ࡱ>  z|qrstuvwxy 1'bjbj! ! .,CfCf#"h+22)) * * *4*4*4*8l*L+4*fT. ;;<<]T.bc$W 2 *d^]|]dd))<<Z!ooodx)R< *<~nodoo6T)"ԉ<Fc5`NfP4j0P;nj;8ԉԉ; *zddodddddoddddddd;ddddddddd2 R(: Warning! A Comprehensive Model of the Effects of Digital Information Security Warning Messages Mario Silic1, Jordan Barlow2, Dustin Ormond3 1University of St Gallen 2 California State University 3 Creighton University Research-in-progress Abstract Despite existing countermeasures to combat malicious actions, users are the last line of defense to protect personal and organizational assets. Given that users often ignore warning messages that motivate compliant behavior, the issue of protecting personal and organizational assets is exacerbated. Messages that are largely ignored cannot have any impact on attitudes, motivation, or behavior. Therefore, crafting messages that increase attention and comprehension regarding specific threats and ways to cope with these threats is vital. This research combines the communication-human information processing (C-HIP) model with protection motivation theory (PMT) to assess how warning message content affects adherence especially when users pay attention to the content of the warning message. In essence, this study considers a holistic view of examining the channel (warning message), attention, comprehension and their influence on attitudes and beliefs, motivation, and behavior. Additionally, we propose including alternative courses of action in digital warning messages to increase secure attitudes, beliefs, and behavior. We test this holistic model through a series of field and lab experiments to evaluate message comprehension, attitudes, and beliefs and capture actual attention and secure behavior. Introduction Cybercrime is increasing, targeting individuals, organizations, and governments at a rapid rate. The estimated cost of cybercrime for the global economy is around $445 billion each year, where 800 million people in 2013 were affected by cyber espionage and loss of private information  ADDIN EN.CITE McAffee201419(McAffee, 2014)191927McAffee,Net Losses: Estimating the Global Cost of Cybercrime2014http://www.mcafee.com/us/resources/reports/rp-economic-impact-cybercrime2.pdfApril 2015( HYPERLINK \l "_ENREF_35" \o "McAffee, 2014 #19" McAffee, 2014). Despite many existing countermeasures aiming at protecting users integrity (e.g., antivirus software, firewalls, operating system mechanisms such as password protection when installing new software, etc.), in practice users represent the last line of defense against malicious actions. Such actions can be either directed against themselves (e.g., malware destroying users hard drive) or against organizational assets where the user is used as the backdoor by cybercriminals. Information security research has examined several different methods and techniques for persuading users to behave securely in organizations and other settings (e.g., deterrence techniques, anti-neutralization techniques, SETA training programs, etc.). However, these techniques have not been evaluated when examining pop-up warning messages. Warnings represent communication designed to prevent users from hurting themselves or others  ADDIN EN.CITE Wogalter200620(Wogalter, 2006b)202017Wogalter, Michael SPurposes and scope of warningsHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJ3-92006( HYPERLINK \l "_ENREF_53" \o "Wogalter, 2006 #20" Wogalter, 2006b) and as such, physical warnings have been shown to be very effective in preventing hazards or criminal incidents  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_12" \o "Coleman, 2007 #26" Coleman, 2007;  HYPERLINK \l "_ENREF_18" \o "Goldstein, 2008 #27" Goldstein, Cialdini, & Griskevicius, 2008;  HYPERLINK \l "_ENREF_44" \o "Schultz, 2009 #28" Schultz & Tabanico, 2009). Less is known about digital or computer warning messages. Digital warnings are unique from other security measures in that they are usually not the first line of defense for users. According to the hazard control hierarchy  ADDIN EN.CITE Wogalter200620(Wogalter, 2006b)202017Wogalter, Michael SPurposes and scope of warningsHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJ3-92006( HYPERLINK \l "_ENREF_53" \o "Wogalter, 2006 #20" Wogalter, 2006b), the first step to control or remove risk is an attempt to eliminate or minimize the hazard as much as possible. The second step strives to minimize the interaction between the user and the hazard. Finally, the third step provides warning messages to the user which may reduce risk by enabling better decision-making. In other words, warning messages are unique in that they are provided only when other, potentially more powerful, security measures are not able to keep risk from the user. An example of non-digital warnings is that of the tobacco industry. Users are constantly informed about the health risks of smoking and its consequences. However, warnings are quite often ignored by users and may even produce the boomerang effectthat is, warnings have the potential to increase harmful behavior by drawing attention to such behavior  ADDIN EN.CITE Bushman20067649(Bushman, 2006)7649764917Bushman, Brad JEffects of warning and information labels on attraction to television violence in viewers of different agesJournal of Applied Social PsychologyJournal of Applied Social Psychology2073-207836920061559-1816( HYPERLINK \l "_ENREF_10" \o "Bushman, 2006 #7649" Bushman, 2006). A similar phenomenon seems to be happening in the digital world. For example,  ADDIN EN.CITE Egilman20067650Egilman and Bohme (2006)7650765017Egilman, DavidBohme, SA brief history of warningsHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJHandbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJ35-482006 HYPERLINK \l "_ENREF_16" \o "Egilman, 2006 #7650" Egilman and Bohme (2006) argue that people do not read digital warnings, as they are habituated to them. Other studies found that users ignore web browser SSL warnings and simply skip them  ADDIN EN.CITE Akhawe20131e.g. (e.g. Akhawe & Felt, 2013; Sunshine, Egelman, Almuhimedi, Atri, & Cranor, 2009a)113Akhawe, DevdattaFelt, Adrienne PorterAlice in Warningland: A large-scale field study of browser security warning effectivenessProceedings of the 22nd USENIX Conference on Security257-2722013USENIX Association9781931971034Sunshine200978897889788910Sunshine, JoshuaEgelman, SergeAlmuhimedi, HazimAtri, NehaCranor, Lorrie FaithCrying Wolf: An Empirical Study of SSL Warning EffectivenessUSENIX Security Symposium399-4162009(e.g.  HYPERLINK \l "_ENREF_1" \o "Akhawe, 2013 #1" Akhawe & Felt, 2013;  HYPERLINK \l "_ENREF_46" \o "Sunshine, 2009 #7889" Sunshine, Egelman, Almuhimedi, Atri, & Cranor, 2009a). Research on computer warning messages indicates that HCI elements are integral parts of these messages  ADDIN EN.CITE Bravo-Lillo20118(Cristian Bravo-Lillo, Lorrie Faith Cranor, Julie S Downs, & Saranga Komanduri, 2011a)8817Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie SKomanduri, SarangaBridging the gap in computer security warningsIEEE Security and Privacy18-269March/April2011( HYPERLINK \l "_ENREF_8" \o "Bravo-Lillo, 2011 #8" Cristian Bravo-Lillo, Lorrie Faith Cranor, Julie S Downs, & Saranga Komanduri, 2011a); however, it does not assess the psychological effects of these messageswording of warning messages appears to be based on trial and error rather than persuasion or communication theories  ADDIN EN.CITE Modic20147890(Modic & Anderson, 2014)7890789017Modic, DavidAnderson, RossReading this may harm your computer: The psychology of malware warningsComputers in Human BehaviorComputers in Human Behavior71-794120140747-5632( HYPERLINK \l "_ENREF_37" \o "Modic, 2014 #7890" Modic & Anderson, 2014). However, it is important to test theory-based communication in the unique warning messages context that is less direct than the previously-tested theory-based security trainings and other security information. The few research studies that have addressed computer security warning content have neglected that users commonly ignore warnings in the first place  ADDIN EN.CITE Akhawe20131e.g. (e.g. Akhawe & Felt, 2013; Sunshine et al., 2009a)113Akhawe, DevdattaFelt, Adrienne PorterAlice in Warningland: A large-scale field study of browser security warning effectivenessProceedings of the 22nd USENIX Conference on Security257-2722013USENIX Association9781931971034Sunshine200978897889788910Sunshine, JoshuaEgelman, SergeAlmuhimedi, HazimAtri, NehaCranor, Lorrie FaithCrying Wolf: An Empirical Study of SSL Warning EffectivenessUSENIX Security Symposium399-4162009(e.g.  HYPERLINK \l "_ENREF_1" \o "Akhawe, 2013 #1" Akhawe & Felt, 2013;  HYPERLINK \l "_ENREF_46" \o "Sunshine, 2009 #7889" Sunshine et al., 2009a). For example,  ADDIN EN.CITE Egelman20082Egelman, Cranor, and Hong (2008)223Egelman, SergeCranor, Lorrie FaithHong, JasonYou've been warned: An empirical study of the effectiveness of web browser phishing warningsProceedings of the ACM CHI Conference on Human Factors in Computing Systems1-1020081595931783 HYPERLINK \l "_ENREF_14" \o "Egelman, 2008 #2" Egelman, Cranor, and Hong (2008) manipulated the content of the malware warning to understand the effects on the users behavior but did not take into account the initial warning ignorance where users do not read warning at all. On the other hand, recent research on warnings  ADDIN EN.CITE Anderson201416(B. B. Anderson, Vance, Kirwan, Eargle, & Howard, 2014a, 2014b)161647Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethWhy users habituate to security warnings: Insights from fMRI2014 IFIP 8.11 Dewald Roode Security Workshop2014Anderson201418181810Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethUsers aren't (necessarily) lazy: Using neuroIS to explain habituation to security warnings2014 International Conference on Information Systems (ICIS)2014( HYPERLINK \l "_ENREF_2" \o "Anderson, 2014 #18" B. B. Anderson, Vance, Kirwan, Eargle, & Howard, 2014a,  HYPERLINK \l "_ENREF_3" \o "Anderson, 2014 #16" 2014b) focuses on why computer warning messages are largely ignored and ways (e.g. polymorphic warnings) to have people pay more attention to them  ADDIN EN.CITE Anderson201416(B. B. Anderson et al., 2014a, 2014b)161647Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethWhy users habituate to security warnings: Insights from fMRI2014 IFIP 8.11 Dewald Roode Security Workshop2014Anderson201418181810Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethUsers aren't (necessarily) lazy: Using neuroIS to explain habituation to security warnings2014 International Conference on Information Systems (ICIS)2014( HYPERLINK \l "_ENREF_2" \o "Anderson, 2014 #18" B. B. Anderson et al., 2014a,  HYPERLINK \l "_ENREF_3" \o "Anderson, 2014 #16" 2014b). However, even if people actually read the warning messages, they may reject them based on their content. Therefore, research should assess how warning message content, based on theory, may affect adherence especially when users pay attention to the content of the warning message. Essentially, an empirical research study is needed to understand both the attention and content aspects of computer warning messages and their effects on users. In this study, we measure the total time people spend (attention) reading a variety of warning messages (content). Then we examine the effect of the content for only those who actually paid attention to the warning message. As a foundation for this research, we evaluate the Communication-Human Information Processing (C-HIP) Model to understand and test the process and interactions of attention, comprehension, attitudes, beliefs, and motivation on ultimate user behavior when they encounter computer security warning messages. Further, based on the C-HIP model and other related theories of communication and persuasion, such as the Health Beliefs Model, this study proposes a new content element of computer security warning messages (i.e., suggesting alternative secure courses of action) that users who pay attention to warning message content may be more persuaded to behave securely. Thus, this leads to our research questions: RQ1. What aspects of warning messages are most powerful in keeping individuals from performing potentially insecure IT behavior, particularly considering the attention and comprehension of the user toward the message? RQ2. When warning messages include content directing users to alternative courses of action, is the likelihood to heed the warning increased? Theoretical Background Research on warnings in the physical world has been categorized into the Communication-Human Information Processing (C-HIP) Model  ADDIN EN.CITE Conzola200117(Conzola & Wogalter, 2001; Wogalter, 2006a)171717Conzola, Vincent CWogalter, Michael SA Communication-Human Information Processing (C-HIP) approach to warning effectiveness in the workplaceJournal of Risk ResearchJournal of Risk Research309-322442001Wogalter20061515155Wogalter, Michael SWogalter, M SCommunication-Human Information Processing (C-HIP) ModelHandbook of Warnings51-612006Mahwah, NJLawrence Erlbaum Associates( HYPERLINK \l "_ENREF_13" \o "Conzola, 2001 #17" Conzola & Wogalter, 2001;  HYPERLINK \l "_ENREF_52" \o "Wogalter, 2006 #15" Wogalter, 2006a), shown below in Figure 1. This framework shows that in order to communicate a message (such as a warning), you have to consider the source, the channel, and multiple aspects of the receiver. These aspects start with gaining and retaining attention and then proceed to comprehension, attitudes, beliefs, motivation, and ultimately behavior. Source refers to the person or entity delivering the message. In the case of digital information security warnings, the source could be anti-virus software, an organizations IT department, or others. However, the source is often hidden from the userthey only see warning messages as appearing on the screen out of nowhere. Although the source is an important attribute in successfully communicating a message through a warning channel, for the sake of simplicity and brevity we do not focus on source characteristics in this study. Future research should address this issue. Figure 1. C-HIP model. Adapted from  ADDIN EN.CITE Conzola200117Conzola and Wogalter (2001)171717Conzola, Vincent CWogalter, Michael SA Communication-Human Information Processing (C-HIP) approach to warning effectiveness in the workplaceJournal of Risk ResearchJournal of Risk Research309-322442001 HYPERLINK \l "_ENREF_13" \o "Conzola, 2001 #17" Conzola and Wogalter (2001) and  ADDIN EN.CITE Wogalter200615Wogalter (2006a)15155Wogalter, Michael SWogalter, M SCommunication-Human Information Processing (C-HIP) ModelHandbook of Warnings51-612006Mahwah, NJLawrence Erlbaum Associates HYPERLINK \l "_ENREF_52" \o "Wogalter, 2006 #15" Wogalter (2006a) Channel is the method of delivering the communication. In the case of digital warning messages, the warning message itself is the channel. Previous research has indicated that the source of the warning message communicates the presence of a hazard through some media channel to a recipient  ADDIN EN.CITE Chen201439(Chen, Stepan, Dick, & Miller, 2014)393917Chen, Teh-ChungStepan, TorinDick, ScottMiller, JamesAn anti-phishing system employing diffused informationACM Transactions on Information and System Security (TISSEC)ACM Transactions on Information and System Security (TISSEC)1616420141094-9224( HYPERLINK \l "_ENREF_11" \o "Chen, 2014 #39" Chen, Stepan, Dick, & Miller, 2014). For instance,  ADDIN EN.CITE Bravo-Lillo20139Bravo-Lillo et al. (2013)9910Bravo-Lillo, CristianCranor, Lorrie FaithDowns, JulieKomanduri, SarangaReeder, Robert WSchechter, StuartSleeper, ManyaYour attention please: Designing security-decision UIs to make genuine risks harder to ignoreSOUPS '13 Proceedings of the Ninth Symposium on Usable Privacy and Security1-18attractorsdialogshabituationusable securitywarnings2013978145032319210.1145/2501604.2501610 HYPERLINK \l "_ENREF_7" \o "Bravo-Lillo, 2013 #9" Bravo-Lillo et al. (2013) designed attractors (i.e. user interface modifications) to draw attention to the most important and pertinent information to aid in decision-making. Another study tracked eye movements and found that users paid no attention to SSL icons  ADDIN EN.CITE Grier200840(Grier, Tang, & King, 2008)404010Grier, ChrisTang, ShuoKing, Samuel TSecure web browsing with the OP web browserSecurity and Privacy, 2008. SP 2008. IEEE Symposium on402-4162008IEEE0769531687( HYPERLINK \l "_ENREF_19" \o "Grier, 2008 #40" Grier, Tang, & King, 2008). However, these studies addressed the effect of digital warning channel attributes either on (1) behavior, without investigating the receiver attributes that mediate or moderate this relationship, or (2) attention, without considering that attention is only one aspect of the receiver affecting their ultimate behavior. This study focuses on the various stages of C-HIP involving the receiver (i.e., the person toward whom the warning message is addressed) and the effects of channel content on the receiver. As shown in the gray box in the figure above, there are several steps involved in communicating a message, and several different factors that could affect the ultimate behavior of the receiver. The first step is attention. If receivers are not paying attention to the message, it cannot have any further impact on their behavior. Attention can often be gained through simple visual aspects (e.g., size, colors, graphics)  ADDIN EN.CITE Laughery200619(Laughery & Wogalter, 2006)191917Laughery, Kenneth RWogalter, Michael SDesigning effective warningsReviews of human factors and ergonomicsReviews of human factors and ergonomics241-2712120061557-234X( HYPERLINK \l "_ENREF_29" \o "Laughery, 2006 #19" Laughery & Wogalter, 2006). Another aspect that can have high impact on the users attention is the environment itself, which can be cluttered and noisy. Thus, to attract attention, a warning has to be conspicuous or salient relative to its context  ADDIN EN.CITE Sanders198721(Sanders & McCormick, 1987)21216Sanders, Mark SMcCormick, Ernest JHuman factors in engineering and design1987McGRAW-HILL book company0070449031( HYPERLINK \l "_ENREF_42" \o "Sanders, 1987 #21" Sanders & McCormick, 1987). According to  ADDIN EN.CITE Wogalter199620Wogalter and Laughery (1996)202017Wogalter, Michael SLaughery, Kenneth RWarning! Sign and label effectivenessCurrent Directions in Psychological ScienceCurrent Directions in Psychological Science33-3719960963-7214 HYPERLINK \l "_ENREF_55" \o "Wogalter, 1996 #20" Wogalter and Laughery (1996), a users attention will be driven by (1) spatial and temporal factors such as novelty, size, illumination, and contrast, (2) signal words such as DANGER, (3) signal icons such as an exclamation point, (4) color such as red which signals danger in many cultures, and (5) pictures such as a pictorial sign displaying smoking consequences. One study on web browser warnings, such as those that appear when users visit suspected phishing websites, showed that altering text and color led to a significant increase of users attention  ADDIN EN.CITE Egelman20137894(Egelman & Schechter, 2013)789478945Egelman, SergeSchechter, StuartThe importance of being earnest [in security warnings]Financial cryptography and data security52-592013Springer3642398839( HYPERLINK \l "_ENREF_15" \o "Egelman, 2013 #7894" Egelman & Schechter, 2013). Because the effects of channel aesthetics on attention are complex and have been studied extensively in the literature, we do not focus on this aspect for the purposes of empirical testing in our model. The next step is comprehension. Comprehension evaluates an individuals level of understanding of the message itself and the consequences associated with disregarding the message. Given the variety of knowledge among users, warning messages should be crafted to target the least-skilled user to ensure all recipients understand the messages  ADDIN EN.CITE Wogalter199620(Wogalter & Laughery, 1996)202017Wogalter, Michael SLaughery, Kenneth RWarning! Sign and label effectivenessCurrent Directions in Psychological ScienceCurrent Directions in Psychological Science33-3719960963-7214( HYPERLINK \l "_ENREF_55" \o "Wogalter, 1996 #20" Wogalter & Laughery, 1996). When examining messages related to information security, technical jargon may increase the difficulty of comprehension for some users and should be avoided where possible  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_8" \o "Bravo-Lillo, 2011 #8" Bravo-Lillo et al., 2011a;  HYPERLINK \l "_ENREF_9" \o "Bravo-Lillo, 2011 #2833" Cristian Bravo-Lillo, Lorrie Faith Cranor, Julie S. Downs, & Saranga Komanduri, 2011b). The last three stages in the receiver portion of C-HIP are attitudes and beliefs, motivation, and behavior. Once a user pays attention to and comprehends a warning, their attitudes and beliefs can be changed, which is essential to affect their motivation to behave in a certain way. C-HIP postulates that in order for the receiver to change their behavior based on the communication from the source, they must be motivated by attitudes and beliefs. This part of the C-HIP process corresponds to several research models regarding the effects of attitudes and beliefs on ultimate behavior (e.g., Theory of Reasoned Action, Theory of Planned Behavior, Protection Motivation Theory). Motivation, or intentions, explains the connection between attitudes/beliefs and behavior in these models. To examine the effects of attitudes and beliefs on the success of warning messages, we utilize Protection Motivation Theory (PMT), which is based on the Health Beliefs Model (HBM). Although some information security research has been conducted using HBM  ADDIN EN.CITE  ADDIN EN.CITE.DATA (e.g.  