Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

East is not west : is there validity in cross-cultural usability? Haddad, Shathel Yacoub 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_haddad_shathel.pdf [ 4.52MB ]
Metadata
JSON: 24-1.0052193.json
JSON-LD: 24-1.0052193-ld.json
RDF/XML (Pretty): 24-1.0052193-rdf.xml
RDF/JSON: 24-1.0052193-rdf.json
Turtle: 24-1.0052193-turtle.txt
N-Triples: 24-1.0052193-rdf-ntriples.txt
Original Record: 24-1.0052193-source.json
Full Text
24-1.0052193-fulltext.txt
Citation
24-1.0052193.ris

Full Text

EAST IS NOT WEST: IS THERE VALIDITY IN CROSS-CULTURAL USABILTIY?  by Shathel Yacoub Haddad  B.A., Macalester College, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Computer Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2013  © Shathel Yacoub Haddad, 2013  Abstract  This work reports on the design and evaluation of culturally appropriate technology. We investigated cultural differences related to attitudes toward uncertainty between Western Caucasians (more tolerant) and East Asians (less tolerant). Using theory triangulation of cultural attitudes toward uncertainty, we designed information-minimal and information-rich interfaces and hypothesized they would be culturally appropriate for Caucasians and East Asians respectively. Our design context was Cognitive Testing on a Computer (C-TOC): a home-based computerized test under development, intended to screen older adults for cognitive impairments in the absence of a health professional. Using the two interfaces we designed for one C-TOC subtest, we ran an experiment with 36 participants to investigate the effects of cultural attitudes toward uncertainty on performance, preference and experience of anxiety. We found that East Asians preferred the information-rich interface augmented with security elements and learning support: they found it easier to use and felt less anxious with it. By contrast, Caucasians preferred the simpler information-minimal interface with only elements essential for the primary task. Based on our findings, we provide cultural design guidelines for Western Caucasians and East Asians in interaction contexts characterized by uncertainty, such as cognitive testing. We also provide guidelines for using a short uncertainty avoidance questionnaire as a low-cost method for creating adaptive interfaces that cater to varying cultural attitudes toward uncertainty.  ii  Preface  The study described in this thesis was conducted under the approval of the University of British Columbia’s Behavioral Research Ethics Board (BREB), certificate H09-02293 C-TOC.  iii  Table of Contents  Abstract .......................................................................................................................................... ii	
   Preface ........................................................................................................................................... iii	
   Table of Contents ......................................................................................................................... iv	
   List of Tables................................................................................................................................ vii	
   List of Figures ............................................................................................................................. viii	
   Acknowledgements ....................................................................................................................... ix	
   Dedication ..................................................................................................................................... xi	
   Chapter 1: Introduction............................................................................................................... 1	
   1.1	
   The Power of Culture: An Illustrative Story ...................................................................... 1	
   1.2	
   Technology Design and Culture ......................................................................................... 3	
   Chapter 2: Related Work ............................................................................................................ 7	
   2.1	
   Cognitive Testing On a Computer (C-TOC) ...................................................................... 7	
   2.2	
   Culturability in HCI ........................................................................................................... 9	
   2.3	
   Hofstede’s Uncertainty Avoidance .................................................................................. 12	
   2.4	
   The Theory of Uncertainty Orientation ............................................................................ 13	
   2.5	
   The Theory of Behavioral Inhibition and Activation ....................................................... 14	
   Chapter 3: Cultural Design for Uncertainty............................................................................ 16	
   3.1	
   Design Approach .............................................................................................................. 16	
   3.2	
   Low Fidelity Prototyping ................................................................................................. 17	
   3.3	
   High Fidelity Prototyping................................................................................................. 18	
   Chapter 4: Experiment .............................................................................................................. 21	
   iv  4.1	
   Methodology .................................................................................................................... 21	
   4.1.1	
   Primary Task ............................................................................................................. 21	
   4.1.2	
   Distractor Task .......................................................................................................... 21	
   4.1.3	
   Dependent Measures ................................................................................................. 22	
   4.1.4	
   Participants ................................................................................................................ 24	
   4.1.5	
   Design........................................................................................................................ 25	
   4.1.6	
   Procedure ................................................................................................................... 25	
   4.1.7	
   Apparatus .................................................................................................................. 27	
   4.1.8	
   Hypotheses ................................................................................................................ 27	
   4.2	
   Results .............................................................................................................................. 28	
   4.2.1	
   Initial Analysis: Caucasians vs. East Asians ............................................................. 28	
   4.2.2	
   Primary Analysis: Caucasians vs. East Asian Strangers ........................................... 32	
   4.2.2.1	
   General Anxiety ................................................................................................. 32	
   4.2.2.2	
   Overall Interface Preference............................................................................... 33	
   4.2.2.3	
   Usability Preference (Effort Expectancy) .......................................................... 35	
   4.2.2.4	
   Usability Preference (Trust/Confidence) ........................................................... 36	
   4.2.2.5	
   Performance ....................................................................................................... 37	
   4.2.2.6	
   Use of Supportive Buttons ................................................................................. 37	
   4.2.2.7	
   Attitude toward Uncertainty ............................................................................... 37	
   4.2.3	
   Secondary Analysis with Uncertainty Avoidance Score........................................... 39	
   4.2.3.1	
   Usability Preference (Security) .......................................................................... 40	
   4.2.3.2	
   Usability Preference (Information Richness) ..................................................... 40	
   4.2.3.3	
   Interface-Specific Anxiety ................................................................................. 40	
   v  4.3	
   Discussion ........................................................................................................................ 42	
   4.3.1	
   Reflection on Results ................................................................................................ 42	
   4.3.2	
   Design Guidelines ..................................................................................................... 44	
   4.3.3	
   Reflection on Cultural Attitude toward Uncertainty ................................................. 45	
   4.3.4	
   Reflection on Cultural Research ............................................................................... 46	
   Chapter 5: Conclusion and Future Work ................................................................................ 48	
   Bibliography ................................................................................................................................ 50	
   Appendices ................................................................................................................................... 53	
   Appendix A ............................................................................................................................... 53	
   A.1	
   Study Recruitment Poster ............................................................................................ 54	
   A.2	
   Study Consent Form .................................................................................................... 55	
   Appendix B ............................................................................................................................... 59	
   B.1	
   Usability Questionnaire ............................................................................................... 60	
   B.2	
   Uncertainty Avoidance Profile Questionnaire ............................................................. 63	
   B.3	
   Behavioral Inhibition Questionnaire ........................................................................... 65	
   B.4	
   State Trait Anxiety Inventory ...................................................................................... 67	
   Appendix C ............................................................................................................................... 70	
   C.1	
   Low Fidelity Minimal Interface. ................................................................................. 70	
   C.2	
   Low Fidelity Rich Interface. ........................................................................................ 71	
    vi  List of Tables  Table 1. Mean uncertainty avoidance and behavioral inhibition scores, ordered from highest to lowest score (N=36). ................................................................................................................ 38	
   Table 2. Uncertainty avoidance (UA) grouping for secondary analysis (N=36).. ........................ 39	
    vii  List of Figures  Figure 1. Screenshots of two of the fifteen subtests from the current C-TOC battery: the pattern reconstruction test (above) and the symbol-digit matching test (below). The C-TOC battery is implemented as an interactive PowerPoint prototype. ............................................................... 5	
   Figure 2. High fidelity Minimal interface of the symbol-digit matching task. The instruction does appear for the first trial, but disappears for all subsequent trials. ............................................ 19	
   Figure 3. High fidelity Rich interface of the symbol-digit matching task. ................................... 19	
   Figure 4. Overall anxiety score for Caucasians (n=20) and East Asians (n=16). ......................... 29	
   Figure 5. Overall anxiety score for East Asian Strangers (n=8) and East Asian Acquaintances (n=8). ........................................................................................................................................ 30	
   Figure 6. Boxplots of overall anxiety scores on the Minimal (above) and Rich (below). .…….. 31 Figure 7. Overall anxiety score for Caucasians (n=20) and East Asian Strangers (n=8).............. 33	
