Behavioral Insights Model

42

Transportation Research Record: Journal of the Transportation Research Board, No. 2322, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 42–50.DOI: 10.3141/2322-05

KiM Netherlands Institute for Transport Policy Analysis, P.O. Box 20901, 2500 EX The Hague, Netherlands. Corresponding author: N. T. W. Schaap, [email protected].

and underlying psychological and behavioral mechanisms are all rel-evant components in the ways in which mobility behavior arises and in the ways in which policy measures lead to effects on mobility and travel demand. The combination of social psychology and behavioral economics is a powerful one, providing insights into mechanisms pertaining to numerous behavioral responses to policy measures.

Psychological mechanisms derived from social psychology and behavioral economics help to explain mobility behavior and their use helps to shape effective policy measures. Policy makers and policy analysts search for ways to use insights from social psychol-ogy and behavioral economics to influence mobility behavior and (qualitatively) understand and predict the effects of policy measures [see, e.g., Anable et al. (6), Ministry of Infrastructure and Environ-ment (7)]. However, information about behavioral mechanisms and insights clarifying behavioral responses to policy measures are currently dispersed.

The best way to fill the aforementioned knowledge gap is by pro-viding an intelligible overview of mechanisms that influence mobility behavior. Such an overview needs to be cogent and comprehensible, yet should also be concise, because an overview of mechanisms that prevents readers from seeing the forest for the trees is dispropor-tionate. A number of existing overviews are largely complete on the level of individual biases and behavioral mechanisms, but they do not incorporate physical or social mechanisms [see, e.g., Prendergast et al. (8)]. Other overviews are so complete that it remains difficult to see the entire situation and apply the mechanisms. These types of overviews are rendered less useful for the intended purpose, which is to be applicable for policy makers and policy analysts. Some insights have already been applied in certain fields, for instance in different models [see, e.g., Ben-Elia and Ettema (9), Flötteröd and Rohde (10), Popuri et al. (11), and Sumalee et al. (12)], but more opportunities exist for using the behavioral mechanisms.

In an attempt to provide a practical, yet overarching, overview of behavioral mechanisms that affect responses to policy measures, this paper presents a framework of psychological mechanisms that contribute to human behavior in general and mobility behavior in particular. This overarching framework of behavioral mechanisms is called the behavioral insights model (BIM). It is based on an exten-sive literature review, analyses of existing cases, and a review session with experts with backgrounds in social psychology, behavioral eco-nomics, and transport policy analysis. The BIM is the result of cat-egorizing a comprehensive set of behavioral mechanisms into three comprehensible clusters. The model seeks to support the view that there are important behavioral and psychological mechanisms that all too often go unrecognized, while encouraging the use of psychologi-cal mechanisms to explain past behavioral effects and to explain why certain policy measures have had the intended effect, whereas others failed to reach their full potential.

Behavioral Insights ModelOverarching Framework for Applying Behavioral Insights in Transport Policy Analysis

Nina T. W. Schaap and Odette A. W. T. van de Riet

The behavioral changes that people exhibit in response to policy measures often differ from what policymakers expected ex ante, and behavioral changes are difficult to realize. However, information about behavioral mechanisms and insights clarifying behavioral responses to policy mea-sures are currently dispersed. This paper is the result of an attempt to gather these insights, starting with mechanisms deriving from social psychology and behavioral economics. An overarching framework con-sisting of three clusters of behavioral mechanisms is presented. This framework can be of assistance in shaping evidence-based policy mea-sures that make optimally efficient use of the available means, as well as helping to explain why certain policy measures have had the intended effect, while others failed to reach their full potential. At the frame-work’s theoretical base lies the insight that behavior can originate from conscious and subconscious decisions. The three clusters consist of numer-ous behavioral mechanisms on the individual, social, and physical levels. The framework was discussed in a session by experts with backgrounds in social psychology, behavioral economics, and transport policy analysis who reviewed the framework’s validity, coherence, and completeness. This discussion resulted in a set of five additional mechanisms that can be used in shaping policy measures and in explaining their behavioral effects. The extended framework, the behavioral insights model (BIM), was subsequently applied in a test case workshop involving prac-titioners from the field of transport policy, who concluded that the BIM was useful for working systematically and helpful in developing policy measures. The paper concludes with a discussion of the BIM’s completeness and applicability.

Behavioral economics and social psychology are receiving increas-ing attention in transport policy analysis (1–4). Social psychology studies the behavioral mechanisms related to social interaction with other people. This social interaction plays an important role in deter-mining the ways in which people react to policy shifts and certain policy measures and how they regard and appreciate policy measures. Behavioral economics, a relatively novel scientific direction, is also receiving increasing attention from both science and practitioners. Behavioral economics combines methods and knowledge derived from social psychology, economics, and neurology and endeavors to study the psychological mechanisms that play a role in choice making and consumer behavior (5). Choice behavior, perception, attitudes,

Schaap and van de Riet 43

Although much work is being done by researchers in different fields to increase the actuality of transport policy analysis, the scope of this paper has been limited to mechanisms derived from behavioral eco-nomics and social psychology. Doing so limits the risk of creating an overview that prevents the reader from seeing the situation as a whole, and as the combined fields cover a substantial portion of behavioral responses, the result is a comprehensible and cogent overview.

Three LeveLs OF BehAvIOrAL And PsychOLOgIcAL MechAnIsMs

The literature review led to the development of three clusters of behavioral and psychological mechanisms that play important roles in shaping and driving (mobility-related) behavior. The clusters relate to the three different levels at which behavior is manifested:

1. Individual-level mechanisms,2. Social-level mechanisms, and3. Physical-level mechanisms.

Understanding that behavior has conscious and subconscious ori-gins is the first step in explaining and clarifying behavior and deci-sions and in shaping evidence-based policy measures. The framework therefore explicitly illustrates that insight. Figure 1 depicts the core of the BIM, representing the three clusters of behavioral mechanisms as well as the previously mentioned basic insight into the conscious and subconscious origins of behavior and decisions.