HYPERLINK \l "_ENREF_27" \o "LaRose, 2005 #28" LaRose, Rifon, Liu, & Lee, 2005a,  HYPERLINK \l "_ENREF_28" \o "LaRose, 2005 #29" 2005b;  HYPERLINK \l "_ENREF_38" \o "Ng, 2009 #26" Ng, Kankanhalli, & Xu, 2009;  HYPERLINK \l "_ENREF_39" \o "Ng, 2007 #25" Ng & Xu, 2007;  HYPERLINK \l "_ENREF_56" \o "Woon, 2005 #30" Woon, Tan, & Low, 2005), recent literature has primarily examined PMT in the information security context. Several studies have evaluated PMT together with the theory of planned behavior  ADDIN EN.CITE Anderson201031(C. L. Anderson & Agarwal, 2010; Ifinedo, 2012)313117Anderson, C LAgarwal, RPracticing safe computing: A multimethod empirical examination of home computer user security behavioral intentionsMIS QuarterlyMIS Quarterly613-6433432010Ifinedo201211701170117017Ifinedo, PrincelyUnderstanding information systems security policy compliance: An integration of the theory of planned behavior and the protection motivation theoryComputers & SecurityComputers & Security83-953112012Feb0167-4048WOS:000301011700008<Go to ISI>://WOS:000301011700008http://www.sciencedirect.com/science/article/pii/S016740481100133710.1016/j.cose.2011.10.007( HYPERLINK \l "_ENREF_4" \o "Anderson, 2010 #31" C. L. Anderson & Agarwal, 2010;  HYPERLINK \l "_ENREF_22" \o "Ifinedo, 2012 #1170" Ifinedo, 2012), deterrence  ADDIN EN.CITE Herath200920(Herath & Rao, 2009)202017Herath, TRao, H RProtection motivation and deterrence: A framework for security policy compliance in organisationsEuropean Journal of Information SystemsEuropean Journal of Information Systems106-1251822009( HYPERLINK \l "_ENREF_21" \o "Herath, 2009 #20" Herath & Rao, 2009), habit  ADDIN EN.CITE Vance201221(Vance, Siponen, & Pahnila, 2012)212117Vance, AnthonySiponen, MPahnila, SMotivating IS security compliance: Insights from habit and protection motivation theoryInformation & ManagementInformation & Management190-198493-42012( HYPERLINK \l "_ENREF_49" \o "Vance, 2012 #21" Vance, Siponen, & Pahnila, 2012), bring your own device  ADDIN EN.CITE Loraas201422(Loraas, Crossler, Long, & Trinkle, 2014)222217Loraas, T MCrossler, R ELong, J HTrinkle, B SUnderstanding compliance with BYOD (bring your own device) policies utilizing protection motivation theory: Bridging the intention-behavior gapJournal of Information SystemsJournal of Information Systems209-2262812014( HYPERLINK \l "_ENREF_31" \o "Loraas, 2014 #22" Loraas, Crossler, Long, & Trinkle, 2014), fear appeals  ADDIN EN.CITE Johnston201028(A. C. Johnston & Warkentin, 2010; A C Johnston, Warkentin, & Siponen, 2015)282817Johnston, A CWarkentin, MFear appeals and information security behaviors: An empirical studyMIS QuarterlyMIS Quarterly549-5663432010Johnston201529292917Johnston, A CWarkentin, MSiponen, MAn enhanced fear appeal rhetorical framework: Leveraging threats to the human asset through sanctioning rhetoricMIS QuarterlyMIS Quarterly113-1343912015( HYPERLINK \l "_ENREF_24" \o "Johnston, 2010 #2" A. C. Johnston & Warkentin, 2010;  HYPERLINK \l "_ENREF_25" \o "Johnston, 2015 #29" A C Johnston, Warkentin, & Siponen, 2015), spyware/malware  ADDIN EN.CITE Gurung200826(Gurung, Luo, & Liao, 2008; Lee & Larsen, 2009)262617Gurung, ALuo, XLiao, QConsumer motivations in taking action against spyware: An empirical investigationInformation Management & Computer SecurityInformation Management & Computer Security276-2891732008Lee200927272717Lee, YLarsen, K RThreat or coping appraisal: Determinants of SMB executives' decision to adopt anti-malware softwareEuropean Journal of Information SystemsEuropean Journal of Information Systems177-1871822009( HYPERLINK \l "_ENREF_20" \o "Gurung, 2008 #26" Gurung, Luo, & Liao, 2008;  HYPERLINK \l "_ENREF_30" \o "Lee, 2009 #27" Lee & Larsen, 2009), and user interface  ADDIN EN.CITE Vanceforthcoming6(Vance, Lowry, & Egget, forthcoming)6617Vance, AnthonyLowry, Paul BenjaminEgget, DennisIncreasing accountability through user-interface design artifacts: A new approach to addressing the problem of access-policy violationsMIS QuarterlyMIS Quarterlyforthcoming( HYPERLINK \l "_ENREF_48" \o "Vance, forthcoming #6" Vance, Lowry, & Egget, forthcoming), among others, to determine its impact on information security behaviors. Despite this extensive amount of research, these theories have not been fully applied to the effects of digital warning messages. According to the Health Beliefs Model, to change users behavior, three conditions have to be met: (1) the individual must be personally susceptible to the health problem; (2) the individual should understand that risk can lead to serious harm; and (3) the individual must understand what actions can be taken to avoid harm and the costs or benefits of those actions  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_23" \o "Janz, 1984 #24" Janz & Becker, 1984;  HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975;  HYPERLINK \l "_ENREF_50" \o "Witte, 1992 #24" Witte, 1992,  HYPERLINK \l "_ENREF_51" \o "Witte, 1994 #25" 1994). PMT examines the same three conditions, respectively named perceived threat susceptibility, perceived threat severity, and perceived response efficacy  ADDIN EN.CITE Rogers197519(Rogers, 1975)191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975( HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975). In addition, PMT introduces a fourth condition, perceived self-efficacy, which is the perception that one can successfully enact a recommended response  ADDIN EN.CITE Maddux198323(Maddux & Rogers, 1983)232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983( HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983). Finally, each condition can be affected by the receivers personal characteristics (e.g., demographics) and by environmental factors. Such factors should always be considered in models of warning effectiveness (Wogalter, 2006). While some studies on digital warning messages have addressed some individual aspects of the C-HIP model, none have examined it in one holistic model. Little research exists that incorporates the C-HIP model and, where research does exist, it mostly explains human processing without directly applying, testing, or adapting the model to the warnings context  ADDIN EN.CITE  ADDIN EN.CITE.DATA (e.g.  HYPERLINK \l "_ENREF_8" \o "Bravo-Lillo, 2011 #8" Bravo-Lillo et al., 2011a;  HYPERLINK \l "_ENREF_11" \o "Chen, 2014 #39" Chen et al., 2014;  HYPERLINK \l "_ENREF_43" \o "Schmuntzsch, 2014 #31" Schmuntzsch, Sturm, & Roetting, 2014;  HYPERLINK \l "_ENREF_54" \o "Wogalter, 2002 #30" Wogalter & Conzola, 2002). For example, one study  ADDIN EN.CITE Egelman20082(Egelman et al., 2008)223Egelman, SergeCranor, Lorrie FaithHong, JasonYou've been warned: An empirical study of the effectiveness of web browser phishing warningsProceedings of the ACM CHI Conference on Human Factors in Computing Systems1-1020081595931783( HYPERLINK \l "_ENREF_14" \o "Egelman, 2008 #2" Egelman et al., 2008) cites the C-HIP model as relevant to computer warning messages, but does not empirically investigate the model regarding the effectiveness of web browser phishing warnings. Most research on computer-based IT security warning messages has focused on the channel  ADDIN EN.CITE  ADDIN EN.CITE.DATA (e.g.  HYPERLINK \l "_ENREF_7" \o "Bravo-Lillo, 2013 #9" Bravo-Lillo et al., 2013;  HYPERLINK \l "_ENREF_14" \o "Egelman, 2008 #2" Egelman et al., 2008;  HYPERLINK \l "_ENREF_15" \o "Egelman, 2013 #7894" Egelman & Schechter, 2013;  HYPERLINK \l "_ENREF_47" \o "Sunshine, 2009 #12" Sunshine, Egelman, Almuhimedi, Atri, & Cranor, 2009b) or the attention aspects of warnings  ADDIN EN.CITE Anderson201416e.g. (e.g. B. B. Anderson et al., 2014a; B. B. Anderson et al., 2014b)161647Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethWhy users habituate to security warnings: Insights from fMRI2014 IFIP 8.11 Dewald Roode Security Workshop2014Anderson201418181810Anderson, Bonnie BrintonVance, AnthonyKirwan, BrockEargle, DavidHoward, SethUsers aren't (necessarily) lazy: Using neuroIS to explain habituation to security warnings2014 International Conference on Information Systems (ICIS)2014(e.g.  HYPERLINK \l "_ENREF_2" \o "Anderson, 2014 #18" B. B. Anderson et al., 2014a;  HYPERLINK \l "_ENREF_3" \o "Anderson, 2014 #16" B. B. Anderson et al., 2014b). Few research studies have considered a holistic view of examining attention together with the warning message content and its influence on attitudes and beliefs. It is important to understand how the content of the message can affect the attitudes, beliefs, and motivation of users, but also consider these effects when measuring the user attention and comprehension of the warning message. This study seeks to fill this gap by manipulating the content of a warning message, using theory-based methods, and measuring its effects on user behavior where the user pays attention to the warning. Additionally, this study examines the impact of providing an alternative course of action to lower compliance costs. In other words, even if users have attitudes or beliefs that certain behaviors could be insecure, they may not have the motivation to behave securely if their attitudes and beliefs about the threat do not reflect a knowledge of alternative courses of action  ADDIN EN.CITE Wogalter200615(Wogalter, 2006a)15155Wogalter, Michael SWogalter, M SCommunication-Human Information Processing (C-HIP) ModelHandbook of Warnings51-612006Mahwah, NJLawrence Erlbaum Associates( HYPERLINK \l "_ENREF_52" \o "Wogalter, 2006 #15" Wogalter, 2006a). Hypothesis development To gain a more holistic understanding of the effect of digital warning messages, we propose a variance model that combines HBM and PMT with the C-HIP model to examine attention, comprehension, attitudes, beliefs, motivation, and behavior. This model is depicted below in Figure 2. We present the associated hypotheses starting with the dependent variable and then working back from there. Figure 2. C-HIP variance model and study hypotheses. Attitudes and Beliefs, Motivation, and Behavior C-HIP postulates that in order for the receiver to change their behavior based on the communication from the source, they must be motivated by attitudes and beliefs. This part of the C-HIP process corresponds to several research models regarding the effects of attitudes and beliefs on ultimate behavior (e.g., Theory of Reasoned Action, Theory of Planned Behavior, Protection Motivation Theory, etc.). Motivation, or intentions, explains the connection between attitudes/beliefs and behavior in these models. Based on this large body of literature, we theorize that, after receiving the warning message, a users attitudes and beliefs will motivate the users ultimate compliance or violation behavior. Specifically, to test the effects of attitudes and beliefs on user behavior, we use Protection Motivation Theory. This theory states that persuasive communication in the form of fear appeals increases motivation to protect oneself based on ones threat appraisal and coping appraisal  ADDIN EN.CITE Maddux198323(Maddux & Rogers, 1983; Rogers, 1975)232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983Rogers197519191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975( HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983;  HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975). The threat appraisal includes perceptions of threat severity and threat susceptibility. The coping appraisal includes perceptions of response efficacy, self-efficacy, and response cost. Witte (1992, 1994) states that threats exist regardless of whether or not they are perceived by individuals. PMT and its extensions have identified that fear appeals should be constructed to not only help individuals identify existing threats but also understand the seriousness of these threats and their probability of occurrence  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975;  HYPERLINK \l "_ENREF_50" \o "Witte, 1992 #24" Witte, 1992,  HYPERLINK \l "_ENREF_51" \o "Witte, 1994 #25" 1994). As individuals become more conscious of a threat, they will establish beliefs about its seriousness and its probability of occurrence  ADDIN EN.CITE Johnston201529(A C Johnston et al., 2015)292917Johnston, A CWarkentin, MSiponen, MAn enhanced fear appeal rhetorical framework: Leveraging threats to the human asset through sanctioning rhetoricMIS QuarterlyMIS Quarterly113-1343912015( HYPERLINK \l "_ENREF_25" \o "Johnston, 2015 #29" A C Johnston et al., 2015). Perceived threat severity refers to the users conviction regarding the seriousness of the problem and its associated consequences  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_38" \o "Ng, 2009 #26" Ng et al., 2009;  HYPERLINK \l "_ENREF_39" \o "Ng, 2007 #25" Ng & Xu, 2007;  HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975). In information security, the risk of any given threat is not only limited to damaging organizational assets (e.g. information, systems, operations), but could include other adverse outcomes such as negative impacts on job performance. Considering the impact these risks may have on organizational or individual performance, users perception of threat severity will vary; therefore, increasing understanding regarding the consequences of insecure behavior may motivate compliance behavior. For example, successful warning messages that increase perceptions regarding a given threat may affect individual attitudes to engage in harmful behavior. Perceived threat susceptibility is the users conviction regarding the probability that a threat will occur  ADDIN EN.CITE Rogers197519(Rogers, 1975)191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975( HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975). The level of a threats severity is known to impact vulnerability  ADDIN EN.CITE Milne200027(Milne, Sheeran, & Orbell, 2000)272717Milne, SarahSheeran, PaschalOrbell, SheinaPrediction and in���������������������������������������������������������������������������������t�e�r�v�e�n�t�i�o�n� �i�n� �h�e�a�l�t�h� r�e�l�a�t�e�d� �b�e�h�a�v�i�o�r�:� �a� �m�e�t�a� a�n�a�l�y�t�i�c� �r�e�v�i�e�w� �o�f� �p�r�o�t�e�c�t�i�o�n� �m�o�t�i�v�a�t�i�o�n� �t�h�e�o�r�y�<�/�t�i�t�l�e�>�<�s�e�c�o�n�d�a�r�y�-�t�i�t�l�e�>�J�o�u�r�n�a�l� �o�f� �A�p�p�l�i�e�d� �S�o�c�i�a�l� �P�s�y�c�h�o�l�o�g�y�<�/�s�e�c�o�n�d�a�r�y�-�t�i�t�l�e�>�<�/�t�i�t�l�e�s�>�<�p�e�r�i�o�d�i�c�a�l�>�<�f�u�l�l�-�t�i�t�l�e�>�J�o�u�r�n�a�l� �o�f� �A�p�p�l�i�e�d� �S�o�c�i�a�l� �P�s�y�c�h�o�l�o�g�y�<�/�f�u�l�l�-�t�i�t�l�e�>�<�/�periodical><pages>106-143</pages><volume>30</volume><number>1</number><dates><year>2000</year></dates><isbn>1559-1816</isbn><urls></urls></record></Cite></EndNote>( HYPERLINK \l "_ENREF_36" \o "Milne, 2000 #27" Milne, Sheeran, & Orbell, 2000).  ADDIN EN.CITE <EndNote><Cite AuthorYear="1"><Author>Rogers</Author><Year>1975</Year><RecNum>19</RecNum><DisplayText>Rogers (1975)</DisplayText><record><rec-number>19</rec-number><foreign-keys><key app="EN" db-id="eazavzpspzff9jeteeo50rpgzdddt9z0wp2z" timestamp="1432326489">19</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Rogers, R W</author></authors></contributors><titles><title>A protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975 HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers (1975) further states that as compared to low-fear appeals, high-fear appeals increase perceptions of threat susceptibility and facilitate attitude change. In this same study, a unique motivator to help smokers become aware of their personal vulnerability was to role-play with them. Similarly, warning messages could resemble role-playing by heightening individual awareness of personal vulnerabilities. This awareness would facilitate attitude change against engaging in harmful behavior. Therefore, warning messages should not only address the severity of threats but also its probability of occurrence. Base on the above rationale, we hypothesize that: H1a. Individual perceptions of threat severity, after experiencing a warning message, are positively associated with compliance behavior. H1b. Individual perceptions of threat vulnerability, after experiencing a warning message, are positively associated with compliance behavior. As part of the coping appraisal, effective fear appeals should cause individuals to form cognitions about efficacy, which include perceptions about the recommended response (i.e., response efficacy) and perceptions of efficacy of the individual performing the response, i.e., self-efficacy  ADDIN EN.CITE Witte199425(Witte, 1994)252517Witte, KFear control and danger control: A test of the extended parallel process model (EPPM)Communication MonographsCommunication Monographs113-134611994( HYPERLINK \l "_ENREF_51" \o "Witte, 1994 #25" Witte, 1994). Response efficacy is the degree to which an individual believes the response to be effective in alleviating a threat  ADDIN EN.CITE Johnston201028(A. C. Johnston & Warkentin, 2010; Rogers, 1975)282817Johnston, A CWarkentin, MFear appeals and information security behaviors: An empirical studyMIS QuarterlyMIS Quarterly549-5663432010Rogers197519191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975( HYPERLINK \l "_ENREF_24" \o "Johnston, 2010 #2" A. C. Johnston & Warkentin, 2010;  HYPERLINK \l "_ENREF_41" \o "Rogers, 1975 #19" Rogers, 1975). Self-efficacy pertains to the level of confidence in ones ability to undertake the recommended preventive behavior  ADDIN EN.CITE Herath200920(Herath & Rao, 2009; Maddux & Rogers, 1983)202017Herath, TRao, H RProtection motivation and deterrence: A framework for security policy compliance in organisationsEuropean Journal of Information SystemsEuropean Journal of Information Systems106-1251822009Maddux198323232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983( HYPERLINK \l "_ENREF_21" \o "Herath, 2009 #20" Herath & Rao, 2009;  HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983). Despite the actual response efficacy against a specific threat, users still cognitively evaluate the efficacy of the response and develop their own perceptions. Ultimately, this evaluation will determine whether the user follows the recommendation  ADDIN EN.CITE Maddux198323(Maddux & Rogers, 1983)232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983( HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983). If the response is considered moderately or highly effective then users are likely to enact a recommended response  ADDIN EN.CITE Maddux198323(Maddux & Rogers, 1983)232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983( HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983). Given that users may not know the recommended response to a particular threat, warning messages serve as an appropriate medium for users to identify the most effective response to overcome the threat. Even if users believe that a response is effective, they also consider whether or not they are capable of enacting the recommended response  ADDIN EN.CITE Maddux198323(Maddux & Rogers, 1983; Witte, 1992)232317Maddux, J ERogers, R WProtection motivation and self-efficacy: A revised theory of fear appeals and attitude changeJournal of Experimental Social PsychologyJournal of Experimental Social Psychology469-479191983Witte199224242417Witte, KPutting fear back into fear appeals: The extended parallel process modelCommunication MonographsCommunication Monographs329-3495941992( HYPERLINK \l "_ENREF_32" \o "Maddux, 1983 #23" Maddux & Rogers, 1983;  HYPERLINK \l "_ENREF_50" \o "Witte, 1992 #24" Witte, 1992). Events that spark high levels of emotional arousal (e.g., perceived security threats) are known to negatively impact self-efficacy (e.g., lower perceptions of ones ability to use a computer)  ADDIN EN.CITE Marakas199832(Marakas, Yi, & Johnson, 1998)323217Marakas, G MYi, M YJohnson, R DThe multilevel and multifaceted character of computer self-efficacy: Toward clarification of the construct and an integrative framework for researchInformation Systems ResearchInformation Systems Research126-163921998( HYPERLINK \l "_ENREF_34" \o "Marakas, 1998 #32" Marakas, Yi, & Johnson, 1998). Therefore, as users perceive threatening events (e.g., viruses, spyware) as severe and probable, they doubt their ability to function adequately within the heightened threat conditions  ADDIN EN.CITE  ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_24" \o "Johnston, 2010 #2" A. C. Johnston & Warkentin, 2010). Therefore, when users view warning messages which convey how to complete a recommended response, doubts may diminish about ones own ability and attitude may increase about enacting the recommended response. Thus, we posit that: H1c. Individual perceptions of self-efficacy, after experiencing a warning message, are positively associated with compliance behavior. H1d. Individual perceptions of response efficacy, after experiencing a warning message, are positively associated with compliance behavior. A third factor of the coping appraisal is response costs. Response costs are those that are incurred when coping with the threat  ADDIN EN.CITE Lee200927(Lee & Larsen, 2009)272717Lee, YLarsen, K RThreat or coping appraisal: Determinants of SMB executives' decision to adopt anti-malware softwareEuropean Journal of Information SystemsEuropean Journal of Information Systems177-1871822009( HYPERLINK \l "_ENREF_30" \o "Lee, 2009 #27" Lee & Larsen, 2009). Contrary to the other aspects of the coping appraisal, response costs preclude individuals from enacting a particular response. Given an extensive amount of time, effort, money, and other requirements to adopt the recommend response, individuals may decide to forego this effort  ADDIN EN.CITE Milne200027(Milne et al., 2000)272717Milne, SarahSheeran, Paschal</author><author>Orbell, Sheina</author></authors></contributors><titles><title>Prediction and intervention in health related behavior: a meta analytic review of protection motivation theory</title><secondary-title>Journal of Applied Social Psychology</secondary-title></titles><periodical><full-title>Journal of Applied Social Psychology</full-title></periodical><pages>106-143</pages><volume>30</volume><number>1</number><dates><year>2000</year></dates><isbn>1559-1816</isbn><urls></urls></record>( HYPERLINK \l "_ENREF_36" \o "Milne, 2000 #27" Milne et al., 2000). When evaluating response costs within the realm of information security, security practices are often considered a hindrance leading employees to neglect security practices  ADDIN EN.CITE Herath200920(Herath & Rao, 2009)202017Herath, TRao, H RProtection motivation and deterrence: A framework for security policy compliance in organisationsEuropean Journal of Information SystemsEuropean Journal of Information Systems106-1251822009( HYPERLINK \l "_ENREF_21" \o "Herath, 2009 #20" Herath & Rao, 2009). Therefore, as the cost to heed warning messages increases, compliance with these warning messages will decrease. We then hypothesize that: H1e. Individual perceptions of response cost, after experiencing a warning message, are negatively associated with compliance behavior. Channel Effects on Attitudes and Beliefs As individuals encounter warning messages, they will formulate attitudes and beliefs about the threat appraisal and coping appraisal, in other words, they will develop cognitions about the repercussions of a given threat and the likelihood of its occurrence together with how to combat this threat. For example, a successful warning message may identify one or more responses to a given threat with steps to enact the recommended response(s). As a result, perceptions related to response efficacy and self-efficacy will increase. Although PMT and the Health Beliefs Model explain some effects that impact attitudes and beliefs, this study expounds on current research by identifying whether warning message content will be strong enough to actually changes attitudes and beliefs. Even when users change their attitudes or beliefs, the C-HIP model shows that these attitudes or beliefs must lead to motivation in order to ultimately change behavior. In this step, it is essential to understand the cost of compliance (e.g. effort, time, money). Often, even if users are aware of the benefits or consequences of compliance/noncompliance, they may not be motivated to comply because they are not aware of alternative safe courses of action  ADDIN EN.CITE Wogalter200615(Wogalter, 2006a)15155Wogalter, Michael SWogalter, M SCommunication-Human Information Processing (C-HIP) ModelHandbook of Warnings51-612006Mahwah, NJLawrence Erlbaum Associates( HYPERLINK \l "_ENREF_52" \o "Wogalter, 2006 #15" Wogalter, 2006a). One challenge with existing warning designs is that they usually do not offer any alternative solutions that may influence users behavior. Most of the web based computer warnings offer two choices to the user (e.g., proceed or cancel), and, as such, present a limited binary decision making process. Only recently, web browsers (e.g., Google Chrome) started to incorporate alternative options to the user such as a help me understand link where the user is presented with a third option  ADDIN EN.CITE Sophos201522(Sophos, 2015)222212SophosGoogle redesigns security warnings after 70% of Chrome users ignore themMay 20152015https://nakedsecurity.sophos.com/2015/02/03/google-redesigns-security-warnings-after-70-of-chrome-users-ignore-them/( HYPERLINK \l "_ENREF_45" \o "Sophos, 2015 #22" Sophos, 2015). In their recent study  ADDIN EN.CITE Felt201531(Felt et al., 2015)313147Adrienne Porter FeltAlex AinslieRobert W. ReederSunny ConsolvoSomas ThyagarajaAlan BettesHelen HarrisJeff GrimesImproving SSL Warnings: Comprehension and AdherenceProceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems2893-29022015Seoul, Republic of KoreaACM270244210.1145/2702123.2702442( HYPERLINK \l "_ENREF_17" \o "Felt, 2015 #31" Felt et al., 2015) introduced opinionated design, defined by the use of visual design cues to promote a recommended course of action, which resulted in substantially increased adherence rates (nearly 30% more total users decided to change their course of action), positively impacting user safety and decision making. Therefore, successful warning messages should (1) increase cognizance about the severity of the threat, (2) identify the likelihood that threat will occur, (3) pinpoint effective responses to a given threat, (4) stimulate ones perception about their ability to complete the recommended response, and (5) diminish costs associated with the recommended response. One unique way to diminish perceived costs of compliance that we propose is that warning messages should include alternative courses of action to the user. That is, if users heed a digital warning telling them not to use a certain software, it could be more effective if the warning message also suggested a way to reduce the costs of compliance by presenting an alternative way to complete the work. By presenting alternative courses of action, users are better able to cope with the threat presented and will better understand the efficacy of compliance and perceive reduced costs of compliance. In summary, we hypothesize: H2a. Successful warning messages that identify the threat severity are positively associated with attitudes and beliefs about compliance. H2b. Successful warning messages that identify the threat susceptibility are positively associated with attitudes and beliefs about compliance. H2c. Successful warning messages that identify the effective response(s) to the threat are positively associated with attitudes and beliefs about compliance. H2d. Successful warning messages that stimulate self-efficacy are positively associated with attitudes and beliefs about compliance. H2e. Successful warning messages that reduce response costs through providing alternative courses of action are positively