INTERACTION MODALITIES OF PERSONAL SECURITY DEVICE DESIGN Design and Evaluation of Perception and Usability in Using Wearable Device Against Robbery
8
0
0
全文
(2) This research not only centers around how to cope with street robbery employing design solutions but goes deep into the perspective of users’ experience. Led by design principles that concluded from filed survey, a wearable personal security device was designed and prototyped. Then for the using method we proposed three different interaction modalities: physical button, voice recognition , and heartbeat recognition, aimed to examine which is more useful and efficient in an emergency situation. The test also investigated whether interaction ways have different impacts on psychological perspective.. distribution of robbery on the street or transportation almost occupied one third(28.1%) of all types of criminal cases[4]. Moreover, an average of 70 reported violent muggings every day in Mexico City in the first four months of 2019. This survey also pointed almost half of the subjects felt unsafety or very unsafety in the urban environment. An interviewee in the semi interview talked about he had trapped into an assault near to the campus, that time he was forced by two offenders to hand in his properties. Even though he did not get physical hurt, he avoided walking in the same route because of the horrible hunting memory. It can be seen that even people may not. 2. Methods We started our research in Monterrey, the third-largest city located in the northeastern state of Mexico, which has a relatively high crime rate, especially the problem of violent crime such as assault and armed robbery. First, we concluded the residence’ perception and essential characteristics of crime by consulting the previous victimization survey in PULI 001 Program [6], in which 115 residents were engaged in Sinaloa, Mexico, to investigate the information about crime. The findings are as follows:(1) The sense of personal safety is much higher in their home or neighborhood than other urban areas(2) People trust their relatives more than the authority(3) When the crime occurred, people tend to contact family as an initial reaction. At the same time, the mobile phone is ranked to be the primary tool. To have an insight into the robbery issues, in this survey, we conducted a semi-structured interview with 10 interviewers(6 females and 4 males), with an average age of 24.1. All of them are students and faculties who lived in Monterrey. This interview primarily concentrated on their experience of dealing with robbery. All the interviews conducted by a similar frame as followed:(1) Personal information(2) Experience about robbery(3) Reaction and perception during the process. Furthermore, the conclusion combined previous survey and interview will be illustrated in Chapter 3. Afterward, the guidance of design against robbery was put forward for the insight generation. Followed by a smart bracelet was prototyped and evaluated by 60 participants from Universidad de Monterrey(UDEM) and Tecnológico de Monterrey(TEC) in a field usability test.. getting increased severely. ENVIPE 2018 reported that the. Journal of the Science of Design. a factor is that most of the robbery occurred quite suddenly, thus there is no enough time for the victims to react then their belongings had been robbed. What is more, most of the victims would not try to resist the robbers to protect themselves. Explained by one of the interviewers, because most of the robbers carried firearms, even sometimes the robbers could be brought to justice after committing the crime, with the severe possibility they would revenge the victims or witnesses after they released. According to the previous survey in PULI 001, when victims trapped into danger, the majority of them chose to contact family members(78.5%) for the first option instead of calling the police [7]. As well as reported by Elpais, when crime happened, victims turned to their families for help instead of police, because of the weak trust on authority, the relationship between individual and family get much closer and integrate into an emergency [8]. Another finding is that the mobile phone always ranked as the target in a robbery. Reported by Independent [9], the robber always ordered victims to hand over their cell phones, and then to extort other properties, which result in the victims would lose tools to connect with others. 3. 3. Living habits When we interviewed residence in Monterey, we found that multigenerational family members always live in one trace the origin of Mexico’s family, in pre-Columbian times,. Among all types of crime in Mexico, robbery cases are. 2. Through observing the real robbery case videos in Mexico,. household or live near to each other in a same block. If we. 3. Background Study Findings 3. 1. Crime situation. 96. always get physically injury; the mental damage still existed. 3. 2. Response to crime. Vol. xx. Journal of the Science of Design. No. x. 20xx. Vol. 4 No. 2 2020. the social structure was composed of communities in which the essential and vital composition is family [10]. Therefore, in current days, the family is still the priority position among.