The following sections first elaborate the conscious and subcon-scious origins of behavior and decision making and then describe and clarify the three clusters of behavioral mechanisms. To ensure that this paper remains readable and concise, only a selection of mecha-nisms constituting the three clusters are presented and explained. For a complete review of the mechanisms included in the clusters, see Berveling et al. (13).

conscious and subconscious Origins of Behavior and decisions

People can make decisions both consciously and subconsciously; consequently, behavior can have either a conscious or a subconscious

origin (14, 15). Subconscious, or automatized, behavior can consist of actions that are performed in a habitual manner, or of reflexes or impulses, or of automatic responses to a certain stimulus. Per-forming certain actions in a habitual manner can have a number of advantages: for instance, a habitual driving style reduces the cog-nitive effort required to function effectively, thus freeing cognitive resources (16). Many aspects of behavior can become automatized with training [see, e.g., Aarts et al. (17 ), Rasmussen (18)], and one’s actions, initially untrained, can establish patterns in a relatively short period of time (19).

The origins of planned behavior have been studied extensively. The influential theory of planned behavior, which was first presented some two decades ago, states that intentions are of major impor-tance for the realization of behavior (20). Intentions are shaped by a person’s attitudes, norms, and locus of control (20). Automatized behavior does not necessarily originate from intentions and atti-tudes, but rather can be trained to become increasingly automa-tized. With training, behavior can move from knowledge-based to rule-based and ultimately to a skill-based level of performance (18). Whereas knowledge-based performance relies on the explicit use of knowledge and reasoning to perform a certain task, rule-based performance can rely on the selection and subsequent execution of behavioral rules (for instance, one has to explicitly decide when to overtake another vehicle, but this action can then be performed more or less automatically). Skill-based performance, conversely, is completely trained and can be done automatically. One might even find that certain stimuli trigger the performance of particular actions, without conscious decisions being taken; for instance, if a daily commuter decides to take a different route home to pick up a package on the way, but then becomes distracted, she or he might end up driving the same route home again, without having explicitly decided to do so (21).

The model of strong and weak habits identifies two different types of habits (21). Strong habits activate a complete set of choices once a certain decision is made; for instance, when commuter mobility choices form part of a strong habit, the stimulus of realizing that “I need to go to work,” can automatically evoke a set of habitu-alized decisions about route choice, transport mode, and travel time, without possible alternatives having been considered. Weak habits, meanwhile, trigger a number of situationally driven alterna-tives; for instance, “I need to go to work” and “It is raining” will result in only a small set of alternatives, such as traveling by car or on foot. The final decision will be the result of this limited consideration (21). Habitual behavior and subconscious decisions explain a large part of human behavior [some estimates say 95%, and other authors even state that completely conscious decisions are impossible (22)]. Consequently, it is very important to use this insight when responses to policy measures are explained. Policy makers often cannot rely only on providing information (appealing to conscious decision mak-ing), but rather must also appeal to the subconscious and habitual characteristics of behavior.

Individual-Level Mechanisms

The first cluster of psychological mechanisms consists of many dif-ferent biases and the fact that drivers have one thing in common that affects behavior regardless of the social or physical context that people find themselves in. A number of exemplary mechanisms for each subcategory in this cluster will be provided, but the level of biases and psychological mechanisms will not be exhaustive.

Conscious and Subconscious Origins of Behavior and Decisions


Individually defined factors

Social proofAuthority

Commitment and consistencyLiking/sympathy

ScarcityReciprocity

ReadabilityEase and comfort

Atmosphere

Simple over difficult choicesBetter safe than sorry

The sooner, the better

Social-Level Mechanisms

Physical-Level Mechanisms

FIGURE 1 Core of the BIM: behavioral mechanisms at three levels.

44 Transportation Research Record 2322

Although each person’s unique characteristics are based on per-sonal perceptions and attitudes, a number of psychological mecha-nisms exist for which most people are, to a certain degree, susceptible. These biases can affect behavior on the individual level. The three main subcategories of mechanisms in this cluster are “simple over difficult choices,” “better safe than sorry,” and “the sooner the better.” These three main subcategories and their underlying mechanisms are described in the following three subsections.

Simple over Difficult Choices

It may seem like a trivial statement to say that people prefer simple choices over difficult or complex ones, but there is more to this than first meets the eye. Making decisions requires energy, and consciously thinking about every choice can even lead to cognitive exhaustion (23, 24). People prefer simple choices over difficult ones, and they deal with this preference in two ways: they use heuristics to simplify the choices they are presented with and they may even avoid mak-ing decisions if the choice set is too complex. The affect heuristic is an example of the first mechanism of coping with complex choice sets (25, 26). This heuristic is a decision-making bias that affects the outcome of choices through feelings or emotion, linking the feeling or emotion that one has about a choice outcome to the expected prob-ability of failure or success, or to the risk involved in the decision outcome. For instance, the presence of a certain stimulus (such as, for example, a scent or sound) in a choice situation can affect the decisions people make in that situation. Another decision-making bias in this subcategory is the availability heuristic [e.g., Gilovic et al. (25), Tversky and Kahneman (27)], which is a bias related to the estimation of risk and frequency. People predict the frequency of a certain event on the basis of the ease with which they can produce an example of such a situation. This is why, for example, “newsworthy” events are seen as more likely to happen than other situations, which can greatly affect people’s decision making.

The second way of coping with (too) many choice options is to make no decision at all; for instance, when presented with 24 types of marmalade, people tend not to buy any marmalade, whereas approximately 30% of consumers buy marmalade when they are presented with only six varieties (28). In other words, people tend to become overloaded and, as a result, can exhibit inaction inertia [e.g., Schwartz and Ward (29)].

Better Safe Than Sorry

In general, people like to know what they can expect, and the same is true for situations related to mobility. To avoid regrets humans have a tendency to choose the option in which the outcome is known and can be anticipated. The pseudocertainty effect is an example of a bias related to this subcategory of mechanisms on the individual level [e.g., Kahneman and Tversky (30), Tversky and Kahneman (31)]. As a result of this bias, guaranteed costs and yields weigh more heavily in the decision than uncertain ones, and options that have an uncer-tain outcome are often ignored or avoided. Another heuristic is the ambiguity effect: the situation that has the clearest or most detailed information available (with all other things being equal) is seen as more attractive [see, e.g., Ellsberg (32)].

