Shikakeology: designing triggers for behavior change

1.1 The strength of a small trigger

It is often the case that a small trigger can change our behavior and result in a big social impact (Andrews 2010). A famous example is the fly target etched on urinals in men’s restrooms at Schiphol Airport in Amsterdam (discussed in Sect. 4.3). When men see the fly target, their behavior is expected to change due to the temptation to aim at the fly. The fly target in urinals is a small trigger that causes a slight behavior change, but it is reported that spillage was reduced by 80 % (Thaler and Sunstein 2009). This eventually reduces cleaning expenses and water usage. Consequently, it had a broader impact on reducing environmental pollution. There are several examples of the “small trigger causing a big difference” phenomenon at the personal level, such as increased exercise and healthier eating. These also manifest in pro-social behavior at the government level (e.g., reducing garbage, waste foods, traffic accidents, and crime).

1.2 The concept of “shikake”

“Shikake” is a Japanese word with various meanings, depending on the context. According to Shogakukan’s Progressive Japanese-English Dictionary, shikake means “a device, mechanism, contrivance, and system” as a noun and “to start, set up, prepare, and challenge” as a verb. In this paper, we use “shikake” as a complex integration of these definitions to describe an approach to trigger behavioral changes that will solve social or personal problems. As the problems that we are targeting result from our own behavior, changing behavior is a straightforward approach to solve them. The aim of a shikake approach was to solve a problem through behavior, not through function. This is the most unique and significant aspect of the shikake approach.

In contrast to the shikake-driven approach, one might consider a high-tech device that automatically solves problems without people’s help. However, such a high-tech-driven approach is feasible if and only if the device can be created. Practically, this is not a realistic choice in most cases, as it requires considerable effort in terms of fabrication cost, expertise, and technology.

Open image in new window

Let us consider a simple example of garbage separation. Figure 1 shows a trash bin with a built-in high-tech device that can automatically separate plastic bottles and cans, and a trash bin with a transparent structure. The transparent trash bin does not seem particularly special, but it encourages people to separate bottles and cans more than a normal, nontransparent trash bin would (IPROS 2004). The items that people throw away are visible to others with a transparent trash bin. This elicits pro-social behavior because people do not want lose face and thereby have their self-esteem compromised. In addition, the garbage in the transparent trash bin becomes a social norm indicating that people have correctly separated their trash. Deviations from the social norm require special circumstances. In this case, the transparency of a trash bin becomes an effective shikake to make people aware of their behavior.

Human behavior is superior to the functionality of current high-tech devices; that is, in terms of vision, operation, and mobility, our eyes, feet, and hands are superior to computer vision, robot hands, and wheels at recognizing objects, moving to a destination, and interacting with objects under various environments. Thanks to these human capabilities, social and personal problems can be solved through behavioral changes. In fact, the Central Japan Railway Company has already replaced about 1,600 ordinary trash bins with transparent trash bins. Thus, the shikake-driven approach is far more feasible and effective choice.

This paper aims to provide a general statement on the concept of Shikakeology as a science of shikake in order to understand the underlying mechanisms. In the following sections, we first define the meaning of shikake in the study of Shikakeology. We then introduce the mechanisms of behavior change from precedence studies, followed by concrete shikake cases. We then explain the patterns of best practices, shikake triggers, and shikake trigger matrix in turn. We finally conclude this paper with future work.

3.1 Psychological triggers

Psychological triggers affect our psychological preference. The power of social pressure has been investigated for decades, primarily in experimental settings. Asch (1955) conducted experiments in visual judgment and revealed that social forces easily alter people’s opinions. Milgram et al. (1969) also reported that conformity increases as a function of group size through experimental results that more people in a crowd looking up at a building let to more passersby joining the crowd. Cialdini (2006) described in his book the way in which people are influenced and, as a result, persuaded. He categorized various cases regarding behavior change into six main factors: reciprocity, authority, social proof, commitment and consistency, liking, and scarcity. Thaler and Sunstein (2009) described in their book “Nudge” the way in which people can be nudged to make the world better, especially in terms of health, wealth, and happiness. They explained various cases in accordance with six principles of good choice architecture: incentives, understanding mappings, defaults, give feedback, expect error, and structure complex choices. Fogg’s (2009) behavior model explained that three elements (motivation, ability, and trigger) must converge simultaneously for a behavior to occur. In his model, a trigger that is a reminder of a habitually associated action should be designed first. If the trigger does not work, six ability factors should be tackled next, namely, time, money, physical effort, brain cycles, social deviance, and nonroutine. If these factors do not work, the final approach is motivation. Fogg (2002) also proposed an approach for persuasion using a computer and revealed persuasive techniques in the domain of software applications and mobile devices. Zichermann and Cunningham (2011) described game mechanics as a technique to encourage people to become involved in activities. Game mechanics stimulate human desires for reward, status, achievement, self-expression, competition, and altruism, and so forth by designing a system that controls points, badges, levels, challenges, leaderboards, goods, quests, etc.