   Figure 8. Overall interface preference for Caucasians (n=20) and East Asian Strangers (n=8). .. 34	
   Figure 9. Perceived effort expectancy for Caucasians (n=20) and East Asian Strangers (n=8). .. 35	
   Figure 10. Level of trust in research for Caucasians (n=20) and East Asian Strangers (n=8). ..... 36	
   Figure 11. Individual behavioral inhibition and uncertainty avoidance scores for Caucasians (n=20) and East Asian Strangers (n=8). …………………………………………………... 38 Figure 12. Perception of information richness as excessive for the different UA groups. …….. 41 Figure 13. Interface-specific anxiety for the different UA groups. ……………………………. 41  viii  Acknowledgements  I start by acknowledging my supervisor, Dr. Joanna McGrenere, for all her guidance and support through the myriad of challenges we faced during this exciting research. I thank her for allowing me to study something I am passionate about, for trusting me to plunge both of us into unknown waters, for providing insight, for triggering creativity, and for waiting patiently. I am grateful for her sticking by my side during the dark days of doubt and the bright moments of revelation. I could not have asked for a better supervisor whose academic rigor is equaled by her immense compassion for her students. I thank the C-TOC (Cognitive Testing On a Computer) team at the UBC departments of Computer Science and Medicine for their continuous support and feedback during the entirety of this project. I thank Dr. Claudia Jacova for providing crucial insight regarding the design, user study and data analysis as well as reviewing this thesis. I thank Mathew Brehmer for guiding me during the humble beginning of this work, and for continuing to provide assistance through its following phases. I thank Dr. Charlotte Tang for providing input and connecting me with a network of participants for the user study. I thank Dr. Robin Hsiung, Sarah Le Hurray, William Wang, Nathan Santos, Emily Faith Corenblith, and James Riggs, for all their advice, feedback and support during this project. I thank my second reader, Dr. Karon MacLean, for taking the time to review this thesis and helping to improve it. I thank my friends in the Human Computer Interaction (HCI) program at UBC: Anna Flagg, my confidant, laughter provider and sanity maintainer; Jessica Dawson, Oliver Schneider, and Hasti Seifi, who piloted the user study; Diane Tam, who helped run the user study, Dr. Rock ix  Leung, who helped with participant recruitment for the user study, and all other comrades on the graduate battlefield of the vigorous and exciting HCI program at UBC. I thank all the community and senior centers in Vancouver who welcomed me to post fliers, set information booths during participant recruitment and run user studies on their premises. I especially thank the West End Seniors’ Network, the Kerrisdale Senior Center and the Collingwood Neighborhood House whose warm hospitality helped advance this research. I thank the people at the GRAND NSERC Network Center for Excellence, who partially funded this work. Last but not least, I thank my dear parents: Yacoub and Diana who supported me during the ups and downs of this work and my life through it; my fantastic siblings: Shatha, Ruba and Rabee who gave me unconditional love; and my two little prince nephews, Saif and Kareem, who always brought me joy, unless I had to change their diapers.  x  Dedication  To my mother, Diana ,whose love and strength kept me afloat during the storm.  xi  Chapter 1: Introduction  I am fortunate to have lived and met people in myriad cultural settings in the last ten years of my life. As those years passed, I had observed that culture diversifies more than cuisine, dress and tradition. Not only do people eat, dress and live differently across the globe, they perceive, think and behave differently as well. I had also begun to understand the deep impact of my own cultural upbringing on my human experience in its totality, and I grew interested in doing research about cultural differences. We begin this work with a story of an unfortunate event that shows the impact of culture on human behavior, with the hope it conveys the importance of honoring cultural diversity when designing global technology.  1.1 The Power of Culture: An Illustrative Story On the dark wet eve of December 22 1999, Korean Air Cargo 8509 went from takeoff at the London Stansted Airport to total destruction in less than sixty seconds [2], killing its four Korean crew members and incinerating more than 60 tons of freight into thin air. This incident stands out among other aerial disasters due to various factors. On one hand, the speed with which the catastrophe unfolded took investigators by surprise. On the other hand, investigators found that the incident was avoidable had the Korean Air crew not committed a bizarre combination of human errors. The main question of course was: how did it happen? Initially, investigators found that the cause of the crash was a faulty pilot device called the Attitude Director Indicator ADI, which displays the plane’s bank. Counting on the ADI is especially crucial in dark wet conditions, like those present on the eve of the disaster, when the 1  crew did not have enough visual cues to determine the plane’s bank. The pilot’s ADI misinformed him about the bank, causing him to steer the plane too severely toward the ground shortly after takeoff. With low visibility, the crew could not see their dangerous proximity to the ground, and seconds later it was over. One might jump to the conclusion that the faulty pilot ADI caused the catastrophe and the case was closed. However, the co-pilot had access to a backup ADI that correctly showed the dangerous bank of the plane. Investigators reached the disturbing conclusion that the disaster was avoidable had the co-pilot notified the pilot, or taken over control to correct the aircraft’s bank as international flight protocol calls for. Alas, the co-pilot remained silent even after a loud audible warning horn started sounding in the cockpit. This interaction between the pilot and co-pilot was unconventional in the eyes of western investigators, who realized it was necessary to deconstruct the cultural interaction among the Korean crew to understand these bizarre findings. The cultural investigation shed light on deeply engrained cultural factors that contributed in a major way to the unfolding of the disaster on KA 8509. Two phenomena stood out as being strongly connected to Korean culture and having direct effects on the events in that cockpit. The first one was the superior-subordinate relationship; investigators noticed a stronger hierarchical structure governing the interaction between Korean pilots and co-pilots. This was gleaned both while observing Korean crew training and while listening to the cockpit recording of KA 8509. Korean work culture exhibits a more pronounced status inequality between superior and subordinate, where superiors tend to have more power and subordinates tend to avoid conflict with superiors. The atmosphere in a Korean cockpit is hence characterized not by distributed authority and teamwork, but rather by central authority and relative autocracy. Junior co-pilots, as a result, tend to maintain a subservient attitude with senior veteran pilots. The second 2  phenomenon was related to cultural attitude toward uncertainty. Observation during Korean crew training showed that it focused on procedure, with little emphasis on freethinking or management of unusual circumstances. Korean crews were instructed to trust instruments with little questioning, and to avoid improvising in uncertain situations. This cultural investigation helped clear the mystery of what happened in that cockpit. On one hand, the pilot trusted his instrument, despite having an overwhelming evidence of its defect (alarm horn). On the other hand, the co-pilot chose to remain silent in order to avoid conflict with the senior veteran pilot, even though his instrument demonstrated the danger they were in. Investigators felt that his fear to lose face with his superior seemed to govern his reaction, or lack thereof. In such an instance, cultural values appear to be so deeply entrenched that they constitute a major guiding force of human behavior even in situations where there is strict international regulation, such as commercial aviation.  1.2 Technology Design and Culture The story above shows the power of culture in shaping human beings and driving their behavior. Research has found cultural differences in a myriad of social and cognitive phenomena. The field of cultural psychology is well established. It claims that the connection between mind and culture is bidirectional and intertwined: cultures arise from the participation of minds in them, and the mind develops from participating in a cultural context [12]. Cultures differ from one another in belief systems, social structure, heritage, and other measures that are tangible to culture observers. Cultural differences even persist at the level of basic cognitive and psychological processes. For example, Japanese have been found to perceive contextual information more readily than Americans [16]. Japanese have also been found to focus more on 3  background items and reason holistically compared to Americans who focus on foreground items and reason analytically [21]. Other inter-cultural differences have been found in perception, lowlevel processing, reasoning, self-concept, concept of others and emotional responses [12]. Since culture is influential in shaping us as individuals, and cultures vary across many dimensions, we need to consider culture seriously when designing technology aimed at user populations spanning different cultural loci. In such cases, interface designers should think globally and design locally. Cross-cultural usability, or “culturability” as suggested by Barber and Badre [1], rejects the dominant one-size-fits-all approach of international technology design in favor of one that promotes culturally adapted interfaces. Culturability delivers a culturally customized user experience to international consumers, and should hence be incorporated in the design vision of global technology. Despite the multitude of software technologies that span the globe today, interface design has largely remained framed in a western-centric perspective. Linguistic translation is not enough to provide cross-cultural flavor; it offers only a superficial solution where western culture remains deeply engrained in the design of international technology [20]. To truly design cross-cultural interfaces, we need to look beyond translation and restructure design to manufacture holistic, culturally appropriate user experiences. Our motivation to investigate culturally informed design stems from developing Cognitive Testing on a Computer (C-TOC) – a screening test for cognitive impairments intended for older individuals (55+) to take independently online at home without the need of a test administrator (Fig. 1) [3]. Currently in Canada, the average wait time for a cognitive consultation in a clinical setting ranges between 6 and 24 months during which cognitive performance can degrade further. C-TOC will assist in the early detection of cognitive impairment and provide at4  risk individuals with medical attention faster. C-TOC provided us with a great context within which to explore cross-cultural design. Canada is a very multi-cultural society, thus C-TOC’s design must accommodate users’ different cultural needs and preferences. Further, cultural differences cannot skew test results or cause serious variations in user experience.  Figure 1. Screenshots of two of the fifteen subtests from the current C-TOC battery: the pattern reconstruction test (above) and the symbol-digit matching test (below). The C-TOC battery is implemented as an interactive PowerPoint prototype.  