associated with attitudes and beliefs about compliance. Attention The main objective for many designers of warning messages is to capture users attention and convey information about the possible hazard  ADDIN EN.CITE Bravo-Lillo20139(Bravo-Lillo et al., 2013)9910Bravo-Lillo, CristianCranor, Lorrie FaithDowns, JulieKomanduri, SarangaReeder, Robert WSchechter, StuartSleeper, ManyaYour attention please: Designing security-decision UIs to make genuine risks harder to ignoreSOUPS '13 Proceedings of the Ninth Symposium on Usable Privacy and Security1-18attractorsdialogshabituationusable securitywarnings2013978145032319210.1145/2501604.2501610( HYPERLINK \l "_ENREF_7" \o "Bravo-Lillo, 2013 #9" Bravo-Lillo et al., 2013). The warning design is composed of several different aspects which may impact its attractiveness so it is more visible, and consequently, has higher impact on the user HYPERLINK \l "_ENREF_34" \o "Wogalter, 1996 #20" . Consequently, in this communication delivery process, if a users attention is switched to the warning message, we can expect to see increased compliance and better decision making. However, this is not a direct effect. In fact, in our model we do not theorize that paying attention to a warning message will have a direct effect on compliance. Rather, following the C-HIP model, we propose that attention is necessary in order for warning messages to affect attitudes and beliefs. Without this attention, the warning message will have no effect. But simply getting a user to pay attention to a warning is not enough in itself to affect ultimate behavior. Thus, we hypothesize the following interaction effect: H3. Warning messages positively influence attitudes and beliefs only when users pay careful attention to the warning. Comprehension Once a warning message captures a users attention, the next step is comprehension. One common mistake of warning designers is to assume that an average user will understand the hazard and its consequences and risks  ADDIN EN.CITE Wogalter199620(Wogalter & Laughery, 1996)202017Wogalter, Michael SLaughery, Kenneth RWarning! Sign and label effectivenessCurrent Directions in Psychological ScienceCurrent Directions in Psychological Science33-3719960963-7214( HYPERLINK \l "_ENREF_55" \o "Wogalter, 1996 #20" Wogalter & Laughery, 1996). Given that users range from novices to experts, they act and behave differently based on their level of warning comprehension. For example, due to technical complexity, novice users may not fully understand what is an SSL warning (a web browser warning that appears when a potentially unverified secure connection is about to be established with the remote server). When general users were presented with terms such as startup disk, encryption, virus, attachment, macro, and certificate, they indicated that they heard of them but had difficulties to make sense of them  ADDIN EN.CITE Bravo-Lillo20118(Bravo-Lillo et al., 2011a; Bravo-Lillo et al., 2011b)8817Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie SKomanduri, SarangaBridging the gap in computer security warningsIEEE Security and Privacy18-269March/April2011Bravo-Lillo201128332833283317Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie S.Komanduri, SarangaBridging the Gap in Computer Security Warnings: A Mental Model ApproachIeee Security & PrivacyBridging the Gap in Computer Security Warnings: A Mental Model ApproachIEEE Security & Privacy18-26922011Mar-Apr1540-7993WOS:000289158800004<Go to ISI>://WOS:00028915880000410.1109/msp.2010.198( HYPERLINK \l "_ENREF_8" \o "Bravo-Lillo, 2011 #8" Bravo-Lillo et al., 2011a;  HYPERLINK \l "_ENREF_9" \o "Bravo-Lillo, 2011 #2833" Bravo-Lillo et al., 2011b). Therefore, the content of warning messages should target the least-skilled users because messages that include complex technical terms are not likely to be comprehended  ADDIN EN.CITE Wogalter199620(Wogalter & Laughery, 1996)202017Wogalter, Michael SLaughery, Kenneth RWarning! Sign and label effectivenessCurrent Directions in Psychological ScienceCurrent Directions in Psychological Science33-3719960963-7214( HYPERLINK \l "_ENREF_55" \o "Wogalter, 1996 #20" Wogalter & Laughery, 1996). For example,  ADDIN EN.CITE Sunshine200912Sunshine et al. (2009b)121210Sunshine, JoshuaEgelman, SergeAlmuhimedi, HazimAtri, NehaCranor, Lorrie FaithCrying wolf: An empirical study of SSL warning effectiveness18th USENIX Security Symposium399-432200910.1016/S1353-4858(01)00916-3 HYPERLINK \l "_ENREF_47" \o "Sunshine, 2009 #12" Sunshine et al. (2009b) found that users who understood the web browser SSL warnings behaved very differently compared to those who did not understand them. Thus, we expect that users, regardless of their technical expertise, who fully comprehend warning messages will behave differently. Thus, following the C-HIP model, we hypothesize a second moderating effect: H4. Warning messages positively influence user attitudes and beliefs only when users fully comprehend the meaning of the warning. Methods General research design To test the hypotheses, we design a series of field and lab experiments where users either download an application from the Web or use a Web browser-based application. The applications used in the studies were created by one of the authors using VB.net. After running the application, users experience a variation of a warning message created for this study. The warning message briefly describes the consequences of using the potentially insecure software and asks the user if they would like to Continue to use the software or Exit. After the user presses continue, the message disappears and they can use the software. If the user presses exit, the application is closed. In each study, three measures are collected when the user opens the application for the first time. First, the users decision is recorded as 0 if he or she clicks on Exit or 1 if he or she clicks on Continue. Second, we record the duration (in ms) from the start of the application until the decision. Finally, we record a unique PC identifier (IP address and/or MAC address) of the user so that we can measure the behavior of the users only the first time they encounter the warning, and potentially measure any interesting behavior that occurs if/when users subsequently open the application another time. We vary the possible warning messages that users can see and test their individual effects on the users decision and time taken to make the decision. Pilot Study Design Pilot data was collected from March to May, 2015. This pilot data was used only to test our procedures, but not our theory (which continues to be refined). The pilot study was also done to analyze general differences between behavior when users see a warning message vs. when they do not. In addition to this pilot data, we also plan to conduct an expert panel review and further validity testing of our instrument before completing a full-scale data collection. In the pilot data, we loosely used the Health Beliefs Model and PMT to guide the design of various types of warning messages. While many organizational programs only focus on negative consequences of engaging in insecure behavior (e.g., deterrence theory, protection motivation theory), they neglect the benefits and costs of NOT performing the insecure behavior. Based on the Health Beliefs Model, users of this study are presented with warning messages that focus on the consequences (both positive and negative) of doing the right thing (e.g. benefits and drawbacks of discontinuing unauthorized software use) rather than the consequences of doing the wrong thing (e.g. sanctions for continuing to use the unauthorized software). As this is a pilot study, the full-scale study will contain different treatments and wording. Participants The participants in the pilot data collection are general Internet users who download the PDF application after finding it on SourceForge. Participants are not directly recruited for the study, nor are they aware of its purpose, which improves the validity of results. Procedures In the pilot study, users downloaded an application from the Web designed to manipulate PDF documents (e.g., split, merge, etc.). The application was created by one of the authors using VB.net and placed on SourceForge. When the user opens the application, either a warning message or a control message popped up. The display of the warning or the control message is controlled by the random function within the application; however, the function was set to favor warning messages over control messages because there were many variations of the warning message but only one type of control message. The particular warning message that any given user sees is also controlled by the random function within the software. The warning message briefly describes the consequences of using unauthorized software and asks the user if they would like to Continue to use the software or Exit. The non-warning message (control group) simply thanks the user for using the application and proposes two options (Continue or Exit), which lead to the same outcomes. The control treatment was used in this pilot study simply to compare the effect of using any warning message against the behavior of users when they do not see a warning message. The control treatment is only for the pilot study to analyze general differences in behavior between seeing a warning message or not. Such a control treatment likely will not be needed in future data collection (instead, we will simply compare different versions of warning messages). Warning Variations In this study, there are 16 (2 x 2 x 2 x 2) variations of the warning message: high/low severity x high/low certainty x benefit listed/not listed x barrier listed/not listed. The warning message starts by describing the threat from using unauthorized software, describing the relative certainty and severity of the threat occurring. High certainty: There is a large chance that using unauthorized software on your machine could allow malware to infect your computer Low certainty: There is a small chance that using unauthorized software on your machine could allow malware to infect your computer High severity: which could result in a large loss of private information, intruder attacks, or other serious effects. Low severity: which may be a nuisance to address. Along with asking the user if they would like to continue using the software, the warning message also includes some combination of listing (or not listing) the benefit and/or barrier of exiting the unauthorized software in order to comply with good practice. Benefit: If you choose to exit now, you will protect your computer and information from harm. Barrier: If you choose to exit now, you will need to find a different, authorized software to perform your work. An example warning message (high certainty / high severity / benefit listed / barrier not listed) is shown in Figure 3 below. Figure 3. Example warning message from pilot study Pilot Study Results Data Validation Because the pilot study was collected as we developed our procedures, there is a range of quality in the data collected. When data collection began, we did not account for users attempting to open the software multiple times. Thus, part way through the collection we started collecting the IP addresses of users. Although the data suggests that most IP addresses are from the same user (since the same IP address accessed the software within minutes), IP addresses are limited in that multiple different users could be identified with the same IP address (in the case of proxy addresses). Thus, we report the results of (1) the full data set with all responses (including those with no IP address recorded and multiple attempts by the same user; n=365, including 253 receiving a warning message) and (2) the results of only the first attempt of a recorded IP address (n=186, including 90 receiving a warning message). Analysis with Continue/Exit Decision as Dependent Variable A summary of responses for users who saw a warning message is shown in Tables 1 and 2 below. Most users chose to continue using the software after seeing the warning message, although there were some differences between warning types. Table 1. Summary Data for Full Data SetFactorn# pressed Continue# pressed Exit% pressed ContinueSeverity-High125883770.4%Severity-Low1281072183.6%Certainty-High123962778.0%Certainty-Low130993176.2%Benefit-Present1361043276.5%Benefit-Absent117912677.8%Barrier-Present129923771.3%Barrier-Absent1241032183.1%Total Warning Treatment2531955877.1%Non-warning1121001289.3% Table 2. Summary Data for First Time Viewing a Warning Message from Unique IP AddressesFactorn# pressed Continue# pressed Exit% pressed ContinueSeverity-High45351077.8%Severity-Low4536980.0%Certainty-High41311075.6%Certainty-Low4940981.6%Benefit-Present52391375.0%Benefit-Absent3832684.2%Barrier-Present45331273.3%Barrier-Absent4538784.4%Total Warning Treatment90711978.9%Non-warning96851188.5% In the full data set, it appears that the severity statement and the barrier statement are more powerful than the other factors. Contrary to expectations, users were more likely to heed the warning and exit the software when a barrier to compliance was listed. We would have expected that listing a barrier to comply would make users less likely to heed the warning, but the opposite occurred. This is confirmed with the multiple logistic regression analysis summarized in Table 3 below. In this analysis, the severity statement had a statistically significant effect (p=0.025), and the barrier statement had a strong effect as well (p=0.051). Table 3. Logistic Regression on Full Data SetBS.E.WaldSig.Exp(B)Severity-0.702.3135.019.0250.496Certainty0.020.3070.004.9481.020Benefit-0.013.3080.002.9660.987Barrier-0.611.3133.796.0510.543Constant1.932.37326.863.0006.907 In the partial data set, there are not strong effects of any of the factors, although the sample size is quite small. The multiple logistic regression on this data set is summarized in Table 4 below. The treatments (particularly pertaining to the certainty variable, which only had one word difference between the high and low conditions) may not have been strong enough in this pilot study. The wording and treatments were not validated and refined for the pilot collection. Table 4. Logistic Regression on Filtered Data SetBS.E.WaldSig.Exp(B)Severity-0.064.5300.015.9040.938Certainty-0.356.5320.448.5030.700Benefit-0.503.5610.805.3700.605Barrier-0.683.5391.607.2050.505Constant2.207.64211.821.0019.089 Analysis with Time as Dependent Variable Next, we examined the amount of time users take to make a decision of whether or not to heed the warning message, and how this is affected by the warning message and its contents. Because we cannot simply compare the average duration, due to the right skewed distribution, we use an event history analysis technique, also known as a survival technique. In these techniques, the time between the display of the warning message (or the control message) until the users choice (Continue or Exit) is referred to as the hazard event, which has a beginning and an end. We timed this event by calculating the difference between the start and the end of the event, which provided us the total duration of the event. Specifically, the survival technique we used is the Kaplan-Meier Survival estimator  ADDIN EN.CITE Kaplan19587712(Kaplan & Meier, 1958)7712771217Kaplan, Edward LMeier, PaulNonparametric estimation from incomplete observationsJournal of the American statistical associationJournal of the American statistical association457-4815328219580162-1459( HYPERLINK \l "_ENREF_26" \o "Kaplan, 1958 #7712" Kaplan & Meier, 1958), which enables dealing with differing survival times (i.e., times-to-event), especially when not all the subjects continue in the study  ADDIN EN.CITE Rich20107713(Rich et al., 2010)7713771317Rich, Jason TNeely, J GailPaniello, Randal CVoelker, Courtney CJNussenbaum, BrianWang, Eric WA practical guide to understanding Kaplan-Meier curvesOtolaryngology-Head and Neck SurgeryOtolaryngology-Head and Neck Surgery331-336143320100194-5998( HYPERLINK \l "_ENREF_40" \o "Rich, 2010 #7713" Rich et al., 2010). The Kaplan-Meier survival estimator is largely used in medical research to measure patients and the amount of time they live after receiving a specific treatment. It is also widely used in many other contexts such as determining the length of time people remain unemployed after losing their jobs. In our context, we are using the Kaplan-Meier survival estimator to understand behavior throughout the duration after being exposed to warning or non-warning messages. In Figure 4 we plot the cumulative survival functions for the control treatment vs. the warning treatment (which includes all different subgroups). The cumulative survival function represents the cumulative survival proportion against time for each treatment. The lengths of the horizontal lines along the X-axis represents the survival duration for that interval. The interval is terminated by the occurrence of the event of interest (Continue =1 or Exit = 0). In other words, the x-axis denotes time (in our case expressed in milliseconds). The y-axis denotes the percentage of subjects who have survived. Also, each drop in the curve represents an event (i.e. one user entry point). In simpler words, the Kaplan-Meier survival estimator enables us to understand if warning messages have any effect on the time taken during the users decision-making process; hence, if a warning is presented to the user, we would expect the user to spend more time reflecting on the action to take. More precisely, if the warning message has any effect on the users behavior, then we should see longer durations of the event where the user spends more time heeding the warning message and trying to making sense of it. We can clearly see that the cumulative survival proportion of the warning group is higher compared to the control group. In other words, users who received the warning treatment have a higher survival rate than those that received the non-warning treatment (i.e., control group). Figure 4. Kaplan-Meier survival function A log rank test was run to determine if there were differences in the survival distribution for the different types of interventions (warning vs no-warning). The survival distributions for the two interventions were statistically significantly different (see Table 4), 2(2) = 14.974, p < .0005. Table 4. Test of equality of survival distributions for the different levels of warning type.Chi-Squarep valueLog Rank (Mantel-Cox)14.974.000Breslow (Generalized Wilcoxon).825.364Tarone-Ware5.581.018 Finally, to understand if the effect of a warning on the duration of the event is significant (i.e. whether warning message really impacts the users behavior) we use Cox proportional-hazard regression  ADDIN EN.CITE Box-Steffensmeier20047714(Box-Steffensmeier & Jones, 2004)771477146Box-Steffensmeier, Janet MJones, Bradford SEvent history modeling: A guide for social scientists2004Cambridge University Press0521546737( HYPERLINK \l "_ENREF_6" \o "Box-Steffensmeier, 2004 #7714" Box-Steffensmeier & Jones, 2004). The results from the estimate Cox model are presented in Table 5. The hazard ratio estimate of the warning type (Exp(B)=1.650, Coefficient 0.131 with p<0.0005) indicates that in the presence of a warning message, users are 1.65 times more likely to exit the software. Table 5. Cox regression model over warning typeBSEWald Sig.Exp(B)95.0% CI for Exp(B)LowerUpperWarning vs Control.501.13114.703<.0011.6501.2772.131 In Figure 5, we present the four different warning types and their corresponding survival distribution times. The cumulative survival proportion of the High Severity (HS) has higher survival rate compared to the Low Severity (LS). The same result can be seen with High Barrier (HBA) vs. Low Barrier (LBA). However, High Certainty (HC) vs. Low Certainty (LC) and High Benefit (HBE) vs. Low Benefit (LBE) have very similar cumulative survival proportions which indicates that the warning message (high vs. low) does not have any impact on the survival rate. However, the Log rank test (Table 6) showed that only the survival distribution of HBA vs. LBA is statistically significantly different, 2(2) = .002, p < .05. For HS vs. LS the Log rank test showed 2(2) = 1.516, p > .05 which means that we do not have a statistically significant result between high severity compared to low severity. This could mean that users could not really differentiate the high vs low severity warning message in terms of the risk behind them. Table 6. Overall comparisons for Severity, Certainty, Benefit and BarrierHS vs LSHC vs LCHBE vs LBEHBA vs LBAChi-SquareSig.Chi-SquareSig.Chi-SquareSig.Chi-SquareSig.Log Rank (Mantel-Cox)1.516.218.001.9771.214.2705.002.025Breslow (Generalized Wilcoxon).379.538.005.9414.479.0342.177.140Tarone-Ware.914.339.013.9083.323.0733.458.063  a) High vs Low Severity  b) High vs Low Certainty  c) High vs Low Benefit  d) High vs Low Barrier Figure 5. Kaplan-Meier survival function for Severity, Certainty, Benefit, and Barrier Overall, we can see that in the presence of a warning message, the duration of the decision-making process is affected. Further, when comparing the four main factors manipulated in the warning messages (severity, certainty, benefit of compliance, and barrier to compliance) and their effectiveness on users decision making, we see that only the manipulation of the presentation of severity in the warning message has a strong impact compared to three other warning types. Full-Scale Data Collection (Proposed) Based on feedback from information security experts in academia and industry on our pilot results, we will implement a full scale data collection that, unlike the pilot study, will also include survey measures to more fully assess the complete theoretical model. Participants We have three potential different samples for the full scale study: (1) students recruited from multiple universities with an extra credit or course credit incentive to participate, (2) Internet users recruited using Mechanical Turk, and/or (3) employees recruited from a large international company. We have gained access to all three potential subject pools, and we hope to collect from all of them. Collecting from multiple sources will strengthen our study in multiple ways. First, the sample size in some of the populations may be small (e.g., in the industry company, we may only be able to access 70-90 responses) so multiple sources will increase the sample size. Second, testing the hypotheses in multiple contexts with multiple types of participants will increase the generalizability of the results. Each of the potential subject pools has particular strengths. We will report the results of the three field/lab studies separately. Procedures Because the participants are aware they are taking part in a research study, different procedures must be used in order to disguise the hypotheses and solicit honest behavior. Participants are first instructed that the study will require them to complete some work that involves either (1) editing PDF documents (similar to the application used in the pilot study) or (2) completing an online task on Mechanical Turk. In the first option, the editing will include merging, splitting, and/or extracting pages from PDFs. Participants are then instructed that in order to complete these requirements, they will need to download special software in order to complete the task. Participants are then given a list of links to software applications that could be downloaded in order to complete the work. Unbeknownst to the participants, the list of links actually all lead to the same software, and they are only able to click on one link. Once the participants download the software and open it, one of several warning messages will appear. Regarding the online task option, participants from Mechanical Turk will use a Web application to rate their preference for a particular product photo. This alternative web application is necessary for collecting data in Mechanical Turk because these users will not be able to download the PDF software. However, the procedures will be similar. The participants will be instructed that the study will require them to rate product photos. When the user clicks Start to open the Web application, a variation of a warning message will appear, warning the user about potential dangers of using the Web application. Using both the downloaded PDF application and the online Web application also strengthens our proposed study in multiple ways. First, using different variations of software allows us to access multiple types of subject pools. Second, our