3.2 Physical triggers

Physical triggers work directly and indirectly. As a direct effect, a physical trigger is realized by good visibility of function that enables people to understand intuitively the usage and expected results. For example, a door with good visibility of function tells people (1) opened or closed, (2) locked or unlocked, and (3) push or pull or slide right/left/up/down. If these functions are obscure, the visibility is inadequate and needs to be improved. A shikake should have good visibility because it has to induce the appropriate behavior. Such a perceived function, which enables people to perform an action, was termed “affordance” by Norman (2002). He later revised the term as “perceived affordance” or “signifier” instead of “affordance” to make the vocabulary more precise; a perceived affordance is some signal in the physical and social world that can be interpreted meaningfully (Norman 2010). Sullivan (1896) proposed a principle known as “form follows function,” where the shape of an object reveals the purpose of the object. The concept was first applied to architecture and then spread to other design fields, such as product design and automobile design. The principle is also considered a direct physical trigger.

A physical trigger has an indirect effect on behavior, in that it activates a psychological trigger. For example, the transparent trash bin works as a physical trigger to cause senses of social proof and thereby changes in self-esteem. In this case, these feelings work as psychological triggers ignited by the physical trigger.

3.3 Psychological and physical triggers

Volkswagen introduced the idea “Fun can change behavior for the better,” developing various products that change behavior to solve social problems. For example, The World’s Deepest Bin makes a long falling sound as if the bottom of the bin was deep underground when garbage is thrown into it (Volkswagen 2009). The sound attracts people’s attention, and as a result, more garbage is expected to be collected by people who observe this feedback and have fun trying it repeatedly. The concept of The Fun Theory emphasizes that “fun” is the most powerful driving force to change behavior for the better. The “fun” products are created by way of physical and psychological triggers.

3.4 Challenges for Shikakeology

As described above, psychological triggers are categorized differently—as persuasion, engagement, nudges, human factors, and biases—in accordance with different perspectives, in order to best describe the fundamental mechanism. Physical triggers are also considered in various ways, such as perceived affordance (or signifier), feedback, and technologies, to specify the component of a trigger. Furthermore, a physical trigger is associated with a psychological trigger, and that, specifically, is what a shikake focuses on, as the combination works strongly to facilitate behavior change.

From a shikake point of view, categories of psychological and physical triggers should be optimized to describe the fundamental mechanism. However, the criteria have not yet been studied. One of the biggest challenges in the establishment of Shikakeology is to uncover the criteria for categories best tailored to describe shikake specifications. These categories are useful not only to understand the mechanism of shikake specifications but also to design new shikakes. Currently, we are collecting hundreds of shikake cases to identify these categories. The preliminary shikake categories, extracted from 120 shikake cases, will be presented later in this paper.

4.1 Cylinder

Figure 2 is a cylinder installed at Tennouji Zoo in Japan. There is no explanation board around it. People are not told what this artifact is, but they can guess how to use it because it has some triggers. First, the cylinder looks like a telescope, so people might imagine that one looks through it. Second, there is a hole in the cylinder. When people find the hole, they instinctively look through it out of curiosity. Third, the cylinder is placed about one meter off the ground. The position is ideal for children to look inside the hole. Because of these triggers, children are interested in the cylinder, approach the cylinder, look inside the cylinder, discover something at the other end of the cylinder, and finally enjoy the discovery. In this case, children find a replica of an elephant’s excrement and are surprized to notice the size and color.

When people observe a child’s behavior for a while, they might notice that once he/she looks inside the cylinder, the behavior attracts other children’s attention and makes them gather around the cylinder. This phenomenon is called as snowball effect, where a small action becomes a trigger that leads to a big effect. As a result, the cylinder succeeds in attracting children and triggers their interest to explore and discover.

Open image in new window

The cylinders are installed along the pathway connecting the animal exhibition areas. If there were no cylinders, children would not stop walking to explore and discover. In addition, the cylinder is easy to make but difficult to break due to its simple structure. All things considered, the cylinder works as a superb shikake for behavior change.

The telescope-like shape and the installation height of the cylinder work as physical triggers for people to infer its utility and attract target people (i.e., children). The hole in the cylinder works as a physical trigger to induce a specific behavior as well as a psychological trigger to excite curiosity. In addition, people looking into the cylinder could act as a psychological trigger for a snowball effect. In this case, the analogy of a telescope, perceived affordance of a hole, curiosity, and social effect are considered the underlying mechanisms of the cylinder on the zoo pathway.