5  A computerized self-administered cognitive test battery such as C-TOC creates an interaction context characterized by uncertainty as we have seen in our early work [3]. First, cognitive testing intrinsically causes uncertainty as it raises personal concerns about cognitive health and aging, especially when cognitive deterioration is a cultural stigma. Second, unassisted test-taking on a computer may contribute additional uncertainty due to the demands of selfadministration or doubts about computer proficiency. Third, cognitive impairment – if present – may create further uncertainty for the test-taker. Fourth, cognitive performance is felt to be highly confidential, and C-TOC users would want their information protected during transmission to health professionals and storage. Fifth, high uncertainty may cause anxiety, particularly when cognitive health (or lack thereof) is in question. While experiencing some anxiety is inherent in the context of C-TOC, excessive levels can reduce the size of working memory and distract from the primary task [9]. This degrades performance and increases the likelihood of false positives. We speculate that cultural differences in attitude toward uncertainty might produce varying user needs for anxiety appeasement and security reassurance when using C-TOC. The contributions of this research include the cross-cultural design of two interfaces for one C-TOC subtest (the digit symbol matching test – Fig. 1), using variation in information richness to appeal to varying cultural attitudes toward uncertainty. We also contribute an evaluation of our designs with Western Caucasian and East Asian users where we observed cross-cultural usability differences. Finally, we provide guidelines for cross-cultural design of interfaces in interaction contexts characterized by uncertainty.  6  Chapter 2: Related Work  We begin by describing C-TOC, the computerized cognitive testing tool for older adults we used as a platform for cross-cultural design. Afterwards, we present prior work in culturability. Next we cover three models related to cultural attitudes toward uncertainty: Hofstede’s uncertainty avoidance, the theory of uncertainty orientation, and the theory of behavioral inhibition. Through triangulation of these three theories, we hypothesized cultural profiles for C-TOC users from the East and West. We also used Hofstede’s uncertainty avoidance to guide cross-cultural design of two interfaces for a C-TOC subtest. Prior to starting our review of cultural theories, it is important to note that such theories describe general cultural tendencies; they do not imply that everyone in a certain culture behaves identically. Nonetheless, those theories can give a sense of different cultural expectations in order to design better culturally adapted interfaces.  2.1 Cognitive Testing On a Computer (C-TOC) Demand for cognitive screening has grown in recent years as older adults have been increasingly seeking early diagnosis of cognitive concerns. This is especially the case where there is a growing aging population such as in Canada, where the average waiting time for a cognitive consultation in a clinic is 6 to 24 months. Such long waiting periods allow cognitive impairments – if present – to progress, and are an indicator of the inability of current health resources to cope with the growing demand for cognitive testing. Current in-clinic instruments, such as the Mini Mental State Exam (MMSE) [11] and the more recent Montreal Cognitive Assessment (MoCA) 7  [22], require the presence of a trained clinician for administration and lack some sensitivity in impairment detection. More sensitive tools such as Neuropsychological Testing (NPT) [14] have higher validity in detecting impairments but require more expert personnel to administer, rendering them only available to selective clinical settings. Computerized cognitive testing provides an innovative solution to this problem: by using widely available technology, we can alleviate some of the pressure off health resources and shorten waiting times for a consultation so that at-risk individuals can be identified and assisted more quickly. Computerized cognitive testing has advantages over traditional approaches in several ways. First, self-administration diminishes the need for health personnel during testing and shortens waiting times as mentioned above. Second, computerized cognitive testing can be transferred to a non-clinical setting such as the home where individuals with mobility issues or those in isolated locales can be accommodated as well. Third, computerized testing allows for a high level of standardization in scoring and more precise measurement of cognitive performance. These benefits of computerized cognitive testing however hinge on careful interface design so that older adults can easily understand and use the test on their own. A successful computerized testing tool should therefore be clinically validated but also easy to understand and use by older users. Cognitive Testing On a Computer (C-TOC) is a computerized testing tool developed by researchers in neuropsychology, behavioral neurology, cognitive and computer science at the University of British Columbia [3]. It was developed to provide high validity in cognitive impairment detection, and is intended for self-administration in a clinical or home setting. The current C-TOC application is developed in PowerPoint and is completed in 30 to 45 minutes, The application contains 15 tests targeting various cognitive functions such as information 8  processing speed, attention, orientation, language, visual and verbal episodic memory, visuospatial skills and executive functioning. The C-TOC interface is augmented with instructions and prompts so it accommodates self-administration with minimal computer skills, and needs only a mouse for interaction. Prior to the research reported in this thesis, the C-TOC application had been reviewed by a panel of experts to validate it clinically and had undergone several usability iterations to ensure it is easy to use for older users. A reasonable next step for C-TOC was to ensure that it is culturally fair; it should accommodate the various needs of the multicultural older user population it is intended to benefit.  2.2 Culturability in HCI Most of the HCI literature in culturability focuses on surveying cultural attitudes towards existing designs rather than designing culturally adapted interfaces and evaluating them. For example, one study investigated different cultural attitudes to electronic products, and identified 10 cultural dimensions relevant to user experience with such products [17]. Another found crosscultural differences in four countries regarding loyalty, trust and satisfaction with local websites [8]. Yet another study evaluated attitudes toward mobile data services in three countries and found that users in a shared cultural context tend to like/dislike the same features [6]. Similar work has been done on preferences for mobile phone design, where one study found differences between Indians and Americans in views of privacy in mobile phones [15]. Another found evidence of cultural differences in five countries in terms of experience and evaluation of a smartphone [30]. Most of this surveying work was aimed at detecting cultural differences or identifying cultural dimensions related to user experience with technology. 9  Outside of the HCI field, more extensive work has been done to identify cultural points of variation in human experience and to build models to describe them. One of the best-known and extensive examples of such models is the Hofstede model [13]. Geert Hofstede is a Dutch cultural anthropologist who created one of the most comprehensive cultural models to date. Between 1967 and 1972, Hofstede worked at IBM, and analyzed survey data from about 116,000 IBM employees in 40 countries in one of the largest studies of its kind. This database grew over the years to include in its current state data from 76 countries. Hofstede’s survey solicited information about how culture influences values in the workplace, and enabled him to identify dimensions where cultures differ organizationally. Initially, four dimensions were identified: Power Distance, Individualism versus Collectivism, Masculinity versus Femininity, and Uncertainty Avoidance. Two more dimensions were added in 1991 and 2010: Long-Term Orientation and Indulgence versus Restraint respectively. All countries that were analyzed were given a score for each dimension between 0 and 100. It is important to note that these scores are relative and are used for comparison purposes only as they do not reflect absolute values. They also describe general cultural tendencies in the investigated countries and do not imply that every citizen will exhibit their country’s cultural profile. While much of the literature about culture and HCI emanates from surveys that serve to identify cultural dimensions, little work exists in the area of cross-cultural interface design. Aaron Marcus, by contrast, extended such survey work by providing heuristics for cross-cultural design [18]. In one study, he investigated how Hofstede’s first five cultural dimensions (Power Distance, Individualism/Collectivism, Masculinity/Femininity, Uncertainty Avoidance and LongTerm Orientation) interacted with five components of user interface design (metaphor, mental model, navigation, interaction, and appearance). By surveying websites in various countries, 10  Marcus investigated how each design component varied with respect to each cultural dimension. Through such observations, he was able to provide a set of heuristics for altering the five design components in order to enhance cultural appeal of web interfaces. For instance, regarding the design component of mental model, Marcus suggests that individualistic societies (e.g. Canada) prefer task-oriented models that highlight personal achievement, while collective societies (e.g. China) prefer role-oriented models that highlight connectedness and harmony. In a follow-up study, Marcus surveyed 57 UI design experts from 21 countries to see what cultural dimensions they thought were relevant to technology design. He concluded that five cultural dimensions are most relevant to cross-cultural design: Context, technological development, time perception, authority conception, and uncertainty avoidance [18]. We focus on uncertainty avoidance in our work due to the relevance of uncertainty in the context of usability of C-TOC. While Marcus extends what was done in previous survey work by providing some guidelines for culturability, he did not discuss how it affects user preference or performance, nor did he run experiments where cross-cultural design was actually carried out and evaluated. Reinecke and Bernstein (2011) carried out an actual evaluation of cross-cultural design [25]. They designed Mocca, a task management application, and evaluated two interfaces for it: a fixed culturally neutral interface and an adapted one based on user response to a cultural screening questionnaire. The neutral interface was created along the guidelines of American design since American designers and companies provide a large portion of websites and software solutions globally. The culturally adapted interface was generated by creating a cultural user profile, and then using it to modify the neutral interface. To create the cultural user profile, the system calculated weighted user scores of the Hofstede cultural dimensions based on the 11  countries a user had lived in and the duration of residence in each. Next, and using design heuristics similar to the ones created by Marcus [18], the system modified the neutral interface to align with the cultural user profile. The findings showed that users preferred the culturally adapted interface to the neutral one and performed better on it. These results point to the importance of culturability when designing interfaces for multicultural user populations, as it does have an impact on user preference and performance. As we have seen with the experience of cross-cultural design for Mocca above, it is important to build a cultural user profile to drive the cultural adaptation of the interface. Since uncertainty is a prominent theme for C-TOC users, we embarked on building our cultural user profiles based on cultural variations in attitude toward uncertainty. In the next sections, we present three theories related to cultural attitude toward uncertainty, and employ theory triangulation to create cultural user profiles for the East and West to drive our cross-cultural design. It is important to clarify that the cultural theories we are about to present do not intend to generalize to all individuals in the observed cultures, or pigeonhole people in rigid cultural categories. Instead, these theories describe observed tendencies in these cultures, similar to how gender differences in behavior for instance also describe general tendencies, without implying that all women or all men behave the same way.  2.3 Hofstede’s Uncertainty Avoidance Our first theory related to cultural attitude toward uncertainty is represented by one of the dimensions of the Hofstede cultural model described in the previous section. Uncertainty avoidance describes society’s attitude to uncertainty and tolerance of ambiguity. Based on 12  Hofstede’s survey, studied countries were given relative uncertainty avoidance scores and behavioral differences were observed between high and low uncertainty avoidance societies. Countries with high uncertainty avoidance scores were found to generally maintain rigid codes of conduct and to be less tolerant of untraditional behavior or new ideas. Countries with a low uncertainty avoidance score were found to be more tolerant of uncertainty, eccentric behavior and unfamiliar ideas. Hofstede found a cultural contrast in uncertainty avoidance between the East and West. He found countries like the USA and Canada to be relatively less uncertainty avoidant, and more tolerant of new untraditional contexts. By contrast, countries like Japan and Korea were found to be more uncertainty avoidant, as they seemed to maintain more rigid behavioral codes and feel less comfortable with novelty.  