"#$#%&!'($&%)*$+,(!-&*.(,/,0+&1!2)3! 4,56#$&%!7*+&(*&8!79)(1&)!:(+;&%1+$<! 79)(1&)8!:=8!7>?!@AA! B6C&1/)53,/*.+)%8!5C/C9+/1,(DE19)(1&)C)*C#F! ! .*7"I*&A;4"
"#$%&'()*!+,! >*%?'*"@'A*BC$A8(?*%"D"3*E"FG"H?A'$7" @/$-.3A3$4:% 9+E'&05$-+G01% :+(186% 9$:3#-%!$4:+":3;$%($?)-151#6% 9&#$-KE#4$I116% G"-"#$,$-.%1A%K3215"4%93:14/$4% =+C'&1.+$9+'DE$+#($!+/"C$%+'('+F6% 7+/.1K6$ !"##$%&'(')'*"+,-.' !"#$%&%%'%% ()$%*+#,$-.$/%0)1223-#%(4155$67%*-%*,83$-.%93:25"6%(1%!41;3/$% 0)122$4:%<3.)%=1-'>8;31+:%!41/+?.%@-A14,".31-% !"#$%&'()$*+&,+$-+.#&/+&01$+#($2,"##1$3"41'56$ !"#$%B%'%% !$4:+":31-%@-'03.+7%0)1223-#%A14%C$"5.)6%D11/%3-%0+2$4,"4E$.:%
7.1$-+..18+,1)$9&/+1.$%6$:/",$+#($;&#"$<&+#6% !"#$%FF%'%% (1<"4/:%"%G1835$%*2253?".31-%.1%H4$".$%0$/$-."46%*<"4$-$::%
=&>5$=1.1&>#51)$?+'0$*+#$@+.01'1#$+#($!+#$A&1/8"BB6% !"#$%FI%'% G>=*JH*7%*%!$4:+":3;$%!$4:1-"5%G1-3.143-#%06:.$,%.1%0+2214.% % % % % % % % $ % % !"##$%&'/')'0-.$1#' !"#$%FL%'% K+35/3-#%!$4:+":31-%!41A35$:%3-%.)$%M35/7%N:3-#%G1835$%9$;3?$:%":% !"#$%O&%'%% =+/#3-#%N:$4:%(1<"4/:%!43;"?6%1-%G1835$%9$;3?$:% 31C1DD+$%+.1C+/")$H1('"$I1"#)$@+J&F$?.FE8&F1(&)$H+0'&D/$-1..1K)$ !"#+08+#$9E4+#)$?.155+#('"$?DLE&50&)$I"''&1$M'+#"'$+#($;"'F+#$ !"#$%OB%'%% D41,%P.)3?:%.1%Q"5+$:%3-%.)$%9$:3#-%1A%G1835$%!@=H%
!+#10$A+,&5R% !"#$%&F%'%% >2214.+-3.3$:%"-/%H)"55$-#$:%3-%G3-3-#%K$)";314"5%P?1-1,3?:%.1% !"##$%&'2')'0&"345' !"#$%&I%'%% S".)$43-#%"-/%!4$:$-.3-#%01?3"5%D$$/8"?E%.1%H)"-#$%91,$:.3?%
P5$?.43?3.6%H1-:+,2.31-% 9+0081N$:0E(.1K)$:&F"#$M8+FC1'5)$3E08$3100&1$+#($-1,&#$%E'D81..6% !"#$%&L%'%% (1<"4/:%P#1?$-.43?%D+$5%PAA3?3$-?6%D$$/8"?E%
<&+4"$M+F+D8")$O&.&G1$PE�+.)$9&D81..1$:D"00)$*+55&.&5$-"50+/"5$+#($Q+#$ !"#$%T&%'%% H4$".3-#%UH115V%G1835$%($?)-151#3$:%(1%J$/+?$%($$-%P-$4#6%N:$%
A+#&1.$O&00"#)$!+#10$31+()$3E551..$%1+.1)$%1#$M"N+#$+#($2E/+#4$=E"6$ !"#$%TB%'%% =+/#3-#%!$125$%".%M14E%"-/%>.)$4%()34/'!"4.6%W1?".31-:R%
9+R$I6$S&.5"#)$A1'1/$O"501')$:8+E#$I+N5"#$+#($:&F"#$T((&5"#% The Augmented Shopping Trolley: An Ambient Display To Provide Shoppers with Non-Obvious Product Information Jon Bird, Vaiva Kalnikaité and Yvonne Rogers
Pervasive Interaction Lab
The Open University
Milton Keynes, MK7 6AA, UK {j.bird, y.rogers}@open.ac.uk, vaivak@gmail.com ABSTRACT The Augmented Shopping Trolley consists of an ambient handlebar display connected to a scanner. When a shopper scans an item the handlebar lights up to provide them with information about the product, such as its nutritional, ethical or environmental attributes, that are not obvious from its packaging or label. The system is designed to seamlessly integrate with a shopping experience: it uses familiar supermarket technologies; it keeps both of a shopper’s hands free; and the simple ambient display facilitates the ‘fast and frugal’ decision-making typically observed in a supermarket. Our initial lab-based study shows that the display can be understood at a glance and used to select items based on a product’s nominal properties (for example, it is organic), ordinal properties (for example, it has low, medium or high food miles), as well as a combination of the two at the same time. Where as usability was the focus of our initial design, ethical issues have come to the fore as we develop the system for use in supermarkets and we discuss how these are influencing our design.
Author Keywords Persuasive technologies, product information, ethics.
ambient display,
shopping, ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous.
INTRODUCTION In a supermarket, shoppers tend to make snap judgments based on just a few salient cues (low price, recognized brand and attractive packaging) and they rarely take time to read product information labels [7]. However, recent consumer surveys indicate that shoppers want more Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
Copyright © 2011 for the individual papers by the papers' authors.
Copying permitted only for private and academic purposes. This volume
is published and copyrighted by the editors of PINC2011..
information about the global consequences of their
consumer decisions [
Our handlebar ambient display consists of just sixteen LEDs. When a shopper scans a product, a few pieces of non-obvious information, such as whether it contains nuts, is fair trade or has low food miles, are displayed as a salient pattern on the display.
Given that information salience influences a person’s behaviour unconsciously [1], rather than through rational reflection, this raises ethical concerns about the Augmented Shopping Trolley, chief of which is that this system could potentially manipulate people into behaving in ways that they would not otherwise do, and furthermore, that they might not be aware that they had been manipulated. This concern, and also issues to do with privacy and clarifying how our system benefits shoppers, form the ethical considerations that are influencing how we deploy the Augmented Shopping Trolley in a supermarket.
The paper is structured as follows: first, we describe the display hardware and how it conveys product information; second, we describe a lab-based evaluation of the system that demonstrates the efficacy of the ambient handlebar display for conveying non-obvious product information; and third, we describe the ethical issues that are informing the development of the system for use in supermarkets.
AMBIENT HANDLEBAR DISPLAY DESIGN The handlebar display was designed to provide shoppers with salient and easy to read information about a scanned product’s nominal properties (for example, whether it is organic or contains nuts), its ordinal properties (for example, if it has low, medium or high food miles), as well as a combination of the two at the same time. We constructed the display by attaching 16 bicolour LED units to a piece of wood inside a transparent plastic tube (Figure 1). This replaced the plastic handlebar in a standard shopping trolley. The LEDs are controlled using 2 TLC5940 chips (Texas Instruments) that are driven by an Arduino microcontroller. In our lab-based study this is attached via a USB cable to a laptop running a Processing application. Each LED unit can be set to red, green or orange (when both the green and red LEDs are on). Each time a product is scanned, the display changes in the following way. First, it goes from an all green background (idle state) to a half second sweeping movement of orange that indicates scanning is in progress. There is then a beep, as typically heard at a checkout counter, to signal that scanning is completed and the display then changes to a new state that provides relevant information about the product. If the display is configured to show a nominal property of the product, then it flashes green if the property is present and shows the idle state if it is not. If the display is providing ordinal information about the product, the display employs a bar graph metaphor, with the number of red pixels indicating the degree to which an item has a property. Specifically, if an item has a low degree of a property then pixels 1-3 turn red and 4-16 turn green; if medium then pixels 1-8 turn red and pixels 9-16 turn green; and pixels 1–13 turn red and 14-16 turn green if the item has a high degree of a particular property. Finally, both these representations can be combined to show the value of a nominal and an ordinal property at the same time. In our study, after a participant selected or discarded an item, the display changed back to the all green idle state.
LAB-BASED SYSTEM EVALUATION 5 adults (1 female, 4 male, aged between 20 and 40) took part in a lab-based evaluation of the Augmented Shopping Trolley. Each participant completed 12 shopping scenarios where they were asked to pick up and scan 5 items of a particular product type and only select those items that met specified criteria. A scanner was attached to the shopping trolley (Figure 1) but was non-functional and the handlebar display was changed using a Wizard of Oz methodology.
On the basis of the changes in the patterns on the handlebar display, participants had to decide whether to select the item and place it in their trolley or discard it and place it on an adjacent table. Since this was an exploratory study, we were intentionally vague about the operation of the ambient display as we wanted to see whether participants could understand it intuitively. We only told participants that the display patterns would change depending on whether a product had a specific property (yes/no), the degree to which a product had some property (high/medium/low) or a combination of the two. Participants were allowed to scan the items as many times as they wanted and in any order, before they made their decision about whether to select a particular item. We used 4 product types: milk; breakfast cereal; wine; and juice. Each shopping scenario used one of the product types and participants were asked to select from 5 different items. For example, select those bottles of wines that meet the specified criterion (fair trade) and put them in the trolley, and place the others on the discarded items table. Each of the items was a real product but we masked any product information on the packaging and told participants to only use the handlebar display to decide whether they should select an item or not. The experimenter playing the Wizard of Oz role sat at a table on which the 20 shopping items were grouped by product type. Each item was individually numbered so that the experimenter could change the display appropriately when the participants scanned a particular item.
In the first 4 shopping scenarios the handlebar display indicated whether a scanned item had a particular nominal property or not: whether a milk product was organic; whether a breakfast cereal contained nuts; whether a bottle of wine was fair trade; and whether a carton of juice contained added sugar. In 2 of these scenarios the participants had to select items that had a particular property and in the other half they had to discard items if they had a particular property. For example, in the first shopping scenario participants had to select a milk product if it was organic and discard it if it was non-organic; in the second shopping scenario participants had to select a breakfast cereal if it did not contain nuts and discard it if it did.
In the next stage of the evaluation, the participants completed 4 shopping scenarios where the display indicated whether a product contained a low, medium or high value of a particular ordinal property. The task was to select items that had a specified property to a particular degree.
Specifically, participants were asked to select milk with a medium fat content, cereals with a high sugar content, wine with low food miles and juice with a medium water content.
In none of these scenarios were participants asked to discard items if they had properties of a particular degree.
The final 4 shopping scenarios tested whether participants could understand the display when it simultaneously showed information about both a nominal and an ordinal property of a scanned item. Participants were asked to select milk that was organic and low fat, cereals that contained nuts and had a medium sugar content, juice that had added sugar and high water content and wine that was not fair trade and had medium food miles. Only in the wine scenario did participants have to reject items on the basis of information about a nominal property of the product.
USABILITY RESULTS 4 out of the 5 participants were able to interpret the ambient handlebar display and complete all the tasks without any mistakes. The other participant made one consistent error in 2 of the first shopping scenarios where the task was to discard items if they had a particular nominal property: they selected, rather than discarded, them, but did not repeat this error in the final shopping scenario which also required an item to be discarded if it had a particular nominal property.
Several participants reported that they found the tasks where they had to discard items with particular properties more difficult and it did seem to increase the cognitive load in all participants, resulting in a slightly slower response time (approximately 2 seconds, rather than 1 second for the other conditions). This could be due to the colours used in the display: a nominal property is indicated by a green blinking display, a colour that many people associate with positive properties, rather than ones that should be avoided.
All participants reported that the display was intuitive to use and were able to quickly read it even though they were not given explicit information on the meaning of the display patterns. Only two participants scanned items more than once and this was exploratory activity at the beginning of the evaluation when they were seeing how the interface worked.
ETHICAL ISSUES AND FURTHER DEVELOPMENT Whereas usability issues informed our initial design, ethical considerations are shaping the development of the Augmented Shopping Trolley for use in supermarkets. This is because our ambient display not only provides salient product information for shoppers, but also potentially influences what they purchase. The use of persuasive technologies raises ethical concerns for many people. For example, Page and Kray [3] used an online questionnaire to investigate people’s views on the ethics of using persuasive technologies to encourage healthy living. 72 participants rated the ethical acceptability of a number of different scenarios which varied in 3 different factors: whether a participant chose to use the technology or an external agency initiated its use; whether there was a clear benefit for the participant or not; and the technology used (text messages to the participant’s mobile phone; public announcements in the participant’s location; Facebook messages; restrictions on the participant’s bank account; and electric shocks). The results indicated that the majority of the participants viewed the use of persuasive technologies in most of the questionnaire scenarios as unethical. When there was no clear benefit to the participant, mobile phone were considered the most ethical persuasive technology. However, approximately the same proportion of participants (40%) considered them very ethical or ethical as the proportion that considered very unethical or unethical when. A large majority of participants found the other technologies very unethical or unethical. In scenarios where the use of a technology would clearly benefit the participant, for example, save their life, then this usage was considered slightly more ethical than the cases where the technology did not benefit the participant. However, it is not clear whether these differences were statistically significant. When people were able to freely choose whether to use a persuasive technology or not, then texts, public announcements and Facebook messages were considered ethical by the majority of respondents, in comparison to the situation where the use of the persuasive technology was initiated by an external entity (for example, the UK’s National Health Service).
Electric shocks and bank account restrictions were considered very unethical or unethical by the majority of respondents, even when a participant chose to use them.
Page and Kray’s findings seem to concur with a central factor identified by applied philosophical analyses of ethical behaviour, for example, the use of persuasion in advertising [5]. Namely, the ethics of an action are determined, to a large degree, by the extent to which that action impacts on an individual’s autonomy, that is, their capacity to choose how to act and determine their own life.
Page and Kray’s research also highlights that privacy and the extent to which a participant benefits are important issues for determining the ethical acceptability of persuasive technologies. All three of these ethical considerations (autonomy, privacy and benefits) are shaping the development of the Augmented Shopping Trolley.
To ensure shopper’s autonomy, they will be free to decide whether they use the Augmented Shopping Trolley and also able to choose which particular non-obvious product information they want to be informed about. Given that users can configure the system to provide different product information, privacy is not compromised, even though the handlebar will be visible to other shoppers, as they will not understand what particular LED patterns mean. Some of the product information that will be provided by the Augmented Shopping Trolley can clearly benefit a participant, for example, nutritional data, whereas other information, such as food miles, may not have direct personal benefits. In fact, trying to minimize food miles may lead, literally, to a personal cost. However, we assume that if participants choose to be informed about a particular type of product information then they do so because it is of benefit to them and in keeping with their lifestyle choices.
We are currently considering how to use the display to provide aggregate information about the contents of a participant’s trolley. The display could indicate how averaged values of all the participant’s purchases relate to some norm(s), for example, is the weekly shop below or above the average shopper’s food miles. Clearly, there are normalization issues to be resolved to enable such comparisons to be made. One ethical consideration with this type of display is that even if an observer did not know what aspect of product information the aggregate display encoded, under certain conditions it could be evident whether a participant was above or below a norm, thereby compromising a shopper’s privacy. For example, if the observer had also used the display themselves and the colour encoding was fixed. One way to ensure privacy is to allow participants to customize aspects of the display, such as the colour encoding used. A second ethical concern with this sort of display is that norms, like salience, typically influence people unconsciously. To ensure that the autonomy of participants is not compromised it seems important to inform them about the methods used in a display and how these typically influence behaviour before they choose to use the Augmented Shopping Trolley CONCLUSIONS Our lab-based study shows that participants can rapidly read a shopping trolley handlebar display to determine both nominal and ordinal properties of a scanned product. Our display is intuitive to use and requires no training.
Participants find it easier to select items when they have desirable properties than to not select them because they have undesirable properties. The Augmented Shopping Trolley makes non-obvious nutritional, ethical and environmental product information salient to shoppers and facilitates the fast and frugal decision making typically used in a supermarket. Some of the global consequences of selecting particular products can now be made salient to shoppers at the point of decision making, potentially facilitating changes in consumer behaviour. We argue that our system is an ethical persuasive technology as it enhances the ability of shoppers to buy choose products in accordance with their individual values.
Persuasion In-Situ: Shopping for Healthy Food in Supermarkets ABSTRACT Healthy lifestyle is a strong trend at the moment, but at the same time a fast growing number of people are becoming over-weight. Persuasive technologies hold promising opportunities to change our lifestyles. In this paper, we introduce a persuasive shopping trolley that integrates two tools of persuasiveness namely reduction and suggestion.
The trolley supports shoppers in assessing the nutrition level for supermarket products and provides suggestions for other products to buy. A field trial showed that the persuasive trolley affected the behaviour of some shoppers especially on reduction where shoppers tried to understand how healthy food products are. On the hand, the suggestion part of the system was less successful as our participants made complex decisions when selecting food.
Author Keywords Shopping, health, persuasive, supermarkets.
ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous.
INTRODUCTION Healthy lifestyles is a hot topic in most Western societies as a rapid growing number of citizens are either over-weight or obese, e.g. more than 50% of the adult population in Denmark are either over-weight or obese [9]. Over-weight problems come from several circumstances, e.g. the lack of exercise or unhealthy food, but in general people buy and consume food that contains a lot of sugar or fat. Thus, we need to alter people’s behaviour and attitude while they shop groceries and other food products in supermarkets.
When supermarket shopping, more studies have shown that consumer behaviour is highly controlled by routine and is not simply changed or altered [8]. In fact, even if shoppers want to change their shopping behaviour and patterns, they find it difficult to understand the nutritious values of many Copyright © 2011 for the individual papers by the papers' authors. Copying permitted only for private and academic purposes. This volume is published and copyrighted by the editors of PINC2011 products, e.g. they cannot understand nutrition labels or how much sugar or fat the product contains [5]. Further, one of the fundamental problems resides in the fact that we are confronted with an overwhelming number of different food products and it is often difficult to identify and choose the more healthy ones. Iyengar and Lepper showed in an experimental study that consumers were more satisfied with their own selections when they have fewer options to select from [5]. Schwartz refers to this as the paradox of choice claiming that the huge number of choices decreases people’s real choice and decision-making [10]. Thus, people are likely to continue their current routine type of behaviour (as illustrated by Park et al. [8]) and this could potentially prevent them from making healthier choices.
Emerging technologies are increasingly being used to alter
people’s opinions or behaviour, e.g. smoking cessation [4]
or promoting sustainable food choices [7]. Fogg refers to
such technologies as persuasive technologies or captology
[3]. Fogg states that contemporary computer technologies
are currently taking on roles as persuaders including
classical roles of influence that traditionally were filled by
doctors, teachers, or coaches [3]. Research studies within
different disciplines are increasingly concerned with such
persuasive technologies that may be used to create or
change human thought and behaviour. As examples, Chang
et al. [
Todd et al. [11] illustrate theoretically how nudging could persuade shoppers to select healthy food products based on simplified information to the shoppers in-situ, but call for empirical understandings of persuasive shopping. We propose a persuasive shopping trolley application called iCART that attempts to motivate change towards more healthy shopping behaviour. First, we outline the idea behind the design of the trolley application and then reports from field studies of use on its effects on behaviour change. iCART: INFLUENCING SHOPPING BEHAVIOUR IN-SITU iCART is a persuasive application mounted on a shopping trolley that attempts to persuade the shopper’s behaviour and awareness. The system was implemented in C# using Windows Presentation Foundation for the interface and a Microsoft SQL server.
From our previous research [6], we learned that many consumers actually attempt to buy healthy products when supermarket shopping, but often they would find it difficult to assess the nutrition value or energy level. In fact, several consumers are actually unsure what a healthy food product is. Shoppers find it difficult to understand the nutrition information labels on the food products and they usually don’t bother consulting this information. Supermarket products and groceries are rather diverse, e.g. ranging from simple non-processed products (e.g. an apple) to more complex processed products (e.g. a pizza). Usually people find it difficult to assess how healthy processed products are. Furthermore, people find it difficult to change behavior and usually choose well-known products while shopping.
The overall idea of iCART is that all food products and items in a supermarket can be classified according to nutrition level and this classification will be presented to the user of the trolley every time the shopper puts an item into the trolley. For our persuasive system, we adapt the nutrition label initiative called Eat Most from the Danish Veterinary and Food Administration. For our purpose, it provides a simple classification of food products based on the nutrition values of a product. The classification label includes a table for calculating the value of all food products. According to the label, all products can be classified as Eat Most, Eat Less, or Eat Least.
The typical use situation could be as follows (illustrated in figure 1): The user walks around the supermarket with the trolley, chooses food products and places them in the trolley (a), the trolley recognizes the product and displays its classification according to the Eat Most label (b), and the system updates the status for the entire trolley on numbers of Eat Most, Less, and Least food products (c).
(a) (b) (c)
Figure 1: Illustrating the process of using iCART Interaction Design We adapted three persuasive design tool principles from Fogg namely reduction and suggestion [1]. The persuasive shopping trolley should 1) present or visualize product nutrition in a simple way and 2) present alternatives to less healthy products. Finally, we decided that the system should be a walk-up-and-use system on a shopping trolley.
Reduction reduces complex behaviour to simple tasks in order to increase the benefit/cost ratio and thereby influence the user to perform the behaviour [3]. As stated above, consumers find it difficult to assess the overall nutrition level for products. The persuasive trolley reduces this nutrition value assessment through the simplification in the Eat Most classification and thereby the assessment now becomes a simple task. This is illustrated in figure 2 where different products have been classified, e.g. milk as eat less (middle picture). We colour-coded the three categories with green, yellow, and red. Figure 3 shows the classification for a cereal product called Havrefras and this product is an eat least product. The implementation in iCART reduces the action of assessing the nutrition value of a product by providing a simple classification of only three categories.
Suggestion means that persuasive technologies have greater power if they offer suggestions at opportune moments [3].
Consumers find it difficult to choice healthier alternatives as they often have limited understanding of the relative levels of nutrition between more products. The persuasive trolley offers suggestions for alternative products (Eat Most) within the same product group when the shopper choices an Eat Less or Eat Least product in the trolley. We consider this an opportune moment as the shopper often will find the alternatives in their present supermarket area (as illustrated in figure 4 where two alternative cereals are suggested for the cereal in figure 3).
FIELD TRIALS We conducted field trials with the shopping trolley at the local supermarket called føtex. It was rather important to us to understand the use of the system in-situ to facilitate the whole shopping experience. 11 shoppers were recruited through public announcements and we required that they shopped for food products on a regular basis. The shoppers were between 27 and 58 years old and represented different kinds of households and worked in diverse job professions. We asked them to fill in a questionnaire on their supermarket shopping experiences prior to the trials. Some of the participants were highly concerned with nutritious food while others were less concerned. The participants were divided into two groups one group used iCART while the other group served as a control group using a regular shopping trolley. We balanced them in the two groups based on their selfreported knowledge and attitudes towards nutritious food.
Before the trials, we carried out a pilot test to verify and adjust the process and our instructions. Participants were not informed about the purpose of the study in order to minimize study impact and iCART participants were told about the system but not its focus on healthy food products.
The trials consisted of a three parts namely an introduction, the actual shopping, and a debriefing. We instructed the participants to shop items from a pre-generated shopping list using their own normal criteria for food selection. Thus, they should try to shop as they normally would. The shopping list contained 12 items, e.g. milk, cheese, pate.
The list included only general product groups (except for one item) leaving the participants to choose within the group, e.g. cheese where they could choose more 20 different cheese products. They were free to choose in which order they would collect the items. 303 food items were entered into a SQL database representing all items in the store within the groups from the shopping list. Data collection was done through 1) a trolley-mounted video camera that captured verbal comments and shopping behaviour and 2) the system logged and time stamped all user interactions enabling to reproduce action sequences afterwards. The sessions were done during normal trading hours and they were not required to check out the collected items.
We evaluated iCART as a Wizard of Oz experiment where one of the authors acted as wizard implementing the actions taken by the participant. When a food product was put into the trolley, the wizard would update this information in the system. Another person observed the participant while shopping in order to facilitate the following interview. The same procedure was used for the control group, but without the trolley-mounted display. The total time spent ranged from 12:08 to 40:28 minutes. Finally, a debriefing session including questionnaires and semi-structured interview was conducted immediately afterwards, e.g. they were asked to assess their own session and the collected items.
OBSERVATIONS AND DISCUSSION The five participants using iCART expressed that they liked the system and they would possibly use it if available in supermarkets. While food products in supermarkets already have different labels for determining the health or nutritious level, iCART became a personal technology that guided the shopper while shopping. This also had the advantage that shoppers always knew where to look for the nutritious information for all products. Today, this information is located on the packaging of the product and thereby distributed in the store.
The reduction element of iCART was quite successful. Out of the 60 food products selected by the participants using the system, 30 were classified as Eat Less or Eat Least.
Thus, half of the selected products were less healthy. In several cases, the participants were surprised to realize that a certain product was less healthy. For example, one of the participants chose a bag of carrot buns and got surprised to see that these buns were Eat Least: “I thought they were healthy as they contain carrots”.
On the other hand, several shoppers chose less healthy food products and were aware of it – even without the help from iCART. But the classification made them reflect upon their choices and several of them started talking about nutrition and healthy food. One participant said: “But the Eat-Least classification makes you think and questions whether you have made the right choice”. From our analysis, it seemed that they acted out of routine behaviour and that they partially knew the consequences of these choices. This confirms the findings by Park et al. [8] on changing shopping routine behaviour. In summary, the reduction element of iCART was quite successful as it raised the awareness of the shoppers on the nutritious level of the chosen products.
The suggestion component of iCART was less successful compared to the reduction. The participants changed their choices 3 times out of 30 (10%). This low number was somewhat surprising, but shoppers gave several reasons for this. Some would not change their choice, as they would rather buy an unhealthy food product that was biodynamic than buy a healthy product that was not. So the shoppers would implement their own classification schemes based on other aspects than nutrition. Also, some shoppers stated that they never bought any light or zero products, which often were the products suggested by our system. They said that they would rather eat less of the unhealthy products than buy a light product.
During the field trials, 18 times did the shoppers take a look at the suggestions made by iCART, but in most situations (14 times) they chose not to follow the suggestion. This indicates that the shoppers are interested in receiving suggestions but the actual suggestions made by the system in the situation were not good enough. As illustrated above, they had different objectives when shopping and perhaps suggestion functionality should be carefully organized.
We identified an interesting observation concerning trust to the system. Some users expressed scepticism towards the suggestion part of the system while none of them really questioned the reduction part. Most of them stated that nutrition labelling whether on the actual product or implemented in an interactive system on the trolley should be controlled and accredited by public authorities. They were more critical when it concerned suggestions than reductions. The problem with suggestion could reside in that it could feel like ads or commercials for other products.
That could be a potential problem when implementing suggestion tools. However, as expressed by one of the female participants: “It is cool to be guide. I don’t mind help or receive suggestions, I’m a grown-up who can make my own decisions”. This could imply that to change behaviour designers should focus on providing reduction in complexity of assessing the food product, but they should perhaps not suggest or give recommendations to the user.
Shopping in supermarkets is noisy and complex and it can be stressing due to several multimodal inputs. We noticed how several participants missed reductions or suggestions on the screen while acting in the environment. Thus, they would actually not receive the information proposed by the system. Also, one participant stated that shopping is private even though it takes place in a public environment.
The participants who shopped without the persuasive guidance appeared to have fewer reflections on nutrition and health. In fact, the iCART participants eventually bought 25 food items classified as Eat Least whereas the other participants bought 34 Eat Least products. The difference cannot only be explained in terms of the suggestion tool implemented in iCART, but the interaction made them reflect.
CONCLUSION We presented the persuasive shopping trolley iCART that guides supermarket shoppers in choosing more healthy food products by classifying all products in three groups namely Eat More, Eat Less, and Eat Least. Field trials with 11 shoppers showed that iCART proved to provide good input on reduction, e.g. reducing the complex task of assessing whether a product is healthy or less healthy. Our participants noticed when the system classified a product as Eat Least and usually they would start reflecting upon this.
Only a few times did this result in change of behaviour where the user changed the original choice. But mostly the suggestion part of the system was less successful. This was mainly due to the fact that several participants had rather specific requirements to their products, e.g. they should be biodynamic or they never bought light-products.
Based on our findings, we see a number of future research avenues. First, rather than optimizing the algorithms behind suggestion tools, we propose that we should design systems that enables shoppers to make their own decisions in-situ.
This could require a different approach to reduction. Also, we need to understand the long-term effects of such systems and we plan to conduct more longitudinal studies.
ACKNOWLEDGMENTS We would like to thank the shoppers from the field trial as well as reviewer comments on earlier versions of the paper. Towards a Mobile Application to Create Sedentary
Awareness Gijs Geleijnse, Aart van Halteren and Jan Diekhoff
Philips Research
Eindhoven, The Netherlands gijs.geleijnse@philips.com , aart.van.halteren@philips.com ABSTRACT Prolonged sitting time is a potential health risk, not only for people with an inactive lifestyle, but also for those who do meet the recommended amount of physical activity. In this paper, we evaluate SitCoach, a mobile application to nudge people from their seats. The application is targeted to office workers. SitCoach monitors physical activity and sedentary behavior to provide timely feedback by means of suggesting sitting breaks. A pilot experiment with a group of 8 users learned that the general awareness of the importance of sitting breaks is low. Combined with the belief that the ability to take sitting breaks is highly dependent on external factors, a strategy of proposing break reminders may not be the most successful for this target group. Future work should focus on raising awareness of the problem and providing insights into personal sitting behavior.
Author Keywords Sitting time, mobile persuasion, sedentary awareness, physical activity.
ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous.
INTRODUCTION In the past years, a substantial amount of research has been devoted to physical activity promotion through mobile devices. Using the accelerometer embedded in a mobile phone or in a dedicated device, the energy expenditure of the user can be estimated. The user may receive feedback on his past physical activity level in minutes or burned calories.
Several strategies have been explored to influence the
UHVX¶EDKLRYGQWHSUPLJKFDOV\WYH
Most notably, the usage of virtual rewards [
Copying permitted only for private and academic purposes. This volume is published and copyrighted by the editors of PINC2011." successful persuasive strategies to establish an increased amount of physical activity.
Recent medical literature reports that not only an inactive lifestyle may lead to adverse health effects, but also sedentary behavior itself is harmful. Prolonged sitting time is also dangerous for people who meet the WHO guidelines of 30 minutes of physical activity per day [4,12]. The reduction of sedentary behavior is hence identified as a target behavior that contributes to a healthy lifestyle.
S6XRUW RW FUHDW UHVQDZ RI H¶RVQ HGVQWDU\ EDHKLRUY may be beneficial. However, as Owen et al. state in [12], HLY³JQ HWK UWHFQ LUHFQRWJ RI WLKV HSKRQP RI WR much sitting, there are not yet any recommended clinical guidelines. Commonsense might suggest that it may be prudent to try to minimize prolonged sitting with 5 minute In this paper, we describe SitCoach, a mobile application that assists the user to create sedentary awareness and to have regular sitting breaks. Such an application can be combined with additional physical activity promotion features. To the best of our knowledge, SitCoach is the first prototype mobile application aimed to reduce sitting time.
Using SitCoach, the goal is to collect insights into SLEHROWV RW HLIOXFQ SHRVO¶ LVWJQ EDHKLRUY XLVJQ D mobile device.
SitCoach targets office workers, a group which is often also assisted by break reminder applications on their PCs. Such applications are developed to prevent their users from repetitive strain injuries. Although such applications show to be successful in reducing complaints [7], they may not always be pleasant to use [10]. Morris et al. [10] introduced SuperBreak, which stimulates break compliance for computer usage. Instead of the usual breaks offered by software packages such as XWrits and WorkRave, SuperBreak offers the possibility to make the break time more productive. By offering the user the possibility to interact with the PC through gestures during the break, break compliance is promoted and the productivity during the break time is increased. Hence, although SuperBreak may increase break compliance for computer work, it does not target a reduction in sitting time. Moreover, neither of the computer packages support break compliance during other sedentary time, e.g. during meetings or while reading.
After describing the SitCoach application in the next chapter, we present a first pilot user experiment to assess the usability of the application. Through a locus of control questionnaire and by means of a semi-structured interview, we gather additional insights on opportunities and techniques to promote sitting break compliance.
INTRODUCING SITCOACH SitCoach is an iPhone application that measures physical activity by means of the built-in accelerometer. The application records active time and sitting time at a granularity of one minute.
To fight sitting time and inspire people to take a break once in a while, the SitCoach reminds users after a configurable number of in-active minutes via visual, acoustic and tactile messages. Users set their goals in terms of maximum number of consecutive sitting minutes and number of active minutes per day.
Identifying Sitting Time Using the built-in accelerometer in the smart phone, the UHVX¶ DFWLY\ VL FODVLIHG QL DQ DFWHLY GDQ LDFWHQY DWHV Every sHFGRQ D DHVUPXQW RI WHK RSKH¶VQ [ \ GDQ ] positioning is taken by the accelerometer. These three values are compared with the previous measurement. When the difference for x,y or z exceeds 0.3 the accelerometer recognizes a movement. The 0.3 was determined empirically: it is low enough to pick up the walking movement of the user without getting a false reading from other possible movements like a small turn with the chair while sitting.
To distinguish walking from other smaller movements like a small turn or just standing up from a chair the movement will be monitored over a certain interval of time. An empirically determined value of 5 seconds proved to be sufficient.
Creating Sedentary Awareness To motivate users to become more active, the application stores the number of active minutes per day for each of the users. This provides a social nudge for users to see how others are doing and to comply with the social norm.
When it is time to take a break, SitCoach emits a tactile (vibration) and an acoustic warning. Users can override the acoustic warning. A visual indicator at the main screen shows when a user is moving, giving the user immediate feedback about their current behavior. Figure 1 provides a screenshot of the main screen of SitCoach. The green circle indicates that the application has detect that the user is currently moving and hence the number of active minutes is increasing while in this state. In the state displayed in the figure, the user is nine inactive minutes away from a break reminder. However, if the user is active for a period equal to the actual time of the sitting break, the break timer will be reset.