results become more generalizable (e.g., they will not be limited to warning messages on software alone or warning messages in Web applications alone). After both scenarios, participants are directed to a survey that informs them the warning message was part of the study and asks several questions about variables of interest in the study. There could also be a pre-survey to ask some of the survey questions before users are directed to the task that results in a warning message. Treatments When the user opens the application or accesses the website to complete the task, a warning message will pop up. Several variations of the warning messages are used as the treatment in this study. The particular warning message that any given user sees is random determined by the random function within the software. Validation of instruments The procedures and treatments of this study will be refined based on (1) discussion at the IFIP IT Security workshop and (2) results of an expert panel to be completed after the workshop. Discussion The present paper examines how theory-based communication, through the use of warning messages, influences information security policy compliance. Specifically, based on the C-HIP model and supported by Health Belief Model and Protection Motivation Theory, we suggest a new content element of computer security warning messages (i.e., suggesting alternative secure courses of action) that users who pay attention to warning message content may be more persuaded to behave securely. Through the preliminary pilot study, we observe some interesting findings. The pilot study results suggest that both the severity and barrier statements are more powerful than the certainty and benefit statements. This is an interesting finding which indicates that users are more likely to heed the warning message and exit the software when a barrier to compliance statement is present. We expected the benefit statement to have higher importance, than indicated by our preliminary results, granted that users who feel their data is susceptible to compromise would engage in behaviors to properly safeguard their data. We believe this issue may be due to the wording that was included in the warning message. In subsequent studies, the warning message will be refined and undergo expert panel reviews to better determine the impact of barrier statements on information security policy compliance. Additionally, the presence of a warning message does not lead to an immediate incident termination. Over 77% of users clicked on Continue and used the application despite the warning message. This is consistent with previous studies showing that users often ignore warning messages  ADDIN EN.CITE Maimon20143(Maimon, Alper, Sobesto, & Cukier, 2014)3317Maimon, DavidAlper, MarielSobesto, BertrandCukier, MichelRestrictive deterrent effects of a warning banner in an attacked computer systemCriminology33-5952CybercrimeDeterrenceExperimentsHoneypotsRestrictive deterrence2014( HYPERLINK \l "_ENREF_33" \o "Maimon, 2014 #3" Maimon, Alper, Sobesto, & Cukier, 2014). However, we can clearly see that in the presence of a warning message, the duration of the decision-making process is affected. As expected, users put more time in trying to make sense of the warning message and better understand how warning content may impact them. Interpretation of this could be that users spend more time in thinking, reading and reflecting about the content of the warning message and the consequences their acts may have if they click on continue vs the exit action. That is, even though a majority chose to disregard the potential consequences presented by the warning message, it may not be only because of a lack of paying attention. While some users were inattentive to the message, many stopped to carefully consider the warning before ultimately deciding not to heed its contents. This suggests that warning message may activate an automatic cautious behavior  ADDIN EN.CITE Bargh199637(Bargh, Chen, & Burrows, 1996)373717Bargh, John AChen, MarkBurrows, LaraAutomaticity of social behavior: Direct effects of trait construct and stereotype activation on actionJournal of personality and social psychologyJournal of personality and social psychology23071219961939-1315( HYPERLINK \l "_ENREF_5" \o "Bargh, 1996 #37" Bargh, Chen, & Burrows, 1996). This further highlights the need for studies such as ours that examine both attention as well as attitudes and beliefs, because users could ignore the warning message for either of these reasons. Further refinement and additional lab and field experiments will be conducted to improve warning message content and reduce issues such as technical jargon, which is quite problematic for some users  ADDIN EN.CITE Modic20147890(Modic & Anderson, 2014)7890789017Modic, DavidAnderson, RossReading this may harm your computer: The psychology of malware warningsComputers in Human BehaviorComputers in Human Behavior71-794120140747-5632( HYPERLINK \l "_ENREF_37" \o "Modic, 2014 #7890" Modic & Anderson, 2014). Specifically, theory-driven communication will be presented to users that identifies risks to which they may be exposed. Given the vast difference in the mental approach between novice and advanced users  ADDIN EN.CITE Bravo-Lillo20118(Bravo-Lillo et al., 2011a)8817Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie SKomanduri, SarangaBridging the gap in computer security warningsIEEE Security and Privacy18-269March/April2011( HYPERLINK \l "_ENREF_8" \o "Bravo-Lillo, 2011 #8" Bravo-Lillo et al., 2011a), such theory-driven communication will be useful in distinguishing between all users. Moreover, a challenge users frequently experience is the binary decision of continuing or exiting based on a warning message rather than alternative actions. Only recently,  ADDIN EN.CITE Felt201531Felt et al. (2015)313147Adrienne Porter FeltAlex AinslieRobert W. ReederSunny ConsolvoSomas ThyagarajaAlan BettesHelen HarrisJeff GrimesImproving SSL Warnings: Comprehension and AdherenceProceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems2893-29022015Seoul, Republic of KoreaACM270244210.1145/2702123.2702442 HYPERLINK \l "_ENREF_17" \o "Felt, 2015 #31" Felt et al. (2015) proposed opinionated design, which was implemented in Google Chrome. Though this is an additional step where the user is invited to think again before deciding, it is still limited due to the lack of a relevant alternative that would be useful and increase compliance. Overall, further studies will increase understanding of how user decisions are impacted by the different stages of the C-HIP model. This model identifies each layer to increase information security policy compliance. References ADDIN EN.REFLIST  Akhawe, D., & Felt, A. P. (2013). Alice in Warningland: A large-scale field study of browser security warning effectiveness. Proceedings of the 22nd USENIX Conference on Security: USENIX Association. Anderson, B. B., Vance, A., Kirwan, B., Eargle, D., & Howard, S. (2014a). Users aren't (necessarily) lazy: Using neuroIS to explain habituation to security warnings. Paper presented at the 2014 International Conference on Information Systems (ICIS). Anderson, B. B., Vance, A., Kirwan, B., Eargle, D., & Howard, S. (2014b). Why users habituate to security warnings: Insights from fMRI. Paper presented at the 2014 IFIP 8.11 Dewald Roode Security Workshop. Anderson, C. L., & Agarwal, R. (2010). Practicing safe computing: A multimethod empirical examination of home computer user security behavioral intentions. MIS Quarterly, 34(3), 613-643. Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of personality and social psychology, 71(2), 230. Box-Steffensmeier, J. M., & Jones, B. S. (2004). Event history modeling: A guide for social scientists: Cambridge University Press. Bravo-Lillo, C., Cranor, L. F., Downs, J., Komanduri, S., Reeder, R. W., Schechter, S., & Sleeper, M. (2013). Your attention please: Designing security-decision UIs to make genuine risks harder to ignore. Paper presented at the SOUPS '13 Proceedings of the Ninth Symposium on Usable Privacy and Security. Bravo-Lillo, C., Cranor, L. F., Downs, J. S., & Komanduri, S. (2011a). Bridging the gap in computer security warnings. IEEE Security and Privacy, 9(March/April), 18-26. Bravo-Lillo, C., Cranor, L. F., Downs, J. S., & Komanduri, S. (2011b). Bridging the Gap in Computer Security Warnings: A Mental Model Approach. Ieee Security & Privacy, 9(2), 18-26. doi: 10.1109/msp.2010.198 Bushman, B. J. (2006). Effects of warning and information labels on attraction to television violence in viewers of different ages. Journal of Applied Social Psychology, 36(9), 2073-2078. Chen, T.-C., Stepan, T., Dick, S., & Miller, J. (2014). An anti-phishing system employing diffused information. ACM Transactions on Information and System Security (TISSEC), 16(4), 16. Coleman, S. (2007). The Minnesota income tax compliance experiment: replication of the social norms experiment. Available at SSRN 1393292. Conzola, V. C., & Wogalter, M. S. (2001). A Communication-Human Information Processing (C-HIP) approach to warning effectiveness in the workplace. Journal of Risk Research, 4(4), 309-322. Egelman, S., Cranor, L. F., & Hong, J. (2008). You've been warned: An empirical study of the effectiveness of web browser phishing warnings. Proceedings of the ACM CHI Conference on Human Factors in Computing Systems. Egelman, S., & Schechter, S. (2013). The importance of being earnest [in security warnings] Financial cryptography and data security (pp. 52-59): Springer. Egilman, D., & Bohme, S. (2006). A brief history of warnings. Handbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJ, 35-48. Felt, A. P., Ainslie, A., Reeder, R. W., Consolvo, S., Thyagaraja, S., Bettes, A., . . . Grimes, J. (2015). Improving SSL Warnings: Comprehension and Adherence. Paper presented at the Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, Republic of Korea. Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A room with a viewpoint: Using social norms to motivate environmental conservation in hotels. Journal of consumer Research, 35(3), 472-482. Grier, C., Tang, S., & King, S. T. (2008). Secure web browsing with the OP web browser. Paper presented at the Security and Privacy, 2008. SP 2008. IEEE Symposium on. Gurung, A., Luo, X., & Liao, Q. (2008). Consumer motivations in taking action against spyware: An empirical investigation. Information Management & Computer Security, 17(3), 276-289. Herath, T., & Rao, H. R. (2009). Protection motivation and deterrence: A framework for security policy compliance in organisations. European Journal of Information Systems, 18(2), 106-125. Ifinedo, P. (2012). Understanding information systems security policy compliance: An integration of the theory of planned behavior and the protection motivation theory. Computers & Security, 31(1), 83-95. doi: 10.1016/j.cose.2011.10.007 Janz, N. K., & Becker, M. H. (1984). The health belief model: A decade later. Health Education & Behavior, 11(1), 1-47. Johnston, A. C., & Warkentin, M. (2010). Fear Appeals and Information Security Behaviors: An Empirical Study. Mis Quarterly, 34(3), 549-566. Johnston, A. C., Warkentin, M., & Siponen, M. (2015). An enhanced fear appeal rhetorical framework: Leveraging threats to the human asset through sanctioning rhetoric. MIS Quarterly, 39(1), 113-134. Kaplan, E. L., & Meier, P. (1958). Nonparametric estimation from incomplete observations. Journal of the American statistical association, 53(282), 457-481. LaRose, R., Rifon, N., Liu, S., & Lee, D. (2005a). Online safety strategies: a content analysis and theoretical assessment. Paper presented at the The 55th Annual Conference of the International Communication Association, New York City. LaRose, R., Rifon, N., Liu, S., & Lee, D. (2005b). Understanding online safety behavior: A multivariate model. Paper presented at the The 55th Annual Conference of the International Communication Association, New York City. Laughery, K. R., & Wogalter, M. S. (2006). Designing effective warnings. Reviews of human factors and ergonomics, 2(1), 241-271. Lee, Y., & Larsen, K. R. (2009). Threat or coping appraisal: Determinants of SMB executives' decision to adopt anti-malware software. European Journal of Information Systems, 18(2), 177-187. Loraas, T. M., Crossler, R. E., Long, J. H., & Trinkle, B. S. (2014). Understanding compliance with BYOD (bring your own device) policies utilizing protection motivation theory: Bridging the intention-behavior gap. Journal of Information Systems, 28(1), 209-226. Maddux, J. E., & Rogers, R. W. (1983). Protection motivation and self-efficacy: A revised theory of fear appeals and attitude change. Journal of Experimental Social Psychology, 19, 469-479. Maimon, D., Alper, M., Sobesto, B., & Cukier, M. (2014). Restrictive deterrent effects of a warning banner in an attacked computer system. Criminology, 52, 33-59. Marakas, G. M., Yi, M. Y., & Johnson, R. D. (1998). The multilevel and multifaceted character of computer self-efficacy: Toward clarification of the construct and an integrative framework for research. Information Systems Research, 9(2), 126-163. McAffee. (2014). Net Losses: Estimating the Global Cost of Cybercrime. Milne, S., Sheeran, P., & Orbell, S. (2000). Prediction and intervention in health related behavior: a meta analytic review of protection motivation theory. Journal of Applied Social Psychology, 30(1), 106-143. Modic, D., & Anderson, R. (2014). Reading this may harm your computer: The psychology of malware warnings. Computers in Human Behavior, 41, 71-79. Ng, B.-Y., Kankanhalli, A., & Xu, Y. C. (2009). Studying users' computer security behavior: A health belief perspective. Decision Support Systems, 46(4), 815-825. Ng, B.-Y., & Xu, Y. (2007). Studying users' computer security behavior using the Health Belief Model. PACIS 2007 Proceedings, 45. Rich, J. T., Neely, J. G., Paniello, R. C., Voelker, C. C., Nussenbaum, B., & Wang, E. W. (2010). A practical guide to understanding Kaplan-Meier curves. Otolaryngology-Head and Neck Surgery, 143(3), 331-336. Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. The Journal of Psychology, 91, 93-114. Sanders, M. S., & McCormick, E. J. (1987). Human factors in engineering and design: McGRAW-HILL book company. Schmuntzsch, U., Sturm, C., & Roetting, M. (2014). The warning gloveDevelopment and evaluation of a multimodal action-specific warning prototype. Applied ergonomics, 45(5), 1297-1305. Schultz, P., & Tabanico, J. J. (2009). Criminal beware: a social norms perspective on posting public warning signs*. Criminology, 47(4), 1201-1222. Sophos. (2015). Google redesigns security warnings after 70% of Chrome users ignore them Retrieved May 2015, from https://nakedsecurity.sophos.com/2015/02/03/google-redesigns-security-warnings-after-70-of-chrome-users-ignore-them/ Sunshine, J., Egelman, S., Almuhimedi, H., Atri, N., & Cranor, L. F. (2009a). Crying Wolf: An Empirical Study of SSL Warning Effectiveness. Paper presented at the USENIX Security Symposium. Sunshine, J., Egelman, S., Almuhimedi, H., Atri, N., & Cranor, L. F. (2009b). Crying wolf: An empirical study of SSL warning effectiveness. Paper presented at the 18th USENIX Security Symposium. _`almo|}űmXF0+h h CJH*OJQJ^JaJmH sH "h CJOJQJ^JaJmH sH (h6=h6=CJOJQJ^JaJmH sH #h h H*OJQJ^JmH sH  h 5H*OJQJ^JmH sH #h h 5OJQJ^JmH sH h 5OJQJ^JmH sH &h h 5H*OJQJ^JmH sH #h5h6=5OJQJ^JmH sH hh6=6hlhi5CJ OJQJ^Jh h 5CJ OJQJ^J2`a #$%&'(5gdigdigdigd/z gdi$a$gdi$a$gd $a$gd6=$a$gd6=$a$gd   "#$%56Ydfuҿ೫yogc\c\c\cUcUcU h@h hD]h h h/z h/z 6h/z h/z 6CJh h 6CJhimH sH h1himH sH hih`j hi6hx$h 6hhi6 hl6 hyXh %h CJH*OJQJ^JaJmH sH h CJOJQJ^JaJ h h CJOJQJ^JaJh h CJOJQJaJ!RS123@ABC'()RSUV j h UmHnHu j h UmHnHujh U juh UmHnHu jh UmHnHujh UmHnHuh mHnHujh Uh 25$%rFVX8Z[[[6cff<$Ifgd5&l gd gd $a$gd l{|z{|}""^%_%`%a%%%%%%%%%%)))L)M)N)e)f)g)h) **_0`0f0g00شأؒ j] h UmHnHu j h UmHnHu js h UmHnHu h{h h jh Uh mHnHujh UmHnHu j h UmHnHu700000000001111114444#5$5%5y5z5{5|5:6;6c9d9e9f99999999;;LAMASATAAAAAAAAAؿتؿؙؿ؈ jh UmHnHu j-h UmHnHu j h UmHnHuh jh U jE h UmHnHuh mHnHujh UmHnHu j h UmHnHu5AAAAAAAAAWEXEYEEEEEEEEFFLLLLMMMQMRMTMUMMMMMMMMNNCTDTETFTwTxTyTTTTTTػتؙ؈ jh UmHnHu jrh UmHnHu jh UmHnHu jh UmHnHuh jh Uh mHnHujh UmHnHu jh UmHnHu4TTTTTTT8Z[[[[9\:\bbbbbbbbbbbcc cccccpdvdfffffFjGjHjyjĽ̶إؔ̏̄}rrjhx$h U hx$h jh^h U h 6 j@h UmHnHu jh UmHnHu h!h hah hm-h 6h jh Uh mHnHujh UmHnHu jXh UmHnHu*ffnw<$Ifgd5&l tkdԪ$$IfTl X t0644 lap yt5&Tyjzj{jjjjjjjmmmnn nnnnnnn%nAoBorrrrsss@sAsBsCsSsTs3x4x5xhxixjxxxxxtyuyvyz|{|||}||콯쯽ֽ콯 jGh UmHnHu jҬh UmHnHujh UmHnHuh mHnHu h 6jӫh Uh jhx$h U hx$h jh Uj^h U6nnn ~Yݗ?U@Wgd gd tkdH$$IfTl}X t0644 lap yt5&T||||||| ~ ~K~S~mn׃؃كԄՄVWXY̸̱̦ؕئ̦؄ئ jh UmHnHu j/h UmHnHujh"0h U h"0h h h 6 h 6 h3~Vh h jh Uh mHnHujh UmHnHu jh UmHnHu-4567MN89:;nopjwΕϕЕѕ !'qΖϖޖߖʿʶᶥÝʿʶᶌÿzrjh Ujh}{;h U h}{;h jh UmHnHuh"0h 6 jh UmHnHuh mHnHuh h"0h jh"0h Uh"0h mHnHujh UmHnHu jh UmHnHu)+,-FGIJחؗٗڗݗ-/9?GMNOoprsعبᡝ읊y jLh UmHnHujh Uh h 6h h}{;h j1h UmHnHu jh UmHnHujh UmHnHuh mHnHujh}{;h Ujh Ujxh U+ڜۜܜ'()679:hij'(ץإ٥ڥ +,./bcdqrsضإ؝؈w jh UmHnHu jh UmHnHuh jh U j h UmHnHu jh UmHnHu j6h UmHnHuh mHnHujh UmHnHu jh UmHnHu.stߩ&'()12ѭҭӭ ڱ۱ܱݱ 6789HIָ׸ظ. jh UmHnHu jgh UmHnHu jh UmHnHu j}h UmHnHujh UmHnHuh mHnHuh jh U6./0XYZ[mn =>?XY[\KLMopqrL]^c%2ػتؙ̒̋̋̋̋̋̋ hEh h hGS jh UmHnHu j;h UmHnHu jh UmHnHuh jh Uh mHnHujh UmHnHu jQh UmHnHu6OPQ]^`arsabcd߯ߞߍ| juh UmHnHu jh UmHnHu jh UmHnHu jh UmHnHu jh UmHnHujh UmHnHuh mHnHuj%h Uh jh U0?@,-<=QRTZ['()MNPQ| jh UmHnHu jwh UmHnHu jh UmHnHu jh UmHnHuj_h U jh UmHnHujh UmHnHuh mHnHuh jh U0123GHIJQRabvwyNOPijlmݺ jh UmHnHu j}h UmHnHu j h UmHnHujh U h>#h jah UmHnHuh jh Uh mHnHujh UmHnHu1VW]^BLM)*+:;<=ػت̣؍̂{ hx$h j7h^h U jh UmHnHu h 6 hCh jOh UmHnHu jh UmHnHuh jh Uh mHnHujh UmHnHu jgh UmHnHu/yrkdL$$IfTlI"N t0644 lap yt5&T<$Ifgd5&l GsO"%,2::{{{gd gd gd tkd$$IfTl}I"N t0644 lap yt5&T#$()*+[\]rsuvsz'(*+Z[~ jYh UmHnHu jh UmHnHujFh U huh jh UmHnHu j\h UmHnHujh UmHnHuh mHnHujh Uh hh 61[\ghjk  +,detužžžžžоŦś jh UmHnHujh U haJh jCh UmHnHu h4r h h huh jh U jh UmHnHuh mHnHujh UmHnHu6 >?@LMNOBC1234defrstu   ) * + 7 8 9 : 鿻鿻鿻 jh UmHnHu jhh UmHnHu jh UmHnHuh jh U j~h UmHnHu j h UmHnHuh mHnHujh UmHnHu5:   6P./0;<=>@V$%&'()*7O^_}""ַ֦֕ hHh j<h UmHnHu jh UmHnHu jRh UmHnHujh UmHnHuh mHnHujh U huh h8h hh)lh h 8"""F"G"H"Z"["]"^"""""""""##&&&&)'*'+'@'A'B'C''' + +++?+@+A+V+W+X+Y+,,22222 jh UmHnHu jh UmHnHuh huh jh U j&h UmHnHu jh UmHnHujh UmHnHuh mHnHu22223333E3F3G3R3S3T3U3447777777 8888888888888&9'9(9H9I9J9K91:2:::;;)?*?+?,?Y?ؿتؿ؎ؿ huh jh UmHnHujh U joh UmHnHuh jh U jh UmHnHuh mHnHujh UmHnHu jh UmHnHu5:F;?KKKN8Tbpcdd#eee5ooobǍ^qAgd gd gd gd gd Y?Z?[?m?n?o?p?@@FFFFDFEFFFXFYFZF[FFFFFF G G GhJiJjJkJJJJJJJJJ>K?KKKPPSSSS"T#T$Tػ̴̴̴أд̜̕؄ j~h UmHnHu ha[Ih h8h j h UmHnHu hyh jh UmHnHuh jh Uh mHnHujh UmHnHu jh UmHnHu4$T3T4T5T6T&V'VXXYY1Y2Y3Y?Y@YAYBYZY[Y]]]]]]]]]]]]]d^^^_bbeeee fff-f.fjfkfk?krkܳܳܳܳܬܥܥܥܥܚjhh U hh h:9h h Zh jhh UmHnHu jh UmHnHuh jh Ujh UmHnHuh mHnHu6rksktkkkkk7l8ljlllll5o6o7o9ooo;pppssssssstttt tuuuuBvCv~~~ɺɰɢɗ؆ؗxmmjhCh U hCh hD7Dh jPh UmHnHujh"0h U h"0h h 0JB6hh 0JB6 hGh jh Uh jhh Uh mHnHujh UmHnHu jh UmHnHu*~~J~K~L~e~f~h~i~~~~~~~~~~ [hi؂قڂۂrst鼵鼵鼵 hh j h UmHnHu jh UmHnHu hOv<h h hCh jhCh U j8h UmHnHu jh UmHnHuh mHnHujh UmHnHu/؋NV^_e͎׎ &^ϑБ `e ໶௨jh 0J@B*Uph hh h h hmh h 6hmh 6 hKVh h(eh 6 h!h hqh h h oh 0JB6 h 0JB6hh 0JB66t}y<$Ifgd5&l . & Fgd gd gd  =KOrw2345?@^¥å#$CDdeͦΦϦ'(FlmïÌÌÌÌÌÌÌÌÝÌïÌ h5&h CJOJQJ^JaJ#h5&h 5CJOJQJ^JaJ&h5&h 5CJOJQJ\^JaJ hh h5&h 5\ hx$h jh^h U hYqh hYqh >*h h,h 73y<$Ifgd5&l rkdi$$IfTl t0644 lap yt5&T345IY*?{{ed$Ifgd5&l gd gd gd tkdi$$IfTl} t0644 lap yt5&T?@GI^onnnnn$d$Ifa$gd5&l wkd/j$$IfTl,fk t0644 lBap yt5&T'kdj$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kdk$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T¥$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l ¥å'kdl$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&Tåҥ֥٥ܥ$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kdm$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kdn$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T#$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l #$'kdo$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T$37:=C$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l CD'kdp$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&TDTX[^d$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l de'kdq$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&Tetx|$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kds$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&T$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kdt$$IfTl,r3Yq ff&ffvf t0644 lBap2yt5&TĦǦͦ$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l ͦΦϦ*%gd kd u$$IfTl,r3Yq &&&&&v& t0644 lap2yt5&TϦ'(/1FWlpVVVVV$$d$Ifa$gd5&l wkdv$$IfTl,fg t0644 lBap yt5&T$d$Ifgd5&l lm'kdv$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&Tm{~$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l 'kdw$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l ŧƧ !@A^_1^_`yzǫȫѫҫܫݫ :;ȶ҉ h6*h h CJOJQJ^JaJ hNh CJOJQJ^JaJ h`ch #hNh 5CJOJQJ^JaJhNh 5\h #h5&h 5CJOJQJ^JaJ hh h5&h CJOJQJ^JaJ6'kdx$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&Tŧ$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l ŧƧ'kdy$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&TƧԧקڧܧ$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l 'kdz$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l 'kd{$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T $$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l  !'kd|$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T!147:@$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l @A'kd}$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&TAPSVX^$$d$Ifa$gd5&l $$d$Ifa$gd5&l $d$Ifgd5&l ^_'kd~$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T_wz}$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 'kd$$IfTl,r3Um ff"ffvf t0644 lBap2yt5&T$d$Ifa$gd5&l $d$Ifa$gd5&l d$Ifgd5&l 1*%%gd kd$$IfTl,r3Um &&"&&v& t0644 lap2yt5&T1_`achmryzybbbbb]Ff$<<d@$If]<^<a$gd5&zkdׁ$$IfTs4  044 saf4p yt5&T d$Ifgd5& ǫȫЫ׫ܫFfFf d$Ifgd5&Ff$d$Ifa$gd5& #(/4:;<Mgd Ff d$Ifgd5&Ff$d$Ifa$gd5& LMNghrs}~ˮ̮֮׮ܮݮ)*+STAloXd+?GNȺȺȺȺȺȺȺȺȺȺȺٳ h'h h6*h 6 hcuh hq:h hqh hvIh h CJOJQJ^JaJ hNh CJOJQJ^JaJ h`ch #hNh 5CJOJQJ^JaJhNh 5\h 6MNOQV[`ghqybbbbb]yFf$<<d@$If]<^<a$gd5& d$Ifgd5&zkdї$$IfTs4m  044 saf4p yt5&T qx}ƮˮѮ֮ܮݮFfFf d$Ifgd5&Ff$d$Ifa$gd5& #)*+Tt+CI<$Ifgd5&l gd gd Ff d$Ifgd5&Ff$d$Ifa$gd5&NnoõĵŵƵHNVWXY8mVnҾ۾CG껰ᰘjh U h%h hM/h hR h j@h UmHnHujhZh U h*vh j˭h UmHnHujh UmHnHuh mHnHuh jh*vh U h"0h .GHrsu69b"8<>$&jlvwxy޴ީޔޔޔ։xjxjh UmHnHuh mHnHujh Uh#$h CJaJ)hQh B*CJOJQJ^JaJphhQh CJaJ)h#$h B*CJOJQJ^JaJph h 5hy`<<d@$7$8$H$If]<^<gd5& d7$8$H$gd gd tkdV$$IfTl}e t0644 lap yt5&T$w[[$<<d@$7$8$H$If]<^<a$gd5&d$7$8$H$Ifgd5&wkd$$IfT4ry~  0~64af4p yt5&T$&RL3<<d@$7$8$H$If]<^<gd5&kd$$IfT4rF y h 0~6    4af4pyt5&TR`jl0kd$$IfT4nF y h 0~6    4af4pyt5&T$<<d@$7$8$H$If]<^<a$gd5&l$<<d@$7$8$H$If]<^<a$gd5&<<d@$7$8$H$If]<^<gd5&L3<<d@$7$8$H$If]<^<gd5&kd+$$IfT4F y h 0~6    4af4pyt5&T0$ d7$8$H$gd kd|$$IfT4nF y h 0~6    4af4pyt5&T$<<d@$7$8$H$If]<^<a$gd5&^nkd@$$IfTlt  0 44 lap yt5&T"<<d@$7$8$H$If]<^<gd5&l gd d7$8$H$gd GHZ[NPn|į̄̄̄}̋rdYR h/AZh h5&h B*phh5&h 5B*\phh/AZh CJaJ hh h@h h CJaJ)h5&h B*CJOJQJ^JaJph hh )h5&h B*CJOJQJ^JaJphh5&h 5h jh Uh mHnHujh UmHnHu jh UmHnHu %',3GHIJKLMNTZ"<<d@$7$8$H$If]<^<gd5&l Ff& <<d@$7$8$H$If]<^<gd5&l %$<<d@$7$8$H$If]<^<a$gd5&l Z[nsx d7$8$H$gd d7$8$H$gd FfX%$<<d@$7$8$H$If]<^<a$gd5&l "<<d@$7$8$H$If]<^<gd5&l Ff  !wRRRR%$<<d@$7$8$H$If]<^<a$gd5&l d$7$8$H$Ifgd5&l mkd$$IflAm%0%644 lap yt5&!"#cd&'()CDEabc}~ՠՋvn`nYR h~h hqh jh 0J@CJUaJh CJaJ(jeh5&h CJUaJmHnHu(j^th5&h CJUaJmHnHu(j>h5&h CJUaJmHnHuh5&h CJaJ(j h5&h CJUaJmHnHuh h6*h )h5&h B*CJOJQJ^JaJphh5&h CJaJ !"#1d$7$8$H$Ifgd5&l kd$$IflNrA"m%0%644 lap2yt5&#.3>CNS^cdz%$<<d@$7$8$H$If]<^<a$gd5&l "<<d@$7$8$H$If]<^<gd5&l Ff%$<<d@$7$8$H$If]<^<a$gd5&l  Ffe"<<d@$7$8$H$If]<^<gd5&l Ff %$<<d@$7$8$H$If]<^<a$gd5&l !&'(*+CDFG`a<$Ifgd5&l d$7$8$H$Ifgd5&l d$7$8$H$Ifgd5&l gd Ff%$<<d@$7$8$H$If]<^<a$gd5&l abde|}rXX>XXXd$7$8$H$Ifgd5&l d$7$8$H$Ifgd5&l kds$$IfTlI0H$. t0644 lapyt5&T>+<$Ifgd5&l kd$$IfTl0H$. t0644 lapyt5&Td$7$8$H$Ifgd5&l d$7$8$H$Ifgd5&l p1~ytyoyoyyyygd gd gd d7$8$H$gd tkd1$$IfTlA$." t0644 lap yt5&T  Ke!,Y?78    V   B0d8^`0gd $a$gd gd gd gd  "&IKefgnoz8 fRcde*+,HIJKȺȩȺȺȊȺ j0h UmHnHu h*h hURlh jh UmHnHujh UmHnHuh mHnHujh U hSh hiyIh h,h hqh h h~h 6678NOPQ IJKL-./0:L"#567ˢ˛z h h h@h $jh h CJUmHnHu h`I2h h Zh jh UmHnHu jh UmHnHuh jh U jh UmHnHujh UmHnHuh mHnHu*7J  D  e v  K   H  E_ 1l6{6$^UA$5f%G t  !!!"