2.4 The Theory of Uncertainty Orientation The theory of uncertainty orientation by Shuper et al. asserts that people differ in how they feel in uncertain circumstances and how they resolve uncertainty [26]. The theory stipulates that individuals tend to be either certainty-oriented or uncertainty-oriented. Uncertainty-oriented individuals feel fairly comfortable and oriented when confronted with uncertainty and tend to actively seek out information independently to resolve it. Uncertainty-orientation was found to be associated with high individualism in a society. In such societies, people tend to lead more independent lives where they take care of themselves, hence they tend to deal with uncertainty independently as well. Certainty-oriented individuals, by contrast, tend to react more apprehensively in uncertain circumstances and seem to rely more on others to resolve the uncertainty. Certainty-orientation was found to predominate in collective societies, where 13  interdependence is more common, hence people tend to resort to others in their network for certainty resolution. There is also evidence of an East-West cultural dichotomy in uncertainty orientation. Shuper et al. found that Canadians are significantly more uncertainty-oriented than Japanese [26]. Western cultures tend to be uncertainty-oriented since they are more individualistic, and tend to promote independence in resolving uncertainty. In contrast, Eastern cultures tend to be certainty-oriented because they are more collective in nature, promoting interdependence to deal with uncertain circumstances. These findings resonate with Hofstede’s findings about the EastWest contrast in uncertainty avoidance, as Japanese are significantly more uncertainty avoidant than Canadians.  2.5 The Theory of Behavioral Inhibition and Activation The theory of behavioral inhibition and activation, by Carver and White, postulates that there are two separate neurological systems that regulate behavior [4]. The first system, the Behavioral Inhibition System (BIS), promotes inhibitive behavior to steer away from undesired situations. It is triggered by cues of punishment, non-reward or novelty. It regulates experiences of anxiety, fear or frustration as a result of such cues. Individuals who possess a heightened BIS are more prone to anxiety and more inclined to be conservative in unfamiliar contexts. The second system, the Behavioral Activation System (BAS), motivates the individual toward coveted goals. It is triggered by signals of reward and non-punishment, and generates positive feelings of hope and happiness. Those with a heightened BAS are consequently more willing to engage in unfamiliar or ambiguous contexts. The two motivational systems reside in separate neurological networks in 14  the brain making them orthogonal. Consequently, an individual may possess any combination of BIS and BAS systems (i.e. both hyper, both dormant, or one hyper and one dormant). Once again, research in cultural psychology found evidence of a contrast between the East and West in behavioral inhibition. Tanaka and Yamauchi found that collectivism is associated with high behavioral inhibition [28]. This implies that collective cultures (e.g. China) have more activated BIS, and tend to be more vulnerable to anxiety in novel or uncertain situations. In comparison, individualistic cultures (e.g. Canada) tend to possess less active BIS, feel less aversion to uncertain situations and experience less anxiety. The three theories presented here provided us with direction to build our cultural user profiles as we show in the following section.  15  Chapter 3: Cultural Design for Uncertainty  Given the importance of culturability for the design of C-TOC, we ventured into the crosscultural design of one of its subtests: the symbol-digit matching test (Fig. 1). This is one of the 15 tests that together form the existing PowerPoint C-TOC testing battery. In this task, the user needs to look at a central large figure, and search for its match out of nine smaller figures arranged in a row below it. The user then submits his/her answer by clicking on the number that corresponds to the matching small figure. This task targets the cognitive functions of attention and information processing speed, and intentionally requires the user to click on a numbered button instead of the matching figure directly. We wanted to design two interfaces of this test so that one would cater to the cultural needs of East Asian users (from Japan, China, and Korea), while the other would address those of Western Caucasian users of European descent.  3.1 Design Approach First, we created cultural user profiles for East Asians and Western Caucasians through triangulation of the three cultural theories described in Chapter 2. We hypothesized that individualistic Western Caucasian users would be generally less uncertainty avoidant according to Hofstede, uncertainty oriented, and less behaviorally inhibited. By contrast, collective East Asian users would be more uncertainty avoidant, certainty oriented, and more behaviorally inhibited. Second, we employed Hofstede’s uncertainty avoidance to guide our design of the two cultural interfaces. Previous research in uncertainty showed that users who are less tolerant of 16  uncertainty prefer efficient layouts of large amounts of information on the screen, clear labeling and secondary information about content [6] whereas those more tolerant of uncertainty prefer less information and see little value in secondary information. Hofstede also found a strong correlation between uncertainty avoidance and anxiety where high uncertainty avoidant cultures are more susceptible to stress [13]. Similarly, one study found Cypriots (high uncertainty avoidance) to be more nervous about e-commerce, and more demanding of security reassurance than British (low uncertainty avoidance) [10]. As a result, we postulated that it would be important to provide high uncertainty avoidance users with further reassurance through learning support, and elements of security.  3.2 Low Fidelity Prototyping We started with low fidelity prototypes using paper and sticky notes (see Appendix C). Based on the findings of the studies above on high uncertainty avoidance users, we wanted to augment the existing PowerPoint C-TOC test interface (Fig. 1) with additional information for East Asian users in order to provide extra guidance in uncertainty, deliver security reassurance and soothe anxiety. We wanted to reach a point of balance in the design for East Asians where we augment the interface to cater to their cultural needs without rendering it too cluttered. We presented the low fidelity prototypes to the C-TOC research team to ensure we were preserving the test integrity and creating a meaningful cultural contrast in the design. The team’s feedback reassured us that test integrity was kept since both interface designs contained the primary task elements (central figure, nine small figures, nine numbered buttons, and instructions) with identical dimensions and in the same layout. To get further feedback on usability, we presented our prototypes in a weekly research meeting of Human Computer Interaction researchers. After these 17  iterations, we created high fidelity web prototypes of two interfaces varying in information richness.  3.3 High Fidelity Prototyping The high fidelity web prototypes (Fig. 2 and 3) were implemented using HTML and JavaScript in lieu of PowerPoint like the existing C-TOC application. We did this because we needed complex logging functionality for the user study, and wanted to extend our reach to remote users if need be. We kept the layout of the task elements as faithful as possible to that in the PowerPoint application (Fig. 1) and used the same shape images. We reviewed the high fidelity web prototypes with the C-TOC research team to ensure the integrity of the task was still preserved. After that, we piloted the high fidelity prototypes with 6 Human Computer Interaction researchers and one female over 55 to check for usability issues. We designed a Minimal interface (Fig. 2) for Caucasian users, who are more tolerant of uncertainty. In this interface, we included only the task elements necessary for the primary task (central figure, nine small figures, nine numbered buttons, and instructions) as they appeared in the existing C-TOC interactive PowerPoint application (Fig. 1). The instruction does appear for the first trial, but disappears for all subsequent trials to reduce information richness further. We hypothesized that the Minimal interface would be a better fit for Caucasians who are more comfortable with uncertainty. By contrast, this interface might seem information-deficient in a manner that could induce anxiety for East Asians.  18  Figure 2. High fidelity Minimal interface of the symbol-digit matching task. The instruction does appear for the first trial, but disappears for all subsequent trials.  Figure 3. High fidelity Rich interface of the symbol-digit matching task.  19  For East Asian users who are less comfortable with uncertainty, we designed a Rich interface (Fig. 3). The Rich interface is basically an augmented Minimal interface; it contains all its elements in addition to: (1) Instruction repetition – instructions continue to appear in each trial of the task; (2) Security elements – a UBC logo and an information security icon are added in the left and right top corners respectively; (3) Background information - a test purpose button which when clicked produces a popup containing background information including test justification, description and scoring scheme; (4) Learning support elements – hovering popups for the central figure, the small nine figures, and the nine numbered buttons. These popups reinforce the instructions by clarifying what the task elements are and how users should interact with them (e.g. “do not click on small figures”, “click on the numbered buttons”). The interface also contains a demo button that links to a video showing how to do the task. We postulated that the Rich interface would provide East Asians with a higher sense of control over the environment, soothing anxiety about uncertainty. By contrast, its richness might feel like a distraction from the task to Caucasians. Once we had designed two interfaces varying in information richness, we wanted to observe how Caucasians and East Asians interacted with them and which they would prefer. Our experiment to evaluate these interfaces is described in the next chapter.  20  Chapter 4: Experiment  We conducted an experiment to investigate the effects of user attitude toward uncertainty for Caucasians and East Asians while taking a self-administered cognitive test. We evaluated our two interface designs of a C-TOC subtest with the two cultural user groups.  4.1 Methodology  4.1.1 Primary Task The primary task was a symbol-digit matching task (Fig. 1). This is one of the 15 tasks that together form the C-TOC testing battery under development. In this task, the user needs to look at a central large figure, and search for its match out of nine smaller figures arranged in a row below it. The user then submits his/her answer by clicking on the number that corresponds to the matching small figure. This task targets the cognitive functions of attention and information processing speed, and intentionally requires the user to click on a numbered button instead of the matching figure directly. In the existing PowerPoint C-TOC application, users do one block of 20 trials of this task.  4.1.2 Distractor Task We favored a within-subject design in this experiment in order to expose each user to both the Rich and Minimal interfaces (as shown in Fig. 2 and 3). Given the task was identical for both  21  interfaces, we sandwiched a distractor task in between interface conditions in order to reduce carryover effects. The distractor task was a 2-back working memory task [24]. In this task, the user needs to monitor a sequence of flashing images and click quickly when the current image repeats what they saw exactly two images ago. For instance if the user sees a cat, then a fish, then a ball, then a fish, they should click on the fish the second time. The 2-back working memory task is an instance of the n-back working memory task. This task was a good candidate for two reasons: First, it imposes a significant demand on working memory, providing a considerable distraction between the two interfaces. Second, it has a fixed duration so that time spent on the distractor task is held constant across participants, avoiding a confounding effect. While this distraction minimized the assumed carryover effect of doing the symbol-digit matching task twice, it did not completely eliminate it.  4.1.3 Dependent Measures We collected data for ten measures related to anxiety, preference, performance, use of supportive buttons and dimensions related to uncertainty. See Appendix B for the complete versions of the questionnaires we describe here.  