A FIRST USER EVALUATION To assess the usability and user acceptance of the application, SitCoach has been evaluated with users. This evaluation also provides insights into HWK SDUWLFQV¶ current sitting behavior and their awareness of the harmfulness of sedentary behavior. The goal of the study is to identify future directions for persuasive applications targeting sedentary awareness.
In the study, the participants are provided with an iPhone with the SitCoach application and are invited to use the application throughout a day at the office. At the end of the day, a semi-structured interview is conducted, to discuss experiences. Moreover, the participants are questioned about current sitting break habits and the awareness of the importance of such breaks is assessed. Apart from the interview, two questionnaires were handed to the participants: one focusing on the utilitarian and hedonic qualities of the application [5,6] and a second one focusing on the locus of control that people perceive with respect to possibilities to reduce their sitting time [13].
Participants Eight participants (four females) were invited to participate in the experiment, during one working day. All participants were knowledge workers with high computer dependability.
Procedure and Design The participants were scheduled on a day they described as a typical office day. Per participant, a day was selected without having appointments outside the office during working hours.
In the morning after arriving at the office, the participants received a fully charged iPhone 3G. SitCoach was the only application installed, apart from the standard software. The participants were instructed not to use the phone for other purposes. No SIM card was installed, limiting the functionalities of the phone.
the functionality of the applications and guided through the features and settings. The standard break timer was set to 60 minutes, prompting for a 5 minute break. The standard activity goal was set to 50 minutes. The participants were free to change the settings throughout the day.
interviewed based on a list of pre-defined questions on their sitting behavior, sedentary awareness and the SitCoach application. Moreover, the two questionnaires were handed.
the pragmatic and hedonic qualities of SitCoach [5,6]. His scores on both qualities are important for the prolonged usage of a product. Specifically, the questionnaire measures perceived pragmatic quality, hedonic quality identification HL GRHV HWK SURFWGX RFWUHLEXQ RW WHK HUV¶X LGHWQ\ LQ D social context?), hedonic quality stimulation (i.e., does the product help to develop skills or knowledge) and attractiveness (is the product good, bad or ugly?). Each of those four categories contains seven word-pairs on a seven point semantic-differential scale (e.g. discouraging vs. motivating, complicated vs. simple).
sitting behavior, a locus of control questionnaire was assessed [13]. The commonly used questionnaire, developed by Wallton et al., is adapted for sitting behavior. The questionnaire measures whether the control over the sitting behavior is determined internally (i.e. self-control; example statement: If I take care of myself, I can avoid long sitting periods), by others (e.g. Whenever I feel I sit too much and too long, I should consult a trained professional .) or by chance (e.g. No matter what I do, I 'm likely to have long sitting periods).
Results All participants indicated that they were not aware of the harmfulness of sedentary behavior itself. When taking a break and getting up from their desk, the participants did so because they were aware of the adverse effects of prolonged computer usage and the healthfulness of physical activity. Half of the participants reported to be unhappy with the amount of sitting time during a day in the office.
annoyance with PC break applications. Only one of the participants is using an RSI prevention program on the PC, which is installed by default. The others have disabled it.
The Locus of Control questionnaire revealed that six out of eight participants scored low on the internality dimension (scores < 18 on a range from 6 to 36), while the other scored moderate (18 score 24). This implies that the office workers participating in the study believe that they have little control over their sitting behavior. With overall higher scores on the powerful others dimension, it is believed that others (colleagues, managers) strongly
on the pragmatic dimension, implying that the participants are generally positive about the interaction with the
inaccuracies of the application. This suggests that the current implementation is well usable to distinguish sitting time from active time. Lower scores were reported on the hedonic dimensions, most notably on attractiveness. ontrol questionnaire.
Low Low Low
Some of the participants reported battery problems with the smart phone. Although the participants received a fully charged phone, the battery time was not enough for the application to run for the whole working day. Hence, in future work, solutions should be researched that take the energy consumption of the phone into account when running such accelerometer-based applications.
Pragmatic Hedonic Hedonic Quality Quality Quality
I dentification Stimulatio
n High High findings of Munson et al. [11], participants did not feel the need to bother their social network with such details.
AttrakDif2 questionnaire. CONCLUSION AND FUTURE WORK In this paper, we presented an application to assist people to control their sitting behavior. The mobile application combines feedback on physical activity with insights on the UHVX¶ LWVQJ SHULRGV SitCoach was developed to gain VWLQJKRSHO¶ awareness of their sedentary behavior and the user acceptance of a break reminder application.
With SitCoach, we have created an application that detects sitting time with fair accuracy. However, the users involved in the trial showed not to be in the right stage of change to be responsive to the strategies applied in SitCoach.
Persuasive strategies to stimulate the user to take sitting breaks are likely to be more successful after having established awareness of the adverse health effects of sitting behavior. This can be done by first providing insights in ¶HRQ LWJVQ EDHKLRUY QDG EHTXVWOQ\ VHXWLJQ opportunities to reduce sitting time. For users who are aware of the problem and the adverse effects of their behavior, the triggers applied in SitCoach may be revisited.
ACKNOWLEDGMENTS This work was funded by the European Commission, within the framework of the ARTEMIS JU SP8 SMARCOS project ± 100249 - (http://smarcos-project.eu).
REFERENCES 1. Consolvo, S., Everitt, K., Smith, I., & Landay, -$ 'HVLJ³Q THLU5XWPQV RUI 7RLVOQJHFK DWK RFHU(DJQX 3LFDVK\O F$WLY´\ 3VJUGRLQFHI ,+& 457-66. 3. 9.
E6WUX +% K¶6)LVQHWS³ RFU(DQXLJ 3LFDOVK\ WFL$Y\KZDQ,WHUFLYS&RXP*DH´ S8ERL&P 2006. with Personal,
-63.
Fujiki, Y (2010). iPhone as a Physical Activity Measurement Platform. In Proceedings of the 2010 ACM Conference on Human Factors in Computing Systems (CHI).
Hamilton, M.T., Healy, G.N., Dunstan, D.W., Zderic, T.W., and Owen, N. (2008). Too little exercise and too much sitting: Inactivity physiology and the need for new recommendations on sedentary behavior. Current Cardiovascular Risk Reports 2(4), 292-298.
Hassenzahl, M. (2006) Hedonic, emotional and experiental perspectives on product quality. In Ghaoui, C. (ed) Encyclopedia of Human-Computer Interaction.
Hershey: Idea group, 226-272.
Hassenzahl, M. (2010) Attrakdiff. Retrieved August 11th 2010 from http://www.attrakdiff.de/en/AttrakDiff-/What-isAttrakDiff/Scientific-Background.
Heuvel, S.G. van den, Looze, M de, Hildebrandt, V.H., and The, K.H (2003). Effects of software programs stimulating regular breaks and exercises on work-related neck and upper-limb disorders. Scand J Work Environ
Health 29(2):106± 116.
MONARCA: A Persuasive Personal Monitoring System to Support Management of Bipolar Disorder
Gabriela Marcu Human-Computer Interaction Institute
Carnegie Mellon University 5000 Forbes Ave, Pittsburgh, PA, USA
gmarcu@cs.cmu.edu ABSTRACT MONARCA is a persuasive mobile phone application designed to support the treatment and management of bipolar disorder. Behavioral data is monitored through both sensing and manual patient input, while timely feedback is provided based on clinical recommendations to help patients adjust their behavior and manage their illness. This paper presents the design process behind the MONARCA system and initial findings on the challenge of designing a persuasive system for the management of bipolar disorder. We discuss how difficult the design of such technology has turned out to be, for two primary reasons: (1) the inherent challenges of using persuasive metaphors with a complex mental illness, and (2) the tradeoffs encountered due to varying, and sometimes conflicting, stakeholder needs.
Author Keywords Bipolar disorder, mental illness management, user-centered design, personal monitoring systems ACM Classification Keywords H.5.2 Information Interfaces and Representation: User Interfaces – User-centered design. J.4 Social and behavioral systems: Psychology.
INTRODUCTION Persuasive personal monitoring systems seem promising for the management of mental illnesses such as bipolar disorder.
Bipolar disorder is characterized by recurring episodes of both depression and mania, with treatment aiming to reduce symptoms and prevent recurrence throughout a patient’s lifetime. By applying pervasive healthcare technologies to the treatment of bipolar disorder, we can monitor patients’ behavioral and mood data, and provide timely feedback to them in order to help them adjust their behavior. This data supports the treatment and management of the illness in a
Jakob E. Bardram
IT University of Copenhagen Rued Langaards Vej 7, Copenhagen, Denmark +45 7218 5311 bardram@itu.dk multitude of ways. For example, patients and their clinicians can use the data to determine the effectiveness of medications, find illness patterns and identify warning signs, or test potentially beneficial behavior changes. Behavioral data collected could be used to predict and prevent the relapse of critical episodes.
Despite the plethora of research into personal monitoring systems targeting behavior change [8], health-related behavior change (e.g., physical activity [5, 1], diet [9], cardiac rehabilitation [6], and others [3]), and even the management of chronic illnesses (e.g., diabetes [7, 11], chronic kidney disease [10], asthma [4]), mental illness has remained relatively unexplored. One explanation for this untapped potential is the complexity and variation of a mental illness like bipolar disorder, which causes uncertainty in how to manage it. Moreover, there is no simple connection between measurable parameters and the course of treatment; mental illness is fundamentally complex and is often tied into physical health problems as well as social problems. In the MONARCA project we aim to overcome this challenge by developing a system that, through pervasive data collection and feedback to the patient, supports the treatment of bipolar disorder.
As such, the MONARCA system can be classified as a
persuasive technology [
In this paper, we describe the user-centered design process and initial findings on the challenge of designing a persuasive system for the management of bipolar disorder. We discuss how difficult the design of such technology has turned out to be, for two primary reasons: (1) the challenges of using persuasive metaphors with a complex mental illness, and (2) the tradeoffs encountered due to varying, and sometimes conflicting, stakeholder needs.
METHOD
gram took part in an in-depth participatory design process. They were instrumental in decision-making about features through collaborative design workshops and iterative prototyping. Patients participated in semi-structured interviews about the treatment and management of their own illness to further inform the design process. Notes and artifacts from these design activities were analyzed for 1) an understanding of each stakeholder's motivations and needs, and 2) indicators of tradeoffs that arose in the design of the system.
Workshops were held every other week for six months. At every workshop, 1-3 individuals attended from each of the following three stakeholder groups: patients, clinicians, and designers. The designers led each three-hour workshop by facilitating discussion about particular design goals and issues; system features and functionality; and feedback on mockups and prototypes of the system. During initial workshops, overall goals of the system were introduced from both clinical and technical perspectives. Sharing these perspectives of the project involved drawing from their respective best practices: both medically and practically, clinicians know what works with patients; and designers are aware of related systems and technologies.
Design activities at workshops began in the early stages with hands-on brainstorming. We provided materials such as documents summarizing the goals of the system, images of existing tools and methods, large poster paper, writing materials, scissors, tape, etc. The sketches that came out of this initial brainstorming formed the basis for the first mockups. For the rest of the process, at each workshop we
SYSTEM DESIGN
Self-assessment Subjective data is collected through a mobile phone using a simple one-page self-assessment form. Less than 10 items are entered by the patient on a daily basis, including mood, sleep, level of activity, and medication. Some items are customizable to accommodate patient differences, while others are consistent to provide aggregate data for statistical analysis. A simple alarm reminds the patient to fill out the form.
Activity monitoring
monitor level of engagement in daily activities (based on
Historical overview of data
Coaching & self-treatment
forcement in two ways. Customizable triggers can be set to have the system notify both patient and clinician when the data potentially indicates a warning sign or critical state.
Data sharing In order to strengthen the psychotherapy relationship data and treatment decisions are shared between the patient and his/her clinician. Similarly, sharing data with family members or other caregivers empowers the patient to support the treatment process. Finally, sharing data among patients helps with personal coping and management efforts by reassuring patients that they are not alone, and helping them see how others manage their illness.
CHALLENGES WITH A PERSUASIVE METAPHOR One of the main original goals of the user-centered design process was to design a persuasive system for bipolar patients, which could help them constantly adjust their behavior to manage their own illness. In particular, the design process revealed the following three parameters were crucial to keeping a bipolar patient stable: 1. adherence to the prescribed medication – i.e., ensuring that the patient takes his or her medication on a daily basis
of the home, meeting with people, going to work.
Now – at first glance, this may seem simple, but numerous studies have shown that each of the above three things are very difficult to achieve for many patients, and achieving all three consistently is inherently challenging in combination with a mental illness. Hence, the core challenge is to create technology that would help – or “persuade” – the patient to do these three things every day.
Most persuasive health-related Ubicomp systems have adopted different metaphors with the goal of motivating the patient to perform healthy behavior. Examples of such metaphors include a garden that grows when the person is physically active; a fish that grows when the person walks more; and a dog that is happier when the person eats healthy meals. Common to these metaphors is a simple-tounderstand relationship between behavior (e.g. exercise) and visualizations in the metaphor (e.g. more flowers in the garden).
In the design of the MONARCA project, we tried to adopt the same strategy of creating a metaphor. In total of 5 different metaphors were tested and tried out in a series of design workshops. These metaphors included the use of an abstract color picture, a landscape with a river, a dartboard, a music equalizer, and a scale. The patients and clinicians rejected all of these metaphors – one after the other.
Why did this happen? First we thought that maybe we were just bad at designing the metaphors, and we kept on trying with new ones. But since it turned out to be a persistent “problem”, we think that something more fundamental was at stake, which was expressed by one of the patients as:
“I do not want my illness to be reduced to a game.” We think that this is an important insight into the design of persuasive technologies for healthcare and selfmanagement. Many of the technologies and metaphors reported so far deal with personal lifestyle related health management, which is fundamentally different from patients with a diagnosed mental illness. We think that the design of feedback to the patient needs to follow another pattern other than using a metaphor.
DESIGN TRADEOFFS During the user-centered design process, we discovered several tradeoffs in the design of the system due to conflicting stakeholder needs and motivations. These tradeoffs relate to the clinical efficacy of the system, the patient’s privacy, sustained use of the system, and other issues. In this section, we highlight two of the primary tradeoffs we dealt with during the design of MONARCA.
Clinically driven vs. patient driven strategies If a system has a strong clinical focus – meaning that it adopts only clinically proven treatment strategies – it could miss out on patient-driven approaches that may be helpful to some patients. In addition, the system may also ignore novel technological solutions that the clinical field has yet to evaluate. Since our system was designed for a clinical context, it was important that it adhere to clinical practices so that it could be evaluated as a valid intervention. In addition, considering clinical practices was crucial in designing a system to be viable for adoption and acceptance into a patient's treatment, which includes everyday use by the patient and occasional use by the clinician.
The clinicians that took part in our design activities shared with us scenarios, anecdotes, and commonalities about the treatment of their patients. We understood the context we were developing the system for by understanding the practices of clinicians with their patients. A recurring theme was clinicians' limited resources. This turned into a limitation for the functionality of the system, because if something took too much time or attention on the clinician's part, the clinicians would reject it. An example of one such feature was the system suggesting that the patient contact the clinic if data collected indicated possible warning signs – and making it easy for the patient to place this call. The motivation behind this feature was to encourage the patient to reach out for help when needed, but the clinicians ultimately rejected the idea because we could not find a reasonable protocol to make the benefits to the patient outweigh the burden on the clinic's resources. Features of the system also couldn't present a liability for clinicians, so they were more likely to reject ideas and limit the role of the system to be on the safe side. Any kind of text messages or notes written by the patient and made available to the clinic were kept out of our design, because we could not ensure that the clinicians would always read these messages, so we could not make them liable for their content.
We therefore realized that designing our system with primarily a clinical focus was limiting. The clinicians we worked with were clearly most comfortable with strategies that they were familiar with, they had evidence for based on their experiences with patients, and were backed by clinical trials. Deviating from these practices somewhat, and pushing our clinicians a little bit out of their comfort zone, enabled us to explore other potential strategies, from the perspectives of the patients and the designers.
An additional example of a debated feature is reported stress level. A stress level scale was strongly rejected by a clinician who argued that stress is not a clinically useful measure, nor is there any clinical definition of stress that would support accurate data collection. Interestingly, a second clinician was the one who suggested the stress level scale, and argued for it from a very patient-centered perspective based in psychotherapy. This clinician found that external stressors play a significant part in the mood of her patients, and it was useful for her to consider a patient's reported stress level when assessing how that patient was doing. She also believed that patients would find it useful to assess their own level of stress, regardless of the fact that they would be interpreting its meaning for themselves in the absence of a clinical definition. The patients tended to agree with her, so although this feature was under debate for several weeks, the designers opted to keep it in the design because enough participants believed there could be personal value in assessing one's stress.
The patients were creative in suggesting strategies based on their personal experiences. Knowing what behavioral changes have worked for them in the past, and imagining what new strategies might work for them, patients explored technological solutions unrestrained by considerations of clinical efficacy. This unrestrained creativity was productive during the design process for two reasons. First, it revealed what would motivate the patients to use the system, which is critical to adoption and acceptance. Second, it helped us realize which measures, though clinically significant, would ultimately fail because they were too intrusive for the patient to collect, or were not interesting enough to the patient to motivate collection.
