial behaviour (Hinde & Groebel, 1991). Theories on its acquisition, its origins and the motives behind it have long dominated interdisciplinary research on prosocial behaviour. The questions: are people altruistic? And if yes, why? Have sparkled decades of debates between and among researchers from diverse social sciences, mainly sociologists, anthropologists, psychologists, economist and philosophes to the extent of questioning the very definition of human rationality (Declerck, Boone, & Edmonds, 2013).

Even though there still to be literature elaborating this, we make a clear distinction between prosocial behaviour and prosocial outcome; a decision may have a positive impact on the society without necessarily stemming from an intention or an action bearing that purpose. Nevertheless, it is expected that if most people act in the collective interest, the group will be better off than when a critical mass acts selfishly (Dawes & Messick, 2000; Declerck, Boone, & Edmonds, 2013). Hinde and Groebel (1991) furthermore make a distinction between short-term harm and long-term benefit, as well as the very nature of an action and its prosocial end and the position of the evaluator, whether she is a beneficiary or not. In policy making and incentives setting in organization, even though the goal is to achieve a socially favourable outcome, the focus tend to be made on triggering it by fostering prosocial decisions.

Even though individuals are generally assumed to be selfish and to act in their own self-interest (self-regarding), empirical observations show that humans display the particular feature of voluntarily engaging in actions of cooperation or giving to unrelated strangers, without the expectation of repayment (Whitaker, Colombo, Allen, & Dunbar, 2016; Declerck & Boone, 2015; Declerck, Boone, & Edmonds, 2013; Van Vugt, Schaller, & H. Park, 2005; Fehr & Fischbacher, 2002). Nevertheless, social scientists and professionals tend to spend more time and resources enquiring about anti-social behaviour and how to prevent it, rather than prosocial behaviour and how to promote it (Bierhoff, 2002; Hinde & Groebel, 1991). In addition to the question of altruistic or selfish motives behind prosocial behaviour (Eisenberg, Fabes, & Spinrad, 2007), one important challenge in the study of these attitudes is how to foster them. From the empirical prediction and control perspective, the identification of the most powerful predictors of prosocial behaviour are crucial to promote it, notably by engaging in social engineering (Batson & Powell, 2003, p. 464).

2.2.1 Social preferences

A person displays social preferences, when she does not only care about the resources allocated to her but also about the positive or negative pay-off of relevant reference agents (Fehr & Fischbacher, 2002; Fong, 2000). The reference agents may differ depending on the situation. Hansson (2005, p. 33) also notes that a rational agent may refrain from minimizing total damage to avoid submitting individuals to high probability risks. The evidence of non-selfish motives for action has been intensively investigated especially in the last 30 years (Fehr & Fischbacher, 2002). Some of the most important types or theories of social preferences are:

- Inequity Aversion

Inequity aversion represents the desire to achieve an equitable distribution of material resources. Inequity averse people are altruistic to people who they feel are below an equitable level and envious towards those who are above that benchmark. Even though reciprocity and inequity averse can interact to drive someone’s decision, as both depend on fair and equitable payoff, reciprocal fairness has been established as quantitatively more important and stronger (Fehr & Fischbacher, 2002).

[...]

• Universidad: HHL Leipzig Graduate School of Management
• Calificación: 1.3
• Autor: Estelle Zanga (Autor)
• Año de publicación: 2017
• Páginas: 74
• No. de catálogo: V536440
• ISBN (Ebook): 9783346133182
• ISBN (Libro): 9783346133199
• Idioma: Inglés
• Etiqueta: Behavioural Economics, Behavioral Economics, Biases, Cognitive Biases, Heuristics, Decision-making, Economics, Management