For anxiety, we measured: (1) General anxiety using the State Trait Anxiety Inventory [27], a 20-item questionnaire through which users self-report their current state by indicating their dis/agreement on a 4-point Likert scale with short statements such as “I feel at ease”, and “I feel indecisive”. See Appendix B.4 for the full questionnaire. 22  (2) Interface-specific anxiety using a 4-item block in the usability questionnaire (described below), adapted from the Unified Theory of Acceptance and Use of Technology (UTAUT) questionnaire [29] such as “The interface was intimidating to me” or “I hesitated while doing the task for fear of making mistakes that I could not correct.”  For preference, we measured: (3) Overall interface preference via a question comparing the two interfaces at the conclusion of the study. (4) Usability preferences using an 8-item usability questionnaire administered after completion of the task block on each interface. This was partially adapted from the UTAUT questionnaire [29], and targeted the usability areas of effort expectancy, information richness, trust/confidence and security using 4-point Likert scale questions. See Appendix B.1 for the full questionnaire.  For performance, we measured: (5) Completion time, the mean trial completion time on each interface averaged over 12 out of 15 trials. The first 3 trials were excluded as practice trials. (6) Response accuracy: the correct selection of the number corresponding to the small figure that matches the central figure. (7) Error clicks on interface elements that should not be clicked, such as the central figure or the small figures. Correct clicks include only the number buttons, the demo button or test purpose button in the Rich interface.  23  For use of supportive buttons, we measured: (8) Clicks on supportive buttons in the Rich interface.  For dimensions related to uncertainty of, we measured: (9) Hofstede’s uncertainty avoidance score, calculated using the uncertainty avoidance portion of the Hofstede 2008 Values Survey Module, which solicits user opinion about statements such as “Children must be taught to be organized and avoid ambiguity”. See Appendix B.2 for the full questionnaire. (10) Behavioral inhibition score, measured using the behavioral inhibition portion of the BIS/BAS scale [4], where users evaluate items such as “I worry about making mistakes”. See Appendix B.3 for the full questionnaire.  4.1.4 Participants We recruited 20 Caucasian participants (mean age = 64.7 yrs., 14 female). They were from Canada (10), the USA (4), the UK (4) and Germany (2), and had lived in Canada an average of 51 years. This meant that all Caucasian participants have lived in western countries their entire lives. We recruited 16 East Asian participants (mean age = 64.4 yrs., 10 female). One participant came from Japan, the rest from Hong Kong or the Chinese mainland. They had been living in Canada for an average of 40 years, and all of them had lived the first 18 years of their lives in East Asia. Having lived in Canada for a long period of time, a concern arises that these East Asians might identify with Western more than Eastern culture. However, research in cultural psychology supports our assumption that our East Asian participants would identify more with Eastern culture since they immigrated after spending their formative years in the East. For 24  example, Cheung et al. found that Canadian immigrants from Hong Kong did identify with western culture the longer they were exposed to it, but only if exposure occurred during formative years [5]. All participants culturally self-identified as Caucasian or East Asian. They were all over 55 years old and comfortable with English. We ensured all participants were free of diagnosed cognitive impairments and assumed their cognitive abilities were normally distributed. We recruited Caucasian participants through flyers (see Appendix A.1) posted around Vancouver, Craigslist advertising, and information booths setup in senior or community centers. These means were not sufficient to recruit enough East Asian participants, which led us to rely on the personal networks of our research colleagues. Eventually, half of the East Asian sample (8/16) consisted of individuals recruited through our research network. The difference in familiarity with the researcher in this sample proved to have some repercussions on our results, as we will explain later.  4.1.5 Design We used a 2x2 design with a single exposure to each interface. Our factors were ethnicity (between subject: Caucasian or East Asian) and interface (within subject: Rich or Minimal). Order of presentation of interface was fully counterbalanced for gender and ethnicity.  4.1.6 Procedure To conduct the study, we met participants at a place of their choosing such as public libraries, coffee shops, community centers, and private homes. We told them that they would go through a 25  self-administered  computer  application  containing  cognitive  tasks  interleaved  with  questionnaires, and it would take about 30 minutes to complete. We encouraged them to go through the application independently and to only ask for assistance as a last resort. After getting informed consent (see Appendix A.2), we began the prototype with instructions for interaction alongside an overview of the various components. Next we administered the uncertainty avoidance and behavioral inhibition questionnaires. After that, we showed a 20-second video of a beach at sunset in order to calm users before exposure to the first interface condition. We then presented the Rich or Minimal interface where users did a block of 15 trials with each interface. We administered the State Trait Anxiety Inventory immediately after exposure to the first interface condition in order to get the most accurate measure of anxiety as a result of interaction. Next we administered the usability questionnaire for the first interface condition. We then presented the distractor cognitive task (users were not told it was a distractor) to refresh the contents of working memory after the first interface. Users did two trials of this task for a total of about 2 minutes. Next, we showed a different 20-second relaxing video before presenting the second interface condition, followed by the anxiety inventory and the usability questionnaire. At the conclusion of the prototype, we administered the interface comparison questionnaire. A short open structure interview was conducted at the conclusion of the study to discuss user experience and comments. After that, participants were provided with a small monetary sum as compensation.  26  4.1.7 Apparatus We used a MacBook Pro laptop with Mac OS X Lion 10.7.5 operating system. The experiment prototype was developed in HTML and JavaScript and was run on Mozilla Firefox 16.0.2. Participants interacted with the prototype using the laptop keyboard and a Logitech optical scroll mouse. Videos were hosted on YouTube. Both the Rich and Minimal interfaces were implemented on an 800x600 pixel layout so no scrolling was needed for a full view. The primary task elements (central figure, nine small figures and nine numbered buttons) occupied identical positions on both interfaces and matched in size, color and resolution. The four additional elements in the Rich interface were placed in between the primary task elements so that spatial relationships were preserved in both interfaces.  4.1.8 Hypotheses Our main hypotheses can be summarized as follows: H1-Anxiety: East Asians will experience less anxiety with the Rich than the Minimal. Caucasians will experience less anxiety with the Minimal than the Rich, or no difference. H2-Preference: East Asians will prefer using the Rich interface and Caucasians will prefer the Minimal interface. H3-Performance: East Asians will perform better on the Rich than the Minimal. Caucasians will perform better on the Minimal than the Rich. H4-Use of supportive buttons: On the Rich interface, East Asians will use supportive buttons elements more than Caucasians.  27  H5-Attitude toward uncertainty: East Asians will have higher uncertainty avoidance and behavioral inhibition scores than Caucasians.  4.2 Results We began our analysis using our original 2x2 design of ethnicity (Caucasian or East Asian) and interface (Rich or Minimal), which we describe in section 4.2.1. However, an unexpected result regarding one of our dependent measures alerted us to a possible nuance in experimental settings for the East Asian sample. To remedy this issue, we performed our primary analysis excluding a part of the East Asian sample, as we describe in section 4.2.2. In section 4.2.3, we present an exploratory analysis using uncertainty avoidance score as a factor in lieu of ethnicity.  4.2.1 Initial Analysis: Caucasians vs. East Asians For all measures, we first performed a 2x2 mixed model ANOVA using ethnicity (Caucasian and East Asian) and interface (Rich and Minimal). We obtained some evidence of cultural differences in overall preference and usability factors that fully or partially met our hypotheses. However, the finding of overall anxiety caught us by surprise: East Asians felt less general anxiety than Caucasians on both interfaces. Scores on the State Trait Anxiety Inventory showed no effect of interface and a main effect of ethnicity (F1,34 = 4.11, p = .050, η2 = .108, power = .505), where East Asians scored lower on both interfaces (Fig. 4). This evidence contradicts cultural theories about uncertainty that suggest East Asians would have felt more anxiety than Caucasians. We reflected on what could have caused this unexpected result. As mentioned earlier, we had to resort to personal networks of our research 28  colleagues to build our East Asian sample, as public recruitment had proved insufficient. Consequently, our East Asian sample consisted of two equal-sized subgroups differing in level of familiarity with the researcher administering the study: East Asian Strangers (those recruited through Craigslist or public ads) and East Asian Acquaintances (those recruited through networking). On the other hand, Caucasians were all strangers to the researcher and none were recruited through personal networks. Since East Asian Acquaintances knew the researcher indirectly, we speculated they experienced relatively less anxiety than East Asian Strangers who did not know the researcher either directly or indirectly. To investigate this, we performed a secondary 2x2 mixed model ANOVA on the East Asian sample using familiarity with the researcher (Stranger and Acquaintance) and interface (Rich and Minimal).  Overall	
  anxiety	
  score	
    10	
   8	
   6	
   4	
   2	
   0	
   Minimal	
    Rich	
    Caucasian	
    East	
  Asian	
    Figure 4. Overall anxiety score for Caucasians (n=20) and East Asians (n=16).  29  Our speculation bore out: East Asian Strangers experienced more general anxiety on the Minimal than the Rich, whereas Acquaintances showed no difference in anxiety. Scores on the State Trait Anxiety Inventory showed an interaction of familiarity with researcher and interface (F1,14 = 3.99, p = .052, η2 = .213). Pairwise comparisons showed that East Asian Strangers were significantly more anxious on the Minimal than the Rich (p = .050), whereas East Asian Acquaintances had a relatively lower level of anxiety on both interfaces with no significant difference (Fig. 5). The boxplots of overall anxiety scores on the two interfaces show the range of scores and provide an indicator of individual differences (Fig. 6). They show that East Asian Strangers’ scores were lower and less spread out on the Rich than the Minimal.  Overall	
  anxiety	
  score	
    10	
   8	
   6	
   4	
   2	
   0	
   Minimal	
    Rich	
    East	
  Asian	
  Stranger	
    East	
  Asian	
  Acquaint.	