This subcategory also includes the human tendency to prefer exist-ing situations over new situations. A study conducted in Israel revealed that people are likely to respond negatively to newly introduced policy measures. The Israeli researchers asked a number of citizens for their

opinions about a certain local issue, with some citizens being told that the local government had already implemented policy on the issue, whereas others were told the government was considering designing policy measures. People who were told that the measures had already been implemented judged the measures more positively and could provide more arguments in favor of implementation than those who were told that the measures might in future become part of a new policy (33).

The Sooner, the Better

Many choices are related to effects that can occur either sooner or later; the decision to either spend or save some money and the deci-sion to either go on a diet today or postpone it for another day are both examples of such choices. It takes will power and self-control to realize long-term gains because often an investment in the short term is required. This decision-making bias, which is also some-times called “hyperbolic discounting,” describes how direct gains are often preferred over future gains [see, e.g., Baumeister et al. (14), Laibson (34), Tangney et al. (35)]; for example, only 14% of smokers in the Netherlands are happy with the fact that they smoke, and the rest want to quit—“but not today” (36).

social-Level Mechanisms

Behavior does not originate only from mechanisms on the indi-vidual level; social influences also play an important role. Parents, friends, employers, and characteristics of the social context in which people live, as well as the norms related to this social context, also determine part of their behavior. There are two types of societal norms: prescriptive and descriptive. Prescriptive norms are often used to stimulate desired behavior—often from a societal point of view, for instance relating to sustainability (37, 38). Descriptive norms describe “normal” behavior—exhibited by role models or peer groups—regardless of whether this behavior is desired from a societal point of view. Influential research into the effects of peer groups and social norms has been conducted by Cialdini, who described six basic principles that affect human behavior through social interaction, therefore, mechanisms at the social level (39, 40). Cialdini’s six basic principles of social influence have been used as umbrella terms for the six subcategories that this cluster consists of (40). As with the other categories, a limited number of psychologi-cal mechanisms will be explicitly mentioned for each subcategory. The overview presented here is therefore not exhaustive [for more information see Berveling et al. (13)].

The umbrella terms describing the six major subcategories of mechanisms in this cluster are social proof, authority, commitment and consistency, liking and sympathy, scarcity, and reciprocity. These will be elaborated on in the following subsections.

Social Proof

It is often thought that the prescriptive norm (how one should behave) has the largest effect on human behavior, but the effects of descrip-tive norms (how others actually behave) are often largely under-appreciated [e.g., Cialdini (41)]. People tend to follow the norms of their social peer groups, a contention supported by the broken window theory (42, 43). If a building has several broken windows, thus establishing the norm that breaking windows is acceptable,


the likelihood is that the result will be more windows being broken (42). The mechanism describing how violations often lead to more violations is called the cross-norm-inhibition effect [see also Keizer et al. (43)]. This mechanism however also works in reverse: norm enforcement appears to be “contagious” (44). In another study dem-onstrating the power of descriptive norms over prescriptive norms, people were asked to lower their energy use in various ways. People who were told that their neighbors had all lowered their energy use (descriptive norm) were more likely to actually lower their own energy use than those people who were told that lowering their energy use would save them money or help save the planet (prescrip-tive norm) (38, 45). A second mechanism related to social influence is social comparison, which holds that people compare the actions of their social group to other groups: When this comparison is found to be positive for their own social group, they will adopt their group’s norms more easily or more strongly or both [see, e.g., the Ministry of Infrastructure and Environment (7 ), Festinger (46), Lyubomirsky and Ross (47 )]. A final mechanism from the subcategory of social proof that is described here is herd instinct, a mechanism that can also affect behavior. A recent study revealed that pedestrians are 1.5 to 2.5 times more likely to cross a busy and dangerous road if a fellow pedestrian starts to walk across it first (48).

Authority

People have respect for (reputed) authority. Parents, teachers, and other professionals, such as doctors and also government institutions, can be regarded as authorities, and people are inclined to follow the norms or rules established by these authority figures. In this context of authority, source credibility theory describes a psychological mecha-nism that is of influence on decisions and behaviors (49). This theory holds that the ability to remember a certain message is partially deter-mined by the potential impact the message will have on the person receiving the message, as well as the believability of a source (such as an authority figure), as based primarily on the perceived expertise and trustworthiness of the source.

Commitment and Consistency

People like to think of themselves as being consistent and reliable, often feeling high levels of both personal and interpersonal pressure to keep their promises and fulfill their accepted commitments (39, 40). Consequently, a person who has committed to a first request will often also comply with an additional request, even if that second request is much larger or more demanding. This result is the basic principle of the “foot-in-the-door” paradigm, which is often used in sales (50, 51). Cognitive dissonance is a powerful related mechanism that also stems from the need to consider oneself as consistent (52). An inconsistency between thoughts or beliefs and (past or current) actions can lead to dissonance, an uncomfortable feeling that is often countered by realigning one’s attitudes, beliefs, and actions. Because changing one’s actions is often difficult or even impossible to do, people often change their minds or rationalize their actions (52).

Liking and Sympathy

People take a more positive approach toward people they find sym-pathetic, and the notion of sympathy is based partly on common characteristics (40). This concept is related to the in-group bias, a

mechanism that compels people to give preferential treatment to other people to whom they attribute characteristics similar to their own [see, e.g., Tajfel (53), Brewer (54)]. These characteristics can be personal traits, externalities (hair styles), membership in certain clubs or groups, as well as other communalities, such as having the same birthdates or carrying a similar type of bag. People also feel sympathy toward other people who mirror or mimic their own actions (55).

Scarcity

“Things that are scarce are probably wanted by others and thus must be good.” That is, in short, the way in which people are affected by an object’s level of availability or scarcity (40). In other words, a social group’s (assumed) preferences can influence personal preferences.