(3) many Mexicans. Besides, living nearby is convenient for family contact and support each other. Just like an interviewee said, he would like to contact his family when he got into trouble because the relatives could inform each other in a short time，which made sure he could get the maximal assistance. According to the field research in Monterrey, we knew the residents’ living habits, their attitudes, and response to crime, this information provides a hint for the follow-up work and reference for other related study. 4.Wearable security device design 4. 1. Security products design principles for the robbery situation Based on the results from the background survey, we outlined the following design principles as guidance for the solutions against robbery in Mexico. 4. 1. 1. Convey perception of safety to users The previous online interview in Puli 001 showed that when people feel unsafe, a call from family or friends could relive their nervousness [8]. It indicates that perception of insecurity could be relived to some degree by connecting with a trustful relationship, making sure there is someone who knows the real-time condition can comfort the sufferer as well. Furthermore, the results showed that people feel least insecurity of the neighborhood compared with the federal or municipal government because of the strong and imitate connection of masses were considered to be more efficient [4]. An interviewee said, “We rarely bother lodging a complaint to the police, believing the authorities will never launch an investigation, meaning that these criminals enjoy impunity.” In the meanwhile, the caregivers’ feedback should also be conveyed to the victim, which informed them they were attentive. The previous survey queried about people in Sinaloa how to deal with the emergency, and most victims would like to call family, when they hear the voice of the family, they felt calm down. However, in consideration of most robbery, the mobile phone is a target, how to contact family or friends without making calls, such as convey the victims’ physical situation or location with some mobile devices. 4. 1. 2. Highly efficiency to use The high efficiency to use includes two aspects, first it should be with the users all the time and easy to find out whenever they need it. Based on the survey about robbery, no matter in the street or public transportation robbery, with or without weapons, majority of the cases occurred instantly with short duration, at the top speed even finished in 10 seconds. In this case, the victims need to take out and activate the device. right the first time. On the other hand, it should be comfortable and natural for the users to arouse the function, especially when they are facing danger, in which case it is nearly impossible to be calm enough to know what to do at once. Thus, we think about how to guide people to interact with the device “instinctively” , even their mind turns to a total blank, the SOS message still could be delivered to contacts. 4. 1. 3. Safety to use Consideration should also be given to the aspect of the users’ safety while using the device in an acute situation. Paul Ekblom enlightened designers when taking up a task about the products or services against crime, prevention should be a crucial consideration [11]. Paul explained that these products required great subtlety design, which not be too attractive nor vulnerable to the offenders. Reviewing some research about the robbery, which indicted the people who looked rich or have some valuables, for example, brand-name watches or handbag appeared to become prey, even calling on phones or listening to Mp3 also might be conspicuous for the robbers [12] [13]. Therefore, the device should be “tricky”, not only the appearance cannot be looked like too valuable nor attractive, the way of activating cannot be aggressive nor too obvious to prevent from being hurt by the offenders. 4. 2. Design features for Famiband security service Based on the design principles described above, a test prototype named Famiband was generated for the field usability study. 4. 2. 1. Convey perception of safety to users by establishing a connection between contacts Based on the first principle, Famiband could support contact with family or friends. An SOS message includes the real-time user’s location information would be sent to emergency contacts when the users trapped in danger. In the meanwhile, the Famiband could inform users their SOS had already been sent and read successfully through vibration. In the form of vibration is an appropriate way in this situation that could be perceived only by the users, but also work inconspicuously, no offensive meaning to the robbery. Feedback in a humanmachine system is crucial valuable. Timely and appropriate feedback could make sure the users know the conditions they are in, then to enhance the sense of controlling and safety about the system [14]. 4. 2. 2. Highly security and efficient to use Followed by the second principle, the device first should be touched always and everywhere. We designed it in the shape of a bracelet. Because a wearable device should be portable and capable in most of the situation, even sometimes. Journal of the Science of Design Vol.Vol. 4 No. 2 2020 Journal of the Science of Design xx No. x 20xx 97 3.