    Figure 5. Overall anxiety score for East Asian Strangers (n=8) and East Asian Acquaintances (n=8).  The different experiences of anxiety between East Asian Strangers and Acquaintances confirmed the confounding effect of familiarity with the researcher on anxiety between East 30  Asian Acquaintances on one hand (knew researcher) and the Caucasians or East Asian Strangers on the other (did not know researcher). Thus, we decided to exclude the Acquaintances from the primary analysis to eliminate this effect of familiarity with the researcher. In this manner, we would compare Caucasians to only those East Asians who shared the same level of familiarity with the researcher.  Figure 6. Boxplots of overall anxiety scores on the Minimal (above) and Rich (below).  31  4.2.2 Primary Analysis: Caucasians vs. East Asian Strangers We performed a 2x2 mixed model ANOVA using ethnicity (20 Caucasian and 8 East Asian) and interface (Rich and Minimal). All results reported here are effects of ethnicity and/or interface. Order of presentation had no significant effect on any of the results reported here. All measures were checked for normality. Completion time and overall anxiety were positively skewed, so we performed a log transform prior to the ANOVA in order to normalize the data. Data of self-reported Likert-scale measures, such as usability questions, were also not normal. For those measures, we performed the Align Transform Rank (ART) procedure prior to the ANOVA in order to normalize the data and enable interaction analysis [31]. Since the data was transformed prior to analysis, the graphs of our raw data results do not show error bars. Pairwise comparisons on significant interactions were performed using a Bonferroni correction to protect against type I error. We report on measures that were significant (p < .05). Additionally, we report observed power and partial eta-squared (η2), a measure of effect size, where 0.01 is a small effect size, 0.06 is medium, and 0.14 is large [7].  4.2.2.1  General Anxiety  East Asian Strangers were less anxious on the Rich than the Minimal, and were less anxious on the Rich than Caucasians. Scores on the State Trait Anxiety Inventory showed an interaction of ethnicity and interface (F1,26 = 7.831, p = .010, η2 = .231, power = .768). Pairwise comparisons showed that East Asian Strangers were less anxious on the Rich than the Minimal (p = .012), and that they were less anxious than Caucasians on the Rich (p = .036). (Fig. 7) This partially supported H1. No difference was found in interface-specific anxiety. 32  Overall	
  anxiety	
  score	
    10	
   8	
   6	
   4	
   2	
   0	
   Minimal	
    Rich	
    Caucasian	
    East	
  Asian	
  Stranger	
    Figure 7. Overall anxiety score for Caucasians (n=20) and East Asian Strangers (n=8).  4.2.2.2  Overall Interface Preference  East Asian Strangers preferred the Rich interface, and Caucasians preferred the Minimal interface or had no preference. Responses to the overall preference question on the interface comparison questionnaire showed a significant main effect of ethnicity (F1,26 = 4.31, p = .048, η2 = .142, power = .516). 5 out of 8 East Asian Strangers preferred the Rich to the Minimal. By contrast, 16 out of 20 Caucasians preferred the Minimal interface or had no preference. (Fig. 8)  33  10	
    #	
  of	
  par@cipants	
    8	
    6	
    4	
    2	
    0	
   Minimal	
   Caucasian	
    No	
  pref	
    Rich	
    East	
  Asian	
  Stranger	
    Figure 8. Overall interface preference for Caucasians (n=20) and East Asian Strangers (n=8).  Qualitative comments revealed that 4 out of 8 East Asian Strangers weighed security elements in their preference for the Rich interface, explaining that they rendered it more “legitimate” and “professional”. By contrast, 3 out of 20 Caucasians mentioned security as a factor influencing interface preference. Almost half (9 out of 20) specifically mentioned lack of perceived “clutter” as one of the reasons for their preference. Altogether the overall preference data and the qualitative comments provide support for H2. In order to understand these preferences, we looked at user feedback on the usability areas of effort expectancy, information richness, trust/confidence and security.  34  4.2.2.3  Usability Preference (Effort Expectancy)  East Asian Strangers found the Rich easier to use than the Minimal. In terms of the Minimal interface, Caucasians found it easier to use than East Asian Strangers. Responses to the question “I found the test interface easy to use” showed a significant interaction of interface and ethnicity (F1,  26  = 18.129, p < .001, η2 = .393, power = .984)(Fig. 9). Pairwise comparisons  showed that East Asian Strangers found the Rich interface easier to use than the Minimal (p < .001), and that Caucasians found the Minimal interface easier to use than East Asian Strangers (p = .003). Thus effort expectancy might be one factor that explains the overall preference finding reported in Section 4.2.2.2.  Perceived	
  effort	
  expectancy	
    2	
   1.6	
   1.2	
   0.8	
   0.4	
   0	
   Minimal	
    Rich	
    Caucasian	
    East	
  Asian	
  Stranger	
    Figure 9. Perceived effort expectancy for Caucasians (n=20) and East Asian Strangers (n=8).  35  4.2.2.4  Usability Preference (Trust/Confidence)  Caucasians were more trusting of the research than East Asian Strangers, and trust was higher on the Rich interface. Responses to the question “I have confidence in the legitimacy of the research team which designed the interface” showed a main effect of ethnicity (F1,26 = 7.501, p = .011, η2 = .211, power = .753), and a main effect of interface (F1,26 = 7.158, p = .012, η2 = .204, power = .733) (Fig. 10). This result proved important in understanding recruitment problems we had faced with East Asians and the variation in attitudes toward uncertainty between the two cultural groups, but cannot be used directly to explain the overall preference finding reported in Section 4.2.2.2. No quantitative difference was found in the usability areas of security or information richness.  2	
    Trust	
  in	
  research	
    1.6	
   1.2	
   0.8	
   0.4	
   0	
   Minimal	
    Rich	
    Caucasian	
    East	
  Asian	
  Stranger	
    Figure 10. Level of trust in research for Caucasians (n=20) and East Asian Strangers (n=8).  36  4.2.2.5  Performance  No effects of ethnicity on performance. Mean completion time showed no effect of ethnicity or interface. Data for response accuracy and error clicks showed a ceiling effect, where participants performed well on both interfaces with very few incorrect answer choices and few error clicks. H3 was not supported.  4.2.2.6  Use of Supportive Buttons  No quantitative evidence of ethnicity effect on use of supportive buttons. On the Rich interface, no participant clicked on the test purpose button, and only 1 East Asian Stranger clicked on the demo button, resulting in no significant difference with regard to use of supportive buttons. H4 was not supported quantitatively. Qualitatively however, 5 out of 8 East Asian Strangers indicated they weighed support buttons in their interface preference, whereas only 2 out of 20 Caucasians indicated appreciation for this additional support.  4.2.2.7  Attitude toward Uncertainty  No cultural difference found in uncertainty avoidance or behavioural inhibition scores. But the two scores were correlated. Contrary to H5, no significant difference was found between the two cultural groups in uncertainty avoidance score (F1,26 = .860, p = .363, η2 = .035, power = .145) or behavioural inhibition score (F1,26 = 2.319, p = .141, η2 = .088, power = .310), although East Asian Strangers scored higher on both (Table 1). We reflect on the low observed power of these tests later. We also computed the Pearson product-moment correlation coefficient for the two scores and found they were positively correlated (r = .468, n = 28, p = .012). (Fig. 11) 37  8	
    Uncertainty	
  Avoidance	
  Score	
    6	
   4	
   2	
   0	
    -­‐2	
    -­‐4	
   -­‐6	
   -­‐8	
    -­‐10	
   -­‐6	
    -­‐4	
    -­‐2	
    0	
    2	
   4	
   6	
   Behavioral	
  Inhibi@on	
  Score	
    8	
    10	
    12	
    Caucasian	
   East	
  Asian	
  Stranger	
   Figure 11. Individual behavioral inhibition and uncertainty avoidance scores for Caucasians (n=20) and East Asian Strangers (n=8).  n  Mean UA score  Mean BIS score  East Asian Strangers  8  2.16  5.50  All East Asians  16  1.00  2.87  All Caucasians  20  0.50  2.65  East Asian Acquaint.  8  0.30  1.30  Table 1. Mean uncertainty avoidance and behavioral inhibition scores, ordered from highest to lowest score (N=36).  38  4.2.3 Secondary Analysis with Uncertainty Avoidance Score Since we did not get the expected mapping between ethnicity and Hofstede’s uncertainty avoidance, we wanted to investigate whether uncertainty avoidance had any effect on our primary dependent measures of anxiety and preference. We performed a secondary analysis using uncertainty avoidance (UA) grouping, ignoring ethnic group. The full sample was divided into groups based on their uncertainty avoidance score resulting in 3 groups for Low, Med and High UA (Table 2). We chose this division over equal-sized groups since the latter would have resulted in individuals with the same UA score being placed in different groups. We performed a 2x2 mixed model ANOVA on the dependent measures of the primary analysis using UA groups (Low, Med and High) and interface (Rich and Minimal). We report on the same parameters as the primary analysis, with the addition of the Scheffé post hoc test when a main effect of UA groups is found. We report only on the measures that showed significant differences. Range  Mean UA score  n  UA score  -18 to 18  -  -  Low UA  UA score < 0  -4.2  10  Med UA  0 ≤ UA score ≤ 2  0.9  12  High UA  2 < UA score  4.1  14  Table 2. Uncertainty avoidance (UA) grouping for secondary analysis (N=36).  39  4.2.3.1  Usability Preference (Security)  High UA users noticed the UBC logo more than low UA users. Responses to the question “I noticed the logo of the University of British Columbia (UBC) on the webpage (of the Rich interface)” showed a main effect of UA grouping (F1,33 = 4.52, p = .018, η2 = .215). Post hoc analysis using the Scheffé test showed that high UA users noticed the UBC logo more readily than the low UA users (p = .024). Most high UA users (11 out of 14) indicated they noticed the UBC logo, whereas most low UA users (7 out of 10) indicated they were either not sure or that they did not notice it at all.  4.2.3.2  Usability Preference (Information Richness)  Low UA users found interfaces to have unnecessary information more so than high UA users did. Responses to the question “The information provided in the interface was {not enough, just right, too much} to do the task” showed a main effect of UA grouping (F1,33 = 3.23, p = .042, η2 = .155). Post hoc analysis using the Scheffé test showed that low UA users felt interfaces had “too much information” more than high UA users did (p = .044). (Fig. 12)  4.2.3.3  Interface-Specific Anxiety  High and Medium UA users felt more anxious during the task than low UA users. Responses to the question “I felt anxious while doing the task” showed a main effect of UA grouping (F1,33 = 3.71, p = .035, η2 = .184). Post hoc analysis using the Scheffé test showed that on both interfaces, low UA users felt less anxious than both high and med UA (p = .045 and .033 respectively). (Fig. 13) No difference was found in general anxiety. 40  Feeling	