Egocentric patient bias vs. clinician generalizations Although patients provide valuable insights into the experience of living with and managing bipolar disorder, their input tends to be egocentric, since their knowledge about the disorder mostly comes from their own personal experience with it. Discussions about the amount and type of data to collect were complex due to the different experiences and motivations of the stakeholders: clinicians were interested in data they knew to be relevant for assessment based on clinical studies or their own experiences treating patients; and patients were interested in data they thought would be useful to themselves personally for self-reflection. To balance these sometimes opposing interests, designers focused on what data would be easy and convenient to collect. Without nonintrusive data collection methods, the system will be overloaded with features and burden the patients, who are responsible for collecting the data every day. Here, the designers play an important role in keeping in perspective the implications of collecting different amounts and types of data.
Patients and clinicians disagreed about how to include customizable personal warning signs, which patients would personalize and track on a daily basis. In addition to the universal warning signs that we selected with the help of clinicians to be applicable to most, if not all, patients, we discussed including personal warning signs that each patient could customize based on personal symptoms. Clinicians argued that there should be as few of these items as possible, even stating that one personal warning sign was difficult enough for patients to attempt to track in their daily life. On the other hand, patients argued that having more flexibility would allow them to explore multiple warning signs at once in order to determine which ones applied to them. One patient, who had difficulty understanding her illness and could not identify any of her personal warning signs, asked for a lot flexibility because she would have no idea what to track, so she would need to try many different items. The designers found a solution by suggesting that the feature be limited but flexible. The agreed upon solution would allow patients the option to include as few as one personal warning sign, but no more than three. Those patients who would only be able to handle one item at a time could customize the system to show only one at a time.
CONCLUSION In the design of a persuasive personal monitoring system for bipolar disorder, we ran into several challenges unique to using persuasive technology for the management of mental illness. Our findings demonstrate that the design of a system for bipolar disorder is quite different from that of systems that have been explored for other health purposes such as nutrition, physical activity, and chronic physical illnesses. In this paper we have highlighted some of the main issues that emerged during our design process, including using a persuasive metaphor, balancing clinical- and patient-centered strategies, and dealing with the biases of patient and clinician participants. Our work revealed major challenges due to the complexity of the illness, stigma surrounding the illness, and the often-conflicting needs of clinicians and patients.
ACKNOWLEDGMENTS This work has been partially funded by the EU Contract Number 248545 - MONARCA under the 7th Framework Programme. We would like to thank our participants for their contributions to this project and enthusiasm for the work.
Building Persuasion Profiles in the Wild: Using Mobile Devices as Identifiers.
Maurits Kaptein
Eindhoven University of Technology / Philips Research
Den Dolech 2, 5600MB, the Netherlands. m.c.kaptein@tue.nl ABSTRACT Tailoring — presenting the right message at the right time — has long been identified as one of the core opportunities of persuasive systems. In this paper we describe a scenario in which an adaptive persuasive system which identifies users by the Bluetooth key of their mobile phone is used to promote energy savings. By describing this simplistic system and its possible implementation we identify several key-criteria of adaptive persuasive systems.
Author Keywords Persuasive Technology, Influence strategies ACM Classification Keywords H.1.2 User/Machine Systems: Software psychology.
INTRODUCTION CHI2010 attendees were presented with a choice on entering the conference hotel: A large revolving door provided access to the hotel while next to it was a sliding door—some things simply do not fit through a revolving door. With the air conditioning in full operation revolving doors are efficient at keeping the heat in.
Sliding doors, however, are not. To help save energy a paper-sign was put up: “Please take the revolving door”.
A brief observation proved the paper-sign to be effective just over half the time: 60% of the visitors took the revolving door. This scenario, the “Revolving Door Problem”, offers a framework to describe adaptive persuasive systems. By further elaborating this scenario and exploring a solution we describe the neccesities and difficulties that arise when designing adaptive persuasive systems.
The Promises of Persuasive Technology There are three reasons why employing a persuasive system might be more effective than the current papersign: (1) Persuasive technologies function as social actors and can use social influence strategies, (2) they can be context aware, and (3) they can adapt to individual users [5, 8]. While the paper-sign is probably located at the right place and at the right time—when visitors make their choice—the current version does not implement social influence strategies and does not adapt to its users.
Social Influence Strategies
Cialdini [
The final promise of persuasive technologies however— adapting influence attempts to individuals—will require some kind of interactive system. While adaptation of persuasive strategies to responses by users is mentioned early on in the literature on persuasive technologies Fogg [5, e.g.] we are unaware of any actual implementations.
Individual Differences There is growing evidence that individuals differ in their responses to influence strategies: Constructs like Need For Cognition [1] predict the response of individuals to the usage of social influence strategies. More concretely, Kaptein et al. [9] show that usage of influence strategies for individuals who are low susceptible to these strategies can lead to backfiring: for a portion of participants in their study compliance to a request was lower when the social influence strategy was presented. Next to this overall tendency to respond to influence strategies, some individuals seem more likely to respond to one specific strategy—e.g. an authority argument—while others are more influenced by implementations of other strategies.
Cialdini et al. [3] shows that there are sizable and stable individual differences in people’s responses to the commitment strategy. Similar results have been obtained when looking at the consensus strategy: Self-reported susceptibility to this strategy highly correlates with behavioral responses to this strategy [10].
These individual differences in susceptibility to different persuasive strategies imply that persuasive systems should personalize the way in which they attempt to influence individuals. Such a class of systems, which we call adaptive persuasive systems, are an unexplored area in that we still need to understand how to model, design and build these systems. This paper takes a concrete but simple example that encapsulates the quintessence of this problem to discuss how to address these challenges.
SOLVING THE REVOLVING DOOR PROBLEM? Returning to the revolving door problem, let us consider what is involved in implementing an adaptive persuasive system. We need to (A) identify the visitors entering the lobby—minimally by giving each a unique ID, and (B) measure the effectiveness of a presented message. The Bluetooth key of visitor’s mobile phone could be used for identification [11]. This will capture around 12% of the visitors entering the lobby. This same identification method can also be used to measure the effectiveness of each persuasive attempt: One Bluetooth scanner next to the revolving door and one next to the sliding door could determine which entrance was used by the current visitor. Based on this knowledge about the visitor and records of earlier decisions a message implementing the right influence strategy can be selected. In the remainder of this paper, we focus on the mechanism by which these strategies can be selected.
Suppose we have only two messages to show, one implementing the authority strategy—“The general manager of this hotel urges you to...” (A)—and one implementing the consensus strategy—“80% of our visitors always use...etc.” (B). The system then needs a mechanism to choose the message that is most likely to be effective for the current visitor. It is intuitive that for a new visitor the system should present the message which has lead to the highest compliance for other, previously observed, visitors. If this message is successful then there is no need to try different messages on subsequent visits.
However, when the selected message is not effective, it might become attractive to present another message on the next visit. This decision logically depends on the initial succes probabilities of the messages under consideration, the variance of effectiveness of messages between visitors, and the number of succes’s or failures observed for the current visitor. A collection of estimates of the effectiveness of different influence strategies for an individual is called a Persuasion Profile and can be used to select the most-likely-to-be effective message on a next visit.
Formalizing the Adaptation Problem The probability of a single visitor taking the revolving door on multiple occasions can be regarded a binomial random variable B(n, p) where n denotes the number of approaches the visitor has made to the doors and p denotes the probability of success: the probability of taking the revolving door. Given M messages one can compute for each individual, for each message, probability pm = km/nm where km is the number of observed successes after representation of message m, nm times to a specific visitor. It makes intuitive sense to present a visitor with the messages with the highest pm.
For a large number of observations N of one visitor this would make perfect sense. However, this will not inform a decision for a newly observed visitor. For a new visitor one would present the message m for which pm is maximized over previously observed visitors1. Actually— given Stein’s result [4]—for every user a weighted average of the pm for an individual user and those of other users—one where the estimated pm for an individual is “shrunk” toward the population mean—will provide a better estimate than an estimate based on observations of a single visitor alone. E.g., if the authority message is effective 70% of the time over all visitors and only 30% percent of the time for the specific visitor under consideration, the best estimate of the (real) effectiveness of the authority message pA for this visitor is a weighted average of these two.
Adapting to Individuals To include both the known effectiveness of a message for others, and a specific visitors previous responses to that same message, into a new estimate of message effectiveness, pm, we use a Bayesian approach. A common way of including prior information in a binomial random process is to use the Beta-Binomial model [12].
The Beta Beta(α, β) distribution functions as a conjugate prior to the binomial. If we re-parametrize the beta distribution as follows
π(θ|µ, M ) = Beta(µ, M ) where µ = α+αβ and M = α + β, then the expected value of the distribution is given by: E(θ|µ, M ) = µm.
In our scenario this represents the expected probability of a successful influence attempt by a specific message.
The certainty of this estimated success probability is represented by:
V ar(θ|µ, M ) = σ2 = µ(1 − µ)
M + 1 After specifying the probability of success µm of message m and the certainty about this estimate σm2 we can treat this as our prior expectancy about the effectiveness of a specific message and update this expectancy by multiplying it by the likelihood of the observation(s) to obtain the distribution of our posterior expectation: p(θ|k) ∝ l(k|θ)π(θ|µ, M ) =
Beta(k + M µ, n − k + M (1 − µ)) 1This is assuming the error costs—the effects of presenting the wrong message—are equal for each message. !"#$%&'("# ) *+&,"-%.* ) /0# ) %-/('+-&%-1 ) &,0%&+.2 )!"#$%&$'()* +-&0*!"..+.)")-'*3+#)0/)*+"-.)45625789) :"-; ) !+#.'".%<+ ) .$#"$+1%+.2 ) .'&, ) ". ) .+(/=*0-%$0#%-12) !+#.0-"(%>"$%0-2)$"%(0#%-12)"-?).0&%"()&0*!"#%.0-)45782)#+(;) 0- ) %-/0#*"$%0- ) "30'$ ) $,+ ) '.+#@. ) &0-$+A$ ) "-? ) "&$%<%$%+.9) B-?++?2 ) $C0 ) 0/ )D+#?%&,+<.E; ) "-? ) F+'+-.&,C"-?+#@.) !#%-&%!(+.)!0%-$)$0)!#'+&,-)".)")<"('+)0/)!"#$%&'("#)&0-&+#-G
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
BI
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
C0'(?)-0$)&0-.+-$)$0)3+)!+#.'"?+?)$0)?09)4L8 Q(( ) $,#++ ) !#%-&%!(+. ) /0&'. ) 0- ) '-"&&+!$"3(+) +-?. ) /0#) !+#.'".%0-9 ) J,+; ) !#0<%?+ ) -0 ) 1'%?"-&+ ) ". ) $0 ) C,"$ ) +-?.) C0'(?)3+)?+.%#"3(+9)Q$$+-$%0-)$0)<"('+.)&"-)(+"?)$0)?+.%#"3(+) +-?. ) /0# ) 3+,"<%0# ) &,"-1+9 ) B-?++?2 ) *'&, ) !+#.'".%<+) $+&,-0(01;),".)+A!(%&%$(;)$"#1+$+?),+"($,)0#)+-<%#0-*+-$"() .'.$"%-"3%(%$;9 ) Q($,0'1, ) $,+.+ ) "#+ ) ("'?"3(+ ) 10"(.2 ) !+#,"!.) C+ ) .,0'(? ) "(.0 ) 3+ ) ?+.%1-%-1 ) !+#.'".%<+ ) $+&,-0(01; ) $,"$) ,+(!.)'.)$0)0<+#&0*+)0'#)#"&%"()3%".+.)Y4#"".(2*4#(2*5'&$I2) &0-$#0()0'#)"-1+#)Y,&12)"$$I2)"-?)(+"#-)$0),+(!)"-?)#+(;)0-) 0'#)-+%1,30#.)Y/#%$/I9)S'#$,+#2)%$)%.)%*!0#$"-$)$0)'-?+#.$"-?) $,+)<"('+.)0/)$,0.+)C+)"#+)?+.%1-%-1)/0#9 3&'(#$/#,)4*,) J,+ ) *0.$ ) 03<%0'. ) <"('+ ) $+-.%0-. ) %- ) KBFP ) $+&,-0(01; ) !%$) ?+.%#+? ) 3+,"<%0# ) &,"-1+. ) "-? ) ) $,+ ) <"('+. ) $,+; ) %*!(%&"$+) "1"%-.$ ) $,+ ) %-$+-$%0- ) $0 ) &,"-1+ ) 3+,"<%0# ) "-? ) *+$,0?.) /0#) ?0%-1).09)J,"$)%.2)+-?.)&"-)3+)%-)$+-.%0-)C%$,)*+"-.9)Z+) .++)!#0*0$%-1),+"($,2)+-<%#0-*+-$"().'.$"%-"3%(%$;2)"-?).0) 0-2)<+#.'.)!#+.+#<%-1)"'$0-0*;2)!#%<"&;2)"-?).0)0-9) [0C+<+#2)$,+.+)"#+)-0$)$,+)0-(;)$;!+.)0/)$+-.%0-.9))S%#.$2)$,+) "&$ ) 0/ ) !+#.'".%0- ) %-,+#+-$(; ) !#%<%(+1+. ) $,+ ) <"('+. ) 0/ ) $,+) !+#.'"?+# ) 0<+# ) $,0.+ ) 0/ ) $,+ ) !+#.'"?+?9 ) D; ) ".E%-1 ) ;0' ) $0) &,"-1+);0'#)3+,"<%0#2)B)"*).";%-1)$,"$)*;)<"('+.)"#+)*0#+) %*!0#$"-$)$,"-);0'#)<"('+.)Y0#)"$)(+".$2)$,+)<"('+.);0').++*) $0 ) 3+ ) "&$%-1 ) 0-I9 ) B- ) $,+ ) 3+.$ ) &".+2 ) ". ) %- ) 3+,"<%0# ) &,"-1+) .'!!0#$).;.$+*.2)$,+)!+#.'"?+#)"-?)$,+)!+#.'"?+?)"1#++)0-)") <"('+ ) .'&, ) ". ) ,+"($, ) 0# ) +-<%#0-*+-$"( ) .'.$"%-"3%(%$;\ ) $,+) !+#.'"?+# ) !#0<%?+. ) %-/0#*"$%0- ) 0# ) .'!!0#$ ) $0 ) ,+(! ) $,+) !+#.'"?+?)"&$)%-)"&&0#?"-&+)C%$,)$,%.).,"#+?)<"('+9) ]+&0-?2 ) !+0!(+ ) *"; ) "1#++ ) 0- ) <"('+. ) 3'$ ) ?%."1#++ ) 0-) !#%0#%$%+.9)Z+)*%1,$)"1#++)$,"$)+-<%#0-*+-$"().'.$"%-"3%(%$;) %.)C0#$,C,%(+N3'$)B)*%1,$)#"$+)$,+)&0*/0#$)0#)+A&%$+*+-$) 0/ ) ?#%<%-1 )". ) *0#+ )%*!0#$"-$9 )B-?++?2 )X0E+"&, )&0*!"#+?) %-?%<%?'"(.@)<"('+).;.$+*.).0(+(;)0-)$,+)3".%.)0/)?%//+#+-&+.) %-)$,+%#)#"-E%-1.)0/)").+$)0/)!#+?+/%-+?)<"('+.)4R589 5&6#$7#"&%4*08$-4++#0#,/$%&'(#) S%-"((;2))!+0!(+)*";)"1#++)0-)")?+.%#+?)3+,"<%0#2)3'$),"<+) ?%//+#+-$ ) #+".0-. ) /0# ) <"('%-1 ) $,"$ ) 3+,"<%0#9 ) S0# ) +A"*!(+2) /%<+ ) !+0!(+ ) *%1,$ ) &,00.+ ) $0 ) ?#%<+ ) 3+(0C ) $,+ ) .!++? ) (%*%$2) !"#$ % &'( % )$!*( % '+, % (!"-',-.% )' % '/!0 % )$! %!"#1 % )' % 2(')!#)% $"%&'(1% )'% 2("#)*#! %')*+%'1 % )' % (!34#! % 3!2!,3!,#! % ', % &'(!*5,% '*6%",3%2(')!#) %,"'+-,"!.$&/0*+'(1%'(%)'%(!34#!%)$!%,!!3%&'(% '*6%3(*66*,5%",3%#',)(*/4)!%)'%&,1+*-,2&,'"!.$0$'"+,"3+!+'(4 7-%8'55%2'*,)-%'4)9%)$!%:'/*6!%2$',!%*-%",%*,)*:")!%3!;*#!% <=>?%@&%*)%3'!-%,')%-$"(!%'4(%5'"6-9%/4)%(")$!(%$"-%5'"6-%'&%*)-% '+,9 %+!%&!!6 %/!)("0!3 %<A>?%B$! %-":!%+'463%-!!: %)'%$'63% &'(%;"64!-?%C422'-!%)$")%:0%$*5$!-)%;"64!%*-%)$!%-"&!)0%'&% :0 % #$*63(!,? % @& % @ % "3'2) % " % :'/*6! % "226*#")*', % )' % $!62 % :!% ";'*3%-2!!3*,59%",3%*)%-$'+-%:!%2*#)4(!-%'&%2'6"(%/!"(-9%@% +*66 % /! % 42-!)? % D!#"4-! % *) % #$"66!,5!- % :0% ;"64!-9 % @ % -!! % )$!% "226*#")*', % "- % " % )$(!") % )' % :0% "4)',':09 % ",3 % @ % !E2!(*!,#!% 2-0#$'6'5*#"6 % (!"#)",#! % <F>G6!"3*,5 % :! % )' % 3(*;! % !;!,% &"-)!(?%@,-)!"39%@%-$'463%/!%(!:*,3!3%'&%:0%;"64!%'&%-"&!)0? @%-!!%)+'%"22('"#$!-%)'%"33(!--*,5%*,3*;*34"6%4-!(-H%;"64!-?% 8*(-)9 %3!-*5,!(-H %;"64! %#'::*):!,)-%-$'463%/!%:"3! %#6!"(% )$('45$%/(",3*,5%",3%)$!%*,&'(:!3%#',-!,)%2('#!--9%-'%)$")% 4-!(-%#",%:"I!%*,&'(:!3%#$'*#!-?%C!#',39%*,)!(&"#!-%-4#$% "-%B'339%J'5!(-9%",3%K"0,!H-%*,&'(:")*;!%5('#!(0%-$'22*,5% #"() % <LM> % -$'463 % /! % )"*6'("/6!? % B$!0 % -$'463 % 42$'63 % 4-!(% "4)',':0% /0% "66'+*,5 % 4-!(- % )' % #$''-! % +$*#$ % *,&'(:")*',% ":',5 % ;"64!N6"3!, % '2)*',- % O!?5?9 % -4-)"*,"/*6*)09% $!"6)$&46,!--9 % ",3 % #'-)P % )' % 3*-26"0 % :'-) % 2(':*,!,)60? % 7% 3",5!(%*-%)$")%4,3*-#6'-!39%*,;'64,)"(0%)"*6'(*,5%:"0%#('--% &(':%2!(-4"-*',%)'%:",*246")*',%<FQ9FR>? !"#!$%&'"# 7))!,)*', % )' % ;"64!- % :"0 % #',)(*/4)! % ,') % ',60 % )'% 4,3!(-)",3*,5%!)$*#"6%*--4!-%'&%:'/*6!%K@ST%)!#$,'6'50G /(*,5*,5 % "))!,)*', % )' % #',#!(,- % /!0',3 % 2(*;"#0 % ",3% 3*-#6'-4(!G/4) % "6-' % )' % *,#(!"-*,5 % )$!*( % -#'2! % ",3% !&&!#)*;!,!--G)$!*(%2'+!(%)'%3'%5''3%*,%)$!%+'(63?%84()$!(% +'(I%-$'463%#6"(*&0%('6!%'&%)$!-!%;"64!-%)$('45$%!:2*(*#"6% ",3%)!#$,*#"6%*,;!-)*5")*',-%'&%K@ST%)!#$,'6'50? ()*)()#!)& F?76!"$:"39%B?9%!)%"6?%8*-$*,5%&'(%-4-)"*,"/*6*)0.%B$!%!&&!#)-%'&% *,3*(!#)%",3%3*(!#)%2!(-4"-*',?%@,.56'4.73$'*"/'$.89:.;<<=9%7TU%
K(!--%OLVV=P? L?7,3!(-',9%C?%T'!(#*',?%@,%>)&.?'",%-*@.5,/(/!-A&@+".-% .