(4) emergency, users could touch it rapidly and quietly. About the material and appearance，the bracelet belt is made of silicon materials, which is soft and fit to wear on the wrist. Besides, to consider of the safety principle，there is no display or screen dial on it. This understated and inexpensive look could avoid arousing the offenders’ attention and being targeted attacks in case of robbery happened. See Figure1 for the conceptual model of Famiband.. smart housekeeper, “Google smart home”. Speech recognition technology sprang from the most initial and natural way of communication among human beings, just like we know how to speak to people, convey our needs by words is much more comfortable and direct than open an App on the mobile phone. As Gregory D. Abowd and Elizabeth D. Mynatt stated that humans’ natural actions should be used to input the HCI systems such as speech, gesture or even using tools to communicate with others [15]. Some other scenes are more incapable or difficult to use a cell phone or other device, such as in driving or cooking, also in the robbery, it is impossible to take out the smartphone while we observed in some real robbery videos. However, people might cry out subconsciously. Then we choose voice interaction as another way to arouse the. Figure 1. The appearance of Famiband bracelet conceptual prototype. 4. 3. Three interaction modalities To explore the relatively useful and efficient way to manipulate the device, we considered about 3 options to interact with the Famiband as displayed in Figure 2.. device. A microphone was connected to Famiband to collect and detect the participants’ voices,. when they said. “Espela”(which means “Wait” or “Hold” on in English) would activate Famiband to send SOS message. 4. 3. 3. Heartbeat recognition For the last option, considering when faced with dangerous situations or extremely nervous, people tend to be too anxious to make the correct decision, let alone operate a device. Therefore, it may be more efficient and reliable if the device could react automatically according to different situations. There have been numerous studies on how to read a person's physiological features to analyze the condition. Vivek P N and his team members utilized the phenomenon of sympathetic response to design a fear sensing system [16], as according to the secretion of sweat glands and the volume of the blood. Figure 2. Famiband conceptual service structure. pump into the capillary body tissues would be improved. Takuji Suzuki utilized the changes of the heartbeat to develop. 4. 3. 1. Physical button The first option we came up with as the physical button, users could press a button on the device more than two seconds(which considered of fault tolerance), then the SOS message would be sent. In the consideration of the users’ habit, physical button is the most common component of the product in our daily life, by this way, the users don not need to pay much attention to learn how to use, what’s more pressing button was assumed to be easy to operate during the robbery situation. The second option was designed as voice recognition. As the speech recognition technology has been applied wildly in daily life, from “Siri”, our personal assistant in iPhone, to the 4. Journal of the Science of Design. actual feasibility of using photoplethysmography to detect the fluctuation amplitude data in daily movements [17]. Inspired by these findings ， for the third option, Famiband could monitor the users’ real-time pulse through a built-in heartbeat sensor, if the heartbeat values were detected above a reasonable threshold it could be considered that an emergency occurred, the device would send SOS messages at once. The working frame of the last option is displayed in Figure 3. The instruction of the prototype utilized in the field test will. 4. 3. 2. Voice recognition. 98. an irregular pulse detection algorithm, and they authorized the. Vol. xx. Journal of the Science of Design. No. x. 20xx. Vol. 4 No. 2 2020. be explained in subsection 5.2..
(5) Figure 4. Implements and components of Famiband test prototype. Figure 3. The framework of the heartrate recognition program. 5. Famiband field study in Mexico Field research had been accomplished in two universities, UDEM in Monterey aimed to explore the following questions： 1. Whether different interaction modalities could impact on users’ evaluation of phycology and functionality? 2. Could the Famiband provide the perception of safe to the users? 5. 1. Participants 60 students and staffs(male 25, female 35, aged 17-54) were recruited from UDEM and TEC. There were 20 participants in each group. Because of the object of this study was to test whether three interaction ways have different influence on users’ evaluation, we divided participants into three groups based on interaction modalities: physical button group, voice recognition group and automatic recognition group. For button group there was 10 female and 10 males, mean aged 20.4, for voice recognition group there was 12 females and 8 males, mean aged 22.4, for heartbeat recognition group there was 13 females and 7 males, mean aged 24.8. The same equipment was utilized in all groups. 5. 