  there	
  was	
  "too	
  much"	
   informa@on	
    2	
   1.6	
   1.2	
   0.8	
   0.4	
   0	
   Minimal	
   Low	
  UA	
    Rich	
   Med	
  UA	
    High	
  UA	
    Figure 12. Perception of information richness as excessive for the different UA groups.  Interface-­‐specific	
  anxiety	
    2	
   1.6	
   1.2	
   0.8	
   0.4	
   0	
   Minimal	
   Low	
  UA	
    Rich	
   Med	
  UA	
    High	
  UA	
    Figure 13. Interface-specific anxiety for the different UA groups.  41  4.3 Discussion  We begin our discussion by reflecting on the results in light of our hypotheses. We then provide some design guidelines based in our results. Finally, we reflect on our experience doing research about cultural attitudes toward uncertainty, and doing cultural research in general.  4.3.1 Reflection on Results We detected a cultural difference in preference which partially met our hypothesis. East Asians did prefer the Rich interface overall, and found it easier to use than the Minimal. Caucasians, by contrast, preferred the Minimal interface or had no preference, and found it easier to use than East Asians. Despite not finding a quantitative difference in information richness and security, we did find some qualitative evidence of a cultural contrast in these usability areas. East Asians indicated weighing security elements in their preference for the Rich interface, even though these elements were iconic of security and not functional. On the other hand, Caucasians indicated that the lack of perceived clutter tipped their preference toward the Minimal interface. This implies that varying information richness to align with cultural attitude toward uncertainty has merit. As we expected, Caucasians perceived additional elements as clutter, whereas East Asians valued security elements more. Our performance hypothesis was not supported, as we found no significant difference in performance between ethnicities or interfaces. This is good news for C-TOC as having cultural variations in performance would have serious implications on its integrity as a cognitive test. On the other hand, doing 30 trials of a cognitive task may not have provided enough time to allow 42  for variations in performance; going through the full cognitive test battery would be more taxing and might create a larger variation in performance, especially with other C-TOC subtests that have a richer answer space than the symbol-digit matching test that we used in our study. Moreover, our participants were not actually preoccupied with the implications of cognitive testing as we focused on usability in the user study rather than the cognitive aspects of C-TOC. Further research is required to understand effects on performance. Our hypothesis about use of support buttons did not receive quantitative empirical support either, as we witnessed only one East Asian participant clicking on the demo button. Our qualitative findings however showed that a greater proportion of East Asians than Caucasians appreciated such elements and weighed them in their preference of the Rich interface. This implies that these elements provided a sort of cognitive cushion [23], reassuring users by their presence. This aligns with our design vision that the richness of the interface provides East Asians with a sense of control over the interaction environment through support elements, soothing anxiety about uncertainty. The challenges we faced recruiting East Asian participants contributed to our unexpected observation that East Asians felt less anxiety than Caucasians. We began recruiting from both cultural groups simultaneously using public advertisements. While we succeeded in reaching our Caucasian quota of 20 after only about 1.5 months, it took 3 months to recruit 8 East Asian participants in a city where they constitute about 25% of the population. Expanding our public recruitment artillery, by setting information tables or participating in activities with seniors to build rapport, did little to help. The futility of public recruitment with East Asians led us to consider contacting acquaintances of our research network. Although we could not reach our East Asian quota of 20 within a reasonable timeframe, we did manage to recruit 8 East Asian 43  participants through networking, bringing the total to 16. This group of East Asian Acquaintances, however, ended up lowering the mean overall anxiety of the entire East Asian sample to a level that was lower than hypothesized.  4.3.2 Design Guidelines We provide some cultural design guidelines for C-TOC, and similar multicultural design contexts characterized by uncertainty. For East Asians, we recommend information-rich interfaces augmented with security elements and additional learning support. Such elements are especially important when anxiety is likely during interaction. For Caucasians, informationminimal interfaces are more appropriate, as these users mainly attend to the primary task elements. Catering to cultural preferences may increase willingness across cultures to use CTOC. We imagine C-TOC could develop to become a cultural adaptive interface similar to Mocca [25], which we described in section 2.2, where a culturally adaptive version would be presented to the user after collecting a modest amount of cultural data. We also provide guidelines for uncertainty avoidance design based on our exploratory analysis with uncertainty avoidance grouping. We present those not as a cultural framework, but rather a low-cost method to create adapted interfaces using a short uncertainty avoidance questionnaire like the one we used. For low UA users, reduce information richness sacrificing if necessary security or authority elements which tend to be ignored. By contrast, such elements are more coveted by high UA users, who are more tolerant of information richness than low UA users. High UA users are also more susceptible to anxiety, so be aware of that if it is likely to be induced during interaction.  44  4.3.3 Reflection on Cultural Attitude toward Uncertainty Our hypothesis about differences in uncertainty avoidance scores between Caucasians and East Asian Strangers was not met. This was surprising, but could be explained partially by the low observed power of our statistical test (.145, given in Section 4.2.2.7). However, we observed a trend where East Asians scored higher on it than Caucasians (Table 1). In addition, our many informal conversations with participants during the study implied that East Asians are less comfortable with uncertainty. First, we learned that East Asians in general have less affinity to participate in research given our findings on confidence in research (Fig. 10). Second, we noticed that East Asian participants were generally more apprehensive of the study context than Caucasians. For instance, three of the East Asian Strangers in the study asked for proof of identity before meeting the researcher, while no Caucasian asked for any such information. Third, our struggle in recruiting East Asian participants over a long period of time indicates that they might indeed avoid participating in research. Altogether, we sensed that East Asians seemed relatively less comfortable with unfamiliar contexts, felt less confidence in research, which made them less willing to participate. This implies a stronger aversion to uncertainty among East Asians than Caucasians. We also faced the problem of self-selection to participate in research as most research does. The issue of self-selection had a larger impact on East Asian recruitment as we felt this group was relatively less accessible in a cultural investigation. The reason we suspect Hofstede managed to detect a difference in uncertainty avoidance scores between the East and West was that he had the rare opportunity of avoiding the problem of self-selection in research; he was an IBM researcher requesting international IBM employees to participate in work-mandated  45  research. We, on the other hand, lacked that privilege and had very limited access to study the phenomenon faithfully. Altogether this suggests that our findings related to Hofstede’s uncertainty avoidance are conservative. This is due in part to the intrinsically challenging dilemma of studying uncertainty avoidance: how can we investigate a phenomenon whose presence in a culture discourages participation in research in the first place? The East Asian Stranger participants we recruited were probably less uncertainty avoidant than the average East Asian person whom we struggled to recruit because the latter avoids participating in research altogether.  4.3.4 Reflection on Cultural Research The process of conducting research on culture is very delicate; one treads on thin ice and risks plunging into the frigid waters of cultural profiling and stereotyping. We were interested in this research because cultural differences had been observed in psychology and we saw value in extending such research to Human Computer Interaction. In discussing related literature, our design and results, we tried to clarify that we were describing cultural tendencies and not generalizations. We attempted to mold our strategies to conduct this research, and our language to describe it, in a manner that is sensitive and respectful to cultural tradition. We did not want to produce our research results about cultural differences from an ivory tower, and we made an effort to formulate our questions and share our observations with individuals from the cultures we studied. We wanted our findings to be grounded in our own observations as well as the introspective views of individuals from those cultures. We found that the cultural observations we made tended to resonate with how individuals in the East and West viewed their own cultures. Many of our speculations about cultural behavior were also gleaned 46  from conversations with people. For instance, our conversations with East Asians validated our quantitative finding that East Asians were relatively less trusting of research (Fig. 10). Our interest to pursue research in culture emanated from the uttermost respect for cultural tradition everywhere, and did not intend in any way to simplify complex cultural diversity through one scientific experiment. Our goal was to describe cultural tendencies and not facts. We wanted to provide recommendations for cultural design rather than implications. We hope that our research was not construed as an offensive attempt at cultural stereotyping, but rather for its intended purpose: bridging technology designers with multicultural users effectively.  