$($$$D%`%%%&&';'q''G(](((&h h 6CJOJQJmHnHu#h h CJOJQJmHnHuVE@h3P=D w $!8""7$$o%0d8^`0gd o%%&F''n())*l+ !""P##w$$%%%%)&&>`gd gd gd d8gd 0d8^`0gd (9*v**4+l+   T!p!!!q""####<$h$$$i%~%%%%%%%%%%%'&(&)&*&&&&&&&·|xpxpxjh5&Uh5&jh 0JPJUhDlh 0JPJjh Uh jh 0J@U h h hYqh jhx$h U)h CJOJPJQJaJmHnHsH uU&h h 6CJOJQJmHnHu#h h CJOJQJmHnHu,Vance, A., Lowry, P. B., & Egget, D. (forthcoming). Increasing accountability through user-interface design artifacts: A new approach to addressing the problem of access-policy violations. MIS Quarterly. Vance, A., Siponen, M., & Pahnila, S. (2012). Motivating IS security compliance: Insights from habit and protection motivation theory. Information & Management, 49(3-4), 190-198. Witte, K. (1992). Putting fear back into fear appeals: The extended parallel process model. Communication Monographs, 59(4), 329-349. Witte, K. (1994). Fear control and danger control: A test of the extended parallel process model (EPPM). Communication Monographs, 61, 113-134. Wogalter, M. S. (2006a). Communication-Human Information Processing (C-HIP) Model. In M. S. Wogalter (Ed.), Handbook of Warnings (pp. 51-61). Mahwah, NJ: Lawrence Erlbaum Associates. Wogalter, M. S. (2006b). Purposes and scope of warnings. Handbook of Warnings. Lawrence Erlbaum Associates, Mahwah, NJ, 3-9. Wogalter, M. S., & Conzola, V. C. (2002). Using technology to facilitate the design and delivery of warnings. International Journal of Systems Science, 33(6), 461-466. Wogalter, M. S., & Laughery, K. R. (1996). Warning! Sign and label effectiveness. Current Directions in Psychological Science, 33-37. Woon, I., Tan, G.-W., & Low, R. (2005). A protection motivation theory approach to home wireless security. ICIS 2005 Proceedings, 31.    HYPERLINK "http://sourceforge.net/projects/pdfsplitextractandmerge/?source=navbar" http://sourceforge.net/projects/pdfsplitextractandmerge/?source=navbar  For the sake of brevity, only the analysis on duration for the full data set is reported. In the full-scale data collection, analysis will be done that validates only unique users of the software.      PAGE \* MERGEFORMAT 1 warning control &&&&&&&&&&''''$'%'&'.'/'0'1'gd $a$gd ,,$a$ dgdcq&&&&&&''''''''/'0'1' h h h hshcqjh}UmHnHuh mHnHuh}jh}Uh5&jh5&U,1h/ =!"#$% uDyK  _ENREF_35uDyK  _ENREF_53* DColeman200726(Coleman, 2007; Goldstein, Cialdini, & Griskevicius, 2008; Schultz & Tabanico, 2009)262617Coleman, StephenThe Minnesota income tax compliance experiment: replication of the social norms experimentAvailable at SSRN 1393292Available at SSRN 13932922007Goldstein200827272717Goldstein, Noah JCialdini, Robert BGriskevicius, VladasA room with a viewpoint: Using social norms to motivate environmental conservation in hotelsJournal of consumer ResearchJournal of consumer Research472-4823532008Schultz200928282817Schultz, PTabanico, Jennifer JCriminal beware: a social norms perspective on posting public warning signs*CriminologyCriminology1201-122247420091745-9125uDyK  _ENREF_12uDyK  _ENREF_18uDyK  _ENREF_44uDyK  _ENREF_53uDyK  _ENREF_10uDyK  _ENREF_16sDyK  _ENREF_1uDyK  _ENREF_46sDyK  _ENREF_8uDyK  _ENREF_37sDyK  _ENREF_1uDyK  _ENREF_46uDyK  _ENREF_14sDyK  _ENREF_2sDyK  _ENREF_3sDyK  _ENREF_2sDyK  _ENREF_3uDyK  _ENREF_13uDyK  _ENREF_52Ddp  s >A?Picture 2"b[Ά9詏6&7n/Ά9詏6&PNG  IHDRt sRGBgAMA a pHYsodIDATx^{Gk?s{sz#8'xmc " l & @d&t@ $H]ajzzRO}zouwsIIIIA.RRREҒs5Q^#.E~*@.PxwYc{qE?w{˻L^Tq]^`5z{ĵ!˃F¥bnYs@mh, s$,/Wr㊐KC.BGԾ=׷è $\*PzEPeJfB0=_k EpC @[֑KEir&.zF`F+1,IB^B 6N:״3b-?I¥rdtnnbf^  o@'#d(cׂ^w ½=|w:=1."W[pVH"jo4P  , G.%36C\C =혛GZ~, JǦQDuDդ w24p3ebZ?wyJJX^6ed.l?QHXT1Cf?⳰mxBJz22'Y(YĜXq,%ힻtɯ5 4{ Z'RaJ`"50pP9q}(jIJ,..1 'q B uH,) xQz&$*n.IXRT{'H,1.; n"IXOF U3&F"ٲd+ V<:u3s^ in~o۟βI,51(Q~.*s]^+^:+#b1Ux~y(͖%+\vҳN͛6#b%g(˅TZ~kY-1d:̓fHXI;육GLm,iao2D6- ,H`}9_e%d8{脥g_cձCl/Z;IaȁXm&ߦ|TE2 I$b%\x۩Ι9%\]G*o]bXȽ_z|k'm؝[hzvOӼqCM4 Goԑ#!" aۼLMbu¯idg.7O.D@f 'gςg,L͢U" i.\fr6"[T'c0pر"-?i^ Oy>8ЙמC&>z,@[yaםG @:4 }01E~ ~W4>8+.kWݯ7Uuir\s< / MU'Lm1ryntzB&l[l9SB-Լ2pd>=r #E*w[/K C'K2=~ j0%b~(}g?A:ӉxylDDi蘆nRO77UBbD %ZvAŒ<iDB.[A)M84 09[ K&)nbDGDbY);l2,bƠ&S4@!0΁ -@^]'pAG4cve;>o=QMPP̿w85dl!pbw[l-y sP i-CX_Ɖ-ee*_?-Ak4OJJ@^&6$O>A{.;y])4.snapvZtj v?$`&MpVg s.;?K+&T.('@Nfʆ-rtNpL-ee5UJiѧEhhhh8i94&}xB_d4MjDězlH*Y\K<݅xC'#NP|nE "yF-2O !VvHH"V^9p4tX z>)b.0F e1p(8Cc<9b=&Sd4emb5<۹LH-$A}dCc >G A2`&02XxqE#J."PPkp)eJCbd_ýks'-9AOBX@ӐxleKHC2EIs( bH L-&1@J cN( 2V.V ЛJ$|IΛwlU+IvVu i'&DȖe m;df sx$pPYGô^ŏ7FpHJB5 cƋTkzTtT6HXL=_7 zYUwvs7HTUAbB6<4^z<ń>ń 4 0lRE.8ۮ\)l$\'!BysxdN:GKG/^&4pP gyI \˭CU;#GZHŒBM½ Z^_?|l,?WٌW^ŪBms| / ?v7zF$_*Us _T6 j1C$baZn]`xuL[rM{{gD,>!p*P`S wf/5R|dQy<-B^md}hi.#êoaUm5,|{{g .\ 87ժ (B[$O*e.0L?dH:68Zd9y3yp7ؿ?,$CGmzx:)slpr4u4x|t`5fpuOKJcHl*5\B&C'|C? 10g RS*E-&TW?klk$ x9ްI;ƜJjHń*:\vʇWlCSVh|{3]d*%/l1W{^sROfK)p s.JCD/l1UtpSw.bc&xfiU`<C9{UAΝsB"Ƌ4[Leqp{Q(g٦3aSF1E-fVaV\ xTEmeGLyH.s Zx 5;\l(~RoNM{ZS,a?͑96v>W e|4d@例=omqY0>Zo7J#5T8f oME ~tj4>iQt f⊰BKAo {V|3kB[RQ5yG`< VvL3%:vE$e0CWXaLj-dۦuMOT$swbz,c'3B&o2sk>Sr,923 f!ޫ4 /Տށ.0Ϯ؆T2uO&3totv) l%_O%:: !3Hwv݇/e.8޻-G>~ZGԨ_T7.q|*K%įWt*^\D+h8i_PCpil?%!A5PPXM6LRV C|xF~ksl.8l+=~yDᅬo؇ N?bPZb2D.%b~& J$^X0UZ}=ץ[=/R |4kjX ,;TkHMn.-/f2,|%gxiRJs\x8Rf3a` yxV?iLtpmN|46zm}|zV,b/ȱNI \'[NLySmh|K+|ksp-􆖎_\S7w<(j^J gsQ D[^W,J`9*5;)6<1 Eoĉxxٸع\K($ hHYs/,_NS8"D3ϧM@I~7(5O_H S<^+}ߺ ,.ڗ}\mIF;iVۓE,6%K; 3$˚ R&Swd~%=[̑ea 6A-?g0܊h'ьQvVzMyy Bu d(ΔI<ˇ@tnIG[FV#^d[SvGl:\8Pd9~?n6qpAP6S kWbU6}q͕* MU^V@e$sH4QdtIOFgYAho8`&D'o1ߦ"S$LEfk/R; -3ZkPЖk;=B(޸43= ;YҜy%dr@tV!uI`r6lY00BV黴p^8g?|KˢzH.q(.&jpI"2uhfL!N?<vÓ7t8i[ AЄ 0VQfjX1hJBϋ6n:dV#';* I/IpZLk8Zug9LoHX(  d|y$PHS:ppF c*c6.(޺4nv\6bRg+*#it95b'.FHm0vD.ns@ 2Z e([`GԸW`dd:<:]*H^DD\6Rv O*![~6 ^M #/|$g(\~}$@`j1?pb kn~|K *նoȡ['NG & ZhZ+7øAo1sʿwH.uU,'0nV=E"2u_ 8Hh$1HsX{y1(XVva rÞmNbzPl|zJB7t RgBXEJRYRle-C9Hk.HϼF9S1bpcñrZbkptPFԨ#B0QM8j5N R/-6L4ix[E{@FJF.  HLXp f0 ,Y``,"b5+OVT#e5L.N u-9D 1R `+R^/pB"F@E@>$ ehn%&@&X`d%MX95bQC~pxRf3(@D SN MO6/"HTFH6̂HYÅLo68j?|"X(kX` QxLBR@D"im47}mn@+$%K P< up .,RBm4B1W_j5=x y"jޯaLU5"֫O#!!ub fn+9VlLR~i\"(()5q틴ȶ" 9jl_nܬLTJ.\FȗT>hj6U-%M~~G߰i~$lפf#vU,]onvəg_Q~-Uz/7Y^g}*bpA]-|`CƝ;{RE-47] \Ggqb,}pP fH $~sYW|"( vO5GB3OI Isӥ)mr OKt<: +EP Ke Ugpjȳ2G4mF*R6F?a%ERhArUWG.:pdL9}&QhO]};XR[[}AVBp?FmptǖC_׵{g¸,Zj穄+1Cd l K* B3%) wć߻L&g7/T\,ԧ˳`q2* eA /{Hd‹KE}4uuj^1DIB#Hj{@ &i3WADRp o> n /E ސ⾊+/t\={JRPY1zᨣ®h\]0Yմnv3h|4}uPHGC<@îiYɥ ČȣُvxIA*{M79]Ϟ<#f XŨↆl nzó>ƭ⟻FKru>]ElL\ove`R f & .u”.g;6=?0>nMI9ANsns{&ӍN)PQJ_?(b:J@YC,q=f[-T[DGumq7nUJ@8+4}rtG;<]|TjMms ?wSΖ"u^P]Kos-H%a5PJ pR%@;=}c 3? {p[7T"|!OoT:O%T7|wotn>{.qWґ _22Fl6249`x泽F2@;<Ή1-YSHN.Wf@/A:.i=~|Zem]τ ZxJ2巨z46w]OT /_P;g,3#K\`1$łT2 //<UM}غbg_}o1$Y,DA6$S ,1̅ O/wwp#ĵnzȂ؍4RERp1QĈ?4HRjWzarb -_bH֖/l.LE};LFFOt/(Qz&a,6tv?15Xgd'KY\!8%XY8>^LdU|*_@hX̄= OĠ54=U-OwY廤ѹu];[\GMD$0yG,Rp45TZ^$E% ml!:1$눁ӕ7~g=`Bwi"G՗z>OtWk`lDhAʬY]R.d U8}&L;zGwd{' w ԱOXeyY dX+{c=l&'+UF.Ĉ>U!$$X1 10 FGI2zk]e}awgQ)=G.~OU*Gx@B#`$hhz0d;n6,y E{m%VL$ Xpa" 2:"Lhl^ܥ';P <&4r5`Ju~ ,i DоJ lBzf0OZ>޽t+}(8(RA1`U=[K'=ς ~ )H*.\!C܊Kψ4tj~Ym3ԢA5p5 + !߶>ߪf{6!=5L{ 9&Zf0= FC iVe%H1$\̯y] 9fBSc"H^#E##` k((q C7Z>cM*a;Y ĀQ߇b"HIR擄URi)C@C@'-! 靉LOPt tI0‰ATn UB0P H,ɔ kC?0Vbh߰ rK*\2h5A:a p39 ր =h'$mE2-;'@ჭZuYŅsؗU{x 4i#k]Oyj)pъ.րVl;d\~jzªE<!1vlw aY"PtѦ+t>,Um.D{ O*e.:XCK 0R^9ZeİM"tw وOo*2 4&6m@L(r?zgW'y r("ɒp5[XXv8sC%i衰̯TƋ4[L(K`c]blJ0LDKpk'*/l147{*b.g+HŜ\x/LmݓIMMi4[L+exr^y]/D.(pJ* \GEr>GūpẕmomӪֶb[8SUXJTp 1e [9vѵ;P.`p 爯'(79ZwH-7l:wݫfp+o?s@:ױK댄K:̈́" .i[.jygQLM˪mJuV6eg-U Qe5.$[8<ȣ\n5bP /ה&)3♚X˟^B˗Ÿ9fL{Vp t2Itpgƶ?m<]ӡ:r'6kttV ˗ص?T M狹/.=E(+g~$+UKAEe1*R]gdnx$Kւ z%\ /"(ãd.@}on=KKk[^~Aڝ`!\.xе$Ne%IϚr3HISƯ,Vla:T߽X |ٷ2/ȑ˛3%iǵr|ʸJw(nCL-`JAT_-B SsS'DFvϮbfU^f+|c 6~jۼv96oQ 6I\r 6kո6ɴRf9}o) e) ޫ;KJw6>QdriC58:Z]>1NkS`~\$.p$h!)sOɷ%nz*1oq6؇ݏ$\d(#\І/-2Qŋ+O5Ky7WAӓʇxKt4o K^QҼ q1.N/}mެxuGG,"U*ے7㕘 M Sp7稪-4ᕃ @s]KAY. C)\k Vgͯ>eӉ'~[BUI#㏢a(2 "ƒp1z..>E:b:"txmmwwMRRQ9=2&]3TNچS$%b\(\7.\h^R.0=NO,e*r$*[]"vb\\Li ,o;N^SKIEMvpabC & /e`|R* .-?uO-%E5 %.Qy Es, Y.GS|j]Ǝ8 ߻E>WlԉoE 94Q4r-R \jco܁?\DX.b{5+ ./Z 4dG4iIRՄiY_X.J[r?h \jUC^r!pDB|}c=s$1U\* 4FQ5&aނcľiˣQIZ&/TW'I͖G?DldE$lkc³uIwIȟŤ?^v%* \ -__'Y.zp{uw|6Q磢̈́ıZ$Q:HVDt7+DrK>%]Cd;F'?[cHq5֤qIvIȟ/i7""?uO힓n*l HLVtP\pkjqYZIQٹ4 7z/& Y$6[cʍnqIraKb I?6YHr]bJ,yC䧝5]Z\QpĘIb :OP>i ޳JO߄mˢ0鶊e]T o%& #@M$Rdkr\r8Ip߯Zo+6J:\e\"{>A/փ SBWP5ߙgM6A] hIڐoGM[\̶" 4.ICh%4$$]HɕyMǜ\'.F KbpYwQU P\pwQFB荔l+%nKIB\wfk40qddMV9Ddȸ?~}LhbIpw=f&='*+\D|'32J@*.{hqolSzRfkLuF@$ِ4]dv3HGxZD"{&B,̒shhْ%J0ILρZ7eωXp>dPE=,BK. ɀUT]g\[Zۮs*W2үo9M/yna9*\\,BK.nkpGfzNJ\Ռ_)W@xξKoKWgY?hS~<ڪܲnB.2" K(7JT;/\$/qbP;qkhR]cA2ߣdѰ Uv7TX+N;R$ T {? Ǭ|[Rpa.Eb/ 4jg2]9C]lmdv0ѡ6K#XA]ePR)Gqp?[뙴 gSQ;ePY8M$49TfTb9EZ_bt4;U\#\vqoQ 4XҪpbp1>L҄ @ ™4A&l-o+M %}p~^?@ ERbp1>LR2QyĀX ɺᲡVm^ . K- 1fA19K.VRPk{NU*M BЀ?[KP]6%ႲBg MH¥ £ Pp E9KE:\hB., |T(lƿ%b|.4! K>*\GV\, 1r)$\*X.( ϱ,͒pj$& K>.Ó󢫅""dIju*;\ b<|*1$|*IK>.3{a۵>.ɤQR&7DrIsT*r,zEWXLCkD&KtB d^~4UŐr P"fjbwa=M2[DDBg’nQJcHŸr ">KEzpSXLW$$;t+OiiWpђ壓<\F+pH%5^ْ dI?"^L@J#EtpɨƋ4[q%l4dAN .ў\Ϩb/.F3^ْ Kmփ˲f|ݘ (Z.F3^ْhsRY .%+<ʘfDIŠ"͖.ו1"l|c bpF (c2Q S&@.%i4p EČG8a\QDpZ.E-)\p#ݷo"d1d;KnxnzոsX=T[ipVղJ dbld1d;PBMRwz**?nyqE.Sq2'uy6\76vpj)"htp+$@s{ w>\ Y -;Owmhl2ď!enuOм.㧅oJfVU+f>_ $/5:է9SV<0>.uFY .bȬ[`*[-LQpY^wg$U:T߽GǗ#`^^A1%e%܏xa.~Zt6f+{_`Arn"ɡi)ɦ6RӅt)yEHKjUV=6#Kt^F}Ltn}덓vtpڥ V-?c V݌%p!b94TnDsψ ۡ)b5=u) k/ITu$\bϡ[pa,I\H)pI-IuO'F' *7\`'A(Z%vS} ' •Vi fd pQիw}˫PY(:;콪EkPᙖ:\P8Vl*7\ģ,.%TE5Y-5^$\#/ OjݶgÓK.9tC3rITV.(|D~HRǝ,yrNO#b?;F}ljC.s˫Ol/qhwmbe Is3ߴKD- _9kFCvDY.'mC=*ڱzJ7$2 ڈ͏[\Tq%i4mt=ndb+.t%ZmRxJ꜋]oV7ª *9+V:<=#ZkE4,UvкS~&UzU1"$S D-Xi[oB+@ByH&~6{%N b2KsV^? 7ɤ(J]GwP຋=bF] "h aeV3Oi*ZZH)?ֵ3FT-P%K]M!6%'o|JjʽRB6ri$Hau]n|YqBY~(\tsp`¶k}jĖ\r|^+իąJWOP$m);7~A*$MRrԻ"FפIu$&}E}CunWU@7כdJ,4ɳZ.DHPEQUDn-l|"텞Fbc1łK@YMW]E6$'Չ9\pN_OW}dQ48XXH EL>Q٫Ή*G} =Βt&kBD4Y;#r%.63*]7JӤLWkӤ8qa<륯m5'Qپ炛}멘; Gf,8xIOrXmE x+ihiWY~"Äb{CkKbԻXzj.dMjWWW4"i>"f V|fT0>]%z!MlҮ䯨$i4ؘ='+.dMH>[m2R\4M닖Vw .L@ pKx {Ge/t&ki.iѯ `qG*WS-.\0}1pYs!ʗshp鬮4M(\{.No%z]|IH?uܪ%] P"L =^U礂E]+`[ilpQlĵ>%R V@V2pp܏_@e&aۖ_*\cY)B-­闫$;H-y& 8҆. \ |jj̲kpXEFPbOqzDIY@_&>U;e.vcYI"C_Wz[,o+5/@R4S>=\pWB΍*Z=omC^7.¥xTN*\aPHvb'.-nRKUOdYbe"/˞|<%\ *ݣ_V!L)~huK]5j~\W|p/KA;v,7&g'bRxhGr%1w-,jk8  /|׷VPm,,Ye`cV889}.zeڞbO(_rMhcgMԦ8{ diuZ.\f^ׄ"ypojݜrY*vTSl:zۡ SZVYYȳ\.:CwG0"6xϡ!%W|i?DsBgXzbom.ynw{@'Expy#:JrE>a$.㋢^r[krnKMʺ[Z[+] f;; Mnx oV;y oEHpz9Vxpfr qCþNÊ8K|\ՌX=|,dŰʎK;eՏ)7q{I&sxge,yHp".GSd Q9s\|mn(ʊPߤ>tKEE罬kUH%`wEEc2dT hkVj[-*,\+cZ0b=xT~.LNY]1)%bԢ 2e`|N ?<9oRdy !c1Q,tﲸE) |T(GYW{}ĤpźW bYJ/ pGxb1)\ 2VłLp(+IK>*\w~c.|^[.#xEYIK>.hlepǚ.|^rk$;tsWCٻ=/E$TF PD<{.l<'ת@piQ.pI(''$Z )G?EȌp)Z c׭yr& >X=I@N*r "^8ޅR*\x DBp"^ E%/l1":F?:yxrjdFpbuXB؍ƤKF2^b\E̳,"&rEoDI%/lJ9Ee@A /Tn@'1e" $bDƋ4[RVp, odɌp)BMB4U.@_ K:$v[dʑp1(aH% ªߛ7\HV /Th$tf>=Y bƋ4[rPFuOq7>2"urru[cp1.`E-)=\NچX{D~o:KU.zSKm߾c9hUv] 9NTp>c!"H/:+rI{.^zk^ǟWY}>NkBS&. }?n}cm Fji1.ZW蝟;@!.bNYv% .gV5?^ UDzdejz N\#b'{rtpaoj ׏|c;~/^JYS̿~|X1 @Z<|:udtpaXZpA|y}뮚.iT1X1vDZ#Zb}9BZ .b7yTTdr뽭 *e6vL37+~dMacCoAF4\1Sp)r Υk!UA-wۮ"|jUmf,|Jk.c&%j 6wr }~g!U{F*_8V5 N)a)` L j|) l>:~DƗT5 /jo)YMU~9{D -0.HT*+]$YjɨMmn_na9tݟ%e.`PÅ $%AnC7+\o!Y=OWt(T}9+J&HH]Ad.o 7\=+fB9@vȀp׷{J-=}ڥ2 ٽd)/+hGR ź/@yÅ&`a;61.iMjIjl2V -}- / VK.0+ʞBe"% iiT\hWp g,ag("B 6 V2TpAgyN ~-d]4¶C信.B"qY)]yDKDR,X=r|dMݗh$Һ-Q{]<<~ok;1mIcמLːfU"1bA(6iĒvUbP%&Qѽ7>~`z[^ ;ѩaqN¥fUۿ[p9kf,Ȍ 6qPE%߳em+ҊD*2[ !rd!&[:+i4-2ݝ.%Jl_0)BG֔>|xLqv5({]g =X9 2xg lNR+-\f=(+^h.i&W(U7yGxj"EMReiaړ1Qug@͏geպnWg%n. X{)w=5X9ɝ'ϒ;Z*.=#E}2[r PP[j:X1ǝk:܆_lzi/?MV (F?sh4:V ib& W.n.]֋.6^9Bz0aTJVL޶AZk7yjmCQtTF`o"%\C^3P] YKŃˊ_:u)"0oA&\tF I)0a/xqW]Ĵnr7._Stp IkCF'~P`d*\ץ*8\ĔqM20>ǎ}6ģ&iT$\Qa"Fgt>̖sz> K>* \'=#儋\QGe2\K:ç'poE"ؽpYC T ŔsT%/lJE <<9c*'\؁ErQ%\*U:E-*g>ӤdPCϥ5^ْru@eK)L`$JFxfK. K1$GSe\rq1E-9([o蕃JP&SrkC&Ky .M6H_'|H+T8\φ@ .|*7\YND׋ .rXƿ~m[eN}:q53`XFF/\ā=? \hK_7Eך?jF =[ ̤ Nw} mE ϱGVJl(QeK)fDөPpA\Xb*ɯ-UXuOf2:la]p _1>QGRw!0e" ?\7>n9|e+ضwzLzhgy_nz .{8o|2\V>p9zsM-%D(j6Qe+[O*'\۹],5\lNe/&-\͉*6fɢSŽF?_/!RDk-\ K- ;yٓ3_\j얏+Uy2QT=$XycԒQ(fsM휏+Uyꆮ**.غ{۩KMsolhN70X 9Ny"&|yTUpAm_KYpt嶺 c3|Dz*\JI4T>.#pAFMvVw;i#`1?=W6GKt<׫.%=D<JֹHt,/MNdh>4bb,W4ڠJyaRY* .X\kzST.7Y&+"i6] Y4i6*mVhchˀmhW^rQ6ӥͨWa7\ K(ifjcRhXNv9cjZ-yrʨJ d@UEUmJOBU9,?pm,FKHjcŋEwfY֧?x{<@v)1SipaP\qp!FfJUo$6X3NټRhB3cOڬ49xㇳnX.ͮ3#^uTQp,Cl0/@K8 HKMXL貊[զ,vt 5痫l4"s9Xy+ .Rie9C6i4͛.)Cd5>}܎n^rrQ}-:Rq\plU|U$\=4fgMt܎|d {L3ǚ.: 쾬Un؛R%>s͋4<'ç'kN{ MY~$: J (RkuYJ}<ڡ_͢2UXv8bf[d>@ut3V\KtN_]1Zzk;J'pz/PjW콪8\DİZ܆F)'\J6%TypQQϻ2RSPoZv/zL;:vyfH`c4^~$*\J?.TapqxFr)"V-l[TuȨQ$S݁ȶk}!$IS pJWz2۽7]ǯR1Yk|Oiunz. صɇbs(<5'TNltnBxFM/k[&C|7u7usQTk*,GѰV6EHr%+-OU}Wp_#TM]@ ^+YHn:7Unռm'=6w%l0.xWV¥BQp[ku1;v2krWzi |]Vzr6\JEBʦOvثO뛔K;Y`b,*(w@E _TKT\ 0aȽV _-%/ڃoontxkk;Qv6Y.%"@.yn` VS!<,"dh.&j*NҸ5:TTSjP֐[kk&0 =|͍e~ _Ńuܽ?b D&4MC@em,% 2zCC3dhL. WrcKS~׺&2o.&!C\"[FÍvn|{^mdWU>r(S~\>:sT*\69j~4Ŏj@ }e߯ZM?wڛ]QJ%^.p./)\TzʣpAӑM0M/mR%ᒏ᝜KY݁Kp! vW9]oonot7D}_? /ʨbeU\W*Kyv+#\ 4c4fm'e4^UO]5~%\ `)i}xR*8\ٲF?*=\QŗϨɡvDX|;UDxfq.f Z:-\ VLjx0/Sboho7;OD"ْRiiWa"&0%=|2m/-.2s) V6:"_yJjH%+.YͲT:0ɜptsVs^Ju߱QdےYIidB4uOبJ6sӥ "=/MRznSFy.Eӡ/;oi"z^"t/lVƆ^v >)\@ _/ ֢/gn +xCY@ RBIxfK*\ MhthJ璥q@9rgɠh@ϫp ?oV;}+,%i9D[f-z`<a&QEVVpк@|-.2u/w:3n:elM22.aH% y1?tzJQS*\BKGwϗ6?kbY/̏/i|%K7KJZSV F\(_=92Ɵ;11 ۛN\`9hRj4uo=t!.Ҋ(N9n{acCoҮE \s DB#X@a?j^wHbJ:/<`?od% .,}ێ9`{qE#{ۜ+vd({@V\?j^wAfnpɊ,PRs 6#/E#oRvq}u6 M!^Gt+2 ?&û{v6ٔu/m}s5Xj 1Se~肫u$% ctpc%cT@|!_y 1Mt;vbiǑ2Q©_}7H  )a4;V4 *8 79^VDEo]&@׷.lu.@x Ye"f+Ey…uc#)Mm3R:T`MpNJcAϻ(>|iPxRVoojk+T .h};iT:@|% pB_pkcGHB#@xUpXmmz`Z@RL:\)H%28]ڏ3p)K S1PƯ2An\ʀ^`VȾ̚'š ()\ VtSթｭN~RKR]ӯgSԬe"o"m̡6Q {S]lU2\l~'e˭xތ*sVp|TrٍZSj /,cp>w[WHREBPQ]*.yu|B$T}g{r~9+4̂BZ.0+ T~RK^VIn `)ҝpN)pUE蛟5%[NӪxiqjơXҗ RbG%|:Z)/.A/TwE[^WVas36{ΝMot6% paJЧ[pQ|"":&/K%㠰n*. +ڙXVRI_+]ypO笫T~*2R{֋?aݣ~}bVq\)8RGE iq^Rl~&MtQ=xU ]"c&"x PT }FW\|Jj{ӛ]&(q͙Dq4{h6EϔJ8"?E$^c_{I*.bŢ종r:k߶Z$DEzfNApӬFŲm$YO XL['mCȻį_yIKi0pT:{U[v][%Eƚs xXtd!v\p)~5>:Kv'Tэhe"^+L .`D0x";:=2cis8N%ݢ t\n'4;_M{\Z3\г`$DR,.<;zqm>85 |6o>>P:t!Md׬&ْry|}KJD6]֖_׵UƻGCO*O|>6 -#EcK"vJyT V_(81aTh(D0b)Fڹf5 M_Of)쑹g GXC/li_4. b 2/ bBۋM=N:0\֭kq$+zMt=⊥3~G hl2/mC,Pytm-xd!{~VPd7wv~bRRT7_TN/ 'yblpv/gT1BMf#].gҪxɛca8;)25 pfn &!ӝr)kp>|bdranUPg$CޑOvyxԒk݃B)Ӱp3#k:HKxq|&_0RLOwhxr?ZW9dp)t#ܕP{"/'LŠQR){!l672~ij;sIW.m1]Lp+#bݺOaw{\Rk}>$_"p3iiy<y<%×n؇?CkC4>C>&"F,pKVi.,|jh7䛥*WY@ΡLי^,p6>fK B~ ޟxw]Ir%تzlV "&=6 #?e4"T0}1,du`݇/JL=vcM|nxr~r6՜]WƘR^+CYw1~q_׵5O$eeu20Nl"\c3 .Ъ/ J-%F&)O'rp NnΗm?9ا#Sse\2\fֳŞ2p['&L[7?wuۛ$F<e\ڛa0OZ>}G=#sg2e2\ A]߅<5A  nƶVJK~޾iX+ +;{xBɌp]c`c%IpK>BBEE~:`8p}}ϵ<|Tkb7M CŰ)NPܓg!pxgB[jZB& ~ʏNgu#xWPP  10j Xs?Wjkc lپ1b LL'CV3b.PPu ܓod#TA8bfȄ*P 'Z?NơG P"s 6W k:k#k.~q?R$J;ŹW|Cj絑a"4Z OYpě//DcG0md0CJrЯA~ *2/?2`;Fӆ M]`ƀ" |A1c"~.@ZSxY .ZOaͅKLqEXU@2El\+v'mCڴ܏6LinZ)sʪpp'=s 7[Y85Pϲ ""erY.Lmݓ7_b_&`dѯ;@5-,׽0?77EY AH/eEU\܏tP^9eL.M0Nچ`^i\P }Nچj]OaHH-A--32{ĸ1a Zp1Ρ UAQRRRio?3KtgIENDB`$$If!vh#v:V l  t065p yt5&TuDyK  _ENREF_13uDyK  _ENREF_52$$If!vh#v:V l} t065p yt5&TuDyK  _ENREF_11sDyK  _ENREF_7uDyK  _ENREF_19uDyK  _ENREF_29uDyK  _ENREF_42uDyK  _ENREF_55uDyK  _ENREF_15uDyK  _ENREF_55FDBravo-Lillo20118(Bravo-Lillo et al., 2011a; Cristian Bravo-Lillo, Lorrie Faith Cranor, Julie S. Downs, & Saranga Komanduri, 2011b)8817Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie SKomanduri, SarangaBridging the gap in computer security warningsIEEE Security and Privacy18-269March/April2011Bravo-Lillo201128332833283317Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie S.Komanduri, SarangaBridging the Gap in Computer Security Warnings: A Mental Model ApproachIeee Security & PrivacyBridging the Gap in Computer Security Warnings: A Mental Model ApproachIEEE Security & Privacy18-26922011Mar-Apr1540-7993WOS:000289158800004<Go to ISI>://WOS:00028915880000410.1109/msp.2010.198sDyK  _ENREF_8sDyK  _ENREF_9DNg200725e.g. (e.g. LaRose, Rifon, Liu, & Lee, 2005a, 2005b; Ng, Kankanhalli, & Xu, 2009; Ng & Xu, 2007; Woon, Tan, & Low, 2005)252517Ng, Boon-YuenXu, YunjieStudying users' computer security behavior using the Health Belief ModelPACIS 2007 ProceedingsPACIS 2007 Proceedings452007Ng200926262617Ng, Boon-YuenKankanhalli, AtreyiXu, Yunjie CalvinStudying users' computer security behavior: A health belief perspectiveDecision Support SystemsDecision Support Systems815-82546420090167-9236LaRose200528282810LaRose, RobertRifon, NoraLiu, SunnyLee, DoohwangOnline safety strategies: a content analysis and theoretical assessmentThe 55th Annual Conference of the International Communication Association, New York City2005LaRose200529292910LaRose, RobertRifon, NoraLiu, SunnyLee, DoohwangUnderstanding online safety behavior: A multivariate modelThe 55th Annual Conference of the International Communication Association, New York City2005Woon200530303017Woon, IreneTan, Gek-WooLow, RA protection motivation theory approach to home wireless securityICIS 2005 ProceedingsICIS 2005 Proceedings312005uDyK  _ENREF_27uDyK  _ENREF_28uDyK  _ENREF_38uDyK  _ENREF_39uDyK  _ENREF_56sDyK  _ENREF_4uDyK  _ENREF_22uDyK  _ENREF_21uDyK  _ENREF_49uDyK  _ENREF_31uDyK  _ENREF_24uDyK  _ENREF_25uDyK  _ENREF_20uDyK  _ENREF_30uDyK  _ENREF_48| DJanz198424(Janz & Becker, 1984; Rogers, 1975; Witte, 1992, 1994)242417Janz, Nancy KBecker, Marshall HThe health belief model: A decade laterHealth Education & BehaviorHealth Education & Behavior1-4711119841090-1981Rogers197519191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975Witte199224242417Witte, KPutting fear back into fear appeals: The extended parallel process modelCommunication MonographsCommunication Monographs329-3495941992Witte199425252517Witte, KFear control and danger control: A test of the extended parallel process model (EPPM)Communication MonographsCommunication Monographs113-134611994uDyK  _ENREF_23uDyK  _ENREF_41uDyK  _ENREF_50uDyK  _ENREF_51uDyK  _ENREF_41uDyK  _ENREF_320 DWogalter200230e.g. (e.g. Bravo-Lillo et al., 2011a; Chen et al., 2014; Schmuntzsch, Sturm, & Roetting, 2014; Wogalter & Conzola, 2002)303017Wogalter, Michael SConzola, Vincent CUsing technology to facilitate the design and delivery of warningsInternational Journal of Systems ScienceInternational Journal of Systems Science461-46633620020020-7721Bravo-Lillo201188817Bravo-Lillo, CristianCranor, Lorrie FaithDowns, Julie SKomanduri, SarangaBridging the gap in computer security warningsIEEE Security and Privacy18-269March/April2011Schmuntzsch201431313117Schmuntzsch, UlrikeSturm, ChristineRoetting, MatthiasThe warning glove–Development and evaluation of a multimodal action-specific warning prototypeApplied ergonomicsApplied ergonomics1297-130545520140003-6870Chen201439393917Chen, Teh-ChungStepan, TorinDick, ScottMiller, JamesAn anti-phishing system employing diffused informationACM Transactions on Information and System Security (TISSEC)ACM Transactions on Information and System Security (TISSEC)1616420141094-9224sDyK  _ENREF_8uDyK  _ENREF_11uDyK  _ENREF_43uDyK  _ENREF_54uDyK  _ENREF_144DBravo-Lillo20139e.g. (e.g. Bravo-Lillo et al., 2013; Egelman et al., 2008; Egelman & Schechter, 2013; Sunshine, Egelman, Almuhimedi, Atri, & Cranor, 2009b)9910Bravo-Lillo, CristianCranor, Lorrie FaithDowns, JulieKomanduri, SarangaReeder, Robert WSchechter, StuartSleeper, ManyaYour attention please: Designing security-decision UIs to make genuine risks harder to ignoreSOUPS '13 Proceedings of the Ninth Symposium on Usable Privacy and Security1-18attractorsdialogshabituationusable securitywarnings2013978145032319210.1145/2501604.2501610Egelman20082223Egelman, SergeCranor, Lorrie FaithHong, JasonYou've been warned: An empirical study of the effectiveness of web browser phishing warningsProceedings of the ACM CHI Conference on Human Factors in Computing Systems1-1020081595931783Egelman20137894789478945Egelman, SergeSchechter, StuartThe importance of being earnest [in security warnings]Financial cryptography and data security52-592013Springer3642398839Sunshine200912121210Sunshine, JoshuaEgelman, SergeAlmuhimedi, HazimAtri, NehaCranor, Lorrie FaithCrying wolf: An empirical study of SSL warning effectiveness18th USENIX Security Symposium399-432200910.1016/S1353-4858(01)00916-3sDyK  _ENREF_7uDyK  _ENREF_14uDyK  _ENREF_15uDyK  _ENREF_47sDyK  _ENREF_2sDyK  _ENREF_3uDyK  _ENREF_52DdpH?-  VfY;AbgA?Picture 6"b9L ^ }l{n L ^ }lPNG  IHDR8gVsRGBIDATx^ Ue33@F" zLl5S$4+֭fմ[7KST25+MSPS4P(d /,{\{K:=gYl     -Ptԇ     @B4     @A*XrsG{;::.s;?\ZNQ)     @OxFEbԉmwRW}Y#'84ԙK~ sq~ 5x.u%KN|?>lcp%?=    yOTIws~ν4G~˧>|ǾjUyCoD@ Ǯyg|ݝ{Lm?Ǿ7廏]u3k*H[)\^YYyʩ{ܱ{?x_L!N@@@@(@zIp)~f)θ3oxuQ}'4W?r)îN7;?GGbGGg>(*-. +PJܧOʵǤ~6e@@ |^K{|! [gM6s{5m[ڲOY1FNu Ϫyr {33k _Vvhzk߼cnŠyMpm>Y^Z4|W7/?k<9vpuզ~rܟs7g+cw.]ޢ6mRC W;WߔqV( G f1SO=瞳t[{<'xB?SH(m.s85>^{=8,os1S@@B&(Cz(=o-.3ܦ}#YoN<`33^vK.PQ{)؟q ךJV7fxo\x?Lf\mܧmDޏ]}6M[Q8o  Nƌd .ؗy˨7:\['@@| l+LFPnMG? nܺm5#zaHo= =TCh*+W澱^g>4J.[Ӛr^C]:k׮՚   QexN<3&534l+⦛n2eϟ(+Z+VԜ٪m3sÍrPIms;Ek:N޶tLUrձk`f!P@@T&?qjEgjՓ?uחnx'уt'y>~W[M.θy>Ucl=L  WEɈ0?|+BNPn͊]P SQ]mׯ\s(%:dP͊lj>Wmfqq*CkD, '3{G[h߰ޢuo:Nƶ䦠ͣo߾yms+y蘙_Lr-f^fϞpO  e%Z@"y{4=gav @g^ym)m&GsA~q]oֽ;4>L.]wIyگOJP@H0`k3ްah"+ĕW^=fE5+&^kgzPeߊqB9o 5&0j^~* mj]o{g`ĉUiEMǘҢu+kt;y?[U4ЁL^z%>_xսztPx[Ǎ9Gח>Yam~Eo)AQ  @nUBmfAOňX l͢fBJӔ̱ZVGiê\M^xq:ʳiiR j[4qUk͵Ax\ [-Zq\Ý(!?$fnS@@=qõN{gRW0nu ;_>4}#9ztX]Ԅ~h-Dԉ SUjf_E!4MgT6 6;USIs枕A7L, UsHٺJ?lnNgӦ&)~Z9kгosC;oU'oCϭcq+yg$ @ +-l g\&dleg{+q3**T$nTX(|Y#'8s3=`"e^M  @qM,f)6"QMI@Q   _ OZ 7޹leϥޱͷ_?Bz  .~@@b)'13(@@Z) @{@@Gt5#E@@@@*Mc     @.f     @Q@@@@@@ ]>sH@@@@@ Tڠ*++q%h@ ,MMMaJ~466t@@@ <#s W(}wEEcJ[ @vhP:ސw/i"@@@-4)8}r;@@@@@ #;ut@@@@[>Ǐ?nܸ~t'.!    yqV#ٳO:1cƄMM(%A@@@@/н{O<СCߙ-:TAg@@@@@ÇO>c9bI    m{gIG,[II'e(xg޽{ggO>Δb'.ϽyKc}#ܥm V{ᙏ-[zנnS>=NnFs\pe*3a„kw1sLS Xw(KKN:uƍ}{һd DRLO<JL6N:,i@ uJWScGPI=a]q#qa?6sl[n~ARa<zPd;=4LfRR <(qP);{SO=zoڳa#>kU%Ճ>hMNƼ=}?Lar7c}zu{OrV%~k/{N2ۨ92   qB.&t`"Q`4jvs۵Hyt^Uۭ[/:sMT7lȸ3x+eVhCY;?mVago@RIEP6{l+kʇ3c:@ge7piWHdITMg*=GQ33Lue mbo&{WӇcG}L8~؇!(:rP.B5]w]ubh:?LNNN#"}oٞCESZ7GԽͦy-^lhbV&D[7Ns8m~m_g|miscT   c*+! xwڕkϞ=JZqg+0n @;ԦEY4G+z)cÝ%z{4vZ-6O`ȱ"shBh_8Ӕ26:yRΟl=qU8(@u)#C[o/)yt{2g k_۹?IRE^)-NcXe5jD*xZD^KnoV}qJkvƱ(k1caEn%1hgU^hCNz(k VzK $vΙb )HP4lqz[jE$1)g?2'N%)K3檰x+f9+oִ>nOToHmvk֔za˛ üB]t֍=9(Ѩ#5kSs˨ ,XX'7Zk08@@@#ZK/WOB/pm}%b/=LbkW`uTxSZ1_ ~;pI-7رcGzK=C9/n8|SV1G[8,6~98p>1RĘj]RJ:6 *{fQ&@xfzkКnIL"#NiNk)zd 8)jO?1=4SޟmfZkCٞC}H`3?Wъ_kTތ/Ԋ.qĝ>W^[Jmů}@@@DkEf&c>Šjm}Xy//r+13foo֞ڳfW^bپH>^ktVzӦMnuϵ*okk{Sb+W\|ˋVa :<{V"W+屹\E-Ǧdɔ_cFiRRLY(uzh{갶yHL*G֔kΈ>0Ft.2oR~Vؔ<`_ln= wA'PouW؛#x+o>ʝ }8Sz6otx}8z[F, ;Lb&~bxh2~%[D dyvٌ:'#OeqV$Z-mʿ>۹7:qNY8H@@@ DZr?kVk-4Kt[lXD]Jmց& Y ]غ5^P=N1 ;Ǎwg _8^y啹sYF38Cwi'ܵkWoe~M{Z4̢%V-M2kµ2ՠg=^D)nA'W \>=3e꘷zµ> -\몰:~bxh.{)>nOTWV8ٮv6{&H ^c O}?m2;a1A^~,Vg :   80_O}ߤ!ڣ?sLgjNTqmݡH7pٓ6{yGQBＫzRƛB$+UH j7e*++O9EMc~CRBۙgsmmWlT_uy OJV>Swi}wG>mZc=Ͼ7?[zנL}y}Ne+ @\VZ-ͦS3cI5OQtƨ)Qy34SL5(iw~U-}P>{p{ȴ?w5e8v5a2=M8T:1z,./[8?R'FQ]ϼ{9N2}Oc4f<3>f@KKT&|eشyg^Jl~5w7ݏ8w=R@@@ +Iہ{GL :\կDr9:^X1ꫯY7niVUŗ2@'$9VEޓa *p2sUa?rr:d.2~C'8NL@@@@`3C o< |R~O*Eo7xcر_~y8/tƓtkA=szx@}P89Eȧw 33@0  G elg| m{DO&SO-_>ϜuYڣ*;>|vM7ykLZjteR6YlZϼ f:ujJl8cp+p69>-$9yF[t`*9 8TŖ-[~ӟ:iN(  !x柆j'^t WN9DI=Jqe9)Yey۶m]tQ"(5{l'˓V峀Y}0ђḪ߁9* 6׿7c`ڴiz?=6A@@<֪gO]n{a<@׮M|cߴGS"g-glPqj"$;(4+Vz4LVIZݷo_k)eJL Õbc%BkϒjfmƍhO+f93(QF\Sa?o1Wo;qēO>ԦO%뮳>{Ĉ){7G    @i@v0^V㏟;wҥK/^=Z  UZ 5Jǡ DdƝ2$KU$uw e)0.ZZl"XYkuZW755ḙ1p@ͦi|`E5KMa9G6pӧjh#NJ?  A t)9Z[M({+fgVqaCz!eOULkhSQMlڴIҞVy?FD,7-2eu/0Mė'L`>эl5(0K5*d'p>J" ;Yw'=E@@ N_|el2FDOލX'>WUUq_WM>]}/5J Z,FM/ܾL wv̘1wV8l0-|ֻwoݓۍ7hX 78W%Zl=&r-4QlYa:BͲ6|šnHR8?ja 嵷gGj\t@@-pڞC 7]#^\j>mmm6{> 0}r%Pҥ?Oٞ={ D@jo%N*;\Yp6e#}LݛkݻyҪa-5,x?ng1?~p2֯fn=ը|vc5M^+n Sŋ;wU2]Ca=H^9sny:;Fds9B@@8HJ^ɱ;rHPԬR:csJZsnRyo4 ):u>hVWeKvޚbtSب鏲?/ZH!rt膕 -NW_ [ upC N]pL^"KkՊ=G @@@ @*R% :/Xi5ZѢ3Sa_miiihhH%ɥ,^:͊);(-\gK:3fmJjH*H]3!'k4R|>sWH%&z>g3osy NAY-I>Z=gܩb`r{;>j{t^}VnhȕC@@ :)8Yvy#('= $D~}l @nSa"!P[^lKVRr@j&+. #  @@mxǹ2` gvuDmdm7o6,fh (@K_pm)y{ @@)8"4Yt5@ޯ$H@Rp((#9);3,ɎtZ_8M1t[>RO {h.jmRNt؜bA=d *EIӚ2o(w=iҤ#F88qX@@@v[U^F{<W\ gKcFBs`+_JJa/ezH<x|We^mϜV>ι':dlsn9sX,GAY\Qu\U7(  ,P@б(B8    @NxG}%9[7"j =|:o睜~N*?)8.,9q+n;#9G}wSOTVgy+7w)9Rp~: *oܸQ{3)86gNܖH=Tx* [A)ܹ{{+kC߱uv>Yxȼa8·LI@@@ !PV˿"jYMwq<G83.]|ƻrW 7x٣KM(]mnuXU)8yqr˦jwPAe\hy?->s|ARO C p6B.YA6gNuݖ6![&Hlzi!*ĠYّʺnՇ x_^Y<_u0?q>dJ"   i/&XU~Vٺk?яH5=ʏaM0k1|fV1[SRR(w" J@IOJ&pU%&nkU=-ڣV#oB0^vVy[3ye-#{FܰVeJu98hnCxX*(^p@@.gt-ķt#@ȝ>9soʷ`6P+BJ4aBd)I'4~^uU_n~2;MsU"V\{Mrt"4oULkFvfit/%yk SnyGWΦv Z8 J $@@< Ty]s$HYR<ǽZ\\*c%06qFGܬbs}ыnEKt@+|unʭD[3f̢ETLm]~_Sz^0Ȼe͵S)tXUZ%ʋ"&k{F"4:   9+(=FZdBÊ+ݹО{w~EU[&~m¾Njٲ(M7P%kv57h˛$֩@@@(R3P.zN5R6eѱ +k T:kB]JZTh[o5Ku[ٟ3..RSخnWy|rnI]QeԺwDN45dBR?    G=EYsBJ |M7T.W)tE&(lRedC7Ժ*Wn 됌-N8Qqmt6a+{N=T{U;DŽSƒWRH4^Mu$/{AiLg&ޝq*@@@@]zY'ׂDWN~?яRsz%#m<&0``x ֳr6c9p@@,@3"/%K819r'%ˡ 2IܥXf,yS'(IA@@UZ^Kهl''DʁӦMS{o$V#_缷*x*Naoǔ҆\s5V'=?S:$$s:ur|?*n⤼+Xon[V^rS1/ٮzlJ? UFx{w2N>E3xcb]J5Tkʽ**f-S8rEj7|>_FQu~+S*3 J%0f V7R^ ><ӓY]&MM['NxG_ӟ23֋W:hƨt@#NTihby_ye+>?75:Sƥ=Ϸ1V={=OM$O!dB͎Y8.5>En)jfO;&96'<}*A@@0KlP+6*@ޡ  [I Fyzf"\>qΝ<ٓQ o'ϑZL?XU˲JmE &`gO ү_?=N4SUnQRԍܗeNO2*ӑ}-k4 0Y WJsse|5Q)^il\i@a@@(V/M.Ֆx-[4S˙MLY mf%rre[-IMĶm.ݻgZ:]k9Zj}E& |Ufn?VUOzs_s\53CJ&&neN|4jknOG&rm~.%ʫ@*o>'v8ku<}\;@-|:Ԑ9ށdaS4DŽx!   @irA&)S ˼A}ijPZmk?ܤ~VI+ 0W{/1 +́78묳2Vx }+_0^]giE!h\K5 Pzc?uɡdz1\e^Bd)zkaeP15dw ^M2v2)qB30|A@@J+PYZj{bb2 rDQZ2N*Ԣ?UfšjEMaN?hoo?wuB6ԧL=n;idjZ&n3 Č5J?K۹[Kf; n\ u2dRx;Qr[ӇvӤٌ堂: Jz@@@ОۋӁkL_>e^K yӸKuBϊzw^zrw2ہpKeЊ9}~{+fgh<>- @@@ vIi^#)So߮u_|qMMM dދOJQVFgD\@_0't9]w7⣡   U/nT~O'UdĐ36l>kݻwJuRt•Ë/g, wXzԶY;&e6+OOeOj_M9鳇_j֖+Xl)bJd]۰<5   @{Kh=[!]m)GZ`&tǎ9IuRis9Ge \ŗt5feݹ߾` ک*>pVKpu¢YmBϋ/652U9N၌Tsx`{d`TnJbKWe'CV9yN@@@!LLÀTioNFٮn~;(2pov~ۏ$ Xmo=4G)Xu̍7nڴ$gg=c W-R+躾hE8|/Ƀ7 |.bcc9TPg8AIR  x`tN)'g@tÒD]*/)lW*b*W@Tn21+M W[J ٔђ.ԣ&^j6et~^qVb'1+Vx㍳:KMh]DžQX)ʸ3L],6oѲLqzScN? p @@@J =4f!CcyeMޒ@D TU}+̘CTL36OcckI"ˠ~ѡ=c٤kТf+Xl s1& TjVTnJRj6k~k-rSg̘ܺ1k,uN6qk׮UZcԠlBǕ3Y!] 9R@?]tU)=I鳽?>}zsVQGF%'RBΫl:o KZ>+ah@@(@hu Y]מ" &SIbRG~>=rw@TLSiz,ڬl 2کȬ9V %ٰW;j(VI_y$Xs(J4KmEeȡKq8p@ (gb&4-Rv;Vt>[ܶkՓr>:$3FEWc.¸<40g-]ؼrO}> ?ZD@@<Uz)[˄(ˆ &0t!]SgL*c"&e6Nq ȶהԽ6+xmRqѢE8+_ 7SԬN2e_|qMMMj\VB0S*I9|6a.WJAfN'sN*=\CE>DVvVb0! ֵ^yMHI@@@h-zp2&-veci`\A 'Jv7cJn:YPuT9גg(,} =tDٙ-jZs`Ak?Vk6?dذaZwy+W*=XaVݐy^2Pf1q;Hwb1[Q LXQcc*aʑmaE5)3<&NObNHf@΍B~a2>F RdD@@p ޵̄uTtƤfuT#t@@$vg.]TKɮ*ժ<~ݰZSݫ^Rfj d[re PI3~ 餲N?l'[ݥke7 IeI߇lY'PC@Ih6#.0ۓ˄NoSҴe^SLja`yD㫴ӝZv*? ΄E*/@@@ V҃"WC @PsPw_ƅEhhMVАU\O-WW+]2eW Ia؃  F hh}|2&?66 $pDȖ(gJ©S\o0@ì9IU8=74z5EGE?