2. Materials The design of the test prototype is displayed in Figure 4. The core component consisted of an Arduino microprocessor; a self-regulating transmissive finger pulse sensor tied on the participants' index finger to measure their heartbeat. A button was set for the button group, a microphone for the voice recognition group, and a small motor used for generating vibration. The prototype was tied on the participant's wrist. During the experiment, in both automatic recognition group and voice recognition group, the main part of the bracelet was covered by a cardboard box, while in the button group, we would unfold the box let the button exposed to the participants.. Since the target of this research was to test the user’s response in an emergency situation, it was necessary to make the participants felt they were really there, a scenario which closed to a real nervous robbery. Inspired by some horrible Virtual Reality(VR) games, we decided to use the VR device to create a robbery story that could invoke the participants threatened but, in a way, not really harmful. We planned a VR experiment to make the participants experience an emergency situation through watching a spherical video recorded by a Kodak spherical panorama camera. In this video, the participant could see the whole environment in 360 degree, so the participants could act and look around more flexibly. At the beginning, the participants could see a dim and empty aisle followed the camera, moved forward slowly to an exit at the end of the aisle, when they almost walked out of the exit, suddenly a robber(acted by one of our group members) dressed in black with a horrible skull mask on his face rushed out from dark corner, threatened the beholder in a terrible voice in Spanish which mean “Don’t move and give me your money”, in a while time, a panic and deafening background music sounded. This plot aimed to make the participants suddenly felt scared in an unexpected situation(Figure 5).. Figure 5. Screenshots of “criminal story” video. 5. 3. Procedure The experiment was implemented in two universities, they provided us observation rooms prepared with recording equipment and see-through mirror, for the whole process aural and video recording avoid disturbing participants. Before the experiment started, participants were arranged to wait in a preparing room to sign an informed consent document, then they could enter the experiment room one by one. The experiment started with introducing the process of the experiment and the instruction about the prototype. Then we fitted the participants with the VR device to watch the “robbery story” and wear the Famiband prototype on their wrists. After these preparations the participants started to watch the video. Journal of the Science of Design Vol.Vol. 4 No. 2 2020 Journal of the Science of Design xx No. x 20xx 99 5.
(6) then reacted according to different groups’ settings. In the button group, when the participants felt scared, they could press the button, for the voice recognition group they held the microphone and say the keyword, and for the heart rate recognition group the participants didn’t need to operate anything. When the video ends, all the participants were asked to finish a questionnaire about the usage experience. 5. 4. Measure During the experiment, we recorded the participants’ pulse two times to compare whether there existed a significant change in heart rate when people were shocked. At the beginning for the normal heart rate(NHR), then at the scaring moment for the shocking heart rate(SHR). As the NHR value ranged from 52.6 bpm to 109.4 bpm, and the SHR ranged from 62.4 bpm to 160.3 bpm. On account of the results exhibited significant variations among individuals ， we adopted the growth rate of heartrate(SHR-NHR)/ NHR as a standard gauge to measure the variance. As shown in figure 6, 44 participants(73.3% of the total) whose heart rate showed a noticeable growth of more than 10% at the “scaring moment”. It also indicates that our experiment meets the target to cause panic to the majority of people.. differences as gender and interaction modalities. SPSS® was applied to the statistical analysis. 5. 5. Results Regards to the usability evaluation to Famiband, the ANOVA results showed there had significant differences in the usage evaluation on the aspect of usability contained efficiency to use, intention to use and difficult to use. While there were no significant differences in other scales. As the result was summarized in Figure 7 and Figure 8.. Figure 6. The histogram of the number of heart rate change. A questionnaire included 10 questions about 2 aspects： perception and usability were applied to measure the evaluation on a 5-point Likert scale. The perception aspect consists of：sense of safety, relieved, trust, risk and anxious. For the usability aspect, efficiency, understanding of the use, intention to use, easy and troublesome to use were included. On account of both the voice recognition group and the heartbeat group, a participant did not get the feedback from the device, so they gave up finishing the questionnaire, result in there were 58 questionnaires recovered. ANOVA was applied to test if there were statistically significant differences among three groups, then an LSD posthoc test was carried out to determine effect differences between every two groups. Multi-way ANOVA was also further performed to test whether interaction effect between 6 100. Journal of the Science of Design. Vol. xx. Journal of the Science of Design. No. x. 20xx. Vol. 4 No. 2 2020. Figure 7. ANOVA results of three evaluation scales. LSD(Least Significant Difference) post-hoc test was run to further compare the differences between every two groups. For the scale of efficiency to use, the automatic recognition group was statistically significantly lower than the voice recognition group(MD=-0.58, p=0.045) and button group(MD=-0.78, p=0.007). The analogous results were found in the scale of intention to use. The score of automatic recognition group significantly less than the other two groups represented by(MD=-1.00, p=0.017) and(MD=-0.95, p=0.022)..