4.2 Tiny shrine gate

Littering is an everyday problem that all societies have in common, and its underlying mechanisms have been studied. The broken windows theory hypothesizes that a small trigger causes a big result, like a snowball effect (Wilson and Kelling 1982). If there is a broken window in a building, the window becomes a sign of disorder and leads to more broken windows. The same is true for littering (Keizer et al. 2008). If someone drops one piece of litter, the litter becomes a trigger to induce more littering. Complete cleanliness would likely be a good trigger to stop littering, but it is not a realistic solution due to its high cost.

Open image in new window

Figure 3 shows a tiny replica of a gate to a Shinto shrine. The tiny shrine gate reminds people of a holy place, and this prevents them from littering. This shikake also prevents bad behavior, such as dog walkers not picking up excrement. In this case, the shikake works if and only if people associate the tiny shrine gate with a Shinto shrine.

Other approaches, such as putting up anti-littering signs or installing high-tech devices like security cameras, are possible. However, these approaches give neighbors a descriptive norm of caution and compromise the trusting atmosphere. In contrast, the tiny shrine gate provides an elegant and simple approach that does not negatively affect the atmosphere.

The shrine gate is a physical object, but it works as a psychological trigger to cause pro-social behavior by giving the impression of a holy place. In this case, the analogy of a shrine gate and the associated social norm are considered the underlying mechanisms. In other words, this shikake might not work in places where nobody understands the symbol and respective behavior associated with shrine gates. Therefore, this shikake is culturally dependent. Culture could be a strong trigger when considering the use of social norms, but we must carefully account for the potential sensitivities in order to avoid provoking people’s adverse reactions.

4.3 Urinal fly

Open image in new window

Figure 4 shows a fake fly etched on a urinal. Once people (men in this case) see the fly, they instinctively aim at it. The location and type of target are designed to determine the “sweet spot” and thus minimize splash back (Smets 1995). As a result, spillage is expected to reduce and thereby increase the cleanliness of the facility. We do not need any reason to aim at the fly aside from the fun. We do not have to aim at it, but there is no reason we should not. Due to the effectiveness of the idea and the simplicity of its implementation, various urinal target stickers, such as a bee, a dartboard, and a flame (the color changes with temperature change), are now used worldwide.

A urinal fly has the physical shape of a fly and causes a psychological trigger to aim at it by instinct. In this case, the analogy of a fly for the target and the playful challenge of hitting the target are considered the underlying mechanisms.

4.4 Speed camera

Cars are convenient for transportation, but every year many people are injured or killed in traffic accidents. Various shikakes have been considered to make traffic safer. Figure 5 shows a speed camera that feeds back the speed of a car with a speed limit sign. This system is not connected to a police office, so there is no force compelling drivers to slow down. However, this system works well in practice. The feedback of cars’ speed to the drivers becomes a good trigger to change from mindless to mindful driving.

In addition to feeding back the speed, one more shikake element could be added. The Speed Camera Lottery is a campaign in which the speed of passing cars is monitored, and lottery prizes are randomly mailed to the houses of drivers who maintained their speed under the limit (Volkswagen 2009). The reward comes from speeding fines. As a result, the average speed was reportedly reduced by 22 % when the Speed Camera Lottery was tested in Sweden.

Open image in new window

A physical speed camera produces feedback and thereby ignites a psychological trigger to control speed. In addition, the lottery system works as a psychological trigger for driving to obtain a reward. Furthermore, when a goal is clearly provided, it becomes a challenge. In this case, feedback to car drivers to reduce their speed, and the challenge to keep under the speed limit using rewards worked as the underlying mechanisms.

Unlike the other cases, this case uses electronic technology. A shikake does not necessarily depend on technology because it heightens the efforts needed to make it due to cost and required expertise. However, when it becomes possible to use technology and it generates a synergistic effect, a technology-based trigger becomes a powerful shikake toward behavior change.

4.5 Piano stairs

Figure 6 shows the stairs decorated like a piano, on which people can actually make a piano-like sound with their feet instead of their fingers. The stairs attract people’s attention and encourage them to use the stairs rather than escalators. As a result, more people are willing to exercise. According to a report on The Fun Theory,¹ the number of people who used the stairs increased by 66 %. The point in this case is that no one was forced to use the stairs. Those who were interested in the piano stairs voluntarily chose to use the stairs.

The piano-like appearance allows people to easily associate the stairs with a piano and to have a pleasant experience playing the piano or listening to music. This psychological association between a piano and an enjoyable experience is the core trigger to drive people to use the stairs. This shikake is implemented using technologies (e.g., sensor devices and sound devices). Thus, the shikake is realized as a combination of design, psychology, and technologies.