B)+!-$-A)(%OLVVWP9%$))2.XX26")'?-)",&'(3?!34X!,)(*!-X#'!(#*', M?D!(3*#$!;-I09%Y?%",3%S!4!,-#$+",3!(9%Z?%B'+"(3%",%!)$*#-%'&%
2!(-4"-*;!%)!#$,'6'50?%878C%[L9%Q%OU"0%FAAAP9%QFNQ=? [?T*"63*,*9%J?D?%:,%!0&,/&D.>)&.?/+&,/&.-%.B&*$0"$+-,9%T'66*,-9%
(!;*-!3%!3?%OFAA=P? Q?T\!-I*-9%7?9%!)%"6?%K"(!,)*,5%&(':%)$!%2'#I!).%]"64!%)!,-*',-%",3% )!#$,*#"6%3*(!#)*',%&'(%-!#4(!%",3%2(*;")!%2"(!,)N)!!,%:'/*6!% -"&!)0?%@,%B*-/4.?EFB?.;<G<9%7TU%K(!--%OLVFVP? W?Y";*-9%^?%Z)$*#"6%3!-*5,%:!)$'3-%&'(%2!(-4"-*;!%)!#$,'6'50?%@,%
B*-/4.B5H?F7?:I5.;<<J9%7TU%K(!--%OLVVAP? R?Z-6":/'6#$*6"(9%K?9%_*6-',9%U?`?9%a"I6!09%@?%K@ST.%K!(-4"-*',9% *,&64!,#!9%,435!9%",3%#'!(#*',%)$('45$%:'/*6!%3!;*#!-?%B'% "22!"(K.56'4.73$'*"/'$.89:.;<GG9%7TU%K(!--%OLVFFP? Y?%Z#I6!-%O!3-?P9%C-3+!&.B&*$0"$+-,K%C)",&'(3%T"2)'6'50%U!3*"%
OLVVRP9%QNFF? A?8'559%D?^?%@,#(!"-*,5%2!(-4"-*',%)$('45$%:'/*6*)0?%@,%D?^?%8'55% ",3%Y?%Z#I6!-%O!3-?P9%C-3+!&.B&*$0"$+-,K%C)",&'(3%T"2)'6'50%
U!3*"%OLVVRP9%FQQNFWM? FV?8'559%D?^?%B&*$0"$+1&.>&/),-!-L(D.F$+,L.8-2A0'&*$.'- .
8)",L&.M)"'.M&.>)+,N.",@.O-K.U'(5",%b"4&:",,%OLVVMP?% FF?8(*!3:",9%D?9%c'+!9%Y?9%",3%8!6)!,9%Z?%@,&'(:!3%T',-!,)%*,% )$!%U'\*66"%D('+-!(.%@:26!:!,)*,5%]"64!NC!,-*)*;!%Y!-*5,?%@,% B*-/&&@+,L$.-%.')&.PQ').9"#"++.:,'&*,"'+-,"!.8-,%&*&,/&.-, .
?($'&2.?/+&,/&$%OLVVLP? FL?8(*!3:",9%D?9%b"$,9%K?c?%^(?9%",3%D'(,*,59%7?%]"64!%C!,-*)*;!%
Y!-*5,%",3%*,&'(:")*',%-0-)!:-.%B$(!!%#"-!%-)43*!-?%@,%K?%d$",5% ",3%Y?%e"66!))"%O!3-?PK.%902",R8-2A0'&*.:,'&*"/'+-,.",@ .
C","L&2&,'.:,%-*2"'+-,.?($'&2$D.S-0,@"'+-,$4%U?%Z?%C$"(2!%
OLVVWP9%M[=NMRL? FM?8(*!3:",9%D?9%",3%S*--!,/"4:9%c?%D*"-%*,%#':24)!(%-0-)!:-?%
78C.>E:?%GT9%MOFAAWP9%MMVNM[R? F[?^'$,-',9%Ye?9%",3%U46;!09%^?U?%7##'4,)"/*6*)0%",3%#':24)!(%
3!#*-*',%-0-)!:-?%878C.P=K.FL%OFAAQP9%Q=NW[? FQ?b"2)!*,9%U?%",3%Z#I6!-9%Y?%C!6!#)*,5%!&&!#)*;!%:!",-%)'%",0%
!,3?%@,%B*-/4.B5H?F7?:I5.;<G<9%`STC%WFMR%OLVFVP9%=LNAM? FW?`!;*,!9%Y?%f-*,5%)!#$,'6'50%)'%2(':')!%-!E4"6%$!"6)$?%@,%D?^?% 8'55%",3%Y?%Z#I6!-%O!3-?P9%C-3+!&.B&*$0"$+-,K%C)",&'(3%
T"2)'6'50%U!3*"%OLVVRP9%FQNF=? FR?U*#$"6-I*9%^?%Z)$*#"6%3",5!(-%'&%:'/*6!%2!(-4"-*',?%@,%D?^?% 8'55%",3%Y?%Z#I6!-%O!3-?P9%C-3+!&.B&*$0"$+-,K%C)",&'(3%
T"2)'6'50%U!3*"%OLVVRP9%FMRNF[L?? F=?a*,"-Nb4II',!,9%c?%D!$";*'(%#$",5!%-422'()%-0-)!:-.%7% (!-!"(#$%:'3!6%",3%"5!,3"?%@,%B*-/4.B5H?F7?:I5.;<<=9%`STC%
QVMM%OLVV=P9%FW[NFRW? FA?a*,"-Nb4II',!,9%c?%",3%c"(g4:""9%U?%%7%-0-)!:")*#% &(":!+'(I%&'(%3!-*5,*,5%",3%!;"64")*,5%2!(-4"-*;!%-0-)!:-?%@,%
B*-/4.B5H?F7?:I5.;<<=9%`STC%WFMR%OLVFVP9%[NF[? LV?K"5!9%J?Z?%",3%b("09%T?%Z)$*#-%",3%2!(-4"-*;!%)!#$,'6'50.%7,% !E26'(")'(0%-)430%*,%)$!%#',)!E)%'&%$!"6)$0%6*;*,5?%@,%B*-/4.S+*$' . :,'4.M-*N$)-A.-,.U0@L&.",@.:,%!0&,/&.')*-0L).C-3+!&.O&1+/&$9% 85FH.M-*N$)-A.B*-/&&@+,L$.VJ<%OLVFVP9%FANLL? LF?J'I!"#$9%U?%>)&.U"'0*&.-%.902",.I"!0&$?%8(!!%K(!--%OFARMP? LL?B$"6!(9%J?%c?%",3%C4,-)!*,9%T?J?%U0@L&D.:2A*-1+,L.O&/+$+-,$ .
"3-0'.9&"!')K.M&"!')K.",@.9"AA+,&$$K%h"6!%f,*;!(-*)0%K(!--%
OLVV=P? LM?B'339%K?U?9%J'5!(-9%h?9%",3%K"0,!9%C?^?%S435*,5%)$!%#"()%*,% )$!%-42!(:"(I!)1%c'+%:4#$%*-%!,'45$%*,&'(:")*',%&'(%&''3% -$'22!(-i%@,%B*-/4.S+*$'.:,'4.M-*N$)-A.-,.U0@L&.",@.:,%!0&,/& . ')*-0L).C-3+!&.O&1+/&$9%85FH.M-*N$)-A.B*-/&&@+,L$.VJ<.
OLVFVP9%LMNLW? L[?_"-$*,5)',9%^?%B$!%,!+%3*5*)"6%3*;*3!?%O&$.C-+,&$.H&L+$'&*.
O^",4"(0%A9%LVFFP9%F779%Q77? LQ?_("09%J?%7&(*#"%-!!-%:"--*;!%5('+)$%*,%:'/*6!%+!/%4-"5!9%
L0"*@+",4/-40N%OY!#!:/!(%L9L%LVVAP9% $))2.XX+++?54"(3*",?#'?4IX)!#$,'6'50XLVVAX3!#XLLX:'/*6!2$', !-N*,)!(,!)? Opportunities and Challenges in Mining Behavioral Economics to Design Persuasive Technology
Min Kyung Lee Human-Computer Interaction Institute
Carnegie Mellon University Pittsburgh, PA 15213 USA
mklee@cs.cmu.edu ABSTRACT Behavioral economics examines people’s decision making processes in everyday situations. I argue that behavioral economics can provide a repertoire of a tool that can inform the design of persuasive technology. In this position paper, I propose strategies drawn from behavioral economics, and identify opportunities and challenges in applying the strategies to the design of persuasive technology. This position paper is a modification of the paper [16].
Author Keywords Persuasive technology, behavioral economics, decision making, decision bias, choices ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous.
General Terms Design INTRODUCTION The role of information technology in people’s daily decision making is steadily growing. For example, we decide which route and transportation to take to visit a friend’s house, which restaurant to go for dinner, or which grocery products to buy based on the information and choices presented in information technology applications.
This change offers tremendous opportunities for humancomputer interaction (HCI) researchers to provide interventions to assist people to make self-beneficial or prosocial choices.
As one way to promote self-beneficial choices, we suggest approaches drawn from the field of behavioral economics.
Behavioral economics examines the gamut of large and small decisions people make about such choices as how much to invest in retirement savings, whether to join a health club, and whether to eat a delicious but caloric candy bar. The persuasive element in this approach consists of presenting choices in a way that leverages people’s decision processes and induces them to make self-beneficial choices [17].
We argue that designs for HCI that leverage behavioral economics theory and research are a highly promising avenue for persuasive technologies. Although widely discussed outside the HCI and design communities in both academic and popular arenas (e.g., [24]), this approach has not yet influenced our field. The message of behavioral economics is simple: people are susceptible to decision bias, which often makes it hard for them to make selfbeneficial choices. Thus, we should present choices in a way that helps people to make self-beneficial choices and understand the implications of their decisions as well—all without restricting their freedom of choice.
In this paper, I explain several behavioral economics theories and discuss opportunities and challenges in applying the theories to the design of persuasive technology.
APPLYING BEHAVIORAL ECONOMICS Departing from the premise of economics that people make rational choices, behavioral economists have shown that people’s decision making processes are biased by various situational factors, such as the manner in which options are presented and the times when the choices are offered, and the emotional or visceral state of the person at the time of choice [1, 12]. This understanding of people’s decision biases provides a rich repertoire of tools that designers can leverage. In this section, we present five decision biases and discuss how these biases can be leveraged in the design of persuasive technology.
Default Bias When people make choices, they tend to favor the default option or the status quo, rather than taking the time to consider and then adopt an alternative state [11, 21]. People tend to take “the path of least resistance,” and keep doing what they have been doing, or doing what comes automatically, even when they can make improvements.
The reasons for this decision bias could have roots in people’s limited attention and tendency to “satisfice” [21], their perception that an organization’s selection of a default option constitutes a recommendation (see [6]), and the implied popularity of the default option.
Default biases have been blamed for a wide range of undesirable outcomes, including Americans’ excessive consumption of fries and large sodas as part of “supersized” meals at McDonald’s [17]. Yet if carefully designed, the default bias can be a powerful tool to propel people toward self-beneficial behaviors (see [5, 23]).
Opportunities Convenience and salience. HCI design can leverage the default bias in many ways, by making healthy choices more convenient and salient physically and cognitively. In the domain of snacking, featured healthy snacks can be made easy to access, e.g., on websites, on vending carts, and so forth. For example, on a website, the checkbox of healthy snacks among available options could be selected as the default, reducing the need to select one of these options explicitly. Or when presenting sale items at a bakery, a system could filter and first offer items that are made with whole grain flours. For a kiosk system, the placement of buttons, the number of clicks or the number of screens a user has to access to choose an item could be decreased or increased to change the perceived priority of a snack or sandwich order.
An eldercare robot working in a nursing home could organize the physical placement of food in a way that the healthy food is placed closer to an elder’s room. In addition, a snack delivery robot might only deliver healthy snacks to people’s offices, but require people to walk to the robot to get unhealthy snacks.
Convenience can be further leveraged using sensing technologies that tell people when they are near healthy snacks. For instance, if shoppers are in a food court in a mall, the system could present healthy choices to them via mobile phone as convenient food options.
Default bias is different with other biases presented in the paper; leveraging default bias can be effective, even with those who are not motivated to change their potentially problematic behaviors, or are not aware of issues with their current behaviors [16].
Attention span. People might be more subject to default bias when their attention spans are limited or when they do not have enough time to do exhaustive search. HCI technology can target moments when people’s attention spans are limited, such as when they are using mobile devices on the move, or when people are making decisions with limited time, such as when they are ordering food in a fast-food restaurant, or making choices in a public kiosk.
Interface components can be also designed to manipulate people’s attention spans. The use of banners or graphic images may be distracting [1], reducing people’s attentiveness and efforts in making decision.
Challenges Depending on the way it is implemented, the default strategy may harm people’s experience of making a choice [16]. Explicitly suggesting a certain options as default may cause people to feel forced to make those choices. Careful design of the strategy and iterative testing of its efficacy and its impact on people’s experiences will be important.
Another caveat in using default strategy might be its lack of educational effect. In comparison to persuasive techniques that use informative messages (e.g., indicating consequences of choices), the default strategy do not provide any information that people can use to reflect on their behaviors and learn the consequences of their choices.
If users are subsequently put in a new environment without the interventions, the changed behaviors may not continue.
Designers using the default strategy should be aware of this potential problem, and consider using them with educational methods. New research is needed to understand the long-term effects of these techniques.
Present-biased preference Present-biased preferences represent people’s tendency to weigh the pros and cons of present choices more heavily than future choices, and to underestimate their needs in the future. This decision bias is also known as “time discounting” [18]. The tendency typically promotes unhealthy eating because the immediate pull of tasty food is likely to eclipse considerations of future health consequences. However, present-biased preferences can be used to encourage healthier choices if people are asked to plan ahead. Read and van Leeuwen [19] gave their participants a choice of snack to be eaten in one week or at the time of eating, the next week. They found that their participants chose far more unhealthy snacks for immediate choice than for advance choice.