Reciprocity

Receiving a gift makes people feel like they want to return the favor: one good turn deserves another. People even feel obliged to do some-thing for strangers or people they find unsympathetic if they have received something from those people in the past (40). This favor can also be in a form other than a gift, as described by the “door-in-the-face” paradigm. When a person receives an extremely large request, and turns it down (with a metaphorical slamming of a door), the sharply negative response creates a sense of debt, and this feel-ing of debt or guilt will often subsequently lead to compliance with a second, smaller request (40).


Physical situations and surroundings also play important roles in shaping behavior and choices through physical-level mechanisms, partly determining the available options and constraints, for exam-ple, through the availability of infrastructure, accessibility, or trans-port modes and the ways in which travelers are guided through the transport system. This cluster of mechanisms consists of three subcategories: legibility, ease and comfort, and atmosphere.

These subcategories are, again, umbrella terms, with each cov-ering a number of mechanisms. These underlying mechanisms are described in greater detail in Berveling et al. (13).

Legibility

The physical environment gives off signals that either consciously or subconsciously lead to reactions in the people receiving those signals [see, e.g., Berger et al. (56), Critcher and Gilovich (57 ), Dijksterhuis et al. (58), Lindenberg and Steg (59)]. This is called the environment’s cue power, which can be used, for example, to help influence driving speeds by rearranging the road setting or by placing trees alongside a road [see, e.g., Institute for Road Safety Research (60)]. Nudges are another type of mechanism related to the readability of the environment. An instance in which the physi-cal environment is organized in such a way that certain types of behavior or choices are made easier than other types of actions or choices is called a nudge (61). Nudges can, for example, affect transport mode choice at an airport when people get off airplanes and want to travel to a nearby city. If, in this situation, transfers by public transport are indicated as the default option, travelers will


(subconsciously) feel a stronger nudge to use those modes than if car rental options are presented as the default option.

Ease and Comfort

Because the ease of navigating through a situation differs, an orderly and comprehensibly organized physical environment can affect choices and behavior differently than disorderly or even chaotic situations [see, e.g., Keizer et al. (43)]. This mechanism is also related to the nudge effect, as described in Thaler and Sunstein (61). The physical environment can set expectations as to what is to come, and as previously stated, people prefer simple rather than complex choices and situations [see, e.g., Schwartz (23), Schwartz et al. (24)]. Further, physical ease involving, for example, jour-ney times, ease of entrance, and comfort affect mobility-related choices, such as transport mode or departure time.

Atmosphere

The ambience of the physical environment also affects how people feel, act, and think [see, e.g., Cialdini (38), Keizer et al. (43), North et al. (62). Dijkstra (63), van Hagen (64)]. It has been shown that different colors and other sensory stimuli such as classical music and specific scents can greatly affect people’s behavior; for exam-ple, the choice of in-store music has been shown to affect the choice of products bought in that store (62). A number of studies conducted in the Netherlands that focused on the effects of scent, music, and colors have researched the possibilities for applying sensory stimuli in railway stations and hospitals, and the effects were promising [see, e.g., Dijkstra (63), van Hagen (64)]. How the transport mode itself is organized also determines travelers’ feelings and actions.

Finally, fun can stimulate certain types of behavior. Making nor-matively desired behavior fun and combining this fun mechanism with the normative norm (taking the stairs is fun and good for your health) can be an effective mechanism (59).

AddITIOnAL PrAcTIcAL MechAnIsMs FOr APPLIcATIOn

The core of the BIM framework was reviewed by eight scientists in the fields of social psychology, mobility behavior, behavioral economics, and policy analysis and was applied to a number of test cases. The test cases and feedback from experts resulted in five additional practical mechanisms that can help explain and influence behavior. The core of the BIM, as it was presented in Figure 1, was thus expanded with the following five mechanisms: make use of discontinuities, follow behavioral change processes, use a coherent set of measures, ensure that desired behavior is sustained, and make use of target groups.

discontinuities

As was previously described, creating and sustaining habits offer a number of advantages, related mostly to the saving of cognitive effort and reverting from complex choices to simpler ones. How-ever, it is not always possible to sustain the same habits in all situa-tions; for instance, when road work blocks commuters’ usual route, they must search for an alternative. This is a discontinuity that, if

only for a short period of time, breaks the habit of taking a certain predetermined route, departing at a predetermined time, and using a predetermined mode of transport to move from one location to another. Other, longer-lasting discontinuities can include family expansion, moving house, changing jobs, or other life-changing situations. Old habits do not always comply with new situations, and because people do like having habits (because they save mental resources through efficient decision making), they will start to cre-ate new habits. This moment can be used to present alternatives that are preferable from the societal point of view. By considering the time at which certain policy measures were implemented or pre-sented to people, this mechanism can also be used to partly explain why certain policy measures did affect behavior in the intended way, whereas others failed to reach their full potential.

Follow Behavioral change Processes

Those who want to change or influence behavior should take a step-by-step approach, linking the measures presented to the ways in which people normally change their own habits. Changing one’s own behavior or habits sometimes occurs consciously, but it can also occur subconsciously (65). When people adopt new behavior, they usually move through the following stages:

1. Renewed recognition of needs,2. Searching for information,3. Weighing of alternatives,4. Process of making a decision,5. Behavior after decision, and6. Reevaluation.

The stage of behavioral change that a person is in determines the degree of openness for certain types of information, stimuli, or policy measures. To comply with these natural steps and ensure that infor-mation and measures are presented at the appropriate time, policy makers should include at least the following steps when they attempt to motivate people to change their behavior:

1. Ensure that the subject recognizes that an undesirable situation exists,

2. Create a viable and attractive alternative,3. Inform the subject about the alternative,4. Give the subject the opportunity to come to an understanding

of the alternative, and5. Motivate behavioral change.