Open image in new window

5.1 Utility of patterns

Lessons from best practices could be a promising approach for success or reducing the risk of failure because they give us cues for future prediction as well as estimations of necessary efforts. Of course, best practices depend on the situations, and the same practices cannot be applied across all situations. However, it might be applicable to similar situations if the pattern can be generalized and applicable to a broad range of situations. Patterns are also used to bridge the gap between people in various domains. Common knowledge for an expert in one domain is unknown to people in other domains. A pattern is considered a simple expression of essential knowledge in a specific domain that allows people to share it.

Patterns, which are obtained from knowledge and procedures of best practices, are used to increase the utility of people’s activities. In the following, we briefly review some ideas of useful patterns in practice in both cases.

5.2 Patterns from knowledge

Alexander et al. (1977) coined the term “Pattern Language” to express a new methodology where a pattern as a unit of best practice is combined with other patterns for designing architecture and surroundings. Pattern Language is described in a unified format, such as pattern name, picture or sketch, explanation, and relations with other patterns. Alexander created 253 patterns for describing good design practices and applied them to various design projects. Gamma et al. (1994) proposed “Design Patterns” as a programming approach based on the elements of reusable object-oriented codes. They proposed 23 patterns of codes as a general reusable solution. By combining the patterns, the manners of code are formalized and the generated code becomes more efficient and readable. A pattern language works as a common terminology between people in an interdisciplinary field. Borchers (2008) presented an approach for applying pattern language to interaction design in order to facilitate cooperation between developers and domain experts. The pattern language obtained from good practices can also be used to teach human–computer interaction. User interface patterns are widely used on websites or desktop applications (Tidwell 2011). Software libraries for designing user interfaces are provided as a toolkit, making it possible to reuse the best practices of user interface for solving design problems. Content management systems such as WordPress or Wiki are also considered as applications of user interface patterns because they publish contents with a standardized format based on the patterns of best practices. Hatamura (2008) studied the causes of failure cases in industries to avoid the repetition of failure. Hatamura collected more than 1,000 failure cases, classified the cases from various points of view, and discovered that the failure cases can be classified into a limited number of essential patterns. These patterns could be used to describe the failure cases as well as to transfer the failure knowledge to others. The database has been released on the Web, so anyone can see the details.

5.3 Patterns from procedures

User-centered design (UCD) is a standardized methodology for designing products in order to reflect what users can use, want, or need (ISO9241-210 2010). The process includes not only designing a product but also validating its usage in a real situation. It can be identified as iterative cycles of four steps: (1) specify the context of use, (2) specify requirements, (3) create a design solution, and (4) evaluate the design. IDEO, a company renowned for the UCD of its products, services, and environments, proposes the concept of design thinking. The creative methodology is crystallized as IDEO method cards, which are 51 cards from four categories: Learn, Look, Ask, and Try (IDEO 2003). David Kelley, the founder of IDEO, has been teaching design thinking at Stanford design school, where the processes of design thinking are designed in five iterative steps: empathize, define, ideate, prototype, and test. In this process, rapid prototyping plays an important role in driving this cycle. A manual is also a structurally composed pattern of best practices. In the case of customer service, for example, the words and their order of the contact person are carefully determined. By following the manual, necessary information is collected efficiently and reasonably regardless of the skills or performance of contact persons. Design patterns are the patterns for codes, whereas the pattern of practices in programming software is also known as extreme programming (XP) (Beck and Andres 2004). The patterns of XP are defined as 19 practices classified into four types of basic activities (coding, testing, listening, and designing) under five values (communication, simplicity, feedback, courage, and respect).

6.1 Shikakes as patterns

As we described in the previous section, a pattern is an essential representation of knowledge and procedures due to its generalized simplicity. To date, shikakes are typically designed by teams of expert engineers and/or designers. However, by applying shikake patterns in design, not only experts but also ordinary people might be able to create new shikakes. A shikake is realized as the combination of physical and psychological triggers. Therefore, a pattern of shikake should be described by combining these types of triggers.

Physical and psychological triggers can be classified into hierarchical categories. However, the categories and the hierarchical structures change depending on the target phenomena perspectives. For example, categories for persuasion (Cialdini 2006) and nudges (Thaler and Sunstein 2009) are different. In this paper, we consider the hierarchical categories of triggers from the shikake perspective. In the following, we first describe the concepts to which we refer for shikake triggers. We then present preliminary shikake triggers that we constructed.

6.2 Shikake triggers

Based on these previous studies and practices, we carefully constructed shikake trigger categories to systematically describe physical and psychological triggers as shown in Fig. 7. The hierarchical relations between categories are created based on our analysis of the cases according to our considerations, although interpretations might differ if different viewpoints are considered. The top-level category “Trigger” in Fig. 7 is clustered according to “Physical trigger” and “Psychological trigger” to reflect the idea that a shikake is composed of both physical and psychological triggers. In the following, we explain our considerations by elucidating each of these categories.

Open image in new window