_M3r_ qJ?,!=&2s뒹sxң0  DW hA%E3&-6%gl1%oP)q :ʕt؊iC8*+Ɨ ]*P4Ags[?L֯[ٌy׻#;u;GrqGa@@@2Dn{(#C*`=I J ')ʋT.jF@@Y[Ԁ     uQA    TtH'n!    Q      @H@tb     uQA    TtH'n!    Q      @H@tb     uQA    T ?(yѯ]vYHni.$o_L'ۚhݸ[4B@T_zĥ|c]:S" zx(dlj̵٣)an~ ug'g0QYSBOַlχ>E$CoԨQtOO:$k&JJso#'̲ՓN;͚8P͖_X{= QBj  Q0g1jU,i ;M2-lmmUG*coZ:B~*?1¤~V֤vjjz!39% SL@@@=0d7nt]+/a 3k~JLWS붰Zx.Big*'O>fA~\V"Uχ&e}l0|@9iKy6b&:+-_>xd.P|CSL4%+`NH   C+tRCʂ @+U` M?M3e^,Ȭ4d&C7rPLAkyYrKc\@+*7lc7 [WCҟTy|L3P#7tywLn4D!}:nT)O>Ǐrȣ&m,)iVy^IB~%2~،'OȡkuL+f|*&\оeR=LM@@@ g 3̣o     xtD6     @'B@@@@@h@@@@@T655b*E@@@@(o.BX@@@@@ `T    F    _\ .@:pR*D@@@@H<@@@@@(肰R)     h@@@@@.+"    @@@@@(肰R)     h@@@@@.+"    @@@@@(肰R)     h@@@@@.+"    @@@@@(@---1 W_}5Y >|<988pǏ=88pN(8Q3@@@@(^kJ /ӛʸ3[.6    Z`HZsݧ{Y>{จ8~NJhN 7gܙ{%}j1GdO)8l(تqX熧VT l{Q->yJcԆx-9 χs&˙9'3ЁOE::'e'蔇{g au.fz|?{x뻓{x;m2)<<JvZ .ɟK65556::(L.N,yֻ'?;ϭݕgv7O;*υw:Òkg}|XtVoVS):iNUbM_੦b5vq_40B=.|Wf(9s&9sNn>S pb kaIYV)gng&T9]qJF 'Xk2@MO-|3iB3 mx^oO?)y;”>v(o;oϦFĉρ7GEP8ۖ&kn[;tD%\'gtʄgÆ wit[8}JrH}Jc7 Ͼ.nذ q vV<ԧiO&޾(m0>NW7EH}뿠iCņs@ʐi~0߼tϟ|r?>{; м.}aeũC[EFS -\F^c# G{qzko,c GlzZ߫ؽ g s솧B&՟>>Y{_| iԴ<{Qkx7bVG3άH>Lr=Q;KDtdʗ>z{9"s9?zΖݟJl:l_[زdme{3kN =+VZ}h3ǘ?vJLɜcn< @F F;`~_xXGqz@Iqݿޟ= o_/9vפ{=. ?ptpizy LWHi kۗDWS1^-ej}**6B}+$v:%r v&vVdPo÷vjxwm-.c͆piW ]߫ (R|?P\He+d**;ί M.κJbc5_r$~@%?P|寭w??,l@J0qggsSTQOcθ3r$Ϋ0.rCvķvZFƉ$ %q)x;9}x;; i{Uۄ~?A3W5utZ2K\[W\`Erf{z[oC8Ѷﰆ3V~xYFyV^+Իi<3p٨G_r 8#J x>JĠ;5y +W++erO6 ΍t<.x\ཨCy>b<)Lu^>Xyç᥌GdgN3Ϙ췭j2\? uun?68.럭@8DUԟZrd Yf(J@wf6b+@@@@'&c@hik%C}\|2#w`:\J'S&@k.6(}O6 ΍t<.x\ s <'`x>χF\ճ?mC' ^j{i;lYCUTguFBBx_ Š    ȚEY49+WUeh}3Czrؙ֕kRkM}L5R8 @mk۳[|8ZI(=)@ #G< x(QG8= Jk.y{jC@@@@@ùީ+qwη;>ة}3lµZm<ԟ}BC*%w*gvNX9:@;1tx˟lm|g6sOb]f ] Tꄮqpފ`RL;3~`_:h쟞'NgQS"2km=ivxU󚅗N[yf.;?F*Nr}P5cE5+7.p͗睂6DG@uttݻwWr{ 'O*N봂  O9kr]\m|h=zH)2)J(1RL/݉ }LKn12    DCxݻ~wg4`%     ?6JamKS@@@@@` @lTlo8ڴ    [F@@@@P xvZHgd @@@@@ 3M gG      30!    tTȑjO> !+]Ammm{ݕܞɓq---Jͯ~v <x @%W"X礟SNhjkkkjj*++ɱJ`۶mΛѣGJᦦF5P D>c{P  @09@'@@H^М]q d, GH)P__G@@R ~T򴋀s-3u9,.h zڵ?v_~)lhhꫵm|D@|&}±     n#zb7sj3׿>s=et颤cǎO\ /uFE >hvþEe'    @ҹ]q>|{_> 3~iU r_W-|GS9//yG~@kkkَ#  uG}}zÇM3Y?ҥkٟu'œO>%ϺҠ<}ꪫ-[l-Pٙ_j2?я)Z%P[ħRQBpzvttJ4]8y[ZZxg@@J"~&ښʒ F)*q=Rkjjjllه{]|#~vTsW[5g͆m~/v1N>iK]y|?>޽)+x68\AP@@@@dQJY^pH맩~=GVk~}eZ C+֭sUÒ%K;0Ϯ(AEZkfIC    @l}fh_z_*{sμ5=7 }kAThWkwC(h%߰NDNyr #  {~`}V E ˧˼3^c19inJIڡ| <=}t]tuIA WhF@@@)pݯ~K8>7Z6m7M+1>9RY86o찆_}pE1?y92v1h?    >gzAe+w`/%֕cԨQsux… nj0#"i"ǠM},OGy@@H ~>G'o/o-kWwE/(1#=0GpZ7ϟOh?\E:R}޾hp1cФp~P@@@@ Š'y?-9*}V:O7Ff'Y8/_ޯ_>OsChjþ)ev}OV6YtP@@@@`R Z3֜-cFәƦ̛7y(irG{KMovX2c1X-E@@@@b=JkN9<95ʜpiC[)(rzCAc=sԀμLg}Oeȶ9]֪g缝L/@A lz塭m޽4]8y~*g\}}z@@/~&7*++?"ZDmNzҢ"jf+kF+*6<=qOϸw_uF}{9rCӧ+}V@d"  `t΄D <+ AkW^23槯}6eu)ܼ` /0Gm>׿2e ! ĜJ xYپg2Fk :jgUq}E?Śq\g:l۸ ۱!KmaI!  @|?˿T Ahg=z)[Ye=oP=_~ѣsצ;JrʠrNlKn۾!.unen$_Amoxy|: ?~P@@@@A?%)yGң.o%N hi>u6lXG}ݻx ~9c=F|kf<}ןyg&nqɕM7- *A@@@@ ;v1b&{nkk[p3%eȐ!%iFV = .p$aBtԀ  @Iz?Ï<%"G@mr@"/+1rFg8   @Dx?щ @Y 3mpD2X@@@@@ l @Kna?     @Y dR%!<-9K24Znooٳ3sRr@:@@@9O}TVWW䀎Й@W-r@s2 B\0=F>}a9A4A@@ c? ЭhbGIanط"     @6c@@@@@P 3t9BNg@}}}HzB7@@@݊Q@ @DפE@@@@VZb d(Cxp"2    @86nooݺu!Bpq @(C1hmmو  #kFi;v\򨣎9rdKK˶m"=:[M Rp=(    @{{%K ֽ{.]y- -Zh:    Pnz^c޳g~-](o 3M =8    իWݻwС < I'g>o?6 A@@x?_n3x%,֭Sg=ׯJġn޼9Z#xgڛG!    _`Ϟ=J@smmmzm555&ok.mQ!zeF:C =qzۙ`酓'996w}ˢ Ԁ  @Iz?Ï8nɭN5% "V@mۜգGMMMNjxr;{}a(a-޾}(@N>d]|:8[ZZ@;<(r0ihh  d<hN34@к˭AlЃ~@@@@@_wq@tupFr{&/)8qTXIK  \%:6V@mF~ᕅ-Xڏנ'y TXRp|h@vj  % ]B|p *s)8#o@et  |![Z7-3(@@@@@ -=ζEtJ6     1+ -B@@`x?'!B h^ԉ     P h!@@@@J~/3a    p^lهmM \7NjHQ]V @&"Mہ  %|ɧ Mo~I Z766>ܮInӷo_s x'mnطx@@@@tFp%aS@@@@ .(/GE h#*s@?(@kkk [i@@#y?z@C1 5553g,Bh 3mq_@@@@賊Y9/@<б*     HN"SQ^{ogGno!@@@@țy# Tvvvܛ7n۶1bP8eȐ!S qֶgϞ ^8y+g\}}z@@/~GuKnuuu555Օ-"S@mۜWңGȬkSrٲeW\q[L=7tٳuc„ \sC[+Ae]O8@ht!T$KN  \t'_ Z"%p$=()8q,     @<[ @T@b"dt  X1\F LA@@@@+ɳcs@@@@@ @<9LC@@@@@+tt9Og   +. #D h$4     `6y9c@@@@@ 3]BPF Z0E@@K  [$ G)8L b    PVA;?SZ%o&Ѧ˖-3ySܫ_L;I[AT @Jv@@@@ hG]s~۷5f3*U.0 3t9< @@@@ Nb7n9sfdK 3mq\2^<pBh  @Hx?d2gm$@<Б:    @Fx"=Ԕ$:b !q @ʠ@@@@}`Ǐ?a„,Vh ,OM Rp̨=z8C@@@ ,2 }6_tE9{l@Q    'ep8Gy\`T@1&G1!*@ggZώ^Fi@@G@e?7>g\E7nhR9+M7ݤha.첔=V #Pi^Zrl/WifA퓏<~*q{lKKː!CEyB(`ݻw;wܱcǫMo\8yaj@@@ ~xDݻu֥KڪV">*++m漒=zVTW@I+I^" g8!  kzz M[La     |z\-3B@@@@@D>y`HD-,  _AG(@<&#Gh@@@@@ @<&-ӎ     PBxKJ     F nh|8wXb   6χmF @@&)88@@@@@ -]rP:     `&7Z[[  Ma .(}t?PL}RpS@@@@@iZֽu=>P8eȐ!S t${޽{Ν;vx 'Oa=LC    P|޽{nݺtR[[[܊?"ZD@eemۜWңGMMMk$[EH (FJOC:$C7|    Rt,A Gc(@|>4i7[Ȩ8 i8@@|YM7E"*tD'n#    qgZ83ru      NxM"p+B%@@p.yVD(@<ХҤ]@@@@@K nh|     @67HQ     q @3 JUy #  @xx?' &g<     PrxIQ      @3 PaI!  @x?? QqQb W]ӧϐ!C\Ba)Бݻ{;wرզ7.<o[ZZC    P|޽{nݺtR[[[܊?"ZD@eemۜWңGMMMk$R:`R 蕘[4ʇSt8^!  PZХ %"yfӌg41q     a޸iӒח8w -q{B_~ljc?s~lM(Y|-"npw!  @Vsr _] k׿fm+VlYr}Q^]:t>GE:xRG4cFŔ)\<_[137'MHykkoڼmv5p      @xkfUf\K W\F@@@@ !?ݺU۪}z{ޝ|VpyyJ"Pe;v  @x? @9@Wݾg9*ytUEG92F@@@@@# ]]{Hq ˨sϙIaߘc@@@@@ @z衍 @v! :Nvþ9*G @@@ x?_ΐ@ @2^~Awx_CEN#    x>rĈGUV]y4Խýbqx;@@@@@Vԣ[.݆lڲb1kA{ĞC7o+]C9$j R t&m͍F&\8?#5   P7-yc@^555UUUɭ$QݳgR oܸ_~k$pMܯcF~7<޼?3vM>Q#Rs޾Q,Ϝ   DWѝ;z@U:m{GۭWZ{֍jWmVywrH]ںt;h*G}{ɣe8 z(A@@0~> @@r x @U88B[    R q-Ơ@@@@@3}    X hm#ich%P__TUԃ  E|i@@&F     %tPԃ     @<&#GRԆ@,Z[[c9.  @9~f1"Qgd@@@@@  @GzJ<     M Rpd     iM RpD/fs   ~>@LB(@bZ#Йz  1L"C@)4)8y+@@@@(+xIQV'1#pE@@ ~4 Pccͯ5/o=m!    XC2ٍ7.[핋ޮܴz}Q:tm?;ܿ_b)FJs}{rҥKmmmUr+h:om漒=znjjjllt^%@ ^]w/u7Zgڣw5=`+ U>o*`nطj@@@@@.s^]%{n\^yt^YՁ&)8\Q@@@@(@"{ǻonֱmVio4    ^zIaoX5&@@@@@|[ݾz_հ⥁& /2P@@&(A(?cm>Ǯv燓ù%@@@@@ @V mm] &E@@@@p+?]WWwPmzSʫ:nl؂m@kkk,Š@@(ϗ,3F@ Ю~ЪGb<9`6@@@@@W=9UUFi     @yk/ֱA(Zs4     .?]YYyh'~ A{ȡj AF=X&xr  @x?)f @ xT6oum4 X @@@@@tmM#]"Q ~HZ}S$ CE@@@@pVK>dGA+Z<@|N(Z[[q@@@ ~>G@pQ G>Qvk2*9ba#P)     3hőF|ꨶO3گ{UGjW]U3)8bv2@@@@ޠ0vG6+>u3yg5mc6zqחj3)8x.g@@@@ߥ!7Q>׉#wcG6ѥ:fR X r\ @@r|92cD.%m\WSջG5|Î>HU钠$ݠQ@@@@@#P`q!M VV@@@@@@|L1   |D'n#e%IQV`@@@@@ 3mq͚     e%tYM!E@@@@)IA pm*!]B@@'wD@ -mog#    [IE@@@@@<ФE7/NC@@(yN @ 3M y@@@@b/t짍"    &{R dȸK     ۷o7o޶mKjmdJE Z0E@@K   pOua-=D@@@@(uVv.\5 ! @@@@@SWcUO~\dCO!hmm y  d< @|ƚm~3=D@@@@(޽{7lz5V/oٸi]lM(4     `6DKm{gbtj^[&)8J84    hZe[[lӞ[vy3  HN"  @9+@/1x -8<]p-9)8JO     Zwf>V߮X֚4    莎V?:55-*(#@N'={   M p^h{ڃԽ/Z|U1onطbN[     );w\qM?^XnӖm۶MܤݰoEk@@@@@U흹?[mf+@G@@@@(Ow6o^y[/6-jVoٻa;ŗ58.-"9#  }C/k׮5[͚={vss駟޿QF*:ܳgO3h{=yV\s0` jTի7˗/8oqϝ;3 Fa#   P(x%BQ/1P ]@s~#Z8J콺PBsϝt;ahT v"P1 x?ϩ ~xN@B"s~?tP)ZcIrtcʔ)Br)ߜ͛7w)jO!z  G nh|9{h3fح[*;mڴH /^gzo- A! j@@P nh|"Tox!M`=citϞ=#jӟ~+Ws+JhT#zm@ ~>sG@[|f"x5k睧DϺƠ.繞8`[~]/~{gt&bb   xIs>"%PXnВcǖlR(ԸNBШrV@@@<ФYI@ܵ+`|Eis#2r{=;ĕ #i   a`к|S :e,P___ƣg%y˖1cȿ9}G?W4j% rχ| _o<nȱ3 fM<㌛o_~طs=;Ts+PBhT/B@@*+?,ngIw6o.ɳc+ZOh@ k׮եtN:) oJ =kW^́ǠQ-j"  %^Ɗ,xwl0wc%(8HNQ ܹS 6,sVJSyK1hT y9@x?_γ@x @޽WyA}#gO:{W_A{_愣@L?7"@k,X칦jcݎ.4״W&dr@@ggŬ~䑇| \=m֭o$9jʹ*'   o;vyu៛w}]wnudq܃j:) ;Q  @i3W^1o8?έ(K⋧hoh2wsAbE5'UgB@@".?}Ajjjv힇pCy=NZ;w޼pi\˹瞻{׮pߡ:) @ns _QuK5vuH-! @:+swt-8?;ěo|=wTj/HeG@@"-([4< u!{ܳ7R:fqx- DQ@ u׬];oܳ>;/Iǎ[UYHőa^zC@@@H 8 @1 辝|?/(;fLlGXIo޼@_=]Ԁ @9+@/.trOY; M`md 4#VLXq'#F8 T"~'5kx-N\2ΩslqDhj^5c զ@@@ @m&F-l]?  tv.]dȑ\QHb?nX+4iE]@|MnSuޖ yu*˖( vioJk;묳Jv>L%_)S8/dU5r8   MN@PtIs}Tw`-sĉ/F!Iɵ+ECƳ>;|py=^8$v#]5ʩ @@@<&#G"P*zʞvPb׮][PV( N>-⋵WK:C594̺fYYDoV=qj피f @:zںC8CT}ꩧڥ5e _7M{kFWǮXz~.Iwϴn6+=n{n[SU?A@9@/~}z OR1SOWOgΜ9i$-UUQWg{W*= X;u+< 㗿V?URO{$W1Ϳoy-TlگuNJv&YU[ZZr$PdVx3(hkvG~۔` 9xeb!ińSOiZ'ꪙ5-WUs:cŠը雎k2NGjFm   @d ?E!l[Z VHK/5pwYl+fmڡć#|Մ"ڊZM&Zf`ޖ-[t_2 tAU]|M_otM௹W!^KI":jMjiQ1 J-=fЭjL[ 8S [{3f'   _M Rped \#!Pd]OȩYl/(Fl9ZlrJ#=YXst$s"QGoK]osdz=xMk.Msf{JsW2U @>~@ -mog=@ ,&EtUimͬ5 JʑEfǴV:=S9sxl0Œ謪ZP ٴkN1x`}K&fg{1SҌ*>GG/ȭ͆  @9(g]0ڃvUPRU J.%M=σUFI! *֙X!&zUٜMUN;Zlh15㦇~J=),WJmes6.L{b)6lu5 VHZR8N    3xIӔ ~[.@"O~@Un*Ĭ&ۆvXb VSv(R$9)UR1\erA(pzRo?Tո̕Zk|uIZbwsUCeW:꡹2dT#gu4){5^͎U)N(  R`ߙRF4(CcP&Бݫ%u;ۂ%+/<ɿr(@"aqEhwnO/~bss9}qkjj*+*ۊ>#LEgjmue`8  zGEڪz:@@uuR癋`2ljjjll{,@/'έ( >Ί#`_zXbHHˠJG@@")%;hYߦ)IaߊC(@rAi6"^R̦c֖#%(P `TG   @d\o߾y#67y֭e\Vi +{9F@ VT :TM,bA-E,0ST=" [! pnZc+^%?S"DZ@ӑ#G$.^kt38@(XX{if͚O{eV@'jw2@@@ @/oY{u5XU@ח(b+Kz&ކ#:9@@@ڍ,.sȪw+޲5Rt@ &Yi_7쒬.? ݾ(,:9@@@12zLb 8r/a2Ƙ hnr'ΙSxc> 7o޼/~&O 7T;Wu_'G  [! >{Czz_6σwo^z\9$pl@!ЙH55U9뮿>.i/Y\]]y@ 1TOP~Uur   uhCm^e:a@ 2uAiƌ&sXӌOoP D,U j9v̝\0Κ5kۛ?5ՕʿU'.8   Gښ~8d<_=\lxɊܜ\ojڤEC qc*CDU+k*uݍZk׮2Tp8䫦]}MOFSSoUbU]IQ@ ~>/@J.(^hpHenzG8|WG9*q}%~ǘ4 @;A6 r-VAFp~QǍ[[e-F8Wu^'%@@@x8 @뻹G8q5l-{^9i~zԌ @@e>Th/ʷ*9/J5k֟ׯ]{I\W]U\3c{nf9|r^"  !.->O{7uxC IQ*yEJ+AUWn}Μ9U[_bŗv=[%X,Σq}   @q\kkǎ{'wܘ1u-[!Gi7r3F ?)t{twD9=w.U @+GKf"Fծjޓ ;Gq.5؊>WVt>AjU Cg@@@B'mnطBxQ' @$82BWSꓟ;O?}޼yN*C S//ɅωR?[^_P-@@@  @Y!Z!S@?^8o_{Ys5k'&{~ϵɴF2 BF(oϗ3z@  @Dјz PHD"d ~<}3u}/~?F&/إ:}6[CL U   g$6=B(t vZSUߣwjWr]P#/Ϲyߡ{ߊ+ Q-j3C   @4}B B 0\0M$H&VZ7:ڏrQnZ*uE=ИqY][]r55UUkZ@I4ی@&)8z>o@hi"tm2GjE?sϿ4w䨣{4[*̙eo~u/ì2o ZH^PK+XµZ,<~?=5?Yξ;C{3f(P>fC ?vs"̼ω'QuD!@ R% @Q%FUUȠA\Ba)Бݫ؎` 'O[]!rWΊΊν{;u#oߟ#hrǝqO?}ԨQeϚ5g=?pi?sλPek66J{x,V>O=!PTK{:: @z?Ï9{rڵkmm.a-NVLgΜy8oƒMMMN 8vxdCBQ W CDzyO[瞶/??{ԘX/VF9&Ib)3U郈jPC=  d A JzPNbR @mC}0z<[]Auݒ֚U]jjik/v9dh2˽Y2ZRS]kԨz=эDwuDq#tdH<ѱD$\ՓatI5z0z  e/rHQ >wypY3HA9W]3U ݨLBMSh[PZW 3[&"+ըb%"nw"#G(CvЕՊ;Wf?U   @"4 tR חiE {aÆtCIZN_NޭQTA h7Mh.t LqOTk;'JSZ>`"Frs:YA>RPjO3=B(KN .v@q@k/ & zab & &BtrreH3SXر*$HtN.vNdNĝ{U=Z|@@""IA ~to,^xȑcǎ5R6Md46ĒTC{3U6kV2s2 :PՋaj`CE   @pq @ | <֭['O|j5D)시U f!Mկas" |䧐 e @Z/l^e@&iӦ{޽Ʀ|З_~I'q2@@@J!+y_]7iڡ=Ej,@ =V _~F@rt8f{c[+V=f%gH {hת*+?˗+G('çuyV\qYmeu&at :T;V~4% @^G%1uڵVo )(O3g^pNXAN+Zޚ'Z + =c7|` XKKa`T@R;w$],X@ٟu{STiõk=cg͚={cƌIZ*꒴|7Ο?__M5kր6a„Ig=l0%#EH֭H;!  A lN|m > Ojc6lF@oX@ ?~|eHʐW_}G=n̞O?o믿K/Nk+B2;wɓ͛73?˿œ9Ni+ U轻^a:}69D83rc@$V(A1 ,_7߬矧T]7b3#Z=eʔo]yl;vӦ ?o~󛋛'E@@`o߾jw&uԎ-[ܼtc@.t^{s%aܧS0TQڪŋM6jԨb]6m(l9sQ$Z-8;^`Atٜ @@ Эn߰QɌn-[#E@ gϞG}}{%8A~{)V~%\Cӧ77/U3N?C٢QhvG@@wZl~%u+7kIA `)x'3Dk׭Ӫ.Z۞@oÜ9/\u7pGR&Q ` {G۷Qb];tB|'@ [/G?ѺN@xZ) Z9iW}vInh@@@;>>=LNN&9"ʱF$рD%ꊊ* .(.  T$A^#J I dt2{_3S4=GWwWwWWkgǞzOդ<󯶶}M[4mO#" `Z[[,Y2sɓ'c@ZtCSSK,YBh  @q:T)c{>y `@ 5SO>ϧrgؾ}e]&<0{k6'H:2x+IesI栙7熶w}W/^{j) &UW]7 On᭷߅ x?s@ ex䐁=}/tR+YPxWT÷U_p=ܣŷgw}beEvr$(+5XMMRA]X/_`ZVA;K  $H~MJp1B@ _gΜ9vTOPEG׌j$9;eg尗]vB޴Yy矯*3xڔК\^l[zYCMI^xYOm6GdYnCp:gfdJjڲim[+B;Sl~8_ /ڃyN]@@H@t@n ܜ8SO=ܜ zj9/޶mۯ~KN~{… ς ttR%*jJ>wq滊ko7niJ`x SBnM &n>WX'-3M9Օ( 1gBݚ=7O+6Ln:{Yu2&WQ@쮪W\qų>ښ5:d_|O@)ڔ݈bN ^Pߕ+WqI s%wsM7%(*Ul*_VjVBoѴS%.r͵4QGe=ǑGN;Mulz:n[5MnkS!ilnc4 (#FcӧU5Zmo^~y@@lx'6e7"7ӧ  Gۗ,Y2uӧ{dJL#Zz͚[U#9a-Vjt}ըQ:Qd=oZp|ujEnGq{BgV C+9 VWNe*2lPkEI^  @L1J@ hsmm-7<|Z0o;ҳ"`Slz *p!_ee)RamZ lbQ|!E%՛ՠQ%vGAEV57LykDYXʹD҈#뮻'ktxK/  fm qp@ ن_`Zwwh9YF&:cJIDfzx&50lԦzF-}Y}Z|%\{UC4p[Yl]cE-V2Fzw+-U>SNQXl"vw;ChV\5L 궓Ǝ_ GHvA2/̟F K)k|@#/GNgz̻Z؊᫪Y}iXSXȂ&5!V1+wVߦDP֗2+54ϫJ-yR=D;VV9 jK=yîتahꙘ o"vw;C4FաECk얮@@@ J*h]Æ@ 76l9w祻fh"yFzȬ+tR-4hPfGyt-7nܦMƎQeHRgqg}u˄:+0g?{Ɯ΅@ 8~G6eLYYYyy[TTR?+yۖUUUvv ˉ)A f +d|I- uUդ91`ރ.:ssbLg{X! d1r@ w@ȝ+"@ P|1ӷ/Ϻ.^J|KuUPÓI! A m˯y[j1@ [o?y>g+W>lQͶ7sfytɒvAg۹c  Y$wdתzWo /V Yt 2T@ 1D׽g͚5lذz`Å^g].`P!W  /6ێ_`M T\+I@Ӛ#ҍuN:)G4)d\l9s^\@Ҫy|2 rR9yڙ4 eq6g :EG[f͙g\dzms^ ܴiVAS#/Uy̕+_T@sU   Ϳ6+}: gQUPPѫOU犊#Ft@:[KKj744a˹sg'?5E3TO/_} ׿뵵f[XPW̙3DŽ=KKJ}ɓ'q4E=dܸq6m"LXqp]wSw}wcG |;t}<6O /W`L?)cf\QQm6Rv8Zxy9F-N)`+N,}6#HCTkjjm^4s@hTVV| /۶m_o3<~zСæNtxflq2 nk-[nB'>k׮_G.:tq `YguE ٝfT?5^ŗ_-)TBp 1c9ZpS ]yzT"Љ)@'>g*&NC N5GطvɣKoW?uL i!s=7|˹7l0 q7_}* ڼy NCG`o8bPa~K>/}ˇ~ao}kYg#ayvq~Wh \(w?7彶{wG4G)U]pӞ}o}۶m=x饗f0}YkTw;s9[O|_z!`O2er}[~}H [".P5Gyd͚yH  @X-bxꡥ%f5@Zo3Nbg^_vsJ:k_|E1 ϔۆdULЯ]AIG/3NSpB@ Z͖X}|$;wl-u䣯 ^8iA@HFt2"xX`_msN?nOL\ 8M6A{tǜrpHuT@jll4{^8rG9#  x'6e7"7G o ~oNYu'z͚SN>oKsD謳κTz5J9rһfϗ zu._NU.A uO-=#8%;0hCے 3e~Lsїq3#LSiV@G豞5+fҥKϟ6Zlr=V3rxŬ66=XvK:7N_=yp$cXB|-w˙` = -$Ag$}րM!; B;n4&Wj 37O3{oۡ b ips5J7h FHOB[*]bbe]f끵SNQ#lu?;N3Q-6M .|XhٵBj4MHQ{eI}Ef_%%irɿpxϜJ& @k dЙJ3 @(,Z*'r-f6=Pum%awW$CC'|>u=%)SzR5O8}h2[-[W=wBJ{_ Y r^=~饗]sZQ>$ԑ]wedf! /cXkw;`ŋ_r%fƯY5zzRWxx91G^8;;ÝQ! @v W_ߠGv]R$y|;`+R4o޼|J^6R..*/`W_[-DU>?/R 480ЦJ6V nC[-ZJl^kzPl4+lDYV߬V=k,}9{lɗ{ٔYk_Ψhj_8R= G9" xL Z32s&65}7u͉17(˔Nszά >[}u GuV+,RJKQV>iim{|mZ2ij(Q2RJEҦ6L'<ciiҗ?АdW쎀͟Dֺ1%8-@D]ˣ+UWfuQG=ӕզ%`T y "f-6OTkՉfe"xc*wlo8~$ߡ#/m=3xx?% ~hdЇNqҔ=}A:~hFS@%3g>s̔WΡKMȥ*YVyɲg)nfVA[܇MK#U4+m9Xuzذa /onӦԈ-1U5[(Of2+3*ġbБϔ~Ĕ,SCt--3:~idwMZI9`hH=-6c0Cgoؾ.0m/iQN:C@yUCgbĈ9wJ[KKKccc}}}CC;wvs]Y4 x^%aarxj;'> PA-mY$.)z3;ETv[1?N"^8c?@ul UX#--G[j9JKy`T}iʿv[ZlcצVZsĸvXO@xWB@@ uެmjjjK)1L\8VJw՗_v˝ S:w>gg~ v䫃38hH L,:nր#ޫ4u/Xyj@sEyFО9LKٴΧ=K n@nT/w7\s56omExT)̙ݷ=z86{S{e5VwUaW7l`}ˑ7QGW3kFT9S%( Gpp F ~>@ i )Y~lͲ_E0:ơ@ mmk֬v$ʦ'1֙3g|JСRl8+r4ֽ4UXQo:Dqn+z^y׾fabf!#=^8@c@q N=~o9~j>@0:Yש ?h 4"ˍ p **׬6y>FKkѴj]i[z}F͗XϨa:vDH@!ZvDIHā@-WnRcذa_\=T :Yrڲd^#&+>zde˖ )J*U3SB2>Ӄj_D(UJj_5SNnުAR4'q;I;Eu\uQlkV#ѤHcw=yyp֬Yn>ϚN5B2N*5lV{!t{klJuƭK4cvG׼p<0 @ ü[')A (8&{O:$˥Nt~K*b%:5SSZBn캖*Yp1iZ븦C/Hxxp@@ mmmknXoVcv ̟Q?bI'誁e`N>W<!:e#KR(3 M*pajq-2)ӪdcRZYj=bI鮃[>r_%&3^jjjT? ZmkkӀae*<^5`I[Z%B\l~瘟:Ѳq^7eO( =ESa/d^876W5#+uy]$^ꫯi^[Juɩ+jw3}bC@[MR=wyH̍@@ 7 W~f~W߬ (%82ϡKg׀mOV[ -v񼡩ldAdsm0c>k])voEF~fumnIDATzZ:JvI%}ٔ >:Rn{D{RjCSpCakO5Y+%.aFak 4{j>:/‘P(XX r>__Ԕ=QH fa.`s GN\Z#KyX;ZO ~p*j&sad&aIk[eaԏ$;‘cG,s1 _ zێ/ٰh}}NϿΖfjȔy睚>/z^24r[]wU*ġ]O+5gIܴ[֎XpVWڥaDM]ݭfPOtcoy3`kflfQXL֌?24t!kS jQF2@_v4YU7%P̺]m=-QZI ׋F2ty&6UZ__t}6u9Ai d@t{{ >vkm/^֮k7P#K Y$g53?<춡Jo͚աI۷O~nvM7)p۬]8 ]xR9$Enˆse~`h6Sńz/Y5fZ%fw=6Mߵbzs;J-u '-֌bӊ{Vn2hȾ?NMխW[D~yH di7`jϚ64ZSv#rKA9.} d6o7n (ΨRСU_z<3Θ1#3gμ [a`zp~Y[XӭfɹjӪsU 7z)luV]f=vVs󑱾5[Cai򽯀օΦuC՞Vzm٦AS/ ]< 5n]EqMIe:[nM\{%݈ؔ뇽@~{Ô)Socl|gJpt&B g?YzF'c햼 I7663F{PȮuV! تТ-|K̗7yГVUqݬRcPI^:Wr7#1zB/e O_^]$@8GGaքm}qԎg>O?}ڵ9d*^8F}p`@t3ϼ@l٩v⫯yC3%Omu' SSS3bĈ4! ҢU76l9w䏥5yŭy'_i9pfGW7ֽSR_X%ᒡovҭ-[6x{=6;H<-Epc+Gy9:{ п' \dӏÏcg‘eS[RZ_ GapÏ.;lʘr]]QQQ~xYX&zxyٙB-N);njjZڛVfTjd:od半 h;hqQG ۂGYzذaIVZqe|LÆؾ:&T6o‘zPb2/@'> er*;`n8)<CƌJzdV@'sU cRYY2LɈ)'-.~8(ߧY$K6 lfh 6qL: өͱR*@Rެ Qd#SNdpZ8p{t,d#GH]w_ou3gΜ[U3ZZˊ bItJԵWP*͏í?1CtC֭[?+[nꪫz&{ &h'v\~dNJ:/vi4 s ήhsD vEL;=IId'wW!Zw% u&M2t۳n M='&N?3f${#?p(~`WMiW_sϥOѳYcXb xpN@@ b)+ /N<./ +j?mG!z 3Bpϖ*k{}˧zjVO#@g_8?~ס*> Y8qqFq^89)t  ;;~&1KiD$ #~)̛4yvEnqR \7o]Q':&uY6mZpdhC^8/^|eu;l[o^;"vmz`8qbdvl.^z`RKӿuG_I%\8B #t7@T 5!WU+jn):""~& k ?fڠGs'N*Ngڟ/-PR;+W돃/O= 'IdeZl=W 0* meĦס&&Vj6ٳM|7ZٱzSYYmx iflٲ_Oj:5k9rBypN@@ l[K )+JZ+RVTvKddOi^ޘ1c+=cWFҗ6ugtg*fNM{})VEЛl9R#)j @}PQ *`mo(Ml 2eF%r+ROFPYgօrK+X-X ڤտY7Df1bGGw2B!ՓZSGQ%6Z$' GK~,F.J#gg=V[/;UuzɰuZܓRԫRbp$^  vaImU쮃4틓['M0z4˚i>(C2#^?/_1?#oQ|Z3s.<.uJZAlhD;pZkf ,#ZZ믻jGɂeصӉ~fR[45k`0 Z-#ܼb-F6 ȆG[57zg4YV[3ЎQl-@"/b'D {alZ‘{BPO@g uC':ik,Ӛ&M;cҢ;O~n%8`W}۷a&0.Xɂ"3=]L`+&0ՏYiR4܄qUw۱򙅛irJkŬc%KO5&xUfmkhd ݟ*)uǦO6%+4ϗɵPX4*Q%2KSͦjE|*`!r`iƩ1 ]2{WMf (LՈԌaf  H$#lW9+~L6}x@#`j@HA=s%K˱' 8@*L6PȿP֓: UFkjjD ^*}6QZi~YUR@5RbڬbJbGYyݵާYQGz ut3T3#hT1`M_]ٴ#u9V](O?~&/?w@Z Lx@#1nt=ꋒS p+UJN\p}q*wϙФqҼnoSQ%*SkzP%|Mhx =Iit+:_xq8a~2ԩA‘1{< <7Jp06p^ \2BԩS\CxGU,Yz?sLφ@Jn!LqoچZY6" 7y\WWfuP_}fWE-"UC5Ֆmر]{p A~饗ZL߬Ҭ*i9PéSc?p$J Gr +:w'3G05o\n9'.֕&t>dUPZbY63z*nǑPd6kZk/U#>Q%8L 0*`虮=>=m2hʵsYѳn@\,߰{I%ڎDxH@@ g|љ *5#Fd<F=7 olrx9.4-FӚv;=lhn`9EguCطo߳>f//?#ڥ|2|SO=Qlf͚/egV鮹eEnAkDu`^Kk{Yqgډai5Jp$Ƣ|O]Kyq~QŸC#.li GS~taSƔ凷d(,,T9/tmY]]]UUegw M＜M!~?啕C#]y͵?:ggc-g\>X-}+`?\BZ"!En6/DÓ5'l?ffz`V@E sL hg%um`CK7O=Yrdvf~YuWڼj?c?Z7]WWgk}YTWOg?Y}~דfi`W[%jN%˗/>a-.T2hGu h׍(W@u'؉"}8/yV@;n1CV:w1*Vv Z_=Y_jb+>3[GwS3gg=y䑦sLCZUCco:o-,ğ;G? TpxÓI! @@Bid@\u}V8uspwܑGkLLYٱBatRe&2+2jj$ C;= #iA5PWJMϫgt?q&z̢|С%GJ:˝z/O^cΜ9 C~#GNj-5:0#F<>h tۻ@^8< 'O7&yoGfx[CzP ~q=پ}{\򊵄YtkAӡIͦ.3>릂Tk}/YzH+fha57*%}hxh3럓d͙oU-0ޥK&=~R?|ȁiwuOhzlH`ƍ.pN1sʌ@@ ^ M JpX Z6naқnW.\-K.d҄wM3^z^3hp~L2M?]ᩥz0e˖7+Qt8yFƳb }yi4Cdm90kFhaHLG ֘wNJ,g՝ׁ􌾫6RUg0spN_MƽypYaL  U>&>khj^mM*9jv K`e%{v%1n h}}}CC;wv#qɏ?ҋ|9gOi]-_t_yIiqAYQVIh t>Ckim/+.O\l =UGBpZO*}gO8相V]O/䪁\cQW5jTdzuA}ܽϯg>~[]n|deo[ul9gQYOzCP"gʕ5 Gw>e%Ņ!gnqhwYIJ̊`ŌA_YYIc+5+ZHVtغa4`  .DbqSCF t:M=uZU=x~w^8=żpxg  8蕯ύ-n[,G[ucMmyWw?QS|ouٳo׾:=3< %8o@mZJ%钯_ 6G_|~YC ,)WEz> -{Њc6 L 5ZlZVqM5UPBbh JjY*x5:0q+#2mSHڔMAY\GAHdeJr*5Y+7u4ŦMq=6ZmͰ,XS#^) S  q#̬hXnpYC'[Zb6dJ@Csܔ tT*7roչNN2U>D37\{k`ޅ7@S=5Vf5&5J`MvMp}[ӕ6ؽYe:LSE3xZP{AC2GQяȹXGԎ] ԕlsSAMZnGՀ6rO(hUD|饗Թ4<\yv]3-/Qw^*@/G UV>ZٷE-yֶ5ا(o`I}ڦSOc[J)W!ƶeӨ<@3̿AL$SOӚ_|N:i͚5)={:WN<,P^*Q=)#:tX.עIt:]M`9s ^8xșv#`+3DrVop@|tv5䛊ϪΒ{;B+v<  C ^U羇zs!Y|[~9糟גgݰ>ŐWw]կp4~;w}Lp‘0G@ x3,Vu:v&9.M@A~^8~ }0%ZG?qu7?C=(|3ƒp s,72dPV@|nZZ9fnd/pd ~>>/Z#dBˑ ONWJҢ|_unK}^|y]?tȠAf鷒wyϮF5 ty8b|覚pyf if6>d"W BbЊǙ,2u=ܓ3%K>'Nɭ[^ue 6^gsX[E%OB} CeCkcݲ/9Ճ56 f_nq\jkksH4^8x뇈  @/^ M Jpp#`S lIlG bEBeJ?g^xiӲʕ+VK7oM~ '3o4q'UOXº矟|f/2kT6O@ZbǏA .%8ԟRa5^8xC ~>㧀 ZmoP#执䠀Y[a,o[~lj'f͚QrZ"+h̘ 0YwbS0`ot͛u9{['N4k끕kmiM_V3v1v1-oV-ӧE ,O(f"k5}i`Mk'fljd\4_u]r5GӉVdPw"Fׯ%OlZG#6px3A@@ ^ @!@0AJ!#jpQBŴK/eyWsuO?:nfz j`5}^}զgLK.ECb1{DϮCҿ0'pBP*})#d4M=M5zLgFo  @ x3G^L (dQ]˳ZŵZ,| 'Ǟr"fϹ]{6e[ 9Ũ寬+OׇP-/DKO u=TyTLO<i`QJk棎:JVaZ쮘X5B<{͛ZC/X@ ԏV^GM*D C+~(7+Q4$:f1-ut{ƍzP{g7>fY5HsRq&MQyHѵG  @.x3G.\H?tujqgJW-{u_{쉧ƎWw)LQܸܔ)…Æ+nK ;+旇fQӕ>&-|`0oԩ[lNoV"k@+?5Ve4"jjUҦVKpΦE\:Q:TkUذgTՑ7:1˟{رcOG^zPGfg.o4yr(co14[W^8*py^ @7h1 @LSTi\.5ˇC>>~{=z)?=[VF_wuʣ/_^[[4P:gQ˱nj+Ա>~y(- MDETv{spe_}QG떕i2eSw< Z,+.Qw*fզP }W+ZӕUΫf)ӼtU]ZengE5f͚u|s4GuXkjjj\ iTI Av;T8rd*&//1h  UgeLUV~1E6٧ 'ux\{%Xe7v>|xbho---ZP__5΍SƬfh"7^l飭=ت`[{Rɼ];wl|gݻ{禍oWz P: @ա]9|d'>uԄIGW<ѬS\XS НcE*˼=-ņ3QdĶk,k]Z*fU- YUêhZl#6̷쨸VX芞UC}FC1[窍Iz:Ί}Y_*5GQlYAPlRcp/uP#ͭ1XM3 @]~_aa]#]v򲲲Ң枙2l /M[-bK/뜵@K7 .֏Iee&NAbplӃ}M^8 z( Dj&(JCrtu XV:Άrg}-zh4 R%|߼k֬'p~}l{Yg5We| C/s M(!R@4y:өͱ)3mB JxUZK C-VOrrG咢P35Z\ 0݆Ǡ{ s<~^R߼esk$5VvN3qҔ~?a2h~/q@98覼amq -oƵ#otE' @ {) Z-mG.TYhOI(VTG?BݚΡc?th--U|Հd}E=h/)-:y&zfs @8'|~Bۗ45IwXe\^ٷUm.cnjF@ pl=siYY. Cp(/C 9Tut/vUfbZo$-*g n!^8Ho@#oiF1Nn~=e2fʔɣ;dڛHƐ}@V`UN nYsG^j<+D:޼?E%yD@X8la-fs, @n8hMtXOM|̌ ӧM7jû~2}): m%8'Α@ ':b Rא<97`x-s 97{&q\tYt=iEJ/^8RK   59_)W4j@䡥]?~ՙȌ  ! ߞh'o_5k?f۟^sB7UW}1N< ?%Yq$ ics?4ùGছoӧ \P( o0@@ x3.Q yy;6QܬL˟ydɗ/枫Y3~  @ x3΢P@@ p=|_QAw"ڻ`˟JBQ 9;@@cY@jn;ULTTtK kwBᙟckOT f?_kE`)<$]#)|h@1Mn7ǐ@$uZ}]pB/M^{}i!|٦  F ԰+ xV |+BЪ̙q=2 ,]3ΜJ B ?s   :oЦ"G̅ qAJwecwf@ccUW_sO~_??n>a @X\ fm*r_" @F"h}1+`8hΛ7ᔩS B{P m"h6@@H7y3 xEoA>3ԘG7yMM)z../aA @@R-ߠGG@ U_ÕA/KE~Gϋw+#x~XshsA@@ ^ M Jp~̅  qD%tڻ\`͚5?;rg}?oP'!Q  @f@gV# @ C(, -o_~}͂$K^KoRPZ  \4%8R~p@otr[>v;D~.) Ϫϙ#@3 J8躺Ilmmm}#  U(\3g~~Y7oNGv˴~3xؑǞ+>ZLL  S`0hg;w^q;)-hЯNIlw;{FTÇٗ6\=h1WCCC}}oל;wvîLz@< Dm`K[}5%'tfSS]/\x5r'*}.g[/XSWENeN,///+++---**o^c~ջxyٙ[-N)`7^qckkv7ٌAͳ^z%&N&@3x@mPܳ/9K/=9Gy_q/3V |}>{3)rY:>sV  }vw>e.?CB ~ 0}@e>Ճ+*c!_x鵯_zPY!7L}fs71H_ @fM6:!ҴMN8VH9 d@() CfK|_TuI's[B7VU:o{ >}"s8}/@@@ v3dEe%i-m@  pZEWk}Gs̒%K<3S/MdG}̱>{?I>>/_f\@@*[xV^̼[)OETTtK @V KqnX(Sgc*>ēe}͞~}v-|\{~? 皢ϥ:}}ҩf. н2@/`+om*Hx2K[}o{)A bD@fl/hͲp1#LJ\|ϛ+P ]T?O)S*t6e7T9Ty#|4 @@ 6&>Ѕť)A {" @X T5PAC+9~YQp饗m޼4 (>ևYl:o yh@@l6OܮqZ @'>PD@>.SYfaGhimh)tү]'_|}9.\t)riXrBEE}}+vР:5υ)i@@@u/1;\` _З[6甕V;j)U?*8iɓ] ;V~9|hٳ S~t١+JKKÛt.^x޼yvmꪪ*;l ܽbWlǨfڿ8u֑3=JL G=炑 @G ,:@ۂʠ@h=n? Z͔GCIiDRsrv<#+qy!Ag~ @ s Zk }?8;Yzw^ΧwV/ =EՉ=@@" C Cr QM\zlJsIaG裨 ?W*y)lX^~>dhj=g6@@p@Я;|?OAU9$QΝ|!ʡԛW|zQuDCtyV@g0@W5jOmE9Bu9p6 ͡@$ZcȆYZn=@ (V@sI BC/O}-ʛ9zǠCtyЁ@g~@g_1+}2I|߃y$N0#d@kt9$     @w M8     @@:d      9 `SlI3@@pvF @WЦFF@@@@ȠwhSv#r +F@@@@N͹D@{   An8 @woЦ@@@@@  M! rh@@@@@oМW@@@@@ x-67( dϖ2N@@< @p@@@@@^ =˜      - )-̸@   $& @  )kc!     @^#    䠀/ ڙ]m|N6mSW6$mWZVTT >f?4C᭥ͷkΝ;;1TVV&=   @z?\Y^^^VVVZZZTT?#@z/^"bs@@@@@Z'#    ^fmpx<1@@@@@ @:Yw20?(GD@@Gx?# TtJ@@@@@@ЦF/@@@@@ EY@:l)"[$4  @N ~>N7ET ,g     @x36u9r2M@@@@@ M]w3*@@@@@ G@c$/@@@ #;EKk)A .#    ZmoP# }"    q x-k4F@   d1l@ @S#.b&    fM w^m @@@@rJtNB&    SKrd[\ֆ~}$U2ƶo}MOE%v_5u#+--u:A@@ 8~zûC/*** opDP7ENϸa))@@@@@#isѸg)Rt     ]Ц%8h    B vms))A T\+Iy1)@@rAp# ;v8X®L^{iߺoe=m@@@@@!;9lĉMm=zXKs^w cF O`vGbn    8(;| 8M婍Km^-fZ5f|oF !G@̔i"  ={2#@{h3™SUMos)Y=u߶zѝ;OC:;ԛFVި&ɃӃ c@@@m܄mgd\f0 _|%o}x }Nat555ޝnDu\JpD0 @@@@edhwQÝQ!    @N x3ΩSdHF $;"  @x?A| `S hSm&@@@@@R! /D@@@@)y @@@@@  ME4Sr8Qg3  H\ fm*r_"    f@@@@@lZmoP#[?ƙq@ 10@@HL󉹱 @:@H5ı@@@@@nb{]k{ך F@@@@@@V7u/Y \     @bЅO͘|⸒*Z{7'G/,-~?[8@@@ J\ @kʠ?m!     m[tVdЦF3@@@@@ ݟAnnc@@@x? Y t8     @/ْABx@@@@@ @gEm qdC#    $@gEͩE;~N3   x?“@ ]ABG@@@@2.x ߠG/,    $@ zXl^(4{np@ pې  M(!dP m;Vm{C/8or1G=$mGڨ9     Г@Rm{E1Iv(6j    8@6}d#    A g}Æ K`v삀[KKKSSSCC׾sg'?NՌC   ~?Fӧ(?F$wҷoߨUUU{%H$vCsbjPSSCm6 v9b  _:.?A /7}fϹy:5     lIU#fF@b"G   c_@@gK,#[z9      @v,J9     n@gQloP c@@@@@ oi.Yiaވ#'\shMMͰa3F@᭥kyܹѩf'?z@@@xz?fLݧOxh@o߾Q@K)夤!&N_ݖ>ǜ9 @@@@@R*;./)8?j3%8Rz9     `G vm9"mT O\8*  vx?oG6 fYS     ЦF F@@@@ȠwhSv#r +F@@@@N͹D~{   An8 @woЦ@@@@@  M! rh@@@@@oМW)l  Mn;#@hS|\     @@@@@@@МQ@@@@@x3K./~A2B@@V\ (?xG8j>s0e@@@@@oC0x c`0t.G/9= iQC@@ \? 2%8^|.M` _^wkwiyH]0Q%8׾#5@@Q=crj@ge#( b _gΓ!!    Y^A%q8S|I+     #"6gO O<=?&  x?@x-ya     @ x3G_L@@@@_Ñg     Y,:OCG ~?h   #@ #?!XlMٍ-ǣw@@@@@^R@wnjRZc@@@@@ B>m |O.4iʸ!$F8   ǰ@'Լwpikm>fa}7]TCWTT :4~P[KKKSSSCCC}}w?wWSSSYY|?  pÏ.1ityyyYYYIIIQQQ~xK8"I I߾љUuuuUUh1bꢩ%9/OY˜HQ@Ȗn,@fp  @V@gib N6Bg%i     "[tnU|R?}"11@@rGs) ^ ز0r@@@@@^ =pJ     oЦ 7@ @ 9  N󩳥g@@:N     @ `@@@@@g|`S[U5***3iho---MMM kyܹ@MMMeee  _?lƤeee%%%EEE-3$) ~'}j\]]]UUeZ"@Ll cN&^ E;R@fw  .yh4@{T2/ L/M    H tE'Uxu0@@@@@YWcwbٶ(_y~<%$B+T8  _H)G`dnf͑ۗg&#     )Ⱥ:oO1-67(N@@@@@x/}vGDl8@@@@@ahUF}vxht?G@@@~>%t ਀ݻkٱsK3Jp$     Лj=A91g0u8^@@@@@H6>p@mݾmHh ;6F@@x?{#C z㖭h1anDn F@@@@H*nii{7P(G}n8x9SF䖩p\@@@@@ $@w}HxaE@@@@@ }ێo)x׮]z>㲦GƇpw !  t+y. @ $a߯+o -6 2;sS#c     @ @۷7Z歿U揖?G=غUoe2}@@@@@ mmm/-~wekϑVy]׷k5}FloP#m(@O#  S@,@F+_YƜyc/}sʗWЦ%8b      @  v6nޟ9vXoHd@@@@@=C>s #>79_f j.cN@mR?7`>b@{9ՀNI     @Z]-)4z'>a¤L󵪆On_Q6wϘ:rH/`tƒ#%8b#@@@,|8䔀ڈW?si3Lm:5 ȡ_ &Nq cˋ r "    8hCA%N2m#["q[&N6}aJ ұTv#    dP !J'?ZeCe 6%Sm-mc@d@ 3r@@r\9~0}@ +5cF;͚1"A"    )T}OYQ(!h*ra$@@@@Yh-v1b*2ua,939r2q@@@@@ IКȡӦ#Z)1/7%  J: @[d‚~)/)̸)A     $@Kpኡ+Gs)A 7 ƀ    9.@ݷdʴC%9NF@ f   s _ZطX7$tl)sH@@@@@ g G ƃ    9(:O$SF@@@@&ڔ݈fxpw  % dV;)eV#    9.:O$Uyf   \dV)đYY@V'D@@@p7hS@@@@7h0    dhS|r1N@@@@hS|d~ @@&Qs @ [ZZ;"    xRZkD  F    9+@s`}Ǯ=A{#q@ p  $y @ 8@oݱ=%[wul!     @ [W(no㴍[2u6z9     %l{5i]mlq@@@@@2(lm׾4}޾k_g¡@ ߟ^  nc@] Z57_cs}] U8zG@@@@Ƞ@RMxo N:ᆑ0@@@@@ g|`ٻw;Xm%[Rn_Ǘ354|9qzՌ~}fС<"B Em---MMMo]sN?&C   ~?C'W凷ψ#".຺:7quuuUUh1bЭm`T``I{yaGf3':hoVvƸATL7(^7***c^4 8M @@4 @ù_@DwQS&MgYjOҐ>k%8rRf     6FFVaӏo&kt/QTӧҿ 継3@@@@@ >u czb>yrm;     Y/T-s'M8jȩS豞{jf268r2w@@@@@ `0 Ta7[n92pҸ~ZmܲkϾ?o,9kB13菫sg :> 7߮9wh??O@@Y?C'W凷e+ 矷?c=6quuuUUh1z7j# 5r fӇF?tĘ#H]s @@ 8M 2&+w;n[@|qkTp*_~DS|     @@|a<ӆK}>_OJ>qCFiAtŏ|h@6 hD61#  x?e ~hU:nD΢ϽL(aZ~F     T }R#    %     ^w Y!:+$nnc@@@x?75߮E #`0 tvSCup@FtOm---MMMo]sܴ$%[X4pX?h\ \ \ h^"yW"8A?C'WPڒ/z@ c=6ZmJTꪪ4!m{9=555޾|sgv sǐ"  @O\6Aw]LD\.3]~    @RTHHNrR#91F@@@@%f F     f+zE@@@@Ck7!QQQMhbnB@@znB t 8cWWWWUU Sktkjjc^4 8M @@, ǰ (N:I@vG3%86     D 8k׮^5_>"    ǰ,@۶~P3l1Fa/9soLK@K  )pq= @Jy92hGHWXֆeG>- mJ䈀'G@& G +~Xs1h@Ԁnn<ƺxM{@@@@@ $@t}ku=,@@@@@@z+76{koD"nڬ^:1%kIlGB@@@p#DzI5xcU/^~׮]---;n sO@5k׫Z~o?)jaG@@@2.@SWnݯӇ,}i{kQ 惭6{ZPϿq @@@@t Sc!@w޴ú)KZ|%{hWC/Nn|]MEF͛9vഉclNŋif̙3~cZ",]@@HR*͛7ϑTWWWUU9 @o|o];Z (Au.Y|M^ҾkϾu+ 2dq#    fh7ƖvM[^_fuUoLlQ+oZyH1l@@@@ Α4)`+-L5C>29t*!IyF9     tO-)./6%/%b     H6Ԑ|     @f /nٙpt@@@@@W 8@ t^~aThWM     _Ze'O;?y@@@@p7o1|     ~g#"    \'!    *@gq#    .Aѫë̊5{7}]n„ vg?ѣGgd$   @uuuUUUN# V@B5w\x7\p oz/.@@@@`tN{\ˊ36 G?Çm~ӟn(**t8:    Э+0p\+?Ͳ>fx)pg>s >bo~wMԷcE @@@@whsz ^ޘ-w8&AS7!8h۴i>ksII5τ7K}K mٲECg(     2n VH O睼`yO?SG;;6@@@@'ZK}_P@@@H@uuuUUUʺcrQF.缼s´޻־OmۺNO[j}{sgF     Э@vYvpKMGc_CӶ&βb'&@x?Oq7p@{߾}=܊+.y90@@@@R# ԸkN @gOu365[fͻﾫ&L0sL=:uƧ@@@@.\8.@8)"   dtV6n'k@{@@@@@ i`     @@ιf     @Z@@@@@ ΝsL@@@@@ @!    #@;皙"    i N+7C@@@@rG sg@@@@@ mN5B@@@@rK/_cn͘"    Eia      @ @9g     @ZA@@@@@tsθ N5F{-ݣseZ    Vo9'LO4Ԟ'zOgϘ7guK7zƇ   xQg}xq~nЊVnѯ4>"Q5kđGv ?iC4Kڭ&p"ws#Th}ۘ\$ @@@PTCЛ{ziA`j]9l`5PFwZ@YY\sWf>'deklKpX Zš|Sرұ?e ({jޔ[ժتI+{¦e.6sgk$J[by13n5MM^߿woԇiIu;    @8e}k͞/|oo3Q#]2G'O¿MY].J' X9hJbxh9sr<,&-@=j}V+c*UCvaMZ=:ze0k#Gy,3Ԩ&j`3Vd c*),Ҷ;JUa]mvSCG>_yx@@@zH] Rd%P#*b=|d}c zl&pL u]wm[kډlՀ6ʝM&̑Fki$@G峑}:@wMo8w {xBϦ$:tm/5@d vȻ^.Q?]]WRGvHXEûɝJ6   H StpR> )t5ΦC5icy;'42q6K"j#-gd;1hlole] O\D뚭~bj0y q-Y>Z˰^FOVG2c%    Z% ̓nGތeux;IQ{pu@;#4i)*^4j5qߵkR={*ҪA\@@@HX`يM<&3;f@hn-:o;]1Q /ɩ)ng}*oD>!k8 fc=>f-6JlTyR"K]UѨK7Xiao{&@M? ǜfНsbt-&fqrU56νg_+{*:]~ +a@@@@6G g2=Sܹ3zRA׷XU+LKf|\ö8jQ{msN*CgxYw&f$    @9JdK lnb Μgbz&攝h cRg'|ˌ :kնĠ] {qFub֜    @OQyJ*v}$]Į^TqD@@@@$pѕ >7m྽oۮn40ip     @ؽ ax0O@@@@@@@@@@ @sE     DOI)봭u{wЃB~ <41':7Z >:??3r_- \    @v>J|= = }mO[Aa ;֢|_~~(-mc[ۃ-~/>k_jaAD 2 [GfwӦ[[jlljO<3g<\0'`9 .l>v ~uy>;ֿ/{ @@@@w xjI>񱅲PlkS+V1fM)n-ק[wlܺ-^[6-&4cЩ*uc}E>%Z3@}FcR{wƂU}y3<&֗m-CCtB;Ju++.lnjRu߽2K[PcА~Ža5y&tÏM3rt䯼k&l½D om§3 ';/3q*;kP~l6@@@@ :)ӨPRL/Rm)2mB_̗f/;g#td^Y'^/8M5&knm o F~9e¸FC8l(nhn۷L@@ssƬ-B|h1v^x_FTDN#}QP;͋ m'G? CS@gwsq-˩׆Oi9hSK[Iquq    iUlϝZZ[Z ՝Y]XX0fmS7}'Ͼ6-h+nkЇ3>1g;y_||Z;jwubDљh;@ 8yl>pΦ)ϫoߞ ?};?B:Sڊ7Z팰M"ܹuFqR+,, :YIt |ڄΜ#a(mL;:1!{l@CG+:GϚ?cs pي;ߞ_Ht 7U֯"oh=tȫ%pp8@@@@poZf٪M1Z=zCϭ}_oߘ;7~k} lۻ:}聾ԓnݵcQͳ|Չ2vѳ5ЇZꌄv@A됒mQ<߷<ﭢpNMi\būfuM)w|XW9JF"}j*HlS^pܱY k.sA~nuɕo]q@ CSbY5^{uZS:TZ7?,Tz6K@+.)to,[EEzѡЭymumum-M-ZߖWܙ8GY8Y9tV]C_ŋ.9W_[Gūkvx@g?xl{͡B8jRQ';BL|[k~`}o߿mP'[rL;3a$澺Q9[>l`Ͷkkߺ.<(s<0׳9cVowk1-;kMXo߻\-l    @ XVYwcW|+Jp :r8I=Х%J,y'a˶-;{`z`0Pװ`/jE;jwurOCkt)E60wClJg솲MoQ/4pl}oUMm; )sx |FwԀ}4C3,:9[h"ZG9|hst(GT%JyLYlhj}׭.lh(/G/(+jhnkmk32{^7?}gOU[/Woݮ:Z7$%#?wXsOKol˲wnW?'U?~۶)}yѹ :FGT(QZD ںZ1    ihk3UϦFCc}{[V@da//|OWuw^%=ZCxhuukSy(S¢y]ǽ}r;4 r/enºsG+;Zsgu okغ>?p/ӂhEPN|BeG)p]HUNW/3g}I윗_Z6۬6a+nۺa FW>_5)(i^xwGA6݊1آւ_fˑ9sKxF[KuBYR\x'=yZأ7fW#jkG}tiev,\Xtݱع۝#Z{Fs%;lļG?\-l    @ :SVy=5Ł>'IF?c4ui~vƜ/~!o Kk:; n3 =z?}^gڋO_y+VjJcV@76574ꁵӞ]|mg8SS:fǾnio jw lWSoO^ssk߾;*쬘RRZG s;{)-S-䒢ʡ7?-[w91V۟`ժ}e>G58:h솞z>"/@^]l|KK@47_iy^~iaغ~Uu@J7mWuש_~sW¶ud;C^[k[o穿 |C~GU[ZښZ .]#?GDo7?v7.[]:sy"~˨KO5S'Ȋ-jaC@@@N:S}Oܵ*yzGgwC}J_/ԥqw|vy]ˬ,[}y>zFmM}t^}}e9Ͼ u+ºV5m/} |Fڰe{6߲yɏ'{p8֟g~*445YlԎs3sL&IM6IH "WP *¾D֟D]\u]. (ۥ BiIڦisLz932'Sص=>LΜ9O><`yUc)=s6lB!԰{̍D _\дҌaAk--Db`P`xXSuY0dC2ê q[T#Su ך3SiLz盙`+dĎΎ^dg?<'U#Q@ȧ[TfGb!+qY˚pIG4ιU>i-Ɋf2+7|DTIҎWzB,}捓_|X"D%;o)"7F D"@ D"p(+:UE'gl;yx5):ubtvq?GwfdWDd ,>0`yXEzi(q]ؖzMYqo880Fs0uE+μ*g4N+, z¡ξ>$A , E@(?xdwﾶc LRD|,`;>Z": uһ1]F,QX4 `7f"sM/XY"0)3 xVI e}F}|oCָ ZǠ539kQ% ЪV%g[1~J\G9՞Ϩ_y9Ӕ_ϞW?;Y?hL"@ D"@ DU\9O6g74WU_qnӼBmS}RuzrvϑB}M};Z7Ot^߶T@V3ӠQHo8 |skNt Blh RH(4=)ݥo3&yi盞Ҳƥo[ǻOtwuw:ۿo u$rwbpLߤ( 7znj\SxMy_1%"Ě!+P agҧYS ߙzr'׈"ÿbO % CS(y3h>CFyК!ip><4ɹr2g'~>nKcM4A-X0sW27)۱ޟne[aF{IЁ-Sw6ihiVMX=Ej=BG:[z :]N gozC15-ӣ<.](9E DtڤRBPq+ -?kWty|"@ D"@ Dx?> 7ܰiӦ ;37n?}&tϭW-oħ9)+Jxc ԩ7j/Z+]if^S5ԴE74_XjRfU܄$8dBO(@{=GO-Uӏ̍F47б9j9(&7J"ӟ{[K:K yQa18D#:m hz374x,: A%ipE]M%p&#s#qNdqp5<͊W[y]F[fz:жv@oO=#%C9/+|9 _E^!ef07Sqyc5eOWy٫?dӧ;ў޿G:QCƘ =nhyvD{̍D7IoBKBV<*Z8jT[{c8]wU]\těŝ²+O=%s+}3k.XT{ݥ>Ir;Zp:vv:l%JQ9nK E'У23лvR.te7oD5 S#8OHutm@݈ D"@ D"@0ܹsoY饫c;o~x_qvxw-kgn,BD( @ Gp^;ZDl& hFb;Ոz1+]W>jWDxSGF<;E NHqnIpC|W90 BQLgn#N:Ȝΰ=btZ3LE2F.MٟY)0\fYgؠ搜 h /2!0h|"OEK 4+š6ۿĮ@D="Fј*pXݼ{H)g,L U.cvS9W@{/3zQC9n R|mY;ʃh,oV `MdˈuW@|VLEoug'>#nV+ǟ2pp# D"@ D"@@Hچ؍g}< [S'k6gMsFݯtwc3?pgIf2F":Φivg)^<֋|Ž'Ċ<5Tq7qOG}&%} $GMCz@X ĺ]y.Nן̵?4młd̝^l).D"H[_w0fİSetxsfۜ*2̈́Kzt4$Gz{|h ـgU`QxiIH63M3i6ss a`uD9+Yv"Q*4hd wOEkt9' hr<5~7ԲTe_.Jc9JGnc 8)@ӷr}# cM dϜu#D"@ D"@K |=<>yZvͯCƖ矼3?ξww\S9%6w8'Cl9rm:#@$Qu K/|{u7ʶ+'.= YN-BOr>;CbpLniD3W!:iF2O*02sem0}AЁMeҔoZ;-wC4j4ǒHyZп A}Ed%NmBW޼cjTգPL %?5"QآuQд -%sg+O XW`NmQiApoY^_Sdm;ʲfØM;_iF&JbSf3=܇#Yy,,zt={_%|[,2P'"@ D"@ Dt see%v`y~W|A zss'~{z}5Нw^s{3[slBa2bx{ǻw7%h\/l Xh]|QCߺ-ߟx㍶w0>;vQT⩀}AgOPta( a18Wp `Iv}=C#5UH8Ͼ mXۤOUtKpMmqy04ƉhH -/~oM?ƿSg|>j/. 1,(#wMgs'ahl>Sb2fsfДu5ˬ%H3i<|2Ak!plЋaqq1I1倶KD;|ڊEȠ6Uas[.-hȓ>M aҳʪE!fLx;gTyr▣X?;~wfUҹL}xTKgTn?fP| q"ixm2:Dp e?7ow }3W^h_3gRD"@ D"@_UU}`.oy wr1&v@^u ma3WTu#a|2+f۾w\E *w|}W<_)hvJ_nZ$'1İD"zF'Ц >TgF8W{* 5 /k6,Z/b}l>X%Mw( |BmZO<573:eyܮt(0HWĢX,ɬɱ#j8d &Ϛ*D|w{imc jO+[pok(.^V{Ȩ\v+Ϩ<Цƞ(^}l e9'}F]_[z[z!'D"@ D"@H0+v>|7g|~ѼwU0Yz f >τhEզ NӈSEf^Q\D輪>YbqB;o;q|U _oswqj_DE6,tzjK}"nMNmȪS+1ŗG<ЗѢ%ְAX{kL ?7TtDQd ,;pS)OkjBn yrB6j6$ѿZl7H-d~F٧݈7aB}뢪N1s$@g]iE뒵Y(}!pg +7Jqצꐦp|`zH\zN-Q,>8O،Nu|Sh* NrX,epm ]N?Ȗ"L,7ghЊDj&z f ҦGTUaMQtoȼSv,hl}%!@CŸR0Cu0vٻϱ7,+sjJ(&rgqD_L뎷A5*| ksΕG}߿|Tdo;;[2DpV͝ `!;|ܓ%8oq8uZVSośnD"@ D"@ c {~tM%%%摗^zG̯Ѓy탿y6}vxKg+P?βaEAo"F򢙕6d8`Ve]+.*(UՠQTXk9z+uQQ.Aú E ٟݘal'0Ky}5xV4#8Xt^fAT^V% ~cS h4wMerSVk<̲XxHF!hȝi9V]?CWڲ'r7~e}毬؃eQV44JjL.ð9\y?}j]jۉA/>tnmf)Zgogb&׌;؃!<t- D"@ D"@L-[6mJ?khWƵ?pYǙC?|/c|9_¿mGIoDt 5K˝0Źܕ‘wb6ף3eS['5t/ <E^6 v f|O~`1xRT;҃љ{M;jSN#z@F =@d,{?aݾBOHYiP$u %;;HE‡b:8|@,1ްŰ%ӽ9X+r(#i3'T[\;䐊`jJ3j9<=|A0/7)ႃg0 Riϭr2*7S9S}OGέɒ7#:7t D"@ D"@r м0m+...g?{_qNA##嬅0JgaJj[ <YmjIع}!BٶYw9__=_g)U1ؘIIO2t ^2rGk!6EΊzYVPF6;#>GpLa]?Vgif Bkf5L!_c ̟NEr$g=y$ 0jq 6`faУY04 9ƲGDQr:mfnc"T)%E)!M\ff̜k#O? qD ERdq77zJf3,khfV艷%$oh]E D"@ D3<СC6|V/Yagj[cPٖ;exba bI1Z4;{؎TCӸqHNy4Y\9V82yέVcz$nB24SˡD]J]ϲ\J)]A۾!j ^f :3x?SaPD,S:Q9I D"@ DTo; L 9Sxfvx;UL4^<6Ս o%kwǯd JiV4nP 4i(@|3&Ƣ Ea׼^_U3j'mw#ۆ=6Vr! PZwi1E'\xf!`e8Z3s"rA,D :7\TXgG;c fGx$eӚ\Un]$/*9cӔ@4:ÈB6L%eBC[9攙V®c߶gpQCY5*\YGbSBsYmȹlFpLԮ ϛu? :@W"@ D"@ D&V 7P]s7Gli2sukdBŞ9ͨJ& Вew@ܰik2}E,-YDQq'.{i0yf]^޷F57+C…gU{7\t4OmCGf7R>\SK^љ'3#5Hn66g,8C~Jn ݞ=cHYm}xJWU]ȶFjBq-l0Ot4E7OI$#l$ݑ`8؟ehco. Srm3=='ʀ6ߜp13*}6 ȹmpG&i߳ϵfoH2%b*"@ D"@ D[~Б052/8wU[hx!Ê:LЦ=y9_^>VtIl>f P%#^-)Rی:}NsxފXیBy-odvdqvFI{_^bl3y++r{A@Vߠ^}\n>5vNqΈ(j2l tcW>u`#nQiy̦h泥9`~NV4oo^~IٴիX5 EƒrlXUFF5ܬ+gd5/ =BҗԲp!. gz_O_pyUN+Y;fv-`š2V$|}_ .VSer+#[co$D"@ D"@'0ݹX^ 莖vߪdCӪ"qIʛO