In that way, the behavioral change intended to result from imple-mented measures has the best chance of actually being adopted. This mechanism can also be used to explain why certain policy measures failed, especially in cases in which the previously mentioned stages of behavioral change were not acknowledged or incorporated in the measures.

coherent set of Measures

Presenting a coherent set of measures simultaneously can be more effective than presenting one single measure at a time, provided that the measures reinforce rather than hinder each other. For example, when a municipality attempts to reduce the number of cars parking


in a certain area, it could be advisable to create new opportunities for public transport in addition to increasing parking tariffs and limiting the number of available parking spaces. When the service levels for parcel deliveries are improved simultaneously, people will feel more inclined to take public transport than if only one of the measures were presented. In the same line of reasoning, this mechanism can be used to explain behavioral responses to single measures versus sets of measures.

desired Behavior sustained

It is often unnecessary to focus on changing everyone’s behavior simply because some people already exhibit the desired behavior. However, it remains vitally important to sustain and stimulate the desired behavior, instead of focusing solely on the unwanted behavior. A case in point is the “Spitsmijden” pilot project con-ducted in the Netherlands (drivers regularly driving in certain con-gested areas were rewarded for changing their means of transport or time of departure). This project revealed that people who already used public transport or already avoided traveling at heavily con-gested periods of the day were upset that they could not participate in the pilot project and were not subsequently rewarded for their good behavior, whereas other people were rewarded for behaving in exactly the same way. Such side effects could undermine the ultimate effect of measures taken with good intent.

Target groups

Generic policy measures can, in certain situations, have limited effects because only a small portion of the group that the policy mea-sure is intended to affect is actually affected by the imposed mea-sures. This result is often caused by differences in living situations, demographics, attitudes, and motivations and therefore of differ-ences between groups of people in susceptibility to certain measures. For measures intended to appeal to a specific group of people, it can be useful to focus on the characteristics of that group, which there-fore requires policy makers to know and engage the intended target groups of the policy measure.

Using target groups is not a completely novel way of approaching policy measures; think of senior or family discounts, high-occupancy lanes, or other measures intended for specific groups. However, it could prove useful to search for characteristics that might influence suscep-tibility to measures beyond just the demographic characteristics or means of travel. For instance, income levels, attitudes, lifestyle types, physical disabilities, views toward government institutions, desired image, and personal values can all individually affect susceptibility to certain changes or measures. A study currently being conducted by the KiM Netherlands Institute for Transport Policy Analysis is focus-ing on these “other” passenger-related aspects with the aim being to link specific types of policy measures to specific characteristics (66).

FInAL BehAvIOrAL InsIghTs MOdeL

These five additional mechanisms were added to the core of the BIM framework. Figure 2 is a visualization of the final BIM, which also serves as a reminder of the main mechanisms when the model is applied. As previously stated, the insight that behavior and decisions have conscious and subconscious origins is key to understanding and explaining behavior and shaping evidence-based policy mea-sures, and this insight is thus given a central position in the model. The left-hand side of the chart (the core of the BIM) lists the three clusters of psychological mechanisms derived from social psychol-ogy and behavioral economics. To provide more direct insights into the content of these three clusters, their main relevant mechanisms and biases have been itemized. By using the additional mechanisms listed on the right-hand side of the chart, the clusters can be put to work in practice, helping to shape effective and efficient policy measures aimed at behavioral change and understanding responses to policy measures. Consistent application of these right-hand-side mechanisms ensures that measures are refocused on the groups that one wishes to address, in ways and at times in which the groups are susceptible to these measures and that the desired behavior already exhibited by others does not get lost in the process. This model can therefore be of assistance in shaping evidence-based policy mea-sures that make optimally efficient use of the available means, as well as helping to explain why certain policy measures have had the intended effect whereas others failed to reach their full potential.

Conscious and Subconscious Origins of Behavior and Decisions


Discontinuities

Follow behavioral change processes

Coherent set of measures

Desired behavior sustained

Target groups

Individually defined factors

Social proofAuthority

Commitment and consistencyLiking/sympathy

ScarcityReciprocity

ReadabilityEase and comfort

Atmosphere

Simple over difficult choicesBetter safe than sorry

The sooner, the better

Social-Level Mechanisms


FIGURE 2 Complete BIM.


PuTTIng The BIM TO The TesT

To properly and easily apply the presented BIM, it is important that the clusters and the underlying mechanisms be comprehensible for policy makers and other professionals and that the model can be applied in cases deriving from policy practice. To determine whether the BIM was able to help policy makers understand the effects of pol-icy measures and shape more effective and efficient policy measures, the model was applied to a real-world case study. The case focused on the ways in which travelers could be triggered to use high-level service public transport in a specific location in the Netherlands (ex ante). The focus in this case was on studying ways to influence future behavior. In the interest of brevity, only a limited number of mechanisms and their effects are described. Berveling et al. provide a more detailed discussion of the case (13).

reAL-WOrLd TesT cAse: PuBLIc TrAnsPOrT LIne WITh hIgh-LeveL servIce

In two workshops involving eight government officials from the Dutch Ministry of Infrastructure and the Environment, the clusters of (psychological and practical) mechanisms for behavioral change in the presented model were discussed and applied. The workshops aimed to investigate the extent to which the insights were helpful for policy-making purposes. The case study used was the introduc-tion of a high-level service public transport line between the cities of The Hague and Rotterdam in the Dutch Randstad, a large urban agglomeration. The hope is that the envisioned public transport line, operated by high-level service buses, will attract many passengers, both new passengers and those who currently use other transport modes. The high level of service on this bus line will be guaran-teed by the following: high frequencies (a bus every 10 min), long service hours (from 6:00 a.m. to midnight), high running-time reli-ability levels, and high comfort levels in the vehicles. However, a major challenge for this public transport line is that buses have the worst public image of all transport modes [see, e.g., Harms et al. (67 ), Berveling et al. (68 )]. Despite this image problem, the Dutch Ministry of Infrastructure and the Environment intends to attract new passengers, while also retaining the current passengers. The success of the bus line will depend partly on the question of whether car drivers will transfer to the bus. Communication and providing information related to prescriptive norms (“using public transport is better for the environment”) will not be enough to make this bus line a success.