(7) In the case of difficulty to use, statistically significantly difference was only found between automatic recognition group and button group(MD=0.88, p=0.011), while the voice group had a relatively mediocre evaluation. The LSD post-hoc declared that automatic recognition has notable less positive evaluation than the other two groups on the aspect of efficiency to use, the intention to use and easy to use. Considering both genders were recruited in this experiment, we wondered that if the evaluation existed a difference between gender. Therefore, a Multi-way ANOVA(MANOVA) was carried out. The results illustrated that the interaction modalities had a significant main effect rather than on gender.. Figure 8. Means for evaluation scales in perception and ability(no significant difference). 6. Discussion In the experiment, the primary finding was that different interaction modalities influenced users’ evaluation of usability as the scale of efficiency and easy to use. What’s more the apparent difference exhibited in the control mode of sending an SOS signal by users or by the automatic system. The results largely elucidated the physical button and voice recognition was more useful and efficient. From the aspect of cognitive burden, physical button has already implanted in users’ minds as a skill, as well as the speech interface has been well-utilized on numerous occasions, what’s more based on the NUI(natural. user interface) theory, screaming out quite consists with the robbery scenario. In regard to these two options, participants did not need to learn and practice how to operate it. However, the automatically identification was rarely familiar for most of the participants, which resulted conceivably possible that they were not acquainted with the working mechanism about this device. Even though the heartbeat recognition reduced the body movements to some extent, in the meanwhile it increased the users’ sense of controlling and understanding of the device status. As some participants talked about their confusion during the experiment, “Am I nervous enough to perform the SOS message?” or “I am so nervous now but whether the device could sense that or not?”, “When my contacts could receive the message?” and so on. Moreover, there have individual differences that occurred in the perception of feedback interval, some of the participants responded that they thought the interval time was a little longer than they expected which may be another factor to affect their final evaluations. These reasons led to the process considered to be not so fluent and smooth to use. With regards to there was no significant effect on the psychological feeling, it still couldn’t conclude that interaction modalities had no influent on users’ perception of safety or trust, because of the complex subjective factors did not be taken into account, such as the personal experience and cultural background difference, even the current physical condition, all of these variables would influent the participants’ psychological activities. What else the experiment setting could also affect the result. As the test had to be arranged indoor for the implementation of the prototype, actually it should be used in outdoor environment. In the future study, our experiment would promote the environment for a real usage scenario test. 7. Conclusion This research was associated with PULI(Post Urban Living Innovation) Program 001which jointly in Japan and central America, aimed to develop innovative insights and solutions for nowadays urban living. This part is aimed to utilize design thinking and technology against crime and promote the living conditions in Mexico. In this study, we investigated the dwellers’ attitude and current solutions for street robbery. The previous research and interview reported residence most trust in family and neighborhood. Thus, it is important to establish a connection between individuals and the community against the fear of crime. Necessary elements are also extracted for building a personal security device: portable, high efficiency and safety. Journal of the Science of Design Vol.Vol. 4 No. 2 2020 Journal of the Science of Design xx No. x 20xx 1017.