5RpqaY57\.wnsΕB¦Ye+WYeuU%Eyn:Fei]rfTvJ#U5-_qṵW\ְafqEYx$:γyVҧg̷6"@ D"@ DGi lw~R_/'Opk?saE47Z~5W,\V~jDl+?saM1}GϹ]Uoz;:uKf-? V1>9c:sXvdT%>y|VP$t힇 Wns'Sκ+r, G#X$(2_MvvITΞkW6XXԮlvGSt6wuɮ9hW)t@7(.ccebnTӷSXT>|oI-ŹUDA2|ͥ*kkg͞> ;8v&s yb9J9 _yI>^v͗4U}{֌Vmf~iOʅ5>i#D"@ D"@%g3kC2!̶9,<ِqT_G7\+Gލa g<a_qnpI׉Yl\,:;VH 0]ԅПY|vssϖFL5|A-s#?'bX︗ݬF0sZQ{Jt~(8D"@ D"@ D#pl͌ȡ{=uw]x5z?yG:vj%T߹f!:㒡8P ea7K$XwBn6G@HIs?ʑ/-zyEUSo̳Y D"@ D"@T %$G639&>O!z:vˉ.OYakdEV`8, T{{ŋ6yv3ÌpNf@'RY3?̂6ÞA˔`~nTt" G@ȬjFOK/! ^ṕoxEՠS'<=us/k~ZS54}3s+1R4b,cIlCeZUstsꂌe78ܴvº9+V1 CwG#`*"וdXurD6;S?ۓ10sOŮ'Rg5sk"D"@ D"@YJ˞0|ʻ olI-;i5[LgIENDB`$$If!vh#v":V l t065Np yt5&T$$If!vh#v":V l} t065Np yt5&TuDyK  _ENREF_32uDyK  _ENREF_41DRogers197519(Rogers, 1975; Witte, 1992, 1994)191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975Witte199224242417Witte, KPutting fear back into fear appeals: The extended parallel process modelCommunication MonographsCommunication Monographs329-3495941992Witte199425252517Witte, KFear control and danger control: A test of the extended parallel process model (EPPM)Communication MonographsCommunication Monographs113-134611994uDyK  _ENREF_41uDyK  _ENREF_50uDyK  _ENREF_51uDyK  _ENREF_25DNg200725(Ng et al., 2009; Ng & Xu, 2007; Rogers, 1975)252517Ng, Boon-YuenXu, YunjieStudying users' computer security behavior using the Health Belief ModelPACIS 2007 ProceedingsPACIS 2007 Proceedings452007Ng200926262617Ng, Boon-YuenKankanhalli, AtreyiXu, Yunjie CalvinStudying users' computer security behavior: A health belief perspectiveDecision Support SystemsDecision Support Systems815-82546420090167-9236Rogers197519191917Rogers, R WA protection motivation theory of fear appeals and attitude changeThe Journal of PsychologyThe Journal of Psychology93-114911975uDyK  _ENREF_38uDyK  _ENREF_39uDyK  _ENREF_41uDyK  _ENREF_41uDyK  _ENREF_36uDyK  _ENREF_41uDyK  _ENREF_51uDyK  _ENREF_24uDyK  _ENREF_41uDyK  _ENREF_21uDyK  _ENREF_32uDyK  _ENREF_32uDyK  _ENREF_32uDyK  _ENREF_32uDyK  _ENREF_50uDyK  _ENREF_34DJohnston20102(A. C. Johnston & Warkentin, 2010)2217Johnston, A. C.Warkentin, M.Johnston, AC Univ Alabama, Sch Business, Dept Management Informat Syst & Quantitat Methods, 1530 3rd Ave S, Birmingham, AL 35294 USA Univ Alabama, Sch Business, Dept Management Informat Syst & Quantitat Methods, 1530 3rd Ave S, Birmingham, AL 35294 USA Univ Alabama, Sch Business, Dept Management Informat Syst & Quantitat Methods, Birmingham, AL 35294 USA Mississippi State Univ, Coll Business, Dept Management & Informat Syst, Mississippi State, MS 39762 USAFear Appeals and Information Security Behaviors: An Empirical StudyMis QuarterlyMis QuartMIS quarterly549-566343information securitycountermeasuresprotection motivation theoryfear appealspersuasive communicationinformation assurancethreat appraisalcoping appraisalparallel process modelattitude-changeself-efficacyprotection motivationvirtual teamstechnologysystemsconstructbeliefsacceptance2010Sep0276-7783WOS:000281129800007<Go to ISI>://WOS:000281129800007EnglishuDyK  _ENREF_24uDyK  _ENREF_30uDyK  _ENREF_36uDyK  _ENREF_21uDyK  _ENREF_52uDyK  _ENREF_45uDyK  _ENREF_17sDyK  _ENREF_7uDyK  _ENREF_55sDyK  _ENREF_8sDyK  _ENREF_9uDyK  _ENREF_55uDyK  _ENREF_47dDd^p  s >A?Picture 1"bcU֡B8ܪ~.cncU֡B8ܪ~.PNG  IHDR2G sRGBcCIDATx^ Wy.\{̪F%Ym[ [m\@HuL׉*ȉ%CD!@f1,1U%Y:{O_UuWWwWuWt|#T:;w[g}R| I)3/"D $4H$IO?Km+(nSjc0J3TT 3šU6S" ڼo !Vx'_EpIqq$MqՑ;{HOe#b~9<4Kc ,V4[Ÿ.Oh.Rn Y:v)F 4_~83V<,)_YEbRy9ST4HIEs%E *pk@2 Su^%-@-c .'}R/MҠQBZ G!+.ȃB:nEfjix {f#3V3ݘ݁EZaTMT_TW$aJ%q]_W.ŝr_QF4Z&Oiטc6S)Fmci>"iMv/znx.hQ;h }snC|T]5_zٚD,K*ꅬ@ Ь@@B*-Lĥ(@V_f:h$Y8-?V0>٧I мLfH"MrHYEV ,t7'Ьd+'xX~OO$˃yB^&D%4 f4GyGʨX*HsiUUDE*+YK-Y{yWr\Nt[VFҊ'R/"U[?)*꡽Z! D)kJKuɗYNJˣ؟x]IYjyW/}z-U F͏~̻[,"T`>'dA J2#Hr*#|dNQZ:04.=+NptVBb~}"jb&Z.HE5"]K1DRH/iF jtPbԠ9ɪ&k,ᬒԨj&_\<UyP%~|2~㽧3'!j=գQes] K5Zh95j#MHQ h\8Htk0E°0) 0 Mߔ}gW'[-p |OǓOL>{W' :yݙ􌚓؄D)0JBEES_n9z#AFV&P4 2jj4$x \q5޾ulkK/ꏳ(3C=P $,dMԙR#%2.H&ϑӴt:B5d({21QV26j[5hK)yt{-_تec|.c̆K Fk~\UmIlʉs9iŘr{E91kۘcN$PQ玼jFI Eń5X̒jVS ft%kmYthș?ߝ]屧'}Ѻu=qj^O0)BaC)5vn;ݔ*6Zʞ-!t44Z5*3zSHX?ZBBYx5}Ӧo r}?OR$ O23=384p՗Q\oOee>/DQ@g]HV_llf<ڰ5Zgle:֦Z #l!pƍמڷ+W⏝;ϳXItJg>J矟8y@x7= 3$LVL)I D:"".;n ɜwpc4SO"y8XO\tG[weo>s +u޸i}@VГObы >, (R::ʷU.^zw5 -A)+UMPCH*,. :aA01*)g!cYQ`S|,>*qFn5=`F5x@/d*lASaS""G"rH?RoH8Q\u^ӽ+2K{_f(B$( iYȓ̨*%+*,:eD5)cz1^\9vٲJ3Dګ)OP2L]l B_`)AUISbI]>^|$#~O^SS!0h8F;S/LޫE25&z6;Z&UBBH]>Y(C0RS@@``BVo`X܃@ (h`, >0NAϘ)EI’H -`}4O@O ,h W}9YbkD[wI(PH}=ݡɀ;SP1֨}R*iB,k kJ1)R\/ r~:)F'sa32a*(*T:2%:'3Œ!9!OtGR蹉 3_}gY! JjsJApBwie!# 0^?Sf\F0>' A@O9?_@syU$*|/w}ïpӼZo >P`tI>*Aa`"|h$W0 1G% c$P?,}aor("i|暶n}Y^PoWA> V#*TnQ[e;sYFB 7_Fb9dGD(JjRˆ/Y L.'O Md\VLgT>0i+w쯿!rr|RK3gg΀x_GI/v:c>i-^6cȒB]/>k0fy1CаN.cyN 4jvH%gxhp*4 m'@W$?.D_gh?zQ_Z9?38&2B"K{m[[$9O7}n<N I5dRJVaL ª;⧘ Vn&]'|'@Ob랍O?A,K_yẺ_Q$F(P5˧,EŒزh$ q 2D'9NF#p3Y PVQd"/D6)xWz}ids9^Ni?\ڭi61!vxE}p1ZOɄX^Jjy1Xat"GJ 2/Doي$ (Qէl^򙬘.Y'(K?ثXzӛR0-'I-˧Ir<K*r?{QT'+#]}`K(4 .35I(ڒn Ċwߞ@ K3aaM͜>4= Ւ09OekYhx}N C3)~tGDXRC׿a雷OHdbo*ӓ9?{b OdR, t8fj jg_`o6M %1͜ɜR/~~Pb٧Ƒo&yRd)h#TD`"PFMe|13$i妯~+i}_}R33D+@P.i%H?SOw> wPe0a`Gr lpa#1eAtrg0GG}SD4r6Å&u霏 @1O %~7ҫ4]q7S-tz⚟OIk_;9 O]:}Hk=K%^O9\Olk~}a*>tt O}p:1oɑqdS?{n8t߾I^2 :>Ms_{7M_B|fNfJ${a K)nS^"0TFךqNCe}4 (*5K\>b8yO/jvtZqlTga- <0&_?_DaJ9.,æ|Q_\8\6E_\/ Ʒ:B׼wHÿ|V$?x?ғZX#{fO+XR&` vfIʲA޺L@$H104H烯$3 -KݢDRI2{˃T>M0tC7U?&x+_3~ RXW}/nՆ4Seɿ2S$G"kϝ E kau78V?B_s8u\~)a\ank`6g`Yq>luiG`׆Q`XV^h xIrKM[ፅ4?p"Lw‡&hKỢYQe`j}w|XY6lu6E=a}be}`} 0]W_rZmGm}/E7`FFC!BsTmпieIoО"B4I_O,E2`X30̩c_R6)hFANǃuR~_0t!-1R5a-zay{OxB~0Km%vOw\"t|AV (k:ux'\o~ 1+RtE] Tys!ɰ.v*AT>89]|lrq#XrVHoA\'4.wHq?э~sBT׭Mh04x꧟ڛ} S3!~ǦQ"3&Ne}Q7/ YS+|q#h֨5EnΛS>H@o6K049s…:y /LrPGc>~ ]/ ðA i6Fra+a},:d2=Y&' Q/~\*'{a0zU S8ֱx4OLNezQNΟzӓ}?w$,wi٩~C8 iyK zb͖/A)Fh>3=c>e|:00#VNk 'W:{~/%"T?E$;ͧw=v8E|OH}2O'o}."/e4"μBs^;/9ċ'0> SVJ43F V~pS_'Ppӣ&Xك9: ޡ ^.1X8j z&(mP{!M: ENR2K4 (>u L$)/3PLW'psT2, 1:s>I$`X~(IPb5B /Ϭ fӡ^DL΄FӼiF&lbwRYN @OAIP̓; ,SJMLLAbjbT,ПIP0GE ~.M9A1(krHx¤:H`c8܉8yg1W7SAA_z1pQ/O9yIgo8ɉO?틯~aöe7&@QtV 3X0# InF bY!YhHTx{ZY:Id$Jd&MLL'g2|OeZ8FGѰO@WI`0FQ!y.><"g HH(46w \NO9vLc ܅q9}T8p>]2ttwhlRqMFљm.?w33 B,Q(Ex̊POjĹ(,هF4KXGI<oy n0Jt$b=e?HɥhJyq_>uiv&w.y>&)-@(y kYӘXR E UD s L@d=C8 r Q4 }/|6p$o Cƍ[|"&(9Z6`*b @YsX2|0 asIe|eڸGDXLdJ Qj%蹃TnfVv|`t\< s~M 7IS Y"'|+&_xŷt^5`a-t*6 .2y 8 Nġ KXht3,W!Afjr~NH./z}_~udM/_wX`*]<:/@,- " 1Ω~PAyx!<"PFp14`&0~  zUe>3eJ8 u%{J=}neUh%Y.~;~E峏@7J28*Je}ه~[Y)H2YU]E \8y<|R ,$(+C-pLN0 /2{"P&U}<Kٙ'H$&fJ̰.O'2"ɯzw|%p<./&8;,Th8#&BK2G_O ;;aX̧}B$D"2RVI:6_""2#Li3T.S, Kn8NjOr;~.1G_NKcH|o2)iXP!VXX"+yL 584 :l}r<3g߭?Oqٞ1E[W~V]Ňi!5Hg"v ^01iڷG쇧'X_0?_w7u߾rJ4">uœg~0גi顡cӊ"3& Q!2LER?:!0P})Jf{D$Uaiƈ>3gp,4/?p ?h#"|,L| A+:#\6>]c4WGk`3O!I#,7k)%Wǃ7 "k8 K'fKHWK;yedXEq*%r> M[Aie8"$hMJU4%>h%nN;ՎgU : 2?/oh6;jW<'VIqtisЌ˾B_Shn]mj1z1s|"p:+̚ 9Ү-2zto>H08Gϟ_sq @[T7^l%DyK?˅I8=rl?ZtU_}gK٣ga[&\|v)Q8 ^k\L!e1+ D^k,d՗Aqn.l!P!Oǧ|D}{{NfBR>𯿓yHH ;%IDb_t_V_|"Ӿ+oݺ?~u]Ũ> a)ѠҥJK aJ06LPl?ʓJխjmRB@56buDuVʗkM?jYmnCXf|`I$!h]-[ l+G03X8TgB DAəţp PUN=ӛl.zk~s_3:@Ȱ9V~EH 'GOٙ>S+-])?:3cC|0 |Dw[]=,La!KB?Yo&w?3 K9&m['5h8[Mt93@nrR,lw`I΃!V ¹B tF55srks'f'O\⏿I `W Af|!a$L5osw|<{DG ,rIxM 8JzW'Ȭ 4~ї˄A?ĜLF/M6{N^?oY}M`lFkQ2 wNa:vK 2(hI$l.?x^?[_X{2EMxDz`ni#A2Á̩ɞp Ht"Dh?J l8e$Ry 3xE }I~|\ֽ7Ez4ap:! @!/9& &|Ȅ/C8y<>OrЃ@C)1{,N8)/fixݨŲ<m夺|Z&ǩNbF#V֒0qYuѥT~H&s2 mP!P9 ~Ll~.:rp㮏L&zS2V+3Y.~)y8D50zpU\b* *+Y8Y/JL^6F@S,{F-h&_KHo{QX˞g,IÉt^b*ɀB)CXh>)> 9 `Ld?#?ȅcGVš '8N?V;u'Gᝡoe$pHv02 wK>kƓU/I"PeN*ecHHH[~SODm@Έ%aCř~f"5kP@ #logWOGf. qMaqD8,K^HAMb2kw1`qrv1)̐V@ld4:PrzA9PEcn"6&Wxɉ2d?4cگ`";SmJbI?>.3NETm+ i%e.l|)Mu,w7U,AQ( KɾfPPєiPd yDJIPٟ3бvUG.kֽq0 FNaTF<3XXZx}"N^ Je'g z&թG-y%A=iԯ) 7Or~8\(H&IIUD8α$#,F{ǗEx_NH8OAUtHt& DՅ44jobkt1Fρ+OXPCd1,@Y;Gxd缢 2Z>9?/I 1A5/zrB^=a~*O~1]HHH$U;-[MKBL?n}pBE!#;0F^1 f.Y=;q N. B4/)Usu[!`Z鮎Nv_qRadE)en)YSfyάgql"$m (Vf4^rM9 #l2nV81_p ysxCaLٶ+~l˩;gJ݄`NѢr*WѩL(]HR 7y6hn*%ت3tYpk+ gVTX+ ˱djj**\YIh `MUdƳeKp9BջF RNf9P~ڳl--eS&lm ,.S09H[2^Czy ̊fm U+›%{͎_ו;2s.dU4n׿U@/MOe.ritV8cEFn$wcPH-셞\gxy(fk^@?gU ;iŪflU`E!2uLy7.xi3uVފWڨ)HԨޅ ikTҐݿ)> k,дWѓYKq XDgUyYv,ת=[ݥWd+Qr _Qffl(͒u Q,iH Ѝua 3y0fc27nmʶNݣn#};)U%n-ZJܶN1N s_a!5f Sa1\lVАoGO\Sԣ͈.)7InV+V@[.vB,{ЕJZ][Vx{b8S5jo,wʹ7F[Xֲ{|,OR>I\n[g-Yn̞Flϰ$S)n:jWToQGs+["۹ƺ&\ _vNTtNh-VӄRWҨbVEʦv@S@(Ff{n6ui lGK/[&є~ Uu-*N΃0rԞz]"zCHy;_)<~9aY&$o(jjtM@-Z܀aTiUGp*s::Ew:A:GXc[ >XS[[QUꚠfk*]WTp)7krj3tpo_?)|U_C[צJ )G_Auht ڻؼ}V׍a6 v'1:JFsO8tfw mHmsm/ةo0RD6[Hp ax ~[ϣ#w/KoutR[a, lM ۠B)2+-Ԫtn֭[KM֚=Ufd58f֮, 5F>v`'btr7)Ы-0; fi֬Ӫq [amY]ktS5.;)u;˃t W`[zu=v;k KލΝ:n{`zU/nnI9NE5XTth$צ7W \ Ҫ hQ/kBeZKť`hc{ ].PF=tt]k("̰2m]c긶n$*W#rVswC&sБ0CݺPxW&/{jp];p)!pyY{2#aRsz(KU+T*!ˬW]k@lw6c5oP}{QU fx18XXd6kvﶯsB:XT-cG;Q3clNu=[xāGwnX#FUH;fX#pխkD4 4VD v~P_l oYiT_lhujmAaCT[bgGOXDqg[)d^ʞs&i|%t"wJ4;&vI:fOXF+͘3ҷGu3zkR.l̓-vOœ"@+N}+ĸD`1#};"FTR)}v[X'lnClHm1f-Ӥ{n5ڜ>fsڅދW;4=2[\LJHUoO^{V/)Zw#JcBv\^@yQo!c.D( UR9Ҩ]Dzj uT]t\-IC\^bB^'S&K^Ihz,k8V+0(`Bg\jd2 -v%}Nh8-jZf eVn92ˑAX\PrU:"嵠ԩuAV` -}[:(U[ EB%[#>OhYQ(STe\HU;ꋦ% =*NVc_Gs~}bMmBY]qci_\%QiȺ(w( PJ4UǦD&z%,Is{0jO`)&gKGAgQB>9fvmrYutE?:H&DiE4) m֖jw {rt֔lc\[=ǒOCt \GgԩٷTBm &8V:'6RkWfnTg2xG,<7H;"cݿgm5]dGF ݃@qlt ٰv )4N,j+8c;W_ m.-3z'3 +;=Mf(o9tݍ00PbO| ,XEsGaMtO%Üt;m;m[ aO.}vn6f*S rsN6 HMXw}( 6:SQ=3 a:X>?5630iٌZkf|li/Dh5n=*7QQ%eZ+j ʞ 2جr6[BA{1 ׌5C(`Rjx =kl )rv"XhPZn~jyNre-SB AEd+rȣŶpi9{sd ;6Z}bWkne}UNF? ѽȡA5hs߾8(#n~4ۚ<0Ǵy ; D}vOo^CV b2-sU7ЧQ,Keu%TZ5ܭf0SdL=U| rܤuɲpU}ЪRZ5˪3ȵ2ӺI@$",wy}5xK:ۥoҔ02D@f@*޲eK!bm#yDXtc0Dl3 QK:U܀yD@-GŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%HGŎb/D%iH~(/\fKI{ a,;vUx!P]ΪeխQ8sdEr2R;JD#ƪBC[YkuąCڔ6E y vة߼nMc`hDX,^ިQ {J}zRhߵ(1V| 3B4f{p6[F!f[$e5tKΪQI@f+"7#1~zCNH}ږN^Dz2ƅʜl7P;WQ~pnZX;^`uѢ_ke;4^q1fy(wz+ p(:R R܎(zÿo ] RHz{wh[5;l/Xiu373f)*wy1~En[NW GG1(Zu)#{@/s21rpڕpr\uqEh-fӹ7UppW/#MZZvKl#❊x V(7G˝7 &.2Su[J,Z)Nڵy>j^@ ŧW[7n )-~=/q{R`շndEeٝ6$jqh TNHxǎ&ߎmE.'Vtsyi ,_4[|R)z}]w F6T׬\N9$ԪɎ1U1Bձհ,Ү QEj|AbZYcЊjd-~t!͞d F ȺDCJaG% ^NE=Qĸ [G p~7=aϞcuKN3?S:?dv̶ݴ-ě-K z/m (k#. }z7sꪛޝ'v }3pKZo<<ɜ=t`uJdtsh{/?BÞ; |j<^M1K>bfk֫M@<[ޚmAo4jqƝ 8{*:(>epȅ[.+'?jY`cmFvL\2Ht [Ν樄1 s'aUa*5nt|mX] OMI#%֞;떜U 5{6-u_d-(8In $=Zު̭G- >w( c{w勤 DOT.mCwo}c)*/I'WJUfO!*xcƜ1 X7*E^EnIYO7fp֬}dpc^Kf `89%9α8*4lv.uK;l@9֠8&QY];o-qUUyjnEr^;@ս]ǁ!ZNWAuٚ{#ШamQ X@*޲e(tk(9D@_*}d1%r'F1E,v:22 ۑkܩZ0c"4@ÝfgDXx/0"4@F Z,3E̢Y9D@:lbD+-| < PQ(zn`e^abyNDeP$0ID@*(ШMft]ЛRa-DϊKD@:CD jB&KՕkDX\F-?ob9}DXԤviy\"}Z~w0"=Z~ :袹 GD@i'@ &+1DvP=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#4f1zDvFØ?Dh3HmGnGi=C6#@>}I`"t'[l!5MaDh?:xOS@DkѮu-f @ѹSAEik]CAitnpTDk@ZbD`n@1DZZ.  t`<ϧR^z2yYteUTV7^~NLL|_>vs>M¼,ZV;tn*o+it|7PϗH$>OtS0/nʱQIzzzC1#n:x!ЭUMsSy[Iء߭WU$"G[chD4ZO]4=ʪZU7h$1" ݄@i+_qr "`"ЕU[XDXx5 /<#`m&ZUg*oi."  q/"hǭ-eןoq ^k1"x)a5r^*cfҭS8[o1֧1"@ g|%]}XճKiiwEjYuǣޱ8%^T`D`!x}8o4T7DS呖Fmw߶[ox}[ݻwvwɇGF>:>^"&6jqY1o۽iֱV4~:VHa܈+viTdk?<}(󑣟w>ş=EYoݴ'֦^ Վbq"@3U*"d* fM{_|ׇ47w[:z1Tx{̅oyŒ䖯 V9-Ѷ GLq?c 'QWnٶЉmSᎫ]u@|k;p`ߓzzF[aEc;n$fh7ݪ}q.{իw(h]/5B^m=zMp''?|=j>$"Йrau?w{On'}fMؾrg07濾 j+VӃ}GNތ3V!A_ۣ~a{Ck͏޶jfٲ@[4Bkӡ0ÚoMmYL7nܵkי3gf|vxxxΝvAV#1ŭn;wZu##ЭUI/$" @[ L@B4䪄~l܃"C\ˬkV̭90@Dsh1v5N1uNEKZ@VUbjIHDSZ[6IͩO01D`NF[I "9Fۭ %XUF1FY^lY5k30~D`n֪ZMsSy[+XjۣܔIefff޽Ǐ&ELh+ZUAJ<\yIB0Ǒë[{n!ЕU-ycFD3hhgBD ^@1@ڊh[D@~cD`8UtIENDB`$$If!vh#vr:V l t065p yt5&T$$If!vh#vr:V l} t065p yt5&T$$If!vh#vk:V l, t065k9Bp yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59/ Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59Bp2yt5&T$$If!vh#v#v&#v#vv#v:V l, t06,55&55v59p2yt5&T$$If!vh#vg:V l, t065g9Bp yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59/ Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59Bp2yt5&T$$If!vh#v#v"#v#vv#v:V l, t06,55"55v59p2yt5&T$$If!vh#v:V s4  0,5/ 4 sf4p yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 / / /  / / 4 sf4p<yt5&Tkd$$IfTs4ֈQ^j v      <044 saf4p<yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 /  / /  / / 4 sf4p<yt5&Tkd$$IfTs4ֈQ^j v      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd$$IfTs4ֈQ^j v      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd4$$IfTs4ֈQ^j v      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd$$IfTs4ֈQ^j v      <044 saf4p<yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 / / / /  / 4 sf4p<yt5&Tkd>$$IfTs4ֈQ^j v      <044 saf4p<yt5&T$$If!vh#v:V s4  0,5/ 4 sf4p yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 / / /  / / 4 sf4p<yt5&Tkd$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 /  / /  / / 4 sf4p<yt5&Tkd$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd.$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&Td$$If!vh#v#v #v :V s4 <0,55 5 /  / / / 4 sf4p<yt5&Tkd$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&Tr$$If!vh#v#v #v :V s4 <0,55 5 / / / /  / 4 sf4p<yt5&Tkd8$$IfTs4ֈ0=I Uam      <044 saf4p<yt5&TuDyK  _ENREF_26uDyK  _ENREF_40:Dd$[p  s >A?Picture 3"bS9[ H*\2T/9n'9[ H*\2TPNG  IHDRr?8IDATx PSwqf)qz;0L.88ève[uZt"u0֗e۵TvkYkuuE@D^@AQ@fQDz<<>=7&!g01|$'^]Y@VY@V @V @V U@VGǎ=2nĝR__ 뵟U?8p@a2dT{|$iL&|,+XMHä(RH5jCIOgȷNú)26[#&GZPP@j_~~멧"dYY JVUU dU @VAV*joԨQ핌#GhK҈ۼypf7`DKsnjcRݻ8vMۑEsUuĭՋY%K]cZzPzV ;lfV|Uy09= Cjj$''ĨѣdYYɓkjjf5$$D= ֦d8xtI*[uUW[n5ޯG8l*yVu:6ϭJVǧg@Vϳ^&_e'ZT1n_\gq뮱m Y5)))2#_ COdU$E~QvmYE̪Vgaaa:Nכjߏ?dYEɪǺ{ϯy(x2P%[8 yVQj›'* ݃G.|KF6 yVvb]󮳙ѿ8哳2/SK˷l*j9T9'sKӣey꯳ӬZ}qW֟r:E՚tdc=smj봅4oxAZ j%7YQignqV˪IAV].L]xg/pXel}VOݬ^/(Oyd!9~UwNꉲ$eY/i|qqS||Ze 5IAV=sdYkֻM-߶vv;w6IAV=Q\;:{]l 9:Ȫ'ʯUJVw=϶ `ddUdd՛tmYYAuEl*ȪwUdd YV-Y}AVoÍ'\LW~lZ9m!Y푬xqqmճ-~{rI.5'92 U~Wco:Tʪeu555dz3--^k.jܦ7.Tc͵l}VUŬ_W]죧W5P%9 1GgY%^j%skULgdgb˺?3xc7YY-=#OW]e0@ZRR#""L&S_lg`gu)ϭ@Y5 2CV* ЬZM[Im<٥[ 萙6O]mjjzgFQ\\DGGvرEeoÆ CYYYرc2F(Y(_e^]k2Dvzθq|YiU7lHV%fNYf]ȑ#I U }YGnA2~z1LÆ SJzeFNIwy[gYMV V$Yw|C|cK9?ٖg~CȪ:x Y0jZO[؇+.*vSufgU~ȑ#rȪ?鹟cy6o}?w}_NÍ'Z٥ͭ7ח*FTdG===>)Y''=e`j;Z}lK2/CX6r~lGd pT'fGy COe,:+Yݼ̎(dd"?Odu]rY@VAV%tMT$sddYMdNʟOVUU/d 2`gհ jT\F*ȪwUddk$ 0 ꝃd YY 0 ?JJJzd{QNNȑ#E=x  TINN($$^fkxxܾ}xx]r[8 ?ΪS}v|[[jfذaLlllv p⹍;-YgUt^pbYR[[{~sdYEȪN惂l/;vleeqz\pd~ж6uXz,KVUwVFcJJW`{P+**d\F=Oj& 0 Ϫ]A`dd;YS)`z'*ȪwV1Z@VAVս< z'5|0 V+s* Ꝭ]0qYYNVHVUU/d54=} gZ髥2Ξ:p3wȼYYO8US|IysddscLS~?VUUdiJ zU?Rj {\ ^꣦iAVUd%Ӥ2ÿyYvvٵ[͵-2=jѴ}ZT[% UU uv9 G/Tdd{_yd Y@VAVշ/$/hl*jw:)d* ]iL*UUVJאUdd;YʙMVUU 6*ȪFū* ꅬ~R6:{J9l*jxTKgհ ݒfɔ" vWACql񪷶ͨd#@VAV}}96rdd[\>A`dd;Yn!*ȪwZT!YOݼYYkMVUj 0 ̓om-s7 SqZ:s)+9 ꅬ{^&6rd>jK:Ud>˪Υ@?ʪz]Il*8-n5KVUd}.wILLN糝 YV됐Nϭ ;~'gY@VӬٽ^)((ݻ~W3$e#@Vˬqj򵤤D?~euɼoYEȪ,\0))͛K#""L&Eqqq2:thNNNOg5xժ* gU"Z__ u!55Ufcbbl/BoذC:lذ3yeG8GU22Tׯ?\SSuCBB$2n{_+++m~ W8 U2a{3Hȷ[l jYY,),,L| #sԇذ eVv6/hΝ2S]]b#@Vˬ 4(!!͛]nhhh[[:lt Y=ONΉʙ1e>l*|QFiT2dHFF5)))2#_ EK.ݻwTTTtvg5L ϭJm 1qDw(&NfY͜9SƩgAK%jYEzɒY UVU~cm*CXjFI:wE9 _fuĈڱߡCjr+;Y_ ol*|~s}ci\YYϲ>|p»Y>vG  YVu:]@o}ݬ/?Yϲ"Y]r۷;:::׼xn#9 _d|0wU p`8Sv~Y!gU&.xvdd[[OrDMVUj ƯywډţZ!͒ymYo[ͪ?>~g '%xn ADjzz?O5'ı sPԗgHV?+vd>Ϊ4UUɪV?VV /Y2dijB6xd>{ɒFY}߿&xUF 3oo*jK.JϪQUd˗4Hvƌd֭շ/$Gf@VKY}m_;k,ۗ,K6*4K'^vmYEod5,,L򙕕%oV5qjaaxjZ::{ g@VKY |-SYO?N;,$'f@VYU'G5u۶m֢j>4mYEod5<<\YVV UEjjEosN&67g^{g~xkU Y@VKY,_\[8x5k/ݣY-^EVU^V}`FT\f@VAV=F5UHVy ꅃ{R%YYĚ-fu}evg@VAV=1s۬h46{dd UdUy@.iSuVg.UdAi) 6b־?3o 转ISo޼?^c^ xc{n#G=T۬Y2GgOZ–*O[K.]"M?-YϭvY|R.Y@Vs=rX;:{ [>^֭[%YYYUQq 转8[Oʵ]Rg'*z/:gUɪh{0d@VKY^8LVUUϳj{%-%6~d<5MY-VW[ͪ>UcUՕg׳ ~9VϭvttHVCCC1)/\rny45?Ϊ4UܪdU{RYSEl*FgOI͛ , 2ĝ떔±c|rXK~+YY 荗,immm\`0LrrrLLݥcƌڝeId s]!3pKׯ_i&-OƸ=ߛw:&͒)Htd}:ñ{ON}}}ddzZroŊAAA=t+9j^^ 7Ţkjj?K m/9sܾz'ϭslUlVIÆ :T-ljjrFBCCճ=9vꝬ9Tm<4?>s Y@VYl|:e RYQQaRWTU g{nbB֜pǎՋusUdU̅ӧO_Zt,d=Uq̙#F sq/=jE9*Ȫrf%'fėa/v7iS* ڵY5K~%KV3='2}%dd$~g &ϓ$2e;@VAV=<,͛;&{*{dd] VOY:/8{dd]ƃe/?*Ȫw_e64#@VAV#@VAV=yWig?͛ˎYYuo6ڱ|gFfOYyQ45_\YYu%>)1Cej^b_zxX jŵj*jwjqITZYY}I8bO0O"*lU` [RFgOYY mV#Mӧı;@VAV=jYSEl*y:,2 / 9Tm<Ɩbx *j>ϩuܪX!YFU0U~ v*ꃳ:s^h7_G@VAV]9ugϭڒy= eWNV3=YY<-M3?OW+fv *}l1'Zz$xPfdڼn[W̒Wήeާ=Z-Yʫv5{%?NYYźfw^$YcʻWUgzjٗZ~YYu[uk^qUP ک\w*,Ox%ddY=}2fXj7 UU{s9TD< :7Y5na)I>YYϺ"7Mʟ=Midά M g1I* z9 G_}+Y,*O#2U5ײ@V*Yw{͕+U'ۨ*˗UJV Mu?ʢ߮8q *Ȫ}VWkYS R_plv %oYM b]󜄣\Q*qe7z!ū* jU< ާzgYn?_fmd U޺*_V@V*YUڳt:6۷ Ý;wz +g.$SK7T{嵯 ˼L-߶@Vj mkkSeƸ%p V*Y%F1%%Ef䫌Jm/ə8qjuq>ո/VUdy0t:^o6UelrV7*ԙ ˚*w d빬zIGVdjl**YY [Y=->U:@ZWϭ*CXW{JVhV̝Jdʀ5xGUJV dugjoWKf_Ηo|Y%+;凑2ɞ*{dJSUMl &~Y%d;Y5tMŵjv(*Y@Ya/{|s%S0 JVJVӳjb`;(Z&q=\/Eg8i0@VAVyVxB|y>3 [Rعddv˄<#Y*Y%%=tҟu#9/L+[\[dddzEU[j}b۪UP3=FϿ=M|9@VAVUV!:#ӧO8?ddv7ۿLܢe2nŇw Y=wqNQ$2W*U嵯j[S}A%?HV_<p.ddՏ|b]LP7ZzsKS'cbUsx=wFfU:k}cKLz YsWYc#ZVV~wUK=nk$YYWYO;fCev+?U͵ Yù#%%~/WKddz8xK_*j]I>j|مKdd?d7OjZ{ݒf|=SB+7W**U\+qu:In3}wLŽ/ޭP&UUꊴ6g楾jh+3 F\&*jwIP{sl}2&F v:X/M;Y**Y^AWn\v7dd՟ܢϫjaìrfcy8Y%d^B J*YYܾ96}!3NCuSj4!ro|Ş~4})Ō>%eMl5dU)S} DbW\¨{d/̯O):E7?1Cg~7jj*Y%d8A` گ^ m<(̰# DUU/9~_sIs4~C JVUJV*UUd5x:dd7Z&-,*Ȫwpr`dd }/ƃsWdI\˫ndd{J_1<Ýtk7dgNߙ}O dŹ /5s&& =f]rY%dd. YװşGſ˱[ON:ZŹzIV*Yu%N3n7 ^xJ&j~ge11χ*Ճ+dCP k.6OJVɪ8 pUUվU덖WY`|zU7dU @V1ZRR#""L&Udd *3111/ڿY 萙p[?ߣ>CY t:" Nt:m>((͋* Dhhh[[:+n^DVUU'FcJJWEdYYuBLaaa:Nכ~."*jO!**YJV*UU dYY%@V*Y JVUU dU @VAV* JVUj'|վ=Pt:ݏ~@[ Nm8268}4"oݺ{Maaa@@@ss3wfҥ< ۵k&++K6;wpWUJV*YY%dU JV*Y%ddU JVAV*Y%d?񏥥ky{̽{r?ڲeKyy9N66 @V }III^ 0Lmjj l?Z. αOl63F~QFɟd-!55Ufcbb8gKNAAȑ#U*{Wx  efϞ=#FAꖐikkz$tgKɓkjjl⭻&88 Ȫ[K'ʟY]]l>m⭻R\\h"$ nt|PPGlI?Ȫ sl0A ꃅ7W8ژ|ϖz;iVF?ăd]F1%%EfÇX,jg1qDgKAVuWcwx ?`rrr䯶ZSt{d[w9۽ *d *d Yd Yd Y~Ha^DPPА!C֬Yc~ccT~f@':;v;vl=ǏWiwkdjf͚qV5III9Ԗ۶KY]p㚱v#ܼydYV|GGGAA5eۻU";w^-1c,7o|y׿Ӆ~xS'r][?4@VΪr1Y2qDϲ- RR{޽fںթS:Ul׏ UYɒAyU֭[eȑ#Ւ vڰ5==YG%̶+ۮ_]]ˈdqVBSv)߯VXh6ܽ{ϝ;YR:_W_דU}bg15jԨȪNkħlA?Picture 7"b4oR{>ە(͢4] n4oR{>ە(͢PNG  IHDRr?4sIDATx PTqn720 eq!ƛؔ jѕt/1j\cLjHB1!Mb( IĀJ`%"₀(;DDD 荵)>'/۝Ͳ g0/={vܷ@#x U* U*dU*dU*d *W~~~llz -[V__?"C薑qǭ<::Zÿ ۽{wDY},?dvĉ^zƍmsslʒ x 0  ۦ---߿x/Q0a¬Y[h%KdInn HNNk,\u8YuOeٲe+Lun)ϟ\ݡ`X,?յjժׯy&OTU;Ve .aI:NmͥꞞT=?^ݭ_MΪM||QgUk"EBPPw w$UUV}7nҦMV+W%9992/_e^6pȷxMÁ7ª+L'kfeS]fj-))yɿӻV/EfԝQmU;۷O6dQe_\\zdvd2_PP\MHä(RH5}zCCZ*}:N*~~~UUUrQ[[S !!!of2F"檋9YuׯGFFQOɪUھ*Y~56 jPP:M jM'`` FF(j0F"⭂C1g` ٬oyBBBfe *|'C6>q!6 AV/3dj6 MV *Ȫ6UddUUTߓK6OjJ?x7/9w##'ZYf~Z_Swduԩq-4i##b ó*ۣRP] &?ȑUv U0'''((|WKKKY٬f<*ZǷIPSSSgΜ>pdu$"۳be ϿY>pVmVk@@#Ywidj &MjhhPK"## Y;YN`x?o7oބ z}LLLSS##ONTDq}UUgnG*ȪY=Njբgʎg` fkYY&ϓUUWo:~r}:YLVxxV/_pb ֜zNVbeꅦv óSj^tyɪz~õk{Yƿ6UγOVG/Yg7%m;!Y]g2#vu=39>Y=WiI6o*ȪduӌmddULI:Rs j¸#bl dZ6+/\#Td;zWv|_`lج>xM}'"˪VmDd˲q!vլzŇؐUjނcgܛ}Y2?w:E]ji:ش=9+.//WKz3,[JeaAA,ϗyN7}t̚5KWWڱcGXX̜93 СCdvPY{A?f…d[V6m4D)3Buo\G*i%-[Ȍd,X/CV}*j_I%cOv@/YETflQwsNuBBB}}WwU]TV{>eۅmUۇo:jwwwweTܔ ܌ L;ܪ}ȪOeޤWnblɬ!6WsN9s%YYY{Ӭ5N] XN]N`ۭާ`HJJjjjߐBVk.4e~8ယo iYY Cxb-UWW˒V)lMN>]%oV+**:NnAo/{pU+ל:ujmm-YlN+I*oid𴬂zN'2UU g~6 AV<6N}*Y /{cfue3*Y}IY{<-'n}{mٶ|/g!ܛw|#]?_8>Y^a(Sc=\bpKVW>-t:ߥGOUγOVG/eM뒏twc3SW5u|f'gN~k=cOfՆՑꕫ]j]9OK r>UγOVG/NQ0W&f39>YZJ }pzAӲצ[UxmmUU/jIKk?Vddu jbl dǹ#>@VAV5 >/ jոO70UU [@VAV5jN`ddU1gl k6 N`dd_***BBBԇ'L&M.9N1Ux}V xeF̍7:nY]{qmd^Ui@yժi3yd}Uefٲea>7g CUxV%Na |-//%ǿZ}PD\6 ϪN_4cƌsΩΜ9s^[ݚ*|!AAAVUy7dk1*?F1##Cf`HPkkkeF\G. Z@V Y'SppN 1U}˜?^j*۩UG.o* _0YYYY,; UddU1tG*ȪYͫfkYY&Y'OUddU斝+; jղ:VUUmZc d5糆(c6cYY UdduYTbdUfdW @VAVb_DwdUfd*AK@VAV* aY?'+\hjgx ?:0 dYY]fx dy7 MV*ȪY+;>YYfk5 7 AV> j.*gx ~-7 N[BVUU-C +uqSI / Z}gí odd UQlys}ϾOV_b C%>n7 JVUUjNAú2ݿj^@VAVBSL?'?͟]sddu褩Og2Uաgr32j**Y@VAV=c'ҔyYY Rvze3YY +ߋJMbx dgg?@VAV5ꆂ87 AV7g17 MV|`YY )@VAVꮂ0UU Sr*ȪY:y*Ȫ6Yͫ' oddU j*j* MVU* MVRqWK #YY M3?BVUUm?)F8 du܀t:erxa܇?e pOVu޵61UX޻{[*<.7oLLL~+sy* g󁁁:{_[mvw pVWo"?{tExܒGezk udɪN Y-\5M2ɌL JW3U32r䮮.wK%z>,,d2_ݽb 4iRQQшvĉsss]hȐj0/Zv={dUd>jwwt111PLr+Nb6m ,3k'_>FVUx!KjN^(Z:d pjzzmkU61-[uY; Y;:uT۾I&zWV%uHVUH7n/N`Ss Y;jXKk{?[\R @V),yuVپ!Q9 7dv8gH+Wf @VᆬJ8ɏtw^$ͪo rd M{e̖wd2Y:S{`V7UdnpNQ;ϥ.2U-9վGeu;{* :S{TV_:=;'}9 8d{O/~Ezxa\NAYVT psV}TeOIS3>>.ne pϑ~[?o^'YqrlYW$qmY 2e[N AVUZVu:ݸq9HdG|"ʘ80Y!Yݸq7z{{:u2.KU%`8 SvÉߞ~bd5ʘ]x€@V1Yu#ͳ^^)KsCdduXZOȌ8< ^[VV N\n%Y^[ϪZ_GK>Lq< wNLLxcVE*;Uxkՠ /U/\Ξj Y{_[?+^r_(@V),M8{wgy wd#MZ^ڒGO>^]/|RVcY;_[Mfiʟ}ld54/кy>5= :tGFEU*F#6l?~zYYuG*oa… Y***ެ>yjsxa\BeY@V1JY |7TMe;ԫVɁ'^!fbij%sWXN'/tjqYo@V1YU'E5uj@Vw_Q6*F'ꌌ ۦڲz*@V1Y-..?FiӦM[v;ew+ٌ|d#Uˆ l FѬƵ^k3W]\|bijv#Zjc }l*;UCzjY@VAV+0\"3dYYeO^hjg d)% ~dUyB.N;[W7_@V1Y?~-6~BZ!Y@V1YvuuN(1k](cL9(eMU`V?Czzz|)u~RZsVU/^`-b*۩^[m*ZTRSаck Y:XeuΣ*xmuf[^k/9)jjjd^[y b4:xmU$N!*F5:g|Ս[vC[V @V1YuF. @VAV5 .(Y@VkduZX V*ZPd uU ݝ(Yζ/64 @V1JV*U:ܦg^[-+JI9iU琥'rۊ^f2^!??_]Yms!ʹ8dFjZ]`̔xK{zz"""ܘUUǖʌ{ΰ@V1zޖ^ :\e˖^{͖>d;鬆ƑUd^U8'922Rdgg?}O>߈>[OD&N B϶b4446/hr[7jkUlJ9UhU)::iɶ&MR \ *|uxO1PjN b :eʔN®.Zʝƌ #ݾi *:X 8z^wMRYO^(O`Nb6vY}lͼ bo!_96*F(})uX~9Y.۠>¸Ud#>@LTZZ;.**R }& xg.7oң_>d# St:I&9=Wguuų.Z{ذ0+d#UKf\|LsWʘ3Wv:@VAV֖ Zs*pwUddUoXѷfbIw}r zY{5Wo]u?+dd^[MsfX dYY.>뒏:@VAVUu6ʘYY ̛<5 :Ĭ6vX* MVKZʯ\(w>:fZSC cYS՚sV{js`ކE\UwvmS dddլ0&Y]pwa<ظגlyp~ jկo}Uc͍UzU9+K>Z\Y%dՉԺ8I*[yRnaîBhaQ檋/u@V*YgQEj'0^@VAV* :YMy}UddUͪ3. 8뒏Udd6zr%Y@VAVjBeY@VAV5גx:K֔u AV| U3wp}qa55uY_ dbfUY9>JV*Y@VAVGQxa)\mߝPʪ>f]'_>f8 d::IʺپmZJVUNjYY_ dy%O{, V5__V74@V*Yuj}'y2vDY@V*Yx'ߥ?"vy5~\uQί@V*Y#?[vL/u@V*Yur8KcU:dYYՌ+YdU:xVj\@V*Y$uf5ʘUJV"M4r 6-yuJV갲.K>ʺY%dus-UddUǖUdd ¸C?4QgR@V*YW6ZMLx`V*Y%kK~@=g8_@VAVwYum6dY%d%|hoL}L?+6V?W$>U覼sRU.V7YY%N2$5u=텦vV7YYY /^^ UVQQL&Ef9""B.>}\czMz/* wf`0dffLZZZ||ESL)--ݻwO:c*?o!YY^Z]_w͝;;pxʼn˕ JWIV3?)i)/o=5n^t^yyUdw1~?ެد#M" 稷e;|MefuUۣl~~~}p pMO ,?+^ʁKdY%#%((j2pikk+DL:.$$l6W/:ΎgU*Z@QN:qkjIK* qY=Y}dduLg50n%\2{NV :vZ{>2i>f5A&V=YY[YvW-|+@VAV*Y@V*Y"q I|k1K1%9Oidduley%ouùk'/d|"&l$ Yddu eU36}?YY%dY%duY /K<2%6vXJ]=JV=YYYZw4lN+cUUJVUJV5ʪz+Y@V*Y^KZ"YY%Zf5ʘ-e;=r@VAVJVWaQBBS;k%^7RdF_pddudY@V*Y%*ȪGf\ګdY%duXf[)MJV :ֳhj:/u5^IfXdd(cv먳I$3 kY /S:wUTժ#yo=g^SO$ߪ2Ne߽~r<~ߗCNnfUUYVNŋeK*4>/S<)i Yj~,L|S!y< Yꖝ|@VAVl?PX*@VAVhVM%'OmeM3}ϙ k.$3d Uۮ^d dddWzcK#M dG>Hmd dupMͲzsg'.WdsR2֟z\2k#YY[;uHӝʟݎБUU U1UR2U&]2=yj:Ӫ[%V@fd+7kKV SY}Mu(g5tnUz aUU7d F''.Wֵ7vpK!YY%il7nVQKF'SFC^d *˪W*"dd \9Ie8~]?v'8YϥI#2cCUJVǐkm..*Y%ddoݲ]& dߐUl<_7ؐUU#8@VAVAV JV*@VAV*Y@V*YY***@VAV*Y**Y%d YY%d d dddd dddUddd dddUd^f2\U:a0233e&--->>~~}֭cXU~'00WfVkhh|]wuw0,ݫZKk'@Vz_N`dd Ngs" :dZ՞^w" :a4322dF /"*Ȫd t!!!fʸq]DVUՑBVUU dU @VAV* JVU dd d*Y%@VAVUddY%^ZwY%;N~A'y80pFnȶ'YʿMiiqy(l֮]8w}@͛7y(@V*Y%dUJV*YY%dU JV*YY% dUJVAVGo]YYɿMKK+~g{۶mqijjcZy(0ֳ Yz^f2|mkkg?%ި344tg|p11kO>]$ .1 2lVV֊+디L6MBۇ{L2Tfv=uT$ . jt$Ȝ-: cߟ' ȪKzy$?322?%\KV?8 O|ժU}?hK| Z=?&U opv'LQQ՝Ӳ<33388X}}`B3mڴ:JV *d *d Yd Y Y Y͒IENDB`4Dd$[p  s >A?Picture 9"b84B`UM-O4 ?n 4B`UM-OPNG  IHDRr?3IDATxPuq6 ?o3 eqALWQ{^"))R3򌨈L"5ROADeE'!!p>**d$S1gUe'j5x֮]+ETjɒ%%;;[壬Tr5T/25|l}'V+H${[ر;ʌS>]l`(..uɿś5T/MVwTb#@VpaɔQ.pMəR^o<+e]A&aRuhh2wVSIwgȧ]dnjzH#o(Q_%~'*z{{srrd.;7бxxx(c L ͎};~CgUٕePcCCSY0e8'u+uc\\@j'N-IIIyyy7nܰUV+nQfˤU^5ơ|GEUUgyAz&;ӧ$I>~YFve׮]v;WRRb|INAA@jTFcjjȳ:_trDZq㔋'AofUs7nS||GfExA&Rf>\AM_:xWx5g>3eu޶mr*3Ħ_ԤCfqɒ%>>IIIA{ْI&544(ڤO-^bVea^UU%Qgɪ<O=/AAAzY-n-U(. ճ(7n %ΒUy0M8222d%%%%::ZVɓdYuM6mܸ:KVgϞ`1>>>AČGV<#/RYF8:t=SpV-fD\7~fD} ͳj)am[>| ldnU___elq7LaVk>lp~$YgU*%5=*՝ 0 wΪl@@gTTԍ7YcVtYf|YآAiӦ1*YoVStdgURP]ᙕ_)STUU)[iii1>Y5s+o[:Y:{*7zzzU:zY=x,$oóf$P>HVZ xdՎY-kr6g?-OwGDD˜u۶mk׮%xdՎY PÇ<<Y 0jr}ꘗ,386|EvSeUd|7mNUY$w{ߚ*'8<9+ iu?=j &6RpZT]]-[ڤ|rI&544X̪z*JP$T~#}嚁dսh_q6*Ȫ^V tGyd+cYY>Mŭ7(NY lY-˹:uV6Ɣ%8?ReAٲsk Yuz~cgi9>Yu4ʄU+VSV9>YZt~,A`eZ=CaqYVT/9S񪬟rAms}:`\.UeAmyɪ^ <-/g: SSVOˌclpg~V÷> yGUgjMUտaڬ(뗛;- .ՙ1dYYA~50UU*SlUe] rdd Np8Ofo} l dYYEV!/^dl 69_ cYYAV왕 1{el 6:g ZQQﯼy^7`ĉE=lwu'lU~V5MFFDGG^"+1 @Vn޼suuu#jmd.UiAi7퓏.+111cӛu8 *Qprx;f3/el pT*㺇ESLtשSU p  q"k۬{ 峪jeE>j4ӋdZ[[++5552s_V?Օ=,cYgUL~~~*߿_?ʘ~oVj*T(NrY;dulݺb 6;<턖 r8 &t j;HVUՑfuZP :"7;u3YYAV}z7 j,}27 j&L:7 j.