The workshop’s main aim was to provide a test case for apply-ing and understanding the core of the BIM, as well as determining the added value of visualizing the complete BIM for government officials. After completing the workshop, the government officials stated that they completely understood the BIM and the contents of the various clusters, that they could apply the model to a case study, and that they felt that the chart was a good reminder of the BIM’s contents. The participants stated that it was useful and insightful to work with this chart because it gave them the opportunity to system-atically work their way through the large set of mechanisms involved in behavioral responses to policy measures. The workshop moreover resulted in the government officials raising numerous valid sugges-tions, mainly in the form of new ideas that had not previously been raised in the group of practitioners when they were thinking about the case study. The core of the BIM was linked to the case study in the ways described in the following subsections (for reasons of

brevity, a maximum of two underlying mechanisms are mentioned for each cluster).

conscious and subconscious Origins of Behavior

Try to make “taking the bus” a habit by providing free tickets and easy access during discontinuities, that is, when people’s current habits are discontinued because of external reasons (habitual behavior).


• Instead of calling this new high-level service bus a “bus,” a word that has negative associations and thus reminds people of their negative attitudes toward this transport mode, focus instead on the advantages the new line offers, thus creating a positive attitude toward the transport mode (62, 63).• Create one interface for all companies operating the public trans-

port line and make sure there is not too much variety in the types of tickets and vehicles, so that people are not confused and do not have to choose between companies or types of tickets (simple over difficult choices).

social-Level Mechanisms

• Take advantage of the possibilities that new social media offer; for instance, when passengers vote online that they “like” a new line, they will be more strongly inclined to actually use the transport mode when choosing how to travel (consistency and commitment).• When passengers note that their friends or people that have

similar characteristics “like” a new transport mode, passengers will view this agreement as proof that the new bus is actually good (social proof).


• The transport mode should be easily accessible, clearly present, and recognizable, and the default option in transport modes should be to take this new bus; that goal can be accomplished through signs and clearly visible stops and vehicles (legibility).• The comfort level inside the vehicle should be very high, and the

vehicles should be kept very clean (ease and comfort).

The five additional mechanisms were used throughout discussions about determining the form and timing for implementing measures, but were not explicitly used to define the content of new policy measures.

dIscussIOn OF The BIM

The overarching framework presented in this paper, the BIM is the result of a synthesis of behavioral mechanisms derived from social psychology and behavioral economics. By clustering the large set of mechanisms, the whole was made comprehensible and cogent, thereby increasing the opportunity for application in explaining and understanding behavioral responses to policy measures and shaping evidence-based policy measures.

The case study in which practitioners tried to shape new policy measures by using the BIM revealed that the model can be applied


for the ex ante creation of measures intended to influence behavior. The case study showed not only that officials and practitioners could work well with the model and the underlying mechanisms, but also that they originated more and different measures than those previ-ously identified, and these measures proved to be both feasible and appropriately directed at the ways in which people handle mobility. The government officials stated that owing to the insights provided by the framework and the underlying mechanisms, they now have a better understanding of why certain measures succeeded in the past and why others failed. Furthermore, they stated that using the visualization of the BIM to systematically work through the mecha-nisms was insightful and useful. This model can be of assistance in shaping evidence-based policy measures that make optimally effi-cient use of the available means, as well as helping to explain why certain policy measures have had the intended effect, and others failed to reach their full potential.

The BIM is a living document. Other fields that involve behavioral responses to transportation policy measures have yet to be included in the BIM. The fields of social psychology and behavioral econom-ics are also developing, leading to new insights into mechanisms and how they can be applied. The BIM will therefore be further devel-oped in the future, as further research into these topics is necessary to ensure that the set of mechanisms is expanded, critically reviewed, and ultimately completed.

The KiM (Kennisinstituut voor Mobilitsbeleid) Netherlands Insti-tute for Transport Policy Analysis and others have started to use the BIM in policy analysis studies ex ante and ex post, for instance on the topics of efficiently using the infrastructure (69) and cyclist safety (70). The use of the BIM is in shaping evidence-based policy mea-sures that make optimally efficient use of the available means, as well as in explaining (the effects of) behavioral responses to implemented policy measures. The KiM Netherlands Institute for Transport Policy Analysis will continue to critically review the BIM and to adapt or expand it if needed.

reFerences

1. Bonsall, P., M. Conner, A. Darnton, and G. Marsden. Thinkpiece 1: Influ-encing Individual Citizens. The Institute for Transport Studies, University of Leeds for the Department for Transport, Leeds, United Kingdom, 2009.

2. European Platform on Mobility Management (EPOMM). MaxSem. http://www.epomm.eu/index.phtml?Main_ID=2174&ID1=2178&id=2201. Accessed March 10, 2012.

3. KpVV Weblog Reisgedrag. Raalte Fietst: Hoe een kleine gemeente op een eenvoudige manier reisgedrag beïnvloedt en een fietscultuur creëert. http://kpvv-reisgedrag.blogspot.com/2011/12/raalte-fietst-hoe-een-kleine-gemeente.html. Accessed March 12, 2012.

4. Avineri, E., and P. Goodwin. Individual Behaviour Change: Evidence in Transport and Public Health. Centre for Transport and Society for the Department for Transport, 2009.

5. Tiemeijer, W. L., C. A. Thomas, and H. M. Prast. De menselijke beslisser: Over de psychologie van keuze en gedrag. WRR Scientific Council for Government Policy, Amsterdam University Press, Netherlands, 2009.

6. Anable, J., B. Lane, and T. Kelay. An Evidence Based Review of Pub-lic Attitudes to Climate Change and Transport Behaviour. Final report. Department for Transport, United Kingdom, 2006.

7. Structuurvisie Infrastructuur en Ruimte. Nederland concurrerend, bereikbaar, leefbaar en veilig. Ministry of Infrastructure and the Envi-ronment, Netherlands, 2011.

8. Prendergast, J., B. Foley, V. Menne, and A. K. Isaac. Creatures of Habit? The Art of Behavioural Change. Social Market Foundation, 2008.

9. Ben-Elia, E., and D. Ettema. Rewarding Rush-Hour Avoidance: A Study of Commuter’s Travel Behaviour. Transportation Research Part A: Policy and Practice, Vol. 45, No. 7, 2011, pp. 567–582.

10. Flötteröd, G., and J. Rohde. Operational Macroscopic Modeling of Complex Urban Road Intersections. Transportation Research Part B: Methodological, Vol. 45, No. 6, 2011, pp. 903–922.