(8) to use, as well as convey SOS signal to contacts. A test prototype Famiband was implemented to examine in a filed study. The main objective was to study the effect of different interaction modalities on the users’ evaluation of functionality and perception. The results proved that the self-operated system may be not suitable and useful in the emergency compared with user-controlled system, the feeling of control and clear understanding of the system are basic needs for users in street robbery or other crime situation, also indicated that speech recognition could be considered to be an efficient way in this usage scene. The follow-up research should be implemented outdoor where the location and time are closed to the real scenario to engage the subjects’ psychological feelings. Besides, it is necessary to consider the design elements more deeply like the materials or shape. More than this, it is extremely important to construct psychological safety for urban living, even people who live in a community not critical, the sense of risk from surroundings or social media could have an impact on mental. While in the precocious research, mobile phone was considered to be utilized to contact with family in case of emergency, it still needs to determine further whether smart device such as wearable bracelet could strengthen users’ perception of safety as same as a mobile phone. References [1] West, GB., Scale: The Universal Laws of Growth, Innovation, Sustainability, and The Pace of Life in Organisms, Cities, Economies, and Companies, New York: Penguin Press, 2017. [2] Jackson, J., and Gray, E., Functional Fear and Public Insecurities about Crime. Br J Criminol, 50(1), 1–22, 2010 [3] Ruijsbroek, A., Droomers, M., et al., Social Safety, Selfrated General Health, and Physical Activity: Changes in Area Crime, Area Safety Feelings and The Role of Social Cohesion. Health Place, 31, 39–45, 2015 [4] INEGI：http://en.inegi.org.mx/programas/envipe/2018/ [5] Blom, J., Viswanathan, D., et al., Fear and the City – Role of Mobile Services in Harnessing Safety and Security in Urban Use Contexts, CHI 2010: Death and Fear, 10-15, 2010 [6] PULI：http://design-cu.xsrv.jp/puli/category/puli001/ (A ccessed 12 September 2019) [7] Chacon, JC., Nimi, HM., Watanabe M., Ono K., Paskevicius, A., Reducing Fear of Crime Through Design Against Crime, Journal of the Science of Design, 2(1), 2936, 2018. 8 102. Journal of the Science of Design. Vol. xx. Journal of the Science of Design. No. x. 20xx. Vol. 4 No. 2 2020. [8] Elpais : https://www.inegi.org.mx/programas/envipe/20 18(Accessed 27 December 2019) [9] Independent：https://www.independent.co.uk/news/world /americas/mexico-fake-phones-armed-robberies-buses-a8 924516.html(Accessed 16 October 2019) [10] Sanders, WT., Hagen, VW., Pre-Columbian Civilizations, Encyclopædia Britannica,, 2019 [11] Ekblom, P., Gearing Up Against Crime: A Dynamic Framework to Help Designers Keep up with the Ad aptive Criminal in a Changing World, International J ournal of Risk, Security and Crime Prevention, 2(4), 249-265, 1997 [12] Monk, KM., Heinonen JA., et al., Problem-Oriented Guides for Police Problem-Specific Guides Series，4-9, 2010 [13] POLICE UK: https://www.police.uk/crime-preventionadvice/(Accessed 3 December 2019) [14] Wilson, G., Brewster, S., et al., Thermal Icons: Evaluating Structured Thermal Feedback for Mobile Interaction. Proceedings of International Conference on Humancomputer Interaction with Mobile Devices and Services (2012), 309–312 [15] Abowd, GD., and Mynatt, ED.,ACM Transactions on Computer-Human Interaction, Special Issue on HumanComputer Interaction in the New Millennium, 7(1), 29-58, 2000 [16] Vivek, PN., Suhas, RC., et al., Design and Implementation of Smart Wrist band for Safety Measures in Emergency, WiSE National conference on Wireless, Signal Processing and Embedded Systems, 2013, 62-65 [17] Suzuki, T., Kameyama K., et al., Development of t he Irregular Pulse Detection Method in Daily Life U sing Wearable Photoplethysmographic Sensor, IEEE E MBS Annual International Conference, 2009, 2-6.
(9)
図
関連したドキュメント
GoI token passing fixed graph.. B’ham.). Interaction abstract
Two grid diagrams of the same link can be obtained from each other by a finite sequence of the following elementary moves.. • stabilization
Naohiko Hoshino, Koko Muroya, Ichiro Hasuo, Memoryful Geometry of Interaction:.. From Coalgebraic Components fo Algebraic Effects , submitted to
The framework is based on a traced symmetric monoidal category, and it yields a certain compact closed category as a model of linear combinatory algebra, covering as much as
The mGoI framework provides token machine semantics of effectful computations, namely computations with algebraic effects, in which effectful λ-terms are translated to transducers..
In this section we will show that in the case of a generic quadric the variety G(n, Q) is 2-incompressible, and also will formulate the conjecture describing the canonical dimension
4 because evolutionary algorithms work with a population of solutions, various optimal solutions can be obtained, or many solutions can be obtained with values close to the
Finally, in Figure 19, the lower bound is compared with the curves of constant basin area, already shown in Figure 13, and the scatter of buckling loads obtained