~ Q&x t<@VAV* LY?F1UUd} kfIdx 6=s0UUd odd*Ȫ3e/@VAVmժ2UUd5/^`x jp~$YYAV3.ei*Ȫ ZZNVUՑfjiՕEEU]eYYY>\u'$yY'w2UՑ50UU ų:fP*ʅo)~۫O3U8&Ajjcg#_&-#Ym| pƬ޺u+))ɅZ^;9gA=#YT*}nUfW>@V^~pXEVU8><_>VTTȧӦMs~;-¶ϊA6Ke)P@VȬʒ%K6n=`Z wܸqvH p|V%{JD[ZZǁ}}}&##CVRRRM/BoݺO:~x{~DζYe_{ CZ  Ӌdz%-K._?%3]d\xY ӳFAB>ݵkLUeKIIIϽ{R?;otr}pd'=/k\@]{nY A`1hь*1cxyywwwk}}} rXO: ȊO*I&_4v؜VMOOhL/Z|yff:*UdNjOOtCBy0WT*RW>޸q#**J橡g 0 w~ɒ2t*RW2U8l5--8[)fLLeU²f9 Gf500xwܸqc9E%ٺ9 {ƍ7ozYmj=se/OΟ+Y[.&1Ymjjr=^@p _>)*%˪QUN 1Y5w%.J` ‘Yp*'?RY2y0_>[)K%y7+xI0 _M'4UYsd*Ցjy[%9 Gfu$wNY@V଎TN9{\g@VV>վSe5|O p,zMl Y@V٪jtug8@VᰬǩEqk* uOo)yQuŌsd˪KjTu{mp~&o Yò0a¶Yo$9 eURs< GjU}nUlN){".Ҍ%g2U8 pOOORRd5;;ۥ xd~nOrY]0mtI< eU*}nU?Wg p,;֥eV"< dHj0\7?;I3U8df߱ 6PZuEaJp~dSן*F) ^^^OϟVeKrrgUJCjlb ܹ ,0}RAAfwaڬG Y@V1JY|t:YyRS+zl#?m<Ę@V1YUՒO}cccߧpݓ~2 *'Mi| {du76W|Ę@V1Y |VWW"##EO^hdŢ͌ydB(]_7t[jjKgU< ׯ_?:?]:MKZ7Yhdg3MFi= :Ru*Ȫ < q#YYeG*Ȫ <~ri‘= {dUy@(']22ߥLw쬩@V1JY2O^XXﾒyb!Mv%CVU^V+Y=qDooe5v bT*Too1?ϭrẅVGjCͮHY*xnJUddɪNsV*KU%S>^9?ȪxnU\nxUS*F#DVUUjiՕ0mْk`/V'Oǟd VM6ϭ_} pA$jvvduʧK_۝b V$nO뎐UdUi<>դ6Ud{رcj:((HB^^Y@Vq׽dHj0&##CVRRR. qxV:r=Y@VVǧOV$fnٲeǎƬ[.d{=A\. Y@VY5;iii U^lZs/Y@VY-,,ϸ߿aHӋn{ߤGJV/7w @VaJAW'?޸qܸq܈,|4@90ŭzRZZu] dU) ܈VMOOhzcgM5Y@Vaj ޭ[̶Ke~By0T*RuGj>L!*ՁtN1'gEe^`.Y䳶lKdA&jyW8d޽{(/;)((P6YV=MVU+":: ͛7:?eURbU+*00[Ryzz7}]=MOjyUdͪcfV=[f*ȪmA`dd6u_ڌd* j]gׁ*Ȫ.gp/ErNv*객=߲9,T)d] V^{)ѯXY :l'|t~$Ⱥ, {"),;ddu&LK{9"YYߛE ׺9 $^z]e [MUF'A5.1k9K0 j^z)9w^qK< UWAAAzҐhҤIr5Ȫd5.wJ@V(h4mzф JJJdedYY>Y1 ]SI;w=D_'gQV-*//>HsO=sf)2a*U:T*qãn޼hnݺZ[cW_s?&w d|}} rX.mkkmoowWeUYWNϾY%vjeE>ʬA CL~~~*߿_?ʘ~1&\.O\NVU6mۥI'Voix8d9ؔ+e!M@V*YiVuNY%dʬ.9IXkd@V*Yұ?#g-S>y dxzMdl,(k JVGn0KE9!;dOXov UdU+s_ٻ#dY%dzuR=չ OM UjƬn؟{eW%VoRdd+Oǟ f=᪝g$ץYY%s%W*@V*Y%g58?E}@VAVɪe2[xU2[Uv=YY%dY%d ˚u n~xM/y)k>v-9,v>LUye;k YY%z:7UU[fu[-.7+dd?K_/P-ų_dG6v6W**Y6LuǫIn*@VAV*Y@V*YzsEkEV[˕殖kȊ=~]eWϱg*ȪSnN)z(/$N'>@~ YY%d wGV˽cVuK/^蹲Ϭ^QN^(i8h%V/7ZQ^n|祏e˙" ꌲu wꆚ][*K鸹űH.XЩ(7y7rQ|/zQQZNj-*ȪfuU4**Y%YYk:?]kO\JYv}u[d N-tH:x3:ddUdUJV zeQ&YYuì:%Nd .Ysʚֳ"YUV]l4tUipw\~Gf"WVCB|pgd n&8LV JV*@VAVVldd: UUUUCsWKbJV*YunN)5odQ4hRdUvdU ôYUW`?#N'֭[ dUJVAV*Y%dUJV*YY%dUUJV*Y% ߯ި]a~0k׮f8w{V UgRQQVz{cLX⊺y$܅w}r?`JKKCBBǞ4i +o7Ȫ Ws,f}'bccyVMOO('u_v„ MMM"""o7Ȫ WsxCqqJ/--u_|ĉjzƌ[ j.}U*dU*d *d *d Ydw׃؄Zغu|S|dU UdwEVcʖT*0ɖW0TY߻weKKKJ===j)SdB6>$_cN81i$kkk'*JY5 kY5=n,3(cVu: ߾Bn2UeDYonn6T.V֗-[+e=,,j}}}25Adp٪jUUUk׮ M7"u}[n ?dpR+(("55U_VY|b ΛK.ɖdeZOu:]___||YUҢ_Gz^fv8d KfԩS%rf̘ unܸqxyyǛ7oO٨j;PyTII Y*YUTcǎݶmLLWhhh2e\xZ^^̜9s5k ???ٮ[6ʟL81//oU Y Y@VY@V @V @Vp5gĹIENDB`$$If!vh#v:V lI t0655.pyt5&T7Dd$[r  s @A?Picture 10"bA6 v^xEÉ76tn6 v^xEÉ7PNG  IHDRr?5IDATxPuq绷e9}a/88È18:VMRRθHʹ<"#đ)+D\6@9u.|?Xyg\dd Yd Y Y Y@VYL6mСZUgq C1do56l@ue.*@V.1XvǦ&ʒYfyJcV@Vq6XMew^ggg}'NʒzzsʒL+|1jט={ŋUgɓ',::]`7nerY͕tf޻ǎŋ6L1 `7n &^6]WWWh\]z y+rCUnVbD+""bB[VVf(*5"Z_ 00?̙3mn<&&'**d$C1WUe#n?Ik֬"*Zh,Ȑy9y_ˏxaCgݽ̟ 1_{Yߣ8%KX,b;Y2?cƌɌr#ܣ&uX#@V߰~ɐQ&&L=`0;xTsssń4L"WFUu()Gp7l"cSE)?s=ܣ򿜑#GJ??OU'ݼy3333""B6qwIֻe> ϯSĎz{ mm ws{Mg*A,tl/9dvJXl6ՉQvŋPw3f,ɹv횫`D*ov>ʣj3Zܣ">>^y;88'**g8q Gǫ*KBCCyfRKKZcϝ+))Qߒ]<2ԫ{LNNV{'X/Q.j*zĈYʋ7~fs *^RǥK:s*/-?DY| m߿_2T_g!!!z^L{Q_[V{'X/'ƍV[_l6+-~'pkk3es- ])Y{H[YɓF3YU ;X{{N2J-lȣ1n8DJG׮].Ȑ%2jkkeܩ9e+KPP_ y:;;y Y@VY]ggNq~AV=L/ ,..&Bd[5i$JV*UU dYY%@V*Y dd .sƍ؀V;r׻܎IV`guܹ۷oy[, 800vUYA{l `d{Vnjag]vmiӦɼ,=ztss\xqԨQ/499yȑjMrvƍGS* VVVJBBBbcc***Գ.]oϞ=r̬Zj2,}uvvZRG˗/ߺuȭ >gJVYU%&&Μ9SƝiii€eF$3z^f4''evuu:r_23{lMJV`PeUUSS##W5*F#K$w}RS]N)ìu,˰aU >U :TQ jCΘ16bʔ)VRƾ 0Hdɒ͛7+oDLJJZ|ڪ|߾}&LPn߾]efU޸qC]HSRRꝛ5MVe)#TFgX:::M&9;vl]]ŋRJGqU+**䬖3@oz{ppӆ ȪqNiY zʵké*O1c8L?(55Ufdl6nua z\WWڵk:mڴ:Y P^uX,p6::zyOj?c/“:̪k׮]UUUU?G--*yVK W]ڪduJG*yVwiwߓϬVW*yVz'9tU)7}XU ڬ*?,((HYzYj`z$#,/YYuAV}s#Ob .|ӽQ*Ȫ Չ2v oJ*/ꦭ'X~z"9Ǥ5jTff揷~bџrK ˳*MM=P[Un?)nzl~fYU 6n>ꂬfU2ZYYxf'ZRRKV4Y߻kDrv N`whR9,)%0ȳvO!o~hr2?K*Yf#!W!n: 7'OLVfz w;lMMMJV0ƊFj۵vmUUg5YZff@VAVs >CVUUdu\֬3*jZuBX^dmddVWNxh Ud( y|/U~Zrs}꾝-լ6*^;yY*u&gYoV9>YukV_b'0;aqɪyugES/NV8d=;?ĺ k*fo7^gqɪ޲t*o v08d}հmui|= ޑ՜)u{XUUdYYWV Nm5mˋ| tVrdd*[WKV 3d^ZYx[V}\^{*">jXnGdY]xUm;"^դԉO$UyYJGjiЗ OU `pfU]C_bCV}u'~?;;^VϷkhsOwd44p%foFWWm̘1I [2TI?px2c0F)3ӧOʌN6M܎\mΚ5KYo6Ή9,ӄk'̼fk8A/@*^br e[۾}xLLLmmȩzZ^͛dՇwG,fg}VKlG}2N pSRRl2os-۔%rIX,dՇH7lg0̜9Sٝ[TT>}d߾}&LpUS7555 @fMUwײMN0bYelÇeYݲ[|/iiiKCCC+++eŋ|2;vl]]ì5\Q$QFkߦܯ\rdՇ>5m`YYNsb=6 ꂬ>nݣ@VAV]G1ufx}PUv ⵓW YjO[º ׳çQ9>Yu_V5eVyOVݷ;<+p<ţݝ?LV9>YuSVϟ=6fuAOݝ{mws}꾬>Myn;9>YuSV'k}V8d=YY `gU7eN`#}9'5*Xʺ kZ >y`YY>?!*Ȫ+:u +.wkhi ｣_VrddvsX}L|CS+9 wO`'0 ꚬ^~pYYu[^*"e6 ꂬ~, _O8paO!oY@V1hX2@ SNg}PeJF՗Ud!d h4F_ʐ~gJMe*2?pYݚv  Y' zWV%JVڛXUf5#NZff@VAVմ̚FUddYM}JF-լ**;UUoYYuk̮\ f'SYUf94YYuh#OUdd_Y;^aƽyE$2mz'.@'Yy:YUվcEXUUW9٬*Ȫ MYUUVӸSXUf5e߷UddY;f~df@VAVo**z/Udd_Ys8Bn=-K(d%@VAV;Z 7`@VAV]Ո<7 ߬9BVUUdUHV =9E2rd )*m`%@VAV"O'%$Y}|2[Vrdd;ymYYuYV"YUUdu҂G XUUWduYYuYV &\ 5{~]7 ꚝS z .W`'0 ꚬΉ9LVUUzYV%JVO\(g @VAV]y{Y*Ȫ v> .yϤ }w|#V#|fNojng=@VAV{$~EL_=d%sdd_;nIe'0 ꂬi}ea @VAV=Z}6cUddYկy*] =ƊƏ&g7>핰H^^Rͪ:_zOV#YriCvc: +RJVy30 _|s 7& U>ښvZ&N=6_*z݇;ʗ%UddY}mq~VsxyYYuN`*Y@VAVe]R鲄BUdi4ʪ2Z%*մ̚KS* gUf-ZN`iXՆ :;;.\(#Fg'0Y@VJ655NKMM(볤6l꒦9r@}2?1ڮ @Vo޼)#Ne~2ꜹn@@Sf,KHHY2~^/[ :TFZ-1F:Z} YFAB~ܼy U+++eIIII>#xƊFdYey Y۷o wOkJkҥiii2S]]u\UɅzq9k; ϼ)]>.Ĕ)SԾduNav ,c31Sͧ*ܺwf7DVKYYU)P7*j Ug5,/5Y_[ΪZur 3;;;;$>Ud|mK88vUd˪4Uբ ޲4|p꬗dYWeI%MX,>e dYwN.f=+? ]q$#~CGo3 ج:Tq̙ZV$&&bV%:c*|`ٳg[e)??ڦRV; Ο`@V1Y |ו8G wifSYU `VZS *Y߇ŃJSfu^is+YU `VwLi-[dbݬ.|(Y@VၬH>+++SRRԡHMMу~+[Me ˋI74@VYGi͚5%''\VYx$ ߬7]ohob+YIII?U6,>>=Z~4$'G歿!n۟Yf\e JV1;ʭzaZ@VAV{GjX:UddwY=ZJm jʾo* XN`dd5Yl&*ܑUyB(]R>:W&v XVT/~; XV`'pFnUdnΝ;%7oY3_UdȪS 6V>j6cEִRV* ^[ )버B V]UUdjbbd577w8*Yݚv pwV8⣯dYqħ_[U|]MV/ײ @V1Yۉ6Z7 f\hj V6'*<ڪuV}wω9l8L d 'Yd;Nmkb­vuuIV*Uߦcpjso\&Y޲4|pg[VVjCCC rrrܟU,:2Z}0M 7eI%MX,\WӥLRRRTT͹7o7ndmZ&Y[_[핀. ؜~{Oo9V2ƽ뮻>'p> 7j9455+oRy+W s]:C5 pS.1uuu@V%g͚5Kn_y#;mKV~/[dϪTywȑ#Յ#FP8s#ʫrj@y$2Z@xYŀdU)QՅRYe37SRRdFNu:]w4h+d]Zf>pdbӫOyڮ9@VᲬjZ ލ7lKe?x<4MpphtQfu]Rd>ZˋIe*p gXJ>Q<钗 p}Ve|)Y^SS#e ::wenNUd՝;w*o&*))Qvٙ,ܺu ִ2U[*\Uȅ3fp/톬.K(<Ъ.mblU HVEEEѣ hGо>Uꀕ- جz@gf?0Y@VAV [Q UվK}*B77Udd5Ή9HVUU zGV#֨Y尅*j]uIꛁgĆYY#DVUdd_YЧ˘U-]~  GZ ~gϯ&ydd~ƝڴלņYYWVBUddx;ڴm wJVw=wٍlUվg|L=6^ilܩMlUվgu;1<* /1?<+,/R&*Ybh*f:5v0r]oÏ%ddv(+BL/ݟ߲E@VAVh]RnIN.ҳ9@VAVK ?6}z]sduc'lUվ;"Y=*Yzl>[dd_&*ȪkߵcGW-]mZW }4q/DzHqr-dd/%.Xc* 7̨o3Q@VAV* ѬV:^Mudd_2V4ȩK*3׮x w";kU~{* w?FjX^g{4 }5L2SR-U Q UվK}*BNVUU;njX^dC{[dd/5Ή9̟rAYYWVEi }j>mdYYu5/\c>/nϯ >ζ?~? dd/YS}'o5?;l&UUgw*ZP>};dX(M(:pm ZO_OW:!OηIl#U^ȭ{%80g* گVTGV? ˫2/JVɪε):O_}:qv=g޼w78JVj_WEw2I_g9ƊF6*j̉9G|C ꚬ( .VR֨k@VAVչ+s?/5DV@V*Ye JV UJVUU?=~CGgIVSȴ@qI˵l>UJV{፲?>UGgJV<]D35<JVj/khU,ݚvZ]לwjd^o*KVQ*+SJV@V*Yuʾ}s3dU*JVj-K(I郚mϜX6|P~%&*Yu`Vj02ƍƎ+ݬ'W$Ӹig5@VɪtTIJJ>kԨQ%%%2{ѣGVVl:.h'%d.X,!!!]_I[=wu'J#dY%nj[3L/O?Tz;jhTwޯRdY%EѨ~~~~NqovտVd@Ţys/^p–{'pY}Sr*SC{d^OII9QYSLiooޜ}}YO\()kX^dq JV]ILAAA&88h4W/2ĊoeՌܺ9ߜr<$6&*Y.^Ue:kW UJVUU/ )]P`!B*S6 UkZeZp.|UL=6_ʚx&f D{97 ל%M*Y%.*UJV{'#nK=\`Vɋ_dU"DV}_o}k)!JVo\C뜘dY%duzByܩM! ezxL1o) e)?[:uA|hLkr}i~}>۫,T[44)GnW L%˧OV JVK*JVɪ+zʅZf5,/ @VAVj/ UUmV{xyFeKuq~կ_|酓lw JVmhjkxE.ηIѦ tG 斬\^V=gZ[ۮ6_lgUAUEwYQ;*Y.Y]ze r u>#n]R)[.U UUMpǔom6.}P&2ΖoRsefʾ߬%Y2nz޶r9K&+kvV/y驗e^=UWrα*ȪW{mq5d0S9^L2_z19WfMs_|1K=f:OΒI.Cr;Oܠz^Y[huiYY^kN~= JV*@VAV>XSp6w=pozDQβ?!Ug.q*@VAV '8dduo3O=6ddUdUJV JV=+WOwvR6¾ȝ JV}8ݮK*ȭc+ BV*Y|V"dl&dUUw\);Ix&ŝ?w7)^z-*bom{ԃ?2Wu@'DYRT@J*Y%#MugV/I9KV9|g镵A|b rjy,I-)%d)Y{f'0YY%.mc'l.*Y% dNm_ks_yUg.E|UUY@VAV* zf5,/V dOV JV*JV*Y@VAV gZ~dd q65g**Y%YY%/,/r7k$MK~xBO(1op>L+*uGC/dnJO/4㋔|NѦRV6YYS N`UUJV JVrmC{Y**Yvd YY%d db\z{YY%d.Pfzl>YY%ddUJVqkV"Ϥ0Z\mZ;O**Ym-䍬YY%Y% d |U dYY JVUU d_YYYppV 5 NEVUUt:]jj$%%EEE|־}~we˖Z ? Yo{.VKdd'Z|gYYu@Ѩ~~~NEVUU-W< SNYdYYu@LAAA&88h4W2* :P* JVU dd**Y@VAV*UJVUddY%d**UU U d' >&d[u]ھh4/<"V[qdlp "DVNLL^URR2dȐV ҥK}Qk_cUnn87nY%dU JV*Y%ddUJV**Y%ddUJV*YYu?{Ussz`mUUU<Yq, @V ޤ,88XՆ Ƕ _3|pl; c4Ǎ'رcOIN*3IIIQQQݷo… {3FIk }p; 3jԨٽ{ѣy:% Kf,%Dfddmϙ6mZ]]uB\PcQ$YuVu8?(L"fxxxmm?oK|nW=X?&ʓdɩ92}[2ꪇYc0 $Yu^OII9?vԨQfYXL2߷% z(| |'?anooI AAA&88h4?d23FN4I6-YuCcr?aBBBl>_ēd Yd Y Y Y@VYŝ$jCBB6l;@V1={kd2M< #wuu>\$''3<@V/''GL2Tw9lذ'ʒ&k ~@l =:33澲ǎ`uu5dY|)E :ɬZ7Y/L66(**}>dY|)B:UV|CC-ȏrmoDF2dɒ7oʩGDDX_,&&KF74Zw2WXfMxx%ވRN7n?dR (."99Yչ*=U{ZSS*K%ZV~ꊉ9J# %;;;Y6;ՏU>U>DNI/]F^fСrzuY~=( zMU>UF3|7T@]]ѣG߿笚L&2ӧO7r*SSSdo6Ghƌ]G*@VY@VY@V @V U@V U@V 3tA;IENDB`07Dd$[r  s @A?Picture 11"bj65 {oF6n>65 {oPNG  IHDRr?5IDATxPTݨqn(u:0 /88È&jlkШQjCQo$1ƔcmDVaEDFDD؈)w~iO˂ e9g| t2 Y Y@VY@VY@V ]Θ1cСF{MHH*t7naH7Y֩Sd{wZaÆyU Y1b^~lll,3g`\ { eUܽ ڱiSS, .%Cɓ'I٬^x$;;[{Ş/&RRRŒFsmnnOV=*++N|KBBP/lXƏ2ru|^>s͕4L6޻ǫW.]tذar!!!+WqOTU?ȨT6I:;;$ pI͛7\)j(CE܎܂r;Ǐ'd$ҿ ˗+\YjΝ;e&##CY2JDѡjG+Wܱcȭyi3Y%*SSSgϞ-={( CBB dXXfn/ӥڬfgg}ܹs6*YʪFFjUAH,^ʻ^ƥn* r:Æ SLV*}VmСʌRP&)))i֬Y=\YӦM[j2Y3| AggkBP}3f׻jHHT ClQ.XUlV< 3 ᥥČe*'}&EDzb :d2Cf@VAVj XUUJG@VAV n'}nӳ=&ѣ deerIɪnYD>; JS3>st>+n|$>笪K䎆 ={ҥK' $ee{ `gUƣRPg]q|yՒ$epܼys2Oɪ>Yͮ-bPrY|\]߬~r칬jeez,U<$dU0jc!/aV7TWWbe !$:dvbyVFv79?nfuΝSNUVm۶M]u5kBCV\h vN666±cvttoƌI j_{Y%d5ZpUddUKY@VAVu |_f+6 N`wEVK<}Hwӑn -'@V=pΉ_uze#CVyOV ~gqŋ[%dՁja qfh$*'Uo_]he۬1]Ng[OV&S~X|<6AU$n7=*'UN`O9'^jsd1gƿ+3f YYeVYy2IV_qTfr {'pNmddU.(x%6 CV^U0xSu YFof0h܆UubIVˌS=.[%1Zu@fU]w_bCV5NVnmVfcΔvw+ګ/iiiyGƌcZ%11120afSn|ƍ#Gyyy$77W øq㔿iɲP~ʼr۷GFFMĉ%"~+1m/;wK/)y;vD)3EEEBƷn*Ju 6Ȍb5j̜9S+3sƌʵv:u9s(!N  eZ(wsNĺ:ۑʱze$MV)\=ooi_yToo@f;}v6ڃ ܸ:/Tn7==%.zD.)3NSVwxXsLʼ@fW_b3{lewnQQx̙ʒ}M4mVN݌1cSW ^mLu#GU . 6Aշ?:ڟKlZZZrJY,qf\.o0J_ IK\6~(CjY}im6 _yn |nu{~!6 CVd޼Y-"~꯷>5uOW*Ynonv},'-Yy2Ř3gDpϭgz7'4O+I92_pO忷b/<|s=*'ꛯĽuoɧ)*?V=z;9>YNX:YN`OV(\MXvV8dիY=th2){W_^b%Y8dջYT4eWV߬vgzW_.l`%0زW_o[Nu \}`*Ȫda-Gν/{2ow4 }ʇvuYY!'6 CVg/ѻɬ*ȪYӖynd@VAVuʼԇ.d@VAVujB~]YUUߛ*ȪY}%I̬*ȪY}&o;UUv ʗ$X,YcǎU9N`ddL&SFF̤i QH~FDDnW[[NOXU}Vʗ9NQF)_'?e^ft!V, J+%jʒ;x.;9@Ȫ`P烂gM0Fĉ꒜-ĺ t:2@V߬va@VY52#?M&,VWWLUU\g>#md~Uy2 *Ce>sS{#^V6wb̙2#Ss*5Ϭʫ_vdUfd** }Qiє%XUUfA攅e YY#;y ^!6 گ>Pcg7}cΔYUվguoUddUea'0 OVԞ!*Ȫ>Yϲn :d˺ ʖjjk[;7 h̸VVoddUdYY!vGd ~OkjVoddUYY- @VAV* :js|fS \P+3vG#+9 O%M?/du'dfUVrdd;*Ȫ>Y- *ȪNY[|YY!eƑUddU>JVoddU:7 CV/1[YUU=/&*j?ݨX^zڒ#/Ȓwn2_YzWwYUU/{ϱz }=gc̙2ML|isddzoûg@VAVUKcquL{yYzXdO`ʏmo5YY?`S*j7Y%dYY!.ʏ%*ȪYmhݗWU{5YY!YqVUUZu*Ȫ>Ymm3'+5YYa'NVUUZ}.*? YV`0WVߜ5Yojh7_RU0qKdY`7".eտS6e/ȒyϷ MV?6\^VsL2/3*|հ0+}7-/IVYU88pr99i$?jeK5Y@V1*3/NIIz;ciXgutt,ZH#G,((h'dY*S"ب &)##CfgI7n)M5jKVU>7oޔ2a_{\7$$D)3N3""B{_kjjK~uCHWdYŠjhr 9m۶`VVVʒw2Z~,^ʗ qV=v:/t9kΝ2SWW=;'&*|!C :411իnhhTv=ݡʪHVU 7N}҈#=lNOOi2g-_|Ϟ=2S]]uʪP^[JǯvttHDž6m'W'\`0_yڵ9s85::̺g*A%edG/LjYUxcu*C̄Ȫv:iݦ'`%@Vˬ3F;rHu_eurҊ *|NaÆ\~}i:Ȭs* gf1 eˀuVrd-Kէ˓9~! eU}w 掷|6fe+YU hV% tw\+ZZט W>h/Y޵豂_9vY;{ً?~+;xdzk; d?<duǞYϲڟC^j{˒j_k^ɲXxXcYVxdEq5Z,*:q}mV7g@VwFV_z)++<j`j_UՕ_ ЬFDD='7"Zjns&bj0 ͱu꼒e.+R *jzzduժUׯ_FǒUd>~˒~1\w;1sfU%V{dJPg>u+ͪyc'pjmz*j%ټז\|:V{dCV~"k淋*|ڪ6~keROO+ʫg@Vᛝɒլ,Kg|U b)Y Ǭ$٬AgY_[u=gUd҈#*U,N30c?nc@VVZU{i~a*j{ U tV={h%~+Z\="4׏ nVnݪ}ܹsoY*((Ǭ~@KŘ3YUx7aaaϼ<~RS;%6KSOO<̚»Y5O{-ZdO^(,ӎV*Ȫr[۲tuV]vKVc̙6V~d^jDD䳲2==] =xXUAZRlojl++? /fHknپ}f56=*?vxudYw*҂oIJJR6lݺuK{#ɚ`]1g=6*U| G@VAV(>ʏUOr%dd,ŏfu}Zo@VAVu @M9c- }jT~TpPddlt b*OK*կo}z%1_Y+*աC)U լܠt|پcG| /f5((HzՀ C+Y]RZqal »YݵkdСC7o $٬^䪍1iOfɼLj/9@VVe:lذ!VYMS++y K ^[4Q6څ-*xm\kldYŀf555UH]VY:mUDh|Zа [d^̪mUL.="׷>c" /fՇU-)YA*j_4풪K V"::dYju'0Y@V1;;::%YYY*[b^[씬@Vc..Ҋ lU DVV[f_Rd.\RHVU [FuÍFcddbz\5d[Zr,4!@V,NӓL IKKs9͛ǏaV}^s5lU k)3ለs7lذe5/{}r<O+ʫgCb0fU9466FGG+oRdff|>(({O3*jQQ 7^__du^=kΜ9rk ™7RkFuɋǘ3/\lc[BϬJAw'5J]8rHeaKK'7 +?]>(dmQӏūKj.KV[ 3JAGp8ԅW^*=ٜ.3d2uNc_VEq5Zd~a BF ލ7\Ke=?x< Cxxxiiێ>́Ux%ݍ/jVcKb̙Eev*̪/%.kjjdA#-.Y@K*]7(o;(((Pر#0** gUŹ=rYIVg>u ?b̘1Æ 3 #Gt{hµ*Ϫoj'0Y@VAV+׾jS㼽uыG~4gw֟e lNO<1Ld^&6 *j/r* OVlUվBa/3YY+uƹdm'" fӾeI||Z).;* G-JVj/H)I|V^=dd^rL~Ks~?oo}$9**JYc-dE6**Y@VAV}ֶ5WX]&6*j⻷/-:*ȪY07W@VAV=5 ꩳ'IyaU-:Y%d]st]Ut2Y%d7Y{26@V*YIu5Y@VAV(b̙԰@VAVx- ʀU9+Wejjv"+JVɪӏcܱ/LG\SZq%{UJV%2?2iq$dk8Ȋ | dS=eIK 캣^>VY%d ^xhˬ3Kx6ֿe][ dUWHYWZg:Wٟ[Ƿ dQdMUV dv٩X"mJVY\8t+)kV^/@V*Y7ȼF[f,'eXZq*JVɪ{|IY@V*YJVg>u2dUW-QTmUb̙/n>Y%dՍ%q%+(KYy6FY%d6+NSt^euEJ!/ dUddUoMV)ub̙<ʋlMUh;{m9Umf%-+Y UǬ{qJ [קKm헯\ @VAVҬZX0Y{ wiV dŷ?:KVUJVݰ9HM_E"cӿ:򋕟$'eUZqA&Ǘln ze[/6`EeXf.>8l2X- )r򼽕 @VAVj/tXKV z7fW/UdzQyyyxxhX,ڳJKKǏ/g7N.68ZڐZ.e۪j/)/d5Ɯ c2222d&---..N{ѣKJJdfcƌY m6Ϗb! Uվ łνݣ>z=RV돟< YίSg9@VAV=c4kY֥K̻kc=6kw4rkyԊ™~wåpݶR;YY`P烂^&ƍv'//IVެdU{ddBCCNX]mnn^hQKKˠ}'" }VfszzOj*((6mg%Yy]' 'O0^ZZ_ewLDDV4zuIPW~lEJ!G *<l>dd+'1xFAYY%Q$?ZotF֟oaUUV*ލY-t :}||jUeTnQΦ w]V֬z Zڛ2 /UOKo+_ eJM'*Y%ei,N>YWX%f-z2IVߦi̼扟-;|uβC/Y5=ܚLEdduP&XWIg_|\qW盟˕Iکoox-@VAVNw2_X*JVj_lGx'ǞN<|fwRqL2CV g;U}<`*@VAV*Y*jfuũudY%dUTmɛ*@VAV0.}l%Y@V*YՁU UJV*@VAVYV'̫l65;R?8͖ ^$Aʏ~,> &YY%s dRgƢlY`۴%ed:YyQCG?&o)%M6YYVVclVSr 5Ҝiel EvG^[:A*@VAV@JW6YY%d Ն6vj o!YYAVW'9Yd%*Y%:H&*Y%dddUjV֦7*@VY%ז#M 7.U]N;*@VAVɪYZhd ddd5PrcіYQ2ɌLEp{a@VAVc4|;UU7 :YdYYu#44t*{zeó* lNOOi2<< nȓ),,`0W2* -dYY%@V*Y dd d*Y%@VAV* JVU dd**Y@VAV*UYmU:Xs=ƾ2 -yǁ{+ N8Aj TU%%%C immP-_qA'+΍7x(@V*Y%dUJV*YY%dU JV*YY% dUJVAV[ou)UMMMo7 sϞ=<Z۶mqPvYqN'@V IyyyxxhX,Ƕ pm?r8=<1˟0Ǐ_{ܸq'$YdȐc۷hѢ-;cǎURCN.Œ=Dfv=f$  өz@MdVVV~ߖ3fkCNDŽ'*88' ȪGF$6muuun-&!z=h0 պtR$ 1 |PP]W7777߷%U p Oquɤ|' ꝅ*22ےȪ^?8nz>a-jiiI)ٜ.3SOѣm66m?oK z=0W;$ P\\f0KKKZcǎ5SL ?oK z=h "$ U* U*dU*dU*d *'hظq)>@Ϊjܹދj:u*Y@VqWdU,..1b,پ};Sdorsseɴi^@{Rߵkװa&O,KeTl4'L Ԯ⮷Hʘ1c]СCƍSn2U)KaV|څ;f5//EEE]o_!?@VʪPFfuժU2o۵ '嬠;ވnegy;;;eAdjppYmjj^bڵ]/F2/w]߸q jQQ,Q^+ub)o.K֭[W__WV{X jMMMdd,IMMUF9ٙsVF666Z,Pڳbbb\v!Y2.&N(S.0eʔs#GT/3tPyuY>|pelvVII Y*Y5 #Fشi L O0A3f{ΪjU>*3sL&YFgddrer(;vlnnnw%Yd Yd Y Y@VY@VYo?XgkIENDB`$$If!vh#v:V l t0655.pyt5&T$$If!vh#vh%:V lA t065."p yt5&TuDyK  _ENREF_33sDyK  _ENREF_5uDyK  _ENREF_37sDyK  _ENREF_8uDyK  _ENREF_17jF 666666666vvvvvvvvv666666>6666666666666666666666666666666666666666666666666hH66666666666666666666666666666666666666666666666666666666666666666p62 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~ OJPJQJ_HmH nH sH tH J`J Normal dCJ_HaJmH sH tH `B i Heading 14$$ & F SS@&^S`(CJKHOJPJQJ_HaJmH sH tH `B i Heading 24$$ & F @SS@&^S`'5CJOJPJQJ_HaJmH sH tH `B i Heading 34$$ & F SS@&^S` CJOJPJQJ_HmH sH tH @ iHeading 4,Hypothesis@$$$ & F `SSdT@&^S`a$)B*CJOJPJQJ^JaJmH phsH r@r i Heading 5 $ & Fd<@&a$%B*OJPJQJ^JaJmH phsH t@t i Heading 6 $ & Fd<@&a$(6B*OJPJQJ^JaJmH phsH v@v i Heading 7 $ & Fd<@&a$)B*CJOJPJQJ^JaJmH phsH x@x i Heading 8 $ & Fd<@&a$,6B*CJOJPJQJ^JaJmH phsH | @|  i Heading 9 $ & Fd<@&a$/56B*CJOJPJQJ^JaJmH phsH DA`D Default Paragraph FontRi@R 0 Table Normal4 l4a (k ( 0No List 6U`6 i0 Hyperlink >*B*phlol iAuthorsx^`-B*CJOJPJQJ_HaJmH phsH tH f>@f iTitle$da$/5;B*CJ OJPJQJaJ mH phsH tH\o!\ i Title Char/5;B*CJ OJPJQJaJ mH phsH tHNo1N i Authors CharB*CJOJPJQJaJphloBl i Basic text $xa$1B*CJOJPJQJ^J_HaJmH phsH tH <OAB< iAbstract6aJmH sH tJ`Bt iSubtitle,References$hx@&$CJOJPJQJ_HaJmH sH tH noqn iSubtitle Char,References CharCJOJPJQJaJmH sH ToT iHeading 1 Char CJKHOJPJQJaJmH sH ToT iHeading 2 Char5CJOJPJQJaJmH sH LoL iHeading 3 CharCJOJPJQJmH sH ~o~ iHeading 4 Char,Hypothesis Char)B*CJOJPJQJ^JaJmH phsH ^o^ iHeading 5 Char)B*CJOJPJQJ^JaJmH phsH `o` iHeading 6 Char,6B*CJOJPJQJ^JaJmH phsH ZoZ iHeading 7 Char%B*OJPJQJ^JaJmH phsH \o\ iHeading 8 Char(6B*OJPJQJ^JaJmH phsH dod iHeading 9 Char/56B*CJOJPJQJ^JaJmH phsH j@j p Table Grid7:V!0!*W`!* ,`Strong5\Bo1B ,apple-converted-space(/A( ,username,/Q, , _timestamp^Ob^ , js-tweet-text&ddd[$\$CJOJPJQJaJ\^@r\ yr0 Normal (Web)'ddd[$\$CJOJPJQJaJ8/8 ugu-hiddenvisuallyFV`F 0FollowedHyperlink >*B*ph4@4 +cq0Header * H$6o6 *cq0 Header CharCJaJ4 @4 -cq0Footer , H$6o6 ,cq0 Footer CharCJaJb@b  List Paragraph.d^`m$ OJPJQJB'`B 0Comment ReferenceCJaJZ@Z 1 0 Comment Text0d`CJOJPJQJaJBB 0 0Comment Text CharOJQJDj@D 3 0Comment Subject2 5PJ\N1N 2 0Comment Subject Char5OJQJ\^@B^ 5 0 Balloon Text4d`CJOJPJQJ^JaJNQN 4 0Balloon Text CharCJOJQJ^JaJOb7 EndNote Bibliography Title6$d`a$CJOJQJmHnHuroqr6 EndNote Bibliography Title CharCJOJQJaJmHnHunOn9 EndNote Bibliography8d`CJOJQJmHnHufof8 EndNote Bibliography CharCJOJQJaJmHnHuD`D 0Revision:CJ_HaJmH sH tH 6X`6 @Emphasis]mHnHu`@`= Quote,Table Text <d5OJPJQJmHnHujj< Quote Char,Table Text Char5CJOJQJaJmHnHu\@\ ? 0 Footnote Text>d`CJOJPJQJaJDD > 0Footnote Text CharOJQJ@&`@ 0Footnote ReferenceH*X`X   No Spacing A` CJOJQJ_HaJmH sH tH Pa!P 0Subtle Emphasis,Hypotheses6Z+@2Z D 0 Endnote TextCd`CJOJPJQJaJBAB C 0Endnote Text CharOJQJ>*`Q> 0Endnote ReferenceH*PK![Content_Types].xmlN0EH-J@%ǎǢ|ș$زULTB l,3;rØJB+$G]7O٭V$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3N)cbJ uV4(Tn 7_?m-ٛ{UBwznʜ"Z xJZp; {/<P;,)''KQk5qpN8KGbe Sd̛\17 pa>SR! 3K4'+rzQ TTIIvt]Kc⫲K#v5+|D~O@%\w_nN[L9KqgVhn R!y+Un;*&/HrT >>\ t=.Tġ S; Z~!P9giCڧ!# B,;X=ۻ,I2UWV9$lk=Aj;{AP79|s*Y;̠[MCۿhf]o{oY=1kyVV5E8Vk+֜\80X4D)!!?*|fv u"xA@T_q64)kڬuV7 t '%;i9s9x,ڎ-45xd8?ǘd/Y|t &LILJ`& -Gt/PK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 0_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!0C)theme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] .#gj #   #,, ***-0ATyj|s.[: "2Y?$Trk~NG7(&1' "#$%&'()*+./012356789;Ynrs}5fn:3?¥å#$CDdeͦϦlmŧƧ !@A^_1MqIs$RlZ!#ao%&1'!,-4:<=>?@ABCDEFGHIJKLMNOPQRSTUVWXZ[\]^_`abcdefghijklmopqtuvwxyz{|~R 2 @ B (RUz|^`!!M!e!g! "_(f((((())),,$-y-{-:.c1e11113L9S99999999W=X====>DDEQETEEEEFCLELxLLLLLL9TZZZZZ[[[^FbGbzbbbbeefffAgjjk@kBkSk3p4pipppuqzt|ttttmx{{{{{|VX46M8:o΍Ѝ Ύގ,FI׏ُNor۔(69i'םٝ +.cqsߡ&(1ҥ کܩ68Hװ/XZm >X[LoqԾվ׾P]`ѿֿؿrac?,<RSZ(MP2GIQawxOilV]*:<#(*\ru'*[gj+dt ?LNB13ertRT     f ;=pN  >TVQg!!!""`#w)y)))))***j.l....z////////v255 66687:::; ;;??K?^?`?tEHHHHHJMMMMM NDRFRvRRR[__"`;`=``aa@e`hbhhhhjrrrssOsiskst7w9wmwwww!{"{V{m{o{rtI9KM')Yvx02e|~MwyI[]Pc "#QX̕QX̕QQ@XXX̕QX̕QX̕QX̕QXX̕QX̕QX̕QXX̕QX̕QXX̕QXX̕QXX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QQ@XX̕QQ@XXXXX̕QXX̕QX̕QX̕QX̕QXX̕QXX̕QX̕QQ@XXXX̕QX̕QX̕QQ@XXXX̕QX̕QQ@XXXX̕QXX̕QX̕QXX̕QQ@XXX̕QX̕QQ@XXX̕QX̕QX̕QX̕QX̕QXXt̕QXX̕QX̕QX̕QXX̕QX̕QQ@X̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕XQX̕QXX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕QX̕Q $&-!WjX8@! g (  q  @ FԔYY?6"Straight Arrow Connector 4c"@@ÑPK!8[Content_Types].xmlAN0EH%N@%邴K@`dOdlyLhoDX3'AL:*/@X*eRp208J妾)G,R}Q)=HiҺ0BL):T뢸WQDY;d]6O&8* VCLj"󃒝 yJ.;[wIC_ :{IOA !>Ø4 p;fɑ3׶Vc.ӵn(&poPK!8! _rels/.relsj0 }qN/k؊c[F232zQLZ%R6zPT]( LJ[ۑ̱j,Z˫fLV:*f"N.]m@= 7LuP[i?T;GI4Ew=}3b9`5YCƵkρؖ9#ۄo~e?zrPK!X#8drs/e2oDoc.xmlS0#4i햨 u)[Q$MPx<.fg^YS$La+e{0hkdɯ7rf[+@/zW6WdOdBOfyzC+}|ZR^KNBZ1Ǹf5 ki?@  +N a"lٺVBhi4L8dV|=A5m`m϶"{jԴ5{#GرZ+Ġ%i}i-/ JNgy>]p&8[EЇOv,nJGZ7>p!6SZS mX_jj i8ݐ{ED[=6ǭFvra9X|;Kx J4 WGBt&e2){ mճ>7ϒYB% m2K@1ڵ0p/brPn2㐢;zY|hk@ʓ=Wio~PK!:: drs/downrev.xmlLN0EH5H@nTx>Mam3`y5Wwivwl8o2`u0#>ޟn*`1)4m]{yѨڄK=04՜G=X&LfŹfV {Nj,ܫà&0X՝񢋥=?Tqؾ3R^_w]_~ Zb:eͩ**rF)%femPK-!8[Content_Types].xmlPK-!8! /_rels/.relsPK-!X#8.drs/e2oDoc.xmlPK-!:: }drs/downrev.xmlPK    l&dYY??"l*&Text Box 8C" @ÏPK!8[Content_Types].xmlAN0EH%N@%邴K@`dOdlyLhoDX3'AL:*/@X*eRp208J妾)G,R}Q)=HiҺ0BL):T뢸WQDY;d]6O&8* VCLj"󃒝 yJ.;[wIC_ :{IOA !>Ø4 p;fɑ3׶Vc.ӵn(&poPK!8! _rels/.relsj0 }qN/k؊c[F232zQLZ%R6zPT]( LJ[ۑ̱j,Z˫fLV:*f"N.]m@= 7LuP[i?T;GI4Ew=}3b9`5YCƵkρؖ9#ۄo~e?zrPK![Tdrs/e2oDoc.xmlU]O0}`}4-+1Ӥ} `ڳ85l);lKj_szti5Y 5%mLS/7LŴ5ӓ/n!&VZ] GbJ*CRNֵ,`QXV&E1U\xy> ڋ@tI-Keآq43XL8AʩߠZŝ{ܶ#[׊j/j%D҂n&`#*)eX݈M oct/põYNJ}o{&iĩsU`7/GfAG[ [6kc tdvHGd\qq6ٟN-;;a[%u|Bg>D6l2䩺PZgWd VZ&SdŮYiG tpv0v*qQ EYlÜ3LnV w0b}Y0⨕!woL 6*H56&(5h,Jkx7Ej!gvUOz|DpYX7l0GD!^Lw?!%=υUn k]vDΚ\E4kpMN–1RT#ÕjStNEC_CRr(qGwR!}IgQ?%A:2K>~,7i̤'VPT? iJz+t7I.v.xZREYV"ǖØ4 p;fɑ3׶Vc.ӵn(&poPK!8! _rels/.relsj0 }qN/k؊c[F232zQLZ%R6zPT]( LJ[ۑ̱j,Z˫fLV:*f"N.]m@= 7LuP[i?T;GI4Ew=}3b9`5YCƵkρؖ9#ۄo~e?zrPK!Al;ddrs/e2oDoc.xmlUN0}^,V*Q ^Xrl^NXVU_x.g.>Y 5JJ榑zY))EMN޾9عΨF8#7]v^wg~X!lY-Ʊ Xg,'~ ".}[1I>!I Y`dozɝ {i[ETln:fExf٠vh֣FbG%`sX(-̘u}#!9@7ZRuXߟR!Lfi4Sn['az:T/֗>dQeJg ]+w2Abeb{\)G qVN˃X#2wZ/jGă?vs"/\EU+Z8* S]0x)QĈ_mbr2GjcVAِZo b0i@/- 0e;J1b> s~ dž4.tfXܥ(#];m$p 8 @C焸1ic`((\os lj:SԈ0yY#G9Ra|$#Bb6/aS$O +~ďX`R歩bN: * #Fb1z՟n<'֙ijʃggpqzhc¥|\!qnCVٕ2?KY7*VU8ߊ礵{O~PK! drs/downrev.xmlLAO@&͘xK ,Mce.-{Ǔ罗7+6 G9R\D jg:j=d |dt|MyyQܸ3 VІ0RE Hhuslm/(ZI;/x}O]cz:Tjށ803:tp/zq-8IAp"Io dYPK-!8[Content_Types].xmlPK-!8! /_rels/.relsPK-!Al;d.drs/e2oDoc.xmlPK-! drs/downrev.xmlPK   @ s bJ~jJ8"Straight Arrow Connector 15S"@fPK!8[Content_Types].xmlAN0EH%N@%邴K@`dOdlyLhoDX3'AL:*/@X*eRp208J妾)G,R}Q)=HiҺ0BL):T뢸WQDY;d]6O&8* VCLj"󃒝 yJ.;[wIC_ :{IOA !>Ø4 p;fɑ3׶Vc.ӵn(&poPK!8! _rels/.relsj0 }qN/k؊c[F232zQLZ%R6zPT]( LJ[ۑ̱j,Z˫fLV:*f"N.]m@= 7LuP[i?T;GI4Ew=}3b9`5YCƵkρؖ9#ۄo~e?zrPK!Fdrs/e2oDoc.xmlS0#4v˶U,TZ${N!.=yfvV5f%g 'mk͒J0Ϊ_Uׯ6_;9#2a=w1uQѩye)8!B" ޛ,C Eco3~(6MPO)vwZL4X-5A!/^ t5q&\_Bhcʳr)?X|D%iBO=GvCt9ki}B|XSqz45FOAKmuLPzXUEPϗG1%@!~TgR01sD bژ,lr1{<ؖ30-XD̼3Z= 3~jQ''pE 8q`2=nɩї`%WO@R#%DLeO'YF!5S'MFO|qPK! / drs/downrev.xmlLN0DHH\(q*kSM"uM{'hGo,9%$`s,DZYF JUh\3*P( CSyT ~v*{gP<"NH5z0wsqޟP-66?ֽ?~-a=Eƿ0 *bju`H% &xU_PK-!8[Content_Types].xmlPK-!8! /_rels/.relsPK-!F.drs/e2oDoc.xmlPK-! / Sdrs/downrev.xmlPK]B S  ?/@  # D H t J Dt Ct H Lt8_ENREF_1_ENREF_2_ENREF_3_ENREF_4_ENREF_5_ENREF_6_ENREF_7_ENREF_8_ENREF_9 _ENREF_10 _ENREF_11 _ENREF_12 _ENREF_13 _ENREF_14 _ENREF_15 _ENREF_16 _ENREF_17 _ENREF_18 _ENREF_19 _ENREF_20 _ENREF_21 _ENREF_22 _ENREF_23 _ENREF_24 _ENREF_25 _ENREF_26 _ENREF_27 _ENREF_28 _ENREF_29 _ENREF_30 _ENREF_31 _ENREF_32 _ENREF_33 _ENREF_34 _ENREF_35 _ENREF_36 _ENREF_37 _ENREF_38 _ENREF_39 _ENREF_40 _ENREF_41 _ENREF_42 _ENREF_43 _ENREF_44 _ENREF_45 _ENREF_46 _ENREF_47 _ENREF_48 _ENREF_49 _ENREF_50 _ENREF_51 _ENREF_52 _ENREF_53 _ENREF_54 _ENREF_55 _ENREF_56f.(9p*sMna  ~ k  r :  'RJe t1g3h !!#  !"#$%&'()*+,-./01234567-'8o)rLm`  } j  q 9  &QIds0f2g !!"#[bck!>F_1g111l@t@  ň#.!$|Z]nq$#"#############<Ox ; R 42E2N!N~PPQQoSS[[M]\]vwՈrfpy?ڿ u59Q\~0-0,DIDV Vt``iata@cCcX|`|~d~/<GKbj Ԉ.1Ίώ'7Z1@ -ƙΙ~PZwǢʢҢ>t-.*<14MU2?u|¿ͿfpP\dv.:1>i|Df-.'(89op)*rsLMmn`a    } ~ j k   q r 9 :   &'QRIJde st01fg23g h !!!!"#"##K#W#################3333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333ǾǾPP#"########## lh0,7t?i,c'z)F1HX) p6Ă4@B@.YG-.HhqM ?g w}p3 ^`OJQJo(p^p`OJQJ^Jo(o @ ^@ `OJQJo( ^`OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(P^P`OJQJ^Jo(o  ^ `OJQJo( ^`OJQJo(^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo(^`OJPJQJ^Jo(-^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo(P^`PsH @@^@`.0^`0..``^``... ^` .... ^` ..... ^` ...... `^``....... 00^0`........ ^`OJQJo(^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo(^`o(.^`.pL^p`L.@ ^@ `.^`.L^`L.^`.^`.PL^P`L. ^`OJQJo(^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo( ^`OJQJo(p^p`OJQJ^Jo(o @ ^@ `OJQJo( ^`OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(P^P`OJQJ^Jo(o  ^ `OJQJo( ^`OJQJo(^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo(^`o(.^`.pL^p`L.@ ^@ `.^`.L^`L.^`.^`.PL^P`L. 8^8`OJQJo(^`OJQJ^Jo(o  ^ `OJQJo(  ^ `OJQJo(x^x`OJQJ^Jo(o H^H`OJQJo( ^`OJQJo(^`OJQJ^Jo(o ^`OJQJo( )F1qM7ti,c'p6YG?gw}4@Hlh                   ~                                                                        -Iov*P%4?`?4?N6Y/%1f/%1flZMnoBqlZMnvu ` s]/z  1# T[ Yh\+M\RJ;, "MY%5&K*3r, .5P&7L91R9'=6={@dDYJSL~ENLR":SE:TITVNXYY DZC^A'bjl #lcquv\w]wy}Bo oR?.Z,_zNI,'e ug- C/y Z.c7.|4:dU4gB2a;(R`jviHsyrxH`t4B+(7DJ}V' ;scKo)z9FHN0 8Z;#"%"EN.InstantFormat EN.Layout EN.Libraries<ENInstantFormat><Enabled>1</Enabled><ScanUnformatted>1</ScanUnformatted><ScanChanges>0</ScanChanges><Suspended>0</Suspended></ENInstantFormat><ENLayout><Style>APA 6th</Style><LeftDelim>{</LeftDelim><RightDelim>}</RightDelim><FontName>Calibri</FontName><FontSize>11</FontSize><ReflistTitle>References</ReflistTitle><StartingRefnum>1</StartingRefnum><FirstLineIndent>0</FirstLineIndent><HangingIndent>720</HangingIndent><LineSpacing>0</LineSpacing><SpaceAfter>0</SpaceAfter><HyperlinksEnabled>1</HyperlinksEnabled><HyperlinksVisible>0</HyperlinksVisible></ENLayout>O<Libraries><item db-id="wdwetf057zd5r9ee22opppf1fdd5sr5t0s9z">My EndNote Library Copy LATEST Copy June 2013<record-ids><item>2</item><item>1170</item><item>2833</item><item>7649</item><item>7650</item><item>7712</item><item>7713</item><item>7714</item><item>7889</item><item>7890</item><item>7894</item></record-ids></item></Libraries>@_8:......#`@``@`B`@``@``@`"`H@`@@Unknown G*Ax Times New Roman5Symbol3. *Cx Arial= Arial BoldA. Arial Narrow7.@Calibri9. ")Segoe UI?= *Cx Courier New;WingdingsA$BCambria Math"qhr,Br,B7QQ20 !"KqHP  $Pi2! xx MSMS8         Oh+'0(x   MS Normal.dotmMS2Microsoft Office Word@@+@_@5`@_@5`Q՜.+,D՜.+,D hp  The Western Union Company !  Title  8@ _PID_HLINKSA @9 _ENREF_17 I3 _ENREF_8 B- _ENREF_37 D' _ENREF_5 B! _ENREF_33 G _ENREF_6 E _ENREF_40 C _ENREF_26 E  _ENREF_47 D _ENREF_55 H _ENREF_9 I _ENREF_8 D _ENREF_55 B _ENREF_34 F _ENREF_7 @ _ENREF_17 E _ENREF_45 D _ENREF_52 C _ENREF_21 B _ENREF_36 B _ENREF_30 C _ENREF_24 B _ENREF_34 D _ENREF_50 B _ENREF_32 B _ENREF_32 B _ENREF_32 B _ENREF_32 C _ENREF_21 E _ENREF_41 C _ENREF_24 D _ENREF_51 E _ENREF_41 B _ENREF_36 Ez _ENREF_41 Et _ENREF_41 Bq _ENREF_39 Bn _ENREF_38 Cf _ENREF_25 D` _ENREF_51 D] _ENREF_50 EZ _ENREF_41 ER _ENREF_41 BO _ENREF_32 DI _ENREF_52 BC _ENREF_3 C@ _ENREF_2 E: _ENREF_47 @7 _ENREF_15 @4 _ENREF_14 F1 _ENREF_7 @) _ENREF_14 D# _ENREF_54 E  _ENREF_43 @ _ENREF_11 I _ENREF_8 B _ENREF_32 E  _ENREF_41 D _ENREF_51 D _ENREF_50 E _ENREF_41 C _ENREF_23 E _ENREF_48 B _ENREF_30 C _ENREF_20 C _ENREF_25 C _ENREF_24 B _ENREF_31 E _ENREF_49 C _ENREF_21 C _ENREF_22 E _ENREF_4 D _ENREF_56 B _ENREF_39 B _ENREF_38 C _ENREF_28 C _ENREF_27 H _ENREF_9 I _ENREF_8 D _ENREF_55 @ _ENREF_15 D _ENREF_55 E _ENREF_42 C _ENREF_29 @ _ENREF_19 F _ENREF_7 @y _ENREF_11 Ds _ENREF_52 @m _ENREF_13 Dg _ENREF_52 @d _ENREF_13 B^ _ENREF_3 C[ _ENREF_2 BU _ENREF_3 CR _ENREF_2 @L _ENREF_14 EF _ENREF_46 @C _ENREF_1 B= _ENREF_37 I7 _ENREF_8 E1 _ENREF_46 @. _ENREF_1 @( _ENREF_16 @" _ENREF_10 D _ENREF_53 E _ENREF_44 @ _ENREF_18 @ _ENREF_12 D _ENREF_53 B _ENREF_354cGhttp://sourceforge.net/projects/pdfsplitextractandmerge/?source=navbar  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~     !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWYZ[\]^_abcdefghijklmnop{Root Entry FWc5`}Data 1Table sWordDocument.,SummaryInformation(XDocumentSummaryInformation8`D!CompObjr  F Microsoft Word 97-2003 Document MSWordDocWord.Document.89q