11. Popuri, Y., M. E. Ben-Akiva, and K. Proussaloglou. Time-of-Day Mod-eling in a Tour-Based Context: Tel Aviv Experience. In Transporta-tion Research Record: Journal of the Transportation Research Board, No. 2076, Transportation Research Board of the National Academies, Washington, D.C., 2008, pp. 88–96.

12. Sumalee, A., P. Luathep, W. H. K. Lam, and R. D. Connors. Evaluation and Design of Transport Network Capacity Under Demand Uncer-tainty. In Transportation Research Record: Journal of the Transporta-tion Research Board, No. 2090, Transportation Research Board of the National Academies, Washington, D.C., 2009, pp. 17–28.

13. Berveling, J., H. Derriks, O. van de Riet, R. Smit, and M. Storm. Gedrag in beleid. Met psychologie en gedragseconomie het mobiliteitsbeleid versterken. KiM Netherlands Institute for Transport Policy Analysis, The Hague, Netherlands, 2011.

14. Baumeister, R. F., E. Bratslavsky, M. Muravan, and C. M. Tice. Ego Depletion: Is the Active Self a Limited Resource? Journal of Personality and Social Psychology, Vol. 74, No. 5, 1998, pp. 1252–1265.

15. Pol, L. R., C. E. Swankhuisen, and P. van Vendeloo. Nieuwe aanpak in overheidscommunicatie; mythen, misverstanden en mogelijkheden. Coutinho, Bussum, Netherlands, 2009.

16. Young, M. S., and N. A. Stanton. What’s Skill Got to Do with It? Vehicle Automation and Driver Mental Workload. Ergonomics, Vol. 50, No. 8, 2007, pp. 1324–1339.

17. Aarts, H., R. Meertens, B. Verplanken, and M. van Vugt. Psychologische fundamenten van mobiliteit: vervoerskeuzes als gewoontegedrag in een sociaal dilemma. Maastricht, Rijksuniversiteit Limburg, Netherlands, 1994.

18. Rasmussen, J. The Definition of Human Error and a Taxonomy for Techni-cal System Design. In New Technology and Human Error (J. Rasmussen, K. Duncan, and J. Leplat, eds.), Wiley, Chichester, United Kingdom, 1987.

19. Mehl, M. R., and J. W. Pennebaker. The Sounds of Social Life: A Psy-chometric Analysis of Students’ Daily Social Environments and Natural Conversations. Journal of Personality and Social Psychology, Vol. 84, No. 4, 2003, pp. 857–870.

20. Ajzen, I. The Theory of Planned Behaviour. Organizational Behaviour and Human Decision Processes, Vol. 50, 1991, pp. 179–211.

21. Verplanken, B., H. Aarts, and A. van. Knippenberg. Habit, Information Acquisition, and the Process of Making Travel Mode Choices. Euro-pean Journal of Social Psychology, Vol. 27, 1997, pp. 539–560.

22. Lamme, V. De vrije wil bestaat niet: over wie er echt de baas is in het brein. Bakker, Amsterdam, Netherlands, 2010.

23. Schwartz, B. The Tyranny of Choice. Scientific American Magazine, Vol. 4, No. 4, 2004, pp. 70–75.

24. Schwartz, B., A. Ward, J. Monterosso, S. Lyubomirsky, K. White, and D. R. Lehman. Maximizing Versus Satisficing: Happiness Is a Matter of Choice. Journal of Personality and Social Psychology, Vol. 83, No. 5, 2002, pp. 1178–1197.

25. Gilovic, T., D. W. Griffin, and D. Kahneman. Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge University Press, Cambridge, United Kingdom; New York, 2002.

26. Winkielman, P., R. B. Zajonc, and N. Schwarz. Subliminal Affective Priming Attributional Interventions. Cognition and Emotion, Vol. 11, No. 4, 1997, pp. 433–465.

27. Tversky, A., and D. Kahneman. Judgment Under Uncertainty: Heuristics and Biases. Science, Vol. 185, No. 4157, 1974, pp. 1124–1131.

28. Iyengar, S. S., and M. R. Lepper. When Choice Is Demotivating: Can One Desire Too Much of a Good Thing? Journal of Personality and Social Psychology, Vol. 79, No. 6, 2000, pp. 995–1006.

29. Schwartz, B., and A. Ward. Doing Better but Feeling Worse: The Paradox of Choice. In Positive Psychology in Practice (P. A. Linley and S. Joseph, eds.), John Wiley & Sons, Inc., Hoboken, N.J., 2004, pp. 86–104.

30. Kahneman, D., and A. Tversky. Prospect Theory: An Analysis of Deci-sion Under Risk. Econometrica, Vol. 47, No. 2, 1979, pp. 263–292.

31. Tversky, A., and D. Kahneman. The Framing of Decisions and the Psy-chology of Choice. Science, Vol. 211, No. 4481, 1981, pp. 453–458.

32. Ellsberg, D. Risk, Ambiguity, and the Savage Axioms. The Quarterly Journal of Economics, Vol. 75, No. 4, 1961, pp. 643–669.

33. Marcus, G. Kluge: The Haphazard Construction of the Human Mind. Houghton Mifflin, Boston, Mass., 2008.

34. Laibson, D. Golden Eggs and Hyperbolic Discounting. The Quarterly Journal of Economics, Vol. 112, No. 2, 1997, pp. 443–477.


35. Tangney, J. P., R. F. Baumeister, and A. L. Boone. High Self-Control Pre-dicts Good Adjustment, Less Pathology, Better Grades, and Interpersonal Success. Journal of Personality, Vol. 72, No. 2, 2004, pp. 271–322.

36. Kooreman, P., and H. Prast. What Does Behavioural Economics Mean for Policy? Challenges to Savings and Health Policies in the Netherlands. Presented at Netspar Panel, 2007.

37. Cialdini, R. B., R. R. Reno, and C. A. Kallgren. A Focus Theory of Nor-mative Conduct: Recycling the Concept of Norms to Reduce Littering in Public Places. Journal of Personality and Social Psychology, Vol. 58, No. 6, 1990, pp. 1015–1026.