Opportunities Strategic design of timing of choice. Present-biased preferences can be leveraged by strategically designing the time that technology applications prompt users to make certain choices. Researchers in context-aware technology have been designing applications that can sense the current activity of people and learn their routines over time [4]. A meal planning application or a restaurant reservation system that nudges people to make a choice when they are less likely to be hungry (i.e., 1-2 hours after their lunch) might be as effective as the application that uses persuasive messages or calorie information, and it might be felt to be less intrusive.
Challenges The success of the planning strategy may depend on people’s satisfaction with the choice made previously at the time of consumption. Even when people spontaneously made choices that would have long-term benefits and delayed gratification (e.g., granola bars over more delicious chocolate bar), they may not like their choices anymore at the moment when they experience the outcomes of their choices. If this experience continues, people may stop using the technology or change their minds at the time of consumption. Systems would need to help people stick with their choices and influence them to stay happy with their choices. Messages that remind people of the positive aspects of their choices may mitigate potential negative feelings.
Diversification heuristic Diversification heuristic or naïve diversification means people’s tendency to seek variety when making several choices at once [20, 22]. This bias applies to a lesser degree when people make the same type of choices sequentially over time. For example, when people are asked to pick four snacks for one month at once, they tend to choose four different snacks; on the other hand, when people are asked to pick a snack each week, they tend to choose their favorite snack, having the same four snacks for one month.
Opportunities Diversification heuristic can be leveraged by prompting people to make another choices for the future when they make short-sighted choices. For example, when people order an unhealthy snack to eat immediately, the system can prompt them to make a choice for their next snack. Both diversification heuristic and present-biased preference suggest that people are more likely to choose healthy snacks as their next snack. On the other hand, when people make healthy choices for immediate consumption, the system may not prompt them for future choices, so that they do not choose unhealthy choices for the sake of diversity.
Challenges Providing incentives for people to make choices for future (e.g., a discount) will be important to encourage people to take another step to make a future choice.
Licensing effect Licensing effect refers to people’s tendency to indulge themselves (i.e., making vice choices) after they make choices that activate a positive self-concept (i.e., making virtue choices) [13]. For example, people may feel that they deserve a high-caloric dessert after having a healthy salad for lunch. Some research suggests that prior choices can influence subsequent choices even in different domains. For instance, after donating their money to a charity, people may feel licensed to buy a luxurious item for themselves.
Opportunities Persuasive technology can adaptively change its information presentation to help people avoid licensing effect biases. In a system that tracks people’s previous choices, when they have made virtuous choices (e.g., exercising instead of watching TV on a couch, or carpooling instead of driving), the system may not show or emphasize the tracked behaviors in order not to encourage any licensing behaviors.
Challenges There is little consensus on how people make decisions in responses to their prior choices. Transtheoretical model suggests that the system needs to applaud people making progresses in changing their behaviors in relation to their goals [8]. Licensing effects suggest that emphasizing their previous good behaviors can induce people to feel deserved to deviate from the good behaviors. More research is needed to better understand what factors cause the differences in their subsequent choices [10].
Asymmetrically dominated choices People tend to make choices that are easier to judge as superior than other alternatives. One example of this tendency is the “asymmetric dominated choice” [9], which means placing a choice option next to an inferior option to increase its attractiveness.
Opportunities Asymmetrically dominated choices can be leveraged by intentionally including an inferior option when presenting many options. For instance, consider a cookie as compared to a large, shiny Fuji apple and a small withered apple. By pairing the Fuji with the withered apple, the Fuji’s value seems much higher, and choices of the Fuji will increase.
Challenges Paring only a few options with obviously inferior ones can make users feel suspicious about the systems. In addition, in many choices, finding a clearly inferior option is difficult, which makes this approach practical only to a certain type of choices.
NEEDS FOR SYSTEMATIC DESIGN AND EVALUATION In the previous sections, I described several decision biases drawn from behavioral economics, and opportunities and challenges in leveraging them in the design of persuasive technology. Theory-based design should be implemented through iterative design processes and evaluated systematically to test its efficacy as documented in [16].
Previous research has showed that some design features do not work in the real world, even when theory predicted their effect [1, 16]. In the real world, there might be other factors that may eclipse the power of the intervention strategy.
Characteristics of different design media (website, mobile phone, and/or robot) can influence how theory would work.
CONCLUSION Behavioral economics research suggests that extremely simple changes in user interfaces can have a substantial impact on people’s choices. In this workshop, I hope to have a lively discussion on strengths and weaknesses of design strategies drawn from behavioral economics, and identify domains and situations where these approaches would be most appropriate and useful.
Gathering and Presenting Social Feedback to Change
Domestic Electricity Consumption Matthew Studley and Simon Chambers
University of the West of England matthew2.studley@uwe.ac.uk
Ruth Rettie and Kevin Burchell
Kingston University r.rettie@kingston.ac.uk ABSTRACT This paper describes the CHARM Energy Study in which mobile technology is used to study the impact of social group feedback on household energy consumption. We describe the background and rationale behind the study, the technology which supports the study, and the study’s methodology.
The work described herein builds upon similar studies by using mobile technology and on-line feedback to increase the frequency of accurate social group feedback to the participants.
Author Keywords Nudge, Social Norms, Smart Meters ACM Classification Keywords H.5.2 Information Interfaces and Presentation: User Interfaces—Evaluation and Methodology INTRODUCTION It is widely recognized [10] that lowering domestic energy consumption could make a significant contribution in reducing CO2 emissions and hence mitigate against the risk of anthropogenic climate change and promote economic wellbeing. There are significant challenges to the achievement of this goal; to change a household’s energy consumption the householders must be motivated to change and to have the tools available to enact this change.
CHARM is a three-year EPSRC funded UK project that evaluates the impact of individual and social group feedback on behaviour in three different contexts, including electricity consumption. The research aims to develop, evaluate and understand the ways in which digital technology can be used to shape individual behaviour by informing and thereby challenging ‘normal’ practice. Social norm research suggests that we can influence behaviour by telling people what other people do [14].
Traditional approaches that try to change behaviour by directly influencing attitudes and intentions often prove inefCopyright (c) 2011 for the individual papers by the papers’ authors. Copying permitted only for private and academic purposes. This volume is published and copyrighted by the editors of PINC2011.
PINC2011 fective [1]. Rather than telling people what to do, it can be more effective to use ‘social proof’ [6]; influencing behaviour by showing people what others do. Studies in several related disciplines suggest that everyday practices are malleable, and can be ‘nudged’ in a socially desirable direction by subtle forms of social influence [21]. In particular, research indicates that feedback on an individual’s level of performance (e.g. electricity consumption) can change their behaviour, and moreover, that this effect is enhanced if supplemented by feedback on the performance of a relevant social group.
THEORETICAL FRAMEWORK Writing from a sociological perspective, Shove [18] explores the social organization of normality and argues that patterns of consumption are shaped by the taken-for-granted practices of everyday life: ‘much consumption is customary, governed by collective norms and undertaken in a world of things and socio-technical systems that have stabilizing effects on routines and habits’ (p. 9). Shove emphasises the collective conventions that underlie individual conceptions of basic needs such as cleanliness and comfort. Thus, a yearround indoor temperature of 22◦C has become an accepted standard of comfort that shapes buildings, clothing habits and energy consumption patterns, while daily showering has become an accepted cleanliness practice in the UK, with consequent impact on energy and water consumption. These expectations are taken-for-granted, and treated as inherent aspects of ‘comfort’ and ‘cleanliness’, but their contingency is demonstrated by historical and global variation. Although Shove highlights the complex socio-technical, economic, cultural and symbolic systems that underlie conceptions of ‘normal’ practices, she argues that what people take to be normal is not fixed but ‘immensely malleable’ (p. 199). Consequently, she claims, it is important to understand the ‘dynamics of normalization’, that is, how do the habits and practices of everyday life change and evolve? Whereas Shove avoids a rational choice model with its focus on individual choices, the relatively new field of behavioural economics retains a focus on individual choice, but contests the assumption of a rational economic agent, in the light of research on the psychology of choice. Thaler and Sunstein[21] argue that choices are inevitably influenced by the context or ‘choice architecture’, and that it is legitimate to deliberately ‘nudge’ people’s behaviour in order to improve their lives. A ‘nudge’ is ‘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’ (p. 6). Thaler and Sunstein highlight research in social psychology that shows one can nudge people simply by telling them what other people do.
Whereas earlier research on conformity [5] [12] relied on overt social pressure, more recent research [7] has focused on subtle, indirect influences of which participants may be unaware; these are more analogous to nudges. Cialdini et al. [8] distinguish between two types of social norms, descriptive and injunctive. The former simply state what most people actually do, the latter express an overtly normative message about what people should do. Both can be effective, but descriptive norms are less invasive. Social norm research typically [14] includes descriptive social norms, e.g. ‘70% of students on this campus do not take drugs’, and has been widely used in social-norm marketing campaigns aimed at alcohol and substance abuse among young people. Research suggests that the impact of social norms depends on the extent to which they are focal (i.e. salient) and in alignment [7].
Two field studies are directly relevant to electricity efficiency.
In these studies participants’ electricity meters were read by research assistants who provided feedback on door-hangers.
Nolan et al. [13] tested descriptive social norms such as:
In a recent survey of households in your community, researchers at Cal State San Marcos found that 77% of San Marcos residents often use fans instead of air conditioning to keep cool in the summer. Using fans on energy instead of air conditioning — Your Community’s Popular Choice! The study found that these had significantly more effect on consumption than injunctive appeals to self interest, protection of the environment or social responsibility, although respondents in an earlier study (reported in the same paper) thought that the descriptive norm message would be least motivational. A study using a similar methodology by Schultz et al., [17] again used door-hangers, giving participants feedback on their individual and local neighbourhood electricity usage figures. This research compared a feedback only condition (descriptive social norm) with an intervention than combined feedback with a positive or negative emoticon or ‘smiley’ (descriptive and injunctive social norms). In the feedback only condition, participants who were using more than their neighbours used significantly less after the intervention, but those who were using less moved towards the norm, and started to use more electricity (the ‘boomerang’ effect). In the second condition, when descriptive and injunctive social norms were combined, the ‘destructive’ movement towards the norm was avoided: usage of those below the norm remained stable while the usage of those above declined. Note, these two studies used personal meters readers attached handwritten feedback to respondents’ front doors; this personal element may have enhanced the normative effect of the communication. A large scale year long trial conducted by Cialdini at Positive Energy (O Power) combines descriptive and injunctive social norms in energy bills, with promising results [3].
The study by Schultz et al. combined individual and
social group feedback, but did not distinguish between the
impacts of these two interventions. There is considerable
research on the impact of individual feedback in energy
efficiency. Darby [9] identifies feedback as the single most
promising method for reducing household energy
consumption, and calls for more field testing. Research shows that
more frequent feedback is more effective, and that feedback
can be effectively conveyed through a website [
Methodology We performed two pilot tests, the former involving ten participants recruited from University staff, the latter twenty participants recruited from two coherent geographical areas chosen to represent different socio-economic groups. Due to the small size of the pilots no statistically valid inferences can be drawn from their output; these trials were performed to test technology, recruitment and communications.
The main study includes four hundred and twenty participants professionally recruited in these two target areas. Participants are paid an incentive for their participation. Recruiters administered a pre-trial questionnaire (e.g. ascertaining house type, the number of rooms in the house, heating type, et c..). A matching questionnaire will be administered after the trial to see what change has taken place in the way the participants see themselves and their behaviour.
We believe that the CHARM Energy Study is unique in using mobile technology to study the effect of frequent online social feedback in a UK study large enough to enable statistically-valid conclusions to be drawn.
Households were randomly assigned to one of three conditions; control (no feedback), individual feedback only, or both individual and social group feedback. The control groups have their energy use monitored but receive no communications from the team during the study, and do not receive any feedback on their energy use. We will use the data on the control groups’ usage to account for environmental factors which effect electricity use (cold weather, mass use of TV to watch landmark events, et c.) and to allow us to take into account the fact that simply having an ‘electricity monitor’ in the home may have an effect on the energy behaviours of the household.
In addition to the questionnaires, we will conduct approximately 35 face-to-face semi-structured interviews, with a purposive sample of subjects. Interviews will occur in respondents’ homes and involve as many adult household members as feasible, and will include observation and discussion of home configuration, energy efficiency features, types of energy consumed and appliances used. A number of respondents will be interviewed both before and after the experiments, in order to benchmark conceptions and practices and to facilitate identification of changes (these respondents will be excluded from the field trial analysis). A number of respondents will be re-interviewed at least six months after the trial to identify any long term changes in overall levels and underlying practices. Respondents will receive an additional incentive for their participation in the interviews. In addition, we plan three professionally moderated focus groups, to elicit discussion of the trials and normative discourse in a social context; the focus groups will be reconvened after a period of six months to explore the longevity of any changes in practices.
Technology Each respondent who volunteers to take part in the study is supplied with a box containing three components 1. A current-clamp which attaches to the meter tail and which transmits usage data every two seconds via a 433 MHz wireless link. 2. A monitor which stores this data and sends the data to our
server via GPRS using a roaming SIM. 3. A power adapter which supplies the monitor with power
for operation.
There is no real-time display visible to the individuals in the household. It has been shown [4] that real-time displays are a powerful tool in effecting behavioural change since they promote experimentation to see what effect individual appliances have upon power consumption, but have not been included in this study in order to focus on the effects of social feedback.
The monitor and current-clamp make use of a commerciallyavailable off-the-shelf home energy monitor with a real-time display. We hide the display from view in the box that contains the GPRS modem and microcontroller. Using a COTS solution allowed a significant saving in development time and the time taken to meet regulatory and safety requirements.
As a result of field-testing in the pilot studies, the embedded controller has evolved through several iterations to account for network outages, automatically reloads new versions of firmware as we release them, and can be remotely controlled in situ to trigger recovery from several abnormal conditions.
Usage information is gathered via GPRS upload by the HTTP ‘GET’ mechanism to a web-server where it is logged in a relational database. The web-server provides an passwordcontrolled management interface which allows us to track the performance of each monitoring unit and participant household, to determine for example when participants in a household have not viewed their data on the website, and to track the frequency of data transmission from monitors enabling the team to track network outages, request user interventions such as checking the unit is receiving power, ask the householder to reboot the unit, et c..
Feedback Information is supplied to the participants in the individual and social experimental groups in a number of ways. They can view information about their electricity use on the website (see below). They receive weekly emails which encourage them to maintain their participation in the study. Individuals known to be infrequent visitors to the website may receive SMS text messages prompting them to participate, a mechanism which was shown to be an effective way of encouraging re-engagement in the initial field trial. As previously stated, households are assigned to one of three experimental groups which define the type of feedback they receive. The feedback provided to households in the social feedback category is illustrated in fig. 1. We hope to create the conditions where we may most easily see an effect of social proof in changing behaviour in the following ways.
Firstly, we attempt to increase saliency as recommended in [7] and focus on small geographic areas as recommended in [11]. Secondly, we provide descriptive and injunctive feedback in the form of emoticons after Schultz [17] to reduce the possibility of the ‘boomerang’ effect. Finally, we provide easy access to energy saving tips which we hope will provide householders with the means to lower their energy consumption. The website also provides the user with views of his electricity consumption in a context suited to his experimental group for previous time periods; yesterday, last week, and the whole of the study thus far.
Initial results from the participants in the twenty-strong second test indicate that the feedback is viewed as both interesting and useful, and we look forward to reporting the results of the full trial in the near future. Recruitment for the main trial started in January, 2011, and we expect to present results after the trial in the Autumn of that year.
Novelty The CHARM Energy Study differs from the work reviewed above in the following ways. There have been studies involving more people with monthly feedback on paper-based bills [3], and studies involving small numbers of people with weekly paper-based feedback [17]. We believe that ours is the first study testing the social norm approach with frequent automated data collection and feedback. Further, ours is the first such study in the UK where there may be resistance to the social norm approach [15].
Conclusion It is planned [19] that all UK homes will have Smart Meters installed by 2020, and the EU Smart Meter market has been predicted [16] to be worth 25 Billion Dollars US in the ten years from 2010 to 2020. Although the emerging UK standard [20] mandates that UK Smart Meters will provide bidirectional communications and support in-house displays, we are unaware that there is yet a standard for the type of information that will be displayed to the consumer.
If the study shows a real reduction in domestic electricity use resulting from social feedback methods, we hope that we may influence the emerging Smart Meter standard to provide for this means of change.
Acknowledgements This work was partly funded by the Engineering and Physical Sciences Research Council (EPSRC) and is one of a series of projects operating under the Research Councils UK (RCUK) ‘Digital Economy’ Programme. Data transmission costs have been subsidised by the UK M2M provider, Wyless Plc in partnership with UK mobile telephony provider, O2.
Towards Egocentric Fuel Efficiency Feedback
Ian Oakley Madeira Interactive Technologies Institute (M-ITI)
Funchal, 9000-390, Portugal {tiago, filipe, mscott, vassilis, ian}@m-iti.org ABSTRACT Motivated by anecdotal evidence, we hypothesize that an egocentric approach is more appropriate and relevant to providing fuel efficiency feedback than a systemic approach. In this paper we describe a proposed study to test this hypothesis, and present the design of a fuel efficiency feedback system for public transit bus drivers.
Author Keywords Feedback systems, fuel efficiency, public transit bus drivers ACM Classification Keywords H5.m. Information Interfaces and Presentation (e.g., HCI): Miscellaneous Introduction In 2010 the public transport authority in Madeira, Portugal, installed onboard electronic equipment that gauged driving fuel efficiency by presenting the driver with very simple feedback: 3 green lights progressively suggested that efficiency was increasingly optimum, while 3 red lights progressively suggested that driving efficiency was increasingly sub-optimal. The system was intended to give drivers feedback on their driving and to help them achieve optimal driving efficiency. The result was negative: drivers complained to human resources that the system constantly showed 3 red lights, suggesting that their driving was bad. Human resources complained to operations that the system was bad for morale.
In response, operations attempted to “calibrate” the system by tweaking its thresholds. The result was that the feedback became useless and largely inaccurate, ultimately resulting in the abolishment of the system. In our discussions with the transport authority, it became clear that in addition to the misinterpretation of the feedback by the professional drivers as a rating of their driving, the mountainous terrain of Madeira caused genuinely inefficient driving. There was simply no way to avoid steep hills that took a significant toll on fuel consumption, thereby skewing the feedback towards inefficient driving. The attempts at calibrating the sysCopyright c 2011 for the individual papers by the papers’ authors. Copying permitted only f or private and academic purposes. This volume is published and copyrighted by the editors of PINC2011. tem failed because, effectively, the on-board equipment measured pure fuel consumption which in turn was intricately related to the steep terrain of the environment.
On the other hand, drivers perceived the feedback as a reflection of their skills.
Our anecdotal experience with the public transport authority’s feedback system caused us to hypothesize that providing feedback on specific driver behaviour, as opposed to overall fuel efficiency, may be a more appropriate way for motivating driver behaviour change. Adopting a systemic approach to this issue, we argue that existing feedback mechanisms relating to efficiency provide a view of the complete system, parts of which the driver has simply no way of effecting (such as the steep terrain). Hence we argue that efficiency feedback focusing on parts of the system that the driver can effect (such as acceleration) may result in more efficient driving behaviour. We term this approach to feedback egocentric.
In this paper we describe a fuel efficiency reporting and advisory system that takes advantage of the multisensor and interactive nature of modern smart-phones to present feedback to drivers. More specifically, we are interested in deploying the system in public transit buses to measure its effectiveness on positively influencing drivers’ behavior. By continuously capturing realtime sensor data, we can calculate the Vehicle Specific Power (VSP), a surrogate variable that strongly correlates with both fuel consumption and pollutant emission levels, providing a systemic view of efficiency [11]. Crucially, we are able to manipulate the calculation of VSP to ignore environment variables and provide egocentric feedback. Taking advantage of this manipulation, we propose a study where we intend to test our hypothesis about the benefits of egocentric over systemic feedback. We believe that through the use of our system we can promote not only short-term but also medium/longterm positive changes in public transit bus drivers’ behaviours.
Related Work
Research suggests that it is possible to achieve up to
15% of fuel consumption decrease when appropriate
driving behavior is used [
Of these factors, engine efficiency (physical characteristics of the vehicle) is considered the most important [4].
Still, the driver’s attitude and behavior towards the maneuvering of the vehicle can considerably impact fuel consumption levels. Therefore, it is commonly argued that smoothness of driving leads to higher efficiency of fuel consumption.
Raw fuel consumption levels and pollutant emissions can be calculated through the use of Portable Emissions Measurement Systems (PEMS). These are connected to vehicles through their On-Board Diagnostic (OBD) interface, letting the PEMS system access the vehicle’s on-board computer and calculate multiple parameters [13]. Still, PEMS systems work primarily as a diagnostic/analysis tool, not as a feedback support mechanism. Furthermore, PEMS systems fail to reflect contextual characteristics such as road gradient values.
It is common to augment PEMS with GPS for analysis purposes [13].
The Vehicle Specific Power (VSP) approach is used to approximate and predict actual emissions levels and fuel consumptions [10, 11]. VSP is a model that tries to explain consumption and emission levels from a physical perspective; it corresponds to the Power Demand or Vehicle Engine Load values, therefore correlating strongly with fuel consumption and pollutant emission levels [13].
The VSP model depends on three variable factors: speed, acceleration, and road grade. Through the combination of these factors, along with vehicle specific air and roll resistance coefficients, VSP values are calculated as follows [11]:
V SP = v ∗ (a + g ∗ sin(ϕ) + rcoef ) + acoef ∗ v3 (1) where v is speed in m/s, a is acceleration in m/s2, g is 9.807 m/s2, ϕ is the road gradient value, rcoef is the rolling resistance term coefficient, and acoef is the air drag term coefficient. Another characteristic of VSP is its ability to support payload modeling, especially important in situations where this value has noticeable impact, such as is the case with public transit buses [11]. Still, VSP does require that we calibrate the model for each type of vehicle, as it is necessary to obtain the ground truth for fuel consumption and pollutant emission levels for the model to be effective.
Devices such as smart-phones possess a wide variety of sensors, like GPS and accelerometers, that enable calculation of vehicle dynamics and consequently VSP values.
It is then possible to approximate fuel consumption using solely internal smart-phone sensors. These devices can be easily incorporated into vehicles, and their ability to provide a rich and extensible interaction platform make them a feasible alternative mechanism to provide drivers with fuel efficiency feedback. Furthermore, and comparing with usual commercial systems such as Scania Fuel-Saving Driver Support System1, smart-phones are not restricted to specific vehicles, and can even be device independent, which is the case when using development platforms such as Google’s Android.
Receiving timely feedback is key to motivating behaviour change, people need to be aware of their behaviour in order to change it. Fischer found the most successful feedback was given frequently, clearly presented, used computerised tools and allowed historic or normative comparisons [9]. Our mobile interface reflects these types of feedback. Utilising a mobile display allows frequent opportunities for self-reflection and should increase driver awareness of their behaviour.
Consolvo, McDonald, and Landay [3] suggest a number of design strategies for persuasive technologies that wish to motivate behaviour change. These strategies are based on psychological theories and recent persuasive technology research and we have chosen to follow some of their guidelines.
First, we make use of abstractions rather than counting solely on raw data to display to drivers. Secondly, the data shown should be unobtrusive. This is of paramount importance for safety reasons, as we need the mobile display to support ignorability and not distract the driver unnecessarily. Thirdly, since the data is to be presented in public, we need to present it in a way that the driver will not feel uncomfortable if others are aware of it.
Fourthly, we decided to ensure that only positive feedback is given, not punishing any “bad” behavior. Concretely, we aim at rewarding possible low consumption levels, but not use punishment for poor performance.
This decision is supported by the notion that positive feedback can indeed increase intrisic motivation by affirming competence [5]. The anecdotal evidence from the use of a commercial system by the public transit company also supports this notion. Finally, we have chosen to provide historical feedback. Doing so allows the driver to reflect on past behaviours in order to make more informed decisions on current behaviour.
Research Methodology We propose an experimental approach to study to what extent we can, through the use of egocentric feedback, 1http://www.scania.com/media/featurestories/sustainability/every-drop-of-fuel-counts.aspx
VSP egoVSP influence public transit bus drivers driving behavior. In our study we are interested in the following research questions: • Can we accurately establish driving behavior profiles
for bus drivers through the use of VSP calculations? • To which extent can we positively influence driving behavior through the use of egocentric feedback techniques? • Is the use of real-time more effective than the use of historic feedback, or is a combination of the two approaches most effective? Consequently, and based on the previous mentioned research questions, we raised the following hypotheses: • H1. The use of the VSP surrogate variable (and its derivatives) allows for accurate driving profile characterisation • H2. The use of egocentric driver feedback improves
average fuel consumption levels • H3. The use of real-time feedback does not signifi
cantly influence driving behavior To test these hypotheses we propose to develop an Android based software to continuously collect sensor information so that trip instantaneous parameters, such as speed and acceleration, can be calculated. We will also consider the use of additional variable(s) to model the influence of passenger payload on the overall vehicle weight. Then, we intend to install equipment on-board public transit buses and calibrate the VSP model. The ground truth establishment of instantaneous fuel consumption levels is a necessary condition for the success of the VSP model. This may be achieved through the use of a PEMS system or a similar mechanism. Subsequently, we will develop a derivative of VSP called egoVSP, which ignores road gradient and is defined as follows egoV SP = v ∗ (a + rcoef ) + acoef ∗ v3 Terms of the equation are defined equally as in eq. 1.
These two fuel efficiency models, VSP and egoVSP, are one of the two variables we intend to manipulate in our study. The other variable is the type of feedback to provide: real-time versus historical. Table 1 shows the possible combinations of these two variables.
Ongoing Work As it stands, the system is a working prototype. Targeted mainly at public transit bus drivers, the system
RawInput Sensor Data
Real-time Processing
Pipeline . .
CompNonent Storage
User Interface Real-time feedback
VSP Acc. 10l 50 0.5 Historical feedback
Driver is flexible and extensible enough to provide support for any kind of vehicle.
An overview of the architectural design is seen in Fig. 1, where the mechanism that is used to produce the final output to the driver is visible. Raw sensor data is sampled at several times per second, before it is passed to a real-time processing pipeline. This allows us to execute tasks in parallel that may require some computational complexity, therefore increasing system overall speed and responsiveness. The advantage of such a scheme becomes more evident when, for example, the system is required to perform continuous sensor data integration by means of a Kalman Filter.
The calculation of the vehicle dynamics and the VSP modeling is also included in the processing pipeline. After exiting the pipeline, the transformed output is then fed to the feedback mechanism, which transmits specific information to the driver, according to the type of feedback used. All data is continuously stored in a local database, so that further off-line analysis may be performed. Repeated sampling from sensors will undoubtedly drain the battery in its full in a matter of hours, so there is the need of ensuring that the device is fed continuous power by connecting it to the vehicle’s internal electric circuit.
Drivers initiate interaction through the system’s main menu (see Fig. 2). In order to use the system, drivers must register themselves before receiving a 3 digit PIN code that uniquely identifies them. Vehicles registration and VSP model calibration is also required to be performed, but this may be done by the developers before the system is made available to the drivers. This will be the case when doing the experimental study with the public transit bus drivers. Besides the VSP model calibration, it is also possible to calibrate both the device accelerometer, as well set up the desired orientation of the phone inside the vehicle. This last step has some limitations, as currently we are working with a phone with only one accelerometer and no gyroscope, which limits the phone’s orientation recognition. Just before starting a trip, the driver introduces his PIN code and indicates the vehicle that he is currently using. After this the trip is marked as initiated.
In order to test the effectiveness of the feedback system, we propose using two different types of feedback: realtime and historical. In the first, we will show a real-time VSP graph that represents an approximation to the actual VSP value. The graph is an abstract representation, where it goes from green (low VSP values) to red (high values) with an approximate quadratic function increase. Additionally, actual fuel consumption, speed, and acceleration values are to be represented.
In regard to the historical feedback, our system will make available two modes to the driver. The first will show the distribution of time in the pre-defined VSP bins, and the second will show a heat map of the route, indicating VSP “hot zones”. The use of historical feedback gives the driver a more broad perspective of his driving behavior, as it recalls and identifies potential patterns that may be improved. Furthermore, historical feedback will only take place when the driver is not actively driving.
Conclusion In this paper we have argued that egocentric feedback on fuel efficiency can be more effective than systemic feedback on motivating driving behaviour change. Motivated by anecdotal evidence, we hypothesise that an egocentric approach is more appropriate and relevant.
By re-defining the VSP surrogate metric, we are able to switch between systemic and egocentric feedback while maintaining minimal changes between our experimental conditions. Orthogonal to the manipulation of the efficiency model, we describe our interest in testing the effect of instantaneous versus historic data in the feedback system.
Acknowledgments This work is supported by the Portuguese Foundation for Science and Technology (FCT) grant CMU-PT/ HUMACH/ 004/ 2008 (SINAIS). This work is also supported by the European Union project INTERVIR+ and the local public transit company, Horarios do Funchal, S.A. !"#$%&'()*!++,-).+/&,#)0#12'+,+()0+)4#561#)0##') !"#$%&'()#!
1%#',*2"#' K%9102,%./0+,-12+#'0.3-'*). 80#9,-:#+;.'<.E#-(#0L"1(.
E#-(#0L"1(.
E@M.ANN7.8C.
EH?H!'P10G2:HO"1(H12H*I. ! "#$%&"'%! /0.+"#:.)1),-.P,.)-,:,0+.10.'9,-9#,P.10%.#0#+#1$.P'-I.<-'(. 1. -,:,1-2". )-'T,2+. 2-,1+#0L. U2''$V. ('O#$,. +,2"0'$'L#,:. +'. ,%*21+,. 10%. #0<'-(. +,,01L,-:. #0. '-%,-. +'. -,%*2,. +",#-. ,0,-L;.*:,H.N,,01L,-:.1-,.1$-,1%;.O,2'(#0L.2'0:*(,-:.10%.
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
3/4"#5'6/*'' D,)1-+(,0+.'<.!'()*+,-.Q2#,02,.
QP10:,1.80#9,-:#+;.
Q#0L$,+'0.>1-I.
QP10:,1.QKA.R>>7.8C.
SH3*'G:P10:,1H12H*I. ! "'3!'452267685)6+9!-./0+,12! 5MH(H. /0<'-(1+#'0. #0+,-<12,:. 10%. )-,:,0+1+#'0. 4,HLH7. 5!/6Z.
Y#:2,$$10,'*:H.. :;%&<=>'%:<;! 3'9,-0(,0+:. 12-'::. +",. P'-$%. 1-,. 0'P. 2'((#++,%. +'. -,%*2#0L.![A.,(#::#'0:.10%.'0,.I,;.1-,1.<'-.#()-'9,(,0+. #:.-,%*2+#'0.#0.%'(,:+#2.10%.),-:'01$.+-10:)'-+1+#'0.,0,-L;. *:1L,H. /0. +",. "'(,. ![A. #:. -,$,1:,%. )-#(1-#$;. #0. +",.
L,0,-1+#'0. '<. ,$,2+-#2#+;7. +",. 2'(O*:+#'0. '<. L1:. 10%. '#$. <'-. ",1+#0L.10%.2'(O*:+#'0.'<.<*,$.#0.+-10:)'-+H.\#+"#0.+",.)1:+.
A].;,1-:.+",-,."1:.O,,0.1.:+,1%;.#02-,1:,.#0.+",.0*(O,-.'<. 1))$#102,:.#0.+",.('%,-0."'(,.2'*)$,%.P#+".L-'P+".#0.+",. 'P0,-:"#). '<. ,0,-L;. "*0L-;. %,9#2,:. :*2". 1:. +*(O$,. %-#,-:. 10%.)$1:(1.N^:.10%.10.#02-,1:,.#0.+",.*:,.'<.%,9#2,:.P#+". :+10%O;. <12#$#+#,:. _M`H. Y10;. '<. +",. ,$,2+-#21$. %,9#2,:. 2'0+-#O*+#0L. +'. +",. -#:,. #0. %'(,:+#2. ,0,-L;. 1-,. *:,%. 10%. :'(,+#(,:. 'P0,%. O;. +,,01L,-:H. ?,:,1-2". #0. +",. 8C. "1:. #0%#21+,%.+"1+.aMb.'<.+,,01L,-:."1%.1.N^7.(*:#2.:;:+,(.'-. )"'0,.#0.+",#-.-''(:7.P#+".+P'.+"#-%:."19#0L.1$$.+"-,,._@c`H.