38. Cialdini, R. B. Basic Social Influence Is Underestimated. Psychological Inquiry, Vol. 16, No. 4, 2005, pp. 158–161.

39. Cialdini, R. B., and N. J. Goldstein. The Science and Practice of Persua-sion. Cornell Hotel and Restaurant Administration Quarterly, Vol. 43, No. 2, 2002, pp. 40–50.

40. Cialdini, R. B. Influence: Science and Practice, 4th ed. Allyn and Bacon, Boston Mass., 2001.

41. Cialdini, R. B. Descriptive Social Norms as Underappreciated Sources of Social Power. Psychometrika, Vol. 72, No. 2, 2007, pp. 263–268.

42. Wilson, J. Q., and G. L. Kelling. Broken Windows: The Police and Neighborhood Safety. The Atlantic, 1982.

43. Keizer, K., S. Lindenberg, and E. M. Steg. The Spreading of Disorder. Science, Vol. 322, No. 5908, 2008, pp. 1681–1685.

44. Boluijt, J. Parkeervakken tegen fietschaos. Algemeen Dagblad, 2011, pp. 14.

45. Nolan, J. M., P. W. Schultz, R. B. Cialdini, N. J. Goldstein, and V. Griskevi-cius. Normative Social Influence Is Underdetected. Personality and Social Psychology Bulletin, Vol. 34, No. 7, 2008, pp. 913–923.

46. Festinger, L. A Theory of Social Comparison Processes. Human Rela-tions, Vol. 7, No. 2, 1954, pp. 117–140.

47. Lyubomirsky, S., and L. Ross. Hedonic Consequences of Social Compari-son: A Contrast of Happy and Unhappy People. Journal of Personality and Social Psychology, Vol. 73, No. 6, 1997, pp. 1141–1157.

48. Faria, J. J., S. Krause, and J. Krause. Collective Behaviour in Road Crossing Pedestrians: The Role of Social Information. Behavioural Ecology, Vol. 21, No. 6, 2010, pp. 1236–1242.

49. Gilbert, D. T., S. T. Fiske, and L. Gardner. The Handbook of Social Psy-chology. Oxford University Press, United Kingdom, 1998.

50. Scott, C. A. The Effects of Trial and Incentives on Repeat Purchase Behaviour. Journal of Marketing Research, Vol. 13, 1967, pp. 263–269.

51. Scott, C. A. Modifying Socially Conscious Behaviour: The Foot-in-the-Door Technique. Journal of Consumer Research, Vol. 4, 1977, pp. 156–164.

52. Festinger, L. A Theory of Cognitive Dissonance. Stanford University Press, Stanford, Calif., 1957.

53. Tajfel, H. Experiments in Intergroup Discrimination. Scientific American, Vol. 223, 1970, pp. 96–102.

54. Brewer, M. B. In-Group Bias in the Minimal Intergroup Situation: A Cognitive-Motivational Analysis. Psychological Bulletin, Vol. 86, No. 2, 1979, pp. 307–324.

55. van Baaren, R. B., R. W. Holland, B. Steenaert, and A. van Knippenberg. Mimicry for Money: Behavioural Consequences of Imitation. Journal of Experimental Social Psychology, Vol. 39, No. 4, 2003, pp. 393–398.

56. Berger, J., M. Meredith, and S. C. Wheeler. Contextual Priming: Where People Vote Affects How They Vote. In Proceedings of the National Academy of Sciences, Vol. 105, No. 24, 2008, pp. 8846–8849.

57. Critcher, C. R., and T. Gilovich. Incidental Environmental Anchors. Jour-nal of Behavioural Decision Making, Vol. 21, No. 3, 2008, pp. 241–251.

58. Dijksterhuis, A., P. K. Smith, R. B. van Baaren, and D. H. J. Wigboldus. The Unconscious Consumer: Effects of Environment on Consumer Behaviour. Journal of Consumer Psychology, Vol. 15, No. 3, 2005, pp. 193–202.

59. Lindenberg, S., and E. M. Steg. Normative, Gain and Hedonic Goal Frames Guiding Environmental Behaviour. Journal of Social Issues, Vol. 63, No. 1, 2007, pp. 117–137.

60. SWOV Institute for Road Safety Research. Factsheet Speed Choice: The Influence of Man, Vehicle, and Road. 2010. http://www.swov.nl/rapport/Factsheets/UK/FS_Speed_choice.pdf. Accessed July 21, 2011.

61. Thaler, R. H., and C. R. Sunstein. Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press, New Haven, Conn., 2008.

62. North, A. C., D. J. Hargreaves, and J. McKendrick. In-Store Music Affects Product Choice. Nature, Vol. 390, No. 6656, 1997, pp. 132.

63. Dijkstra, K. Understanding Healing Environments: Effects of Physical Environmental Stimuli on Patients’ Health and Well-Being. University of Twente, Enschede, Netherlands, 2009.

64. van Hagen, M. Waiting Experience at Train Stations. Eburon Academic Publishers, Delft, Netherlands, 2011.

65. Kotler, P., G. Armstrong, J. Saunders, and V. Wong. Principles of Mar-keting, 2nd ed. Prentice Hall, Europe, London, 1999.

66. Berveling, J., and H. Derriks. (forthcoming)67. Harms, L., P. Jorritsma, and N. Kalfs. Beleving en beeldvorming van

mobiliteit. KiM Netherlands Institute for Transport Policy Analysis, The Hague, Netherlands, 2009.

68. Berveling, J., P. Bakker, L. Harms, and E. van de Werff. Imago en open-baar vervoer. KiM Netherlands Institute for Transport Policy Analysis, The Hague, Netherlands, 2009.

69. Berveling, J., and H. Derriks. Opstappen als het kan, afstappen als het moet. Een sociaalpsychologische blik op de verkeersveiligheid van fietsende senioren. Kennisinstituut voor Mobiliteitsbeleid, The Hague, Netherlands (in Dutch), 2012.

70. Berveling, J., and H. Derriks. Safely on a Bike. Old and Wise on a Bike. Presented at International Conference on Traffic and Transport Psychol-ogy, Groningen, Netherlands, Aug. 29–31, 2012.

The Traveler Behavior and Values Committee peer-reviewed this paper.

Behavioral Insights Model

Documents