K.:,)1-1+,.:+*%;.-,)'-+,%.+"1+.d]].:*-9,;,%.+,,01L,-:.1L,%. @c. +'@a. P,-,. 2'$$,2+#9,$;. P1:+#0L. ,0'*L". ,0,-L;. +'. )'P,-. d7e]A. :2"''$:. 10%. 1. +"#-%. '<. +",. ,0,-L;. O,#0L. *:,%. P1:. 1. %#-,2+.2'0:,f*,02,.'<.U:+10%O;V.O,"19#'*-._@`H. [*-. P'-I. '0. +",. U+1I#0L. '0. +",. +,,01L,-:V. )-'T,2+:. 4PPPH(1%d0-LH'-L6. P#$$. ,0L1L,. ;'*0L. ),')$,. 41L,%. @A& @a@6. #0. -,%*2#0L. +",#-. 'P0. ),-:'01$. ,0,-L;. *:,. 10%. (1I,. ............................................................ @./+.#:.)-'O$,(1+#2.-*00#0L.:+*%#,:.P#+".+,,01L,-:.<-'(.1L,. @d&@M. #0. :2"''$:. #0. +",. 8C. P",0. +",;. O,L#0. )-,)1-#0L. <'-. 3!QB. ,W1(#01+#'0:7. P,. +",-,<'-,. 1$:'. +1-L,+. )-,&+,,0:. +'. ,0:*-,. P,. 210. P'-I. +",. :1(,. :*OT,2+:. <'-. +",. c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`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
P",!&"9"#) 2!./) -,,<XYY'"4+700%7%.7$6YIY-&Y+%.,13'#YWJIZAWJ7+,/)
R3%%"++"#)[3'$3!*)JVIIS7) @JB) T"31">) P7) 3'#) Q!""#>) Q7) G22"%,&9") 8',"!3%,&.'X) E.4) "/.,&.'31) 3("',+) 322"%,) $+"!+7) 8',"!'3,&.'31) [.$!'31) .2)
E$/3'5Q./<$,"!)O,$#&"+>)W\)RZS7)\]]5\\W>)JVVZ7) @^B))Q-!&+,."!) G7) TD.!6+6.(>) _&$1&.) [3%$%%&>) `$%&3'.) _3/0"!&'&>) =3,$) U&"/&'"'>=.<&) F&66.13>) Q3!&') =.!+,"'++.'>) 3'#) F3++&/.) T"!,.'%&'&>) aM'"!(*1&2"X) <"!93+&9") "'"!(*) 343!"'"++) 2.!) -.$+"-.1#+>b) &') N!.%""#&'(+) .2) ,-") IJ,-) GQF) &',"!'3,&.'31) %.'2"!"'%") 3#D$'%,) <3<"!+) .') c0&K$&,.$+) %./<$,&'(>) c0&%./<) dIV>) <<7)^WI5^WJ>)JVIV7) ) ) ) ) ) ) @eB) Q!""#>) Q7) 3'#) T"31">) P7>) M'(3(&'() MC<"!&"'%"+) 4&,-)
M/.,&.'31) f&!,$31) =-"!3<&+,+7) &') 8',"!'3,&.'31) H"+&(') 3'#)M'(3(30&1&,*)Q.'2"!"'%")g)U.!#&QE8>)?+1.>)JVVW7) @]B) H"<3!,/"',) .2) M'"!(*) h) Q1&/3,") Q-3'(") RJVVZS)
M'"!(*)Q.'+$/<,&.')&'),-")c'&,"#)i&'(#./7)P",!&"9"#) 2!./) -,,<XYY4447#"%%7(.97$6Y"'Y%.',"',Y%/+Y+,3,&+,&%+Y<$01&% 3,&.'+Y"%$6)Y"%$673+<C)R3%%"++"#)[3'$3!*)JVIIS7) @WB) [3/"+) N&"!%">) H&3'") [7) O%-&3'.>) 3'#) M!&%) N3$1.+>) aE./">) -30&,+>) 3'#) "'"!(*X) "C3/&'&'() #./"+,&%) &',"!3%,&.'+) 3'#) "'"!(*) %.'+$/<,&.'>b) &') QE8dIVX) N!.%""#&'(+) .2) ,-") JA,-) &',"!'3,&.'31) %.'2"!"'%") .') E$/3') 23%,.!+) &') %./<$,&'() +*+,"/+>) U"4) j.!6>) Uj>) cOG>)<<7)IZA]5IZZe>)GQF>)JVIV7) @\B) [.') k!."-1&%->) =343''3) H&113-$',>) N!"#!3() i13+'D3>) ["''&2"!) F3'6.>) O$''*) Q.'+.19.>) T"9"!1*) E3!!&+.'>) 3'#)[3/"+)G7)`3'#3*>)ac0&(!""'X)&'9"+,&(3,&'()3)/.0&1") ,..1) 2.!) ,!3%6&'() 3'#) +$<<.!,&'() (!""') ,!3'+<.!,3,&.') -30&,+>b) &') N!.%""#&'(+) .2) ,-") J\,-) &',"!'3,&.'31) %.'2"!"'%") .') E$/3') 23%,.!+) &') %./<$,&'() +*+,"/+>) U"4) j.!6>) Uj>) cOG>) QE8) dVZ>) <<7IVe^5IV]J>) GQF>)
JVVZ7) @AB) H3'&"1) T"!#&%-"9+6*) 3'#) M!&6) U"$"'+%-43'#"!7) IZZZ7) =.43!#) 3') ",-&%+) .2) <"!+$3+&9") ,"%-'.1.(*7) Q.//$'7)
GQF)eJ>)])RF3*S>)]I5]A>)IZZZ7)) @ZB) F3!+-&'&) Q-",,*>) H39&#) =!3'>) 3'#) P"0"%%3) M7) _!&',"!>) a_",,&'() ,.) (!""'X) $'#"!+,3'#&'() !"+.$!%") %.'+$/<,&.') &') ,-") -./">b) &') N!.%""#&'(+) .2) ,-") IV,-) &',"!'3,&.'31) %.'2"!"'%") .') c0&K$&,.$+) %./<$,&'(>) U"4) j.!6>) Uj>) cOG>))c0&Q./<)dVA>)<<7)JeJ5J]I>)GQF>)JVVA7) @IVB) ?dH.''"11>) i7G7) 3'#) H7`7) L3!#1.4>) G) ,-".!*) .2) ,-") .!&(&'+) .2) %..1'"++>) G#93'%"+) &') Q.'+$/"!) P"+"3!%->)
J\>))<3("+7)I^)5)IA>)JVVV7) @IIB) N!.%-3+63>) [7?7>) U.!%!.++>) [7Q7) 3'#) H&%1"/"',">) Q7Q7)
Q-3'(&'()k.!)_..#7)G9.')T..6+>)U"4)j.!6>)IZZe7) @IJB) =3<<>) G7) 3'#) O7) T&!#>) O.%&31) /3!6",&'() 3'#) ,-") /"3'&'() .2) %..1>) O.%&31) F3!6",&'() l$3!,"!1*>) Ie>) I) <3("+7)IA)5)JZ>)JVVA7) @I^B)=-")O1""<)Q.$'%&1)RJVV\S)[$'6)O1""<X),-")U"4)E"31,-) =-!"3,) ,.) =""'3("!+7) P",!&"9"#) 2!./) -,,<XYY4447+1""<%.$'%&17%./Y[.$!'31&+,+Y<!"++m<3%6+Y[$ '6mO1""<m,-"m+$!9"*m+,.!*7#.%) R3%%"++"#) [3'$3!*) JVIIS7) Nudging People at Work and Other Third-Party Locations
Max L. Wilson1, Derek Foster2, Shaun Lawson2, Simon Eddison1 1Future Interaction Technology Lab
College of Science
Swansea University, UK m.l.wilson@swansea.ac.uk, simon.d.eddison@gmail.com Lincoln Social Computing Research Centre
School of Computer Science
University of Lincoln, UK {defoster,slawson}@lincoln.ac.uk ABSTRACT Nudging people towards positive behaviour change is an important issue recognised by academia, individuals, and even governments. Although much research has been published in this area, little has focused on non-domestic environments such as the workplace. It is widely reported that changing individual behaviour of employees can make a significant contribution to sustainable resource consumption. This position paper focuses on the unique aspects that make nudging consumption behaviour in thirdparty environments like the workplace a very different problem to that of nudging in people’s domestic and private lives. Several studies are discussed to provide context as well as evidence towards our position.
Author Keywords Persuasion, Nudge, Behaviour Change.
Work, Ownership, Sustainability, ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous.
General Terms Theory, Human Factors, Design.
INTRODUCTION The HCI community has recently shown a great deal of interest in the development of interactive systems that facilitate behaviour change for sustainability. Much of this research has exploited ideas recently re-popularised by Thaler and Sunstein [10] in that individuals can be ‘nudged’ to make better lifestyle decisions, given the right information and the environment in which to do so. Much of this work has focused on how individuals might improve their own private and domestic lifestyle, behaviour, and sustainable resource consumption; however such work has rarely taken account of the fact that people spend a significant amount of their waking hours at work where they also contribute towards resource consumption.
Copyright © 2011 for the individual papers by the papers' authors.
Copying permitted only for private and academic purposes. This volume is published and copyrighted by the editors of PINC2011.
A recent report [1] has indicated that if the 17 million UK workers, who regularly use a desktop PC, powered it off at night this would reduce CO2 emissions by 1.3 million tons - the equivalent of removing 245,000 cars from the road.
Similarly, if a UK business with 10,000 computers leaves them on all night for one year, it will cost £168,000 ($220,000) and emit 828 tonnes of CO2. The same report, however, suggested that at least three in ten workers in the UK do not always power off their PC overnight. Further, many more machines are in use or provide services 24 hours a day, all year round.
As an example in our own context, Figure 1 compares the electricity usage at the University of Lincoln campus for the first week in December in 2009 and 2010. There are two compelling features of Figure 1 that characterise the typical energy consumption of a workplace. First, the graph clearly shows how little energy the university uses at the weekend.
Second, this period in 2010 coincided with severe weather that meant that many staff members were unable to travel to the campus. The dramatic reduction in energy consumption can be clearly seen in the first 3 days of the graph and highlights that people can have a significant impact on consumption at work, as well as in their own personal environments.
Figure 1 Campus electricity usage December 2009/10 Despite environmental concerns now playing an established role in the public sector, as well as the corporate and business agenda, there is still much to gain by exploring new ways of persuading people to adopt positive energy usage behaviour. The first and obvious research question is: Do domestic PINC (Persuasion, Influence, Nudge & Coercion) methods simply translate to workplace and other third-party environments? In this position statement we review initial evidence that they do not, and discuss the reasons why. We propose a framework for thinking about Nudge methods in different contexts, and discuss our future work in this area.
RELATED WORK Thaler and Sunstein [10] have recently re-popularised the interest in the idea of Nudge, where the right environments and the right information delivered at the right time can encourage people to adapt and improve their behaviours.
Much research has focused on directly improving one’s own behaviour, whether it be reminders to exercise, or to notably reduce energy consumption. Research into simple home energy monitors [3], for example, suggests that payas-you-go meters typically reduce consumption by only 3%, while those that focus on reducing their payments often reduce their consumption by 0-10%. Having an in-house monitor that provides instant feedback has been shown to reduce consumption by between 5 and 15%. Other prototype systems, such as Kuznetsov and Paulos’s domestic ambient light display [7] successfully encouraged people to reduce their water consumption, by visualising better or worse consumption to their previous average use.
Other research typically provides anonymous averages from a group or community to a user, so that the user can see their own behaviour or consumption in the context of others. In previous work [5], we reduced domestic energy consumption through a carefully designed mixture of online social media and home energy monitors. Our findings suggested that the use of energy feedback delivered in a social context significantly reduced consumption when compared to energy feedback without a social context. We have also shown similar results in a personal fitness/activity domain [4].
A related approach involves facilitating ‘friendly’ competitive behaviour; for instance it has already been shown that the work environment affords powerful opportunities for facilitating such behaviour – for instance Siero et al [8] demonstrated that when a group of employees received information not only about their own energy usage, but also about that of a ‘competing’ group of employees from the same company but a different workplace, they significantly altered their energy usage behaviour compared to a situation in which they only received information about their own usage.
Despite the success of the work by Siero et al some thirteen years ago, little research since has explored energy behaviour interventions based on competition between employees. Therefore, a key question for Nudge researchers going forward is how do differences between the work and domestic leisurely sides of life affect the potential of behaviour change interventions? Also, what theoretical grounding can we draw upon to begin to explore any differences? Stebbins [9] introduced a seminal framework for understanding people’s leisure time. For some, being environmentally friendly is, as Stebbins called it, a Serious Leisure, where people work hard at achieving their goals.
Installing home technology is often a temporary project, and can be seen as Project Leisure, where people take behaviour change to be a new task. The aim of much nudging research, however, is to be embedded in people’s Casual Leisure, so that good consumption is encouraged simply and unobtrusively within our lives. These forms of leisure, however, are very different from our work lives, which are goal-oriented, formalised, and externally driven.
EARLY EXPERIMENTAL FINDINGS Study 1 – Water Consumption in the Work Place One early finding in this space was from Kuznetsov and Paulos [7] who anecdotally saw unexpected results in a work environment, and so proceeded to focus on domestic scenarios. Their anecdotal findings saw consumption increase – double in fact.
One of our recent studies in Swansea University, UK, focused directly on this surprising issue. We created a series of feedback installations, and installed them in a shared work-place kitchen. Like the work by Kuznetsov and Paulos, the installations used a Phidget microphone to track water flow through the pipes. The installations were supported by informational posters, which included a link to a website to provide feedback. Otherwise, we remained as un-intrusive as possible in order to record normal usage as closely as possible. After recording baseline average readings, we first recreated the ambient light display provided used by Kuznetsov and Paulos, which: glowed green with less-than-average consumption; glowed yellow 10% either side of the mean; and glowed red thereafter. Three further displays were installed in subsequent weeks.
The first used similar measures, in respect to average consumption, to create competitive gaming-style textoriented messages on an LED display, such as: “You’re beating most people” and “Sorry, you lost”. The second display converted the light system into a series of audible beeps. The final display tried a different tack altogether, by simply providing environmental information relating to their water consumption, such as the average amount of water available to people in the third world on a daily basis.
Initially, as per the prior anecdotal evidence, the ambient light display did double the average consumption of water during the 2 weeks it was displayed. In comparing studying the additional displays, we saw all but the audio condition increase the consumption. While the increase shown by these alternatives was significantly less than the ambient light display in particular, none were significant. Although the audio feedback did marginally reduce consumption, we also recorded a significant number of opt-out button presses in the audio condition, indicating that people disliked this particular installation. Qualitative comments from an optional online survey confirmed this. Given the surprising increase created by the ambient light display, we concluded the study by reinstalling the ambient light display for a final week. Although not quite double the average consumption, we again saw a significant increase in energy consumption.
In the end, none of the displays managed to significantly decrease consumption of water. It is promising, however, that not all the displays increased consumption significantly. This means that such displays do not simply have the opposite effect in work environments. Instead, the results suggest that people simply do not care for the consumption of the company as a whole, and potentially do not mind entertaining themselves with the resources of the company by using additional resources. The fact that significantly more users opted out of the audio display, which was the only one to reduce average consumption, further indicates that people do not mind avoiding resources in this area; that they do not feel personally motivated to accept the nudging technology.
Study 2 – Energy Consumption in the Work Place Our recently commenced Electro-Magnates study [6] aims to reduce energy usage in the workplace by utilising a suite of social persuasive applications to encourage proenvironmental behaviours. Personal desktop applications (social widgets) and situated displays will be used to deliver energy feedback to individuals, groups and communities about their own – and others’ – energy usage to foster exchange of performance and to support constructive competition to reduce consumption. The workplace in the context of this study is educational and public sector workplace environments in the county of Lincolnshire, UK.
consumption through social norms and social technology. However, designing a similar system for the workplace presents greater challenges across a range of design, ethical and technical issues. From our study focus groups in the domestic environment we discovered that for some people cost was the primary motivating reason to reduce their energy use. In the workplace employees are not typically responsible for paying energy costs, neither are they directly responsible for meeting any governmental carbon policies in place that could lead to institutional ‘carbon’ fines.
To mitigate the absence of financial motivation in employees and to develop workplace energy metaphors, we intend to run a series of focus groups and participatory design workshops to engage and empower the employee in developing an understanding of both the economic and environmental impact of their working practices. The participatory design workshops will provide an opportunity for employees to be directly involved in designing the UX element of Electro-Magnates therefore helping to address ethical concerns over privacy and appropriate disclosure of energy data.
Early work to date includes prototyping a high-impact energy interface for overall energy usage in Figure 3, page viewed on 09/01/2011, as well as a competitive league table for buildings. Both prototypes are designed for large situated displays and are abstracted presentations of what is possible with raw energy sensor data which in itself is intangible and difficult to interpret.
Figure 3 High-impact visualisation of overall energy usage DISCUSSION The workplace, as an example of a non-domestic, nonpersonal environment, creates many unique issues for the ideas behind nudging behaviour. Consequently, we have identified three initial dimensions that differentiate domestic and workplace environments that might be used as a formative framework for thinking about applying nudging technology in different environments:Expression of Self. First, the workplace may be termed a special environment in that there are usually constraints and rules in how employees can interact and carry out activities in the workplace compared to their less inhibited personal life. This is particularly important when considering employee consumption of resources with emphasis on ownership, freedom of choice and sustainable behaviour.
Sense of Responsibility. Second, prior research typically assumes that individuals are trying to change their behaviour, or reduce their consumption, but for many the workplace is not their own and not their responsibility. Consequently, not only is the environment and technology controlled for them, people have a diminished sense of responsibility for the energy costs and environmental impact.
External Constraints. Third, the workplace or type of work has its own requirements – they may need to maintain 24-7365 server support. It may be normal for some businesses to have 3 or more machines running per individual, but unusual for others to have a computer at all. This kind of top-down requirement might make individuals feel out of control of the environment and its consumption, leading to lack of motivation.
Given these limiting and influential factors, it is hard to
consider how we can utilise the same nudging technology
that we typically apply in domestic contexts. The few
successful workplace nudging installations have typically
been dependent on a driven community. The CleanSink
project [
REFERENCES [1] 1E PC Energy Report 2009. online at http://www.1e.com/Energycampaign/Index.aspx [7] Kuznetsov, S. and Paulos, E. UpStream: Motivating Water Conservation with Low-Cost Water Flow Sensing and Persuasive Displays. In Proc. CHI 2010, ACM Press (2010). [8] Siero, F.W., A.B. Bakker, G.B. Dekker, and M.T.C. van den Burg. . Changing organizational energy consumption behavior through comparative feedback. In Journal of Environmental Psychology 16: 235-246. (1996) [10] Thaler, R and Sunstein, C. (2008) Nudge: Improving Decisions About Health, Wealth, and Happiness: Yale University Press.