47  Chapter 5: Conclusion and Future Work  Our design and evaluation of cultural interfaces show that there is validity in culturability. Our design vision to toggle information richness to cater to varying cultural attitudes toward uncertainty is promising. As we hypothesized, we saw that East Asians tend to prefer information-rich interfaces, whereas Caucasians lean toward minimal interfaces. Catering to cultural preferences can increase – in the case of C-TOC at least – willingness to use a technological solution cross-culturally. However, it proved hard for us to attribute the cultural preference to one measure of attitude toward uncertainty, such as Hofstede’s uncertainty avoidance, as it was deficient on its own in explaining the cultural variation. Similarly, we could not unpackage our cultural design to determine how its individual elements (e.g. security elements, additional support, etc.) influenced cultural preference, and we wish to target that in future research. We envision other paths to further this research. Using Galvanic Skin Response (GSR) to measure anxiety can provide a different measure of anxiety than the self-reported inventory we used, as GSR detects levels of arousal physiologically through skin contact [19]. Translating the study prototype might expand the pool of recruitment as our experience showed there is a population of older East Asians who would not have been able to manage the level of conversational English of the prototype. Experimenting with other subtests in C-TOC with richer answer spaces is also warranted as these create a different form of uncertainty. Doing a study in a casual context other than cognitive testing might also attract a larger pool of participants. Cognitive testing is an atypical usability context with different cultural implications on 48  perceptions of mental health and aging, which renders it delicate and stressful. Finally, we are currently working on running the study remotely using Amazon’s Mechanical Turk, a crowdsourcing platform where a large pool of human users can be requested to perform usability studies. We hope this will extend our reach to more cross-cultural users in order to better understand how they deal with uncertainty when interacting with technology.  49  Bibliography 1. Barber, W., & Badre, A. (1998). Culturability: The merging of culture and usability. Proc. of the 4th Conference on Human Factors and the Web, 1–14. 2. Beer, G. (Writer), & Cornish, T. (Director). (2012). Bad Attitude [Television series episode]. In A. Bystram (Producer), Mayday. Discovery Channel Canada. 3. Brehmer, M. M. (2011). Usability and the effects of interruption in C- TOC: SelfAdministered Cognitive Testing on a Computer. M.Sc thesis. University of British Columbia. 4. Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS Scales. Journal of Personality and Social Psychology, 67(2), 319–333. 5. Cheung, B. Y., Chudek, M., & Heine, S. J. (2011). Evidence for a Sensitive Period for Acculturation Younger Immigrants Report Acculturating at a Faster Rate. Psychological Science, 22(2), 147–152. 6. Choi, B., Lee, I., Kim, J., & Jeon, Y. (2005). A qualitative cross-national study of cultural influences on mobile data service design. In Proc. of the SIGCHI conference on Human factors in computing systems (pp. 661–670). 7. Cohen, J. (1973). Eta-squared and partial eta-squared in communication science. Human Communication Research, 28, 473–490. 8. Cyr, D., Bonanni, C., & ilsever, J. (2004). Design and e-loyalty across cultures in electronic commerce. In Proc. of the 6th international conference on Electronic commerce (pp. 351– 360). 9. Darke, S. (1988). Anxiety and working memory capacity. Cognition & Emotion, 2(2), 145– 154. 10. De Angeli, A., & Kyriakoullis, L. (2006). Globalisation vs. localisation in e-commerce: cultural-aware interaction design. In Proc. of the working conference on Advanced visual interfaces (pp. 250–253). 11. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state.” Journal of Psychiatric Research, 12(3), 189–198. 12. Heine, S. J. (2007). Cultural Psychology (1st ed.). W. W. Norton & Company. 50  13. Hofstede, G. (1980). Cultures Consequences  : International Differences in Work-Related Values (Abridged.). SAGE Publications, Inc. 14. Jacova, C., Kertesz, A., Blair, M., Fisk, J. D., & Feldman, H. H. (2007). Neuropsychological testing and assessment for dementia. Alzheimer’s & Dementia, 3(4), 299–317. 15. Jhangiani, I., & Smith-Jackson, T. (2007). Comparison of mobile phone user interface design preferences: perspectives from nationality and disability culture. In Proc. of the 4th international conference on mobile technology, applications, and systems and the 1st international symposium on Computer human interaction in mobile technology (pp. 512– 519). 16. Kitayama, S., Duffy, S., Kawamura, T., & Larsen, J. T. (2003). Perceiving an Object and Its Context in Different Cultures: A Cultural Look at New Look. Psychological Science, 14(3), 201–206. 17. Lee, I., Choi, G. W., Kim, J., Kim, S., Lee, K., Kim, D., Han, M., Park, S., An, Y. (2008). Cultural dimensions for user experience: cross-country and cross-product analysis of users’ cultural characteristics. In Proc. of the 22nd British HCI Group Annual Conference on People and Computers: Culture, Creativity, Interaction - Volume 1 (pp. 3–12). 18. Marcus, A. (2010). Cross-cultural user-interface design for work, ho play, and on the way. In ACM SIGGRAPH ASIA 2010 Courses (pp. 5:1–5:160). 19. McCleary,	
  R.	
  A.	
  (1950).	
  The	
  nature	
  of	
  the	
  galvanic	
  skin	
  response.	
  Psychological	
  Bulletin,	
   47(2),	
  97–117.	
  	
   20. Merritt, S., & Bardzell, S. (2011). Postcolonial language and culture theory for HCI4D. In Proc. of the 2011 annual conference extended abstracts on Human factors in computing systems (pp. 1675–1680). 21. Miyamoto, Y., Nisbett, R. E., & Masuda, T. (2006). Culture and the Physical Environment. Psychological Science, 17(2), 113 –119. 22. Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., Cummings, J., & Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A Brief Screening Tool For Mild Cognitive Impairment. Journal of the American Geriatrics Society, 53(4), 695–699. 51  23. Nekrasovski, D., Bodnar, A., McGrenere, J., Guimbretière, F., & Munzner, T. (2006). An evaluation of pan & zoom and rubber sheet navigation with and without an overview. In Proc. of the SIGCHI Conference on Human Factors in Computing Systems (pp. 11–20). 24. Owen, A. M., McMillan, K. M., Laird, A. R., & Bullmore, E. (2005). N-back working memory paradigm: A meta-analysis of normative functional neuroimaging studies. Human Brain Mapping, 25(1), 46–59. 25. Reinecke, K., & Bernstein, A. (2011). Improving performance, perceived usability, and aesthetics with culturally adaptive user interfaces. ACM Trans. Comput.-Hum. Interact., 18(2), 8:1–8:29. 26. Shuper, P. A., Sorrentino, R. M., Otsubo, Y., Hodson, G., & Walker, A. M. (2004). A Theory of Uncertainty Orientation. Journal of Cross-Cultural Psychology, 35(4), 460 –480. 27. Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait Anxiety Inventory. Retrieved from http://ubir.buffalo.edu/xmlui/handle/10477/2895 28. Tanaka, A., & Yamauchi, H. (2004). Cultural self-construal and achievement goal. Hellenic Journal of Psychology, 1, 221–237. 29. Venkatesh, V., Morris, M. G., Gordon B. Davis, & Davis, F. D. (2003). User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly, 27(3), 425–478. 30. Walsh, T., Nurkka, P., & Walsh, R. (2010). Cultural differences in smartphone user experience evaluation. In Proc. of the 9th International Conference on Mobile and Ubiquitous Multimedia (pp. 24:1–24:9). 31. Wobbrock, J. O., Findlater, L., Gergle, D., & Higgins, J. J. (2011). The aligned rank transform for nonparametric factorial analyses using only anova procedures. In Proc. of the SIGCHI Conference on Human Factors in Computing Systems (pp. 143–146).  52  Appendices  Appendix A User study documents: •  A.1 - Study Recruitment Poster.  •  A.2 - Study Consent Form.  53  A.1  Study Recruitment Poster  The UNIVERSITY OF BRITISH COLUMBIA Department of Computer Science / Medicine University of British Columbia Vancouver, BC, V6T 1Z4  Understanding the Effect of Culture on Computer Use Study Recruitment Are you 55 or older? Do you use a computer? Principal Investigator: Claudia Jacova, PhD Co-Investigators: Shathel Haddad, MSc Student; Joanna McGrenere, PhD Purpose: The purpose of this research is to acquire a more in-depth understanding of cultural differences when users take a cognitive test on a computer. We are investigating whether cultural variation regarding attitudes about uncertainty would affect performance on the test and satisfaction with its interface. Participants: We are looking for participants, aged 55 or older, who: • Are healthy, and have normal or corrected-to-normal eyesight, • Free of diagnosed cognitive impairments, motor impairments to their hands, • Have over six months of computer experience, and • Satisfy a set of cultural criteria Procedure: In this study, you will be asked to do some cognitive tasks on a computer such as matching figures and building sentences. Your performance will be recorded. You will be asked to fill out some questionnaires before using the program as well as answer some interview questions about your experience after. These questions are designed to collect data about your cultural profile, anxiety levels and satisfaction with the testing interface. In all circumstances, you may refuse to answer questions that you are not comfortable with. Photographs/Videos will be taken only with your permission. Objective: The research objective is to inform and refine the design of a computerized program that is intended for cognitive health care purposes and is to be used by individuals with various cultural backgrounds. To achieve this, we need to first understand the effect of cultural differences -if any- on the experience of cognitive testing on a computer. With this understanding we can continue to design effective and usable health care technologies which are culturally appropriate. Commitment: Your participation in the study will take up to one hour using a computer and you will be compensated for your time with a financial honorarium. To participate: Please contact Shathel at shathel@cs.ubc.ca for more information or to sign up. 54  A.2  Study Consent Form  The UNIVERSITY OF BRITISH COLUMBIA Department of Computer Science / Medicine University of British Columbia Vancouver, BC, V6T 1Z4  Consent Form Research Project Title: Development of a Computer-Based Screening Test to Support Evaluation of Cognitive Impairment and Dementia (Part 1D – Understanding the Effect of Culture on Computer Use) Principal investigator: Claudia Jacova, PhD, 604-822-7540 Co-Investigators:  Shathel Haddad, MSc Student, 604-773-8269 Joanna McGrenere, PhD, 604-827-5201 Charlotte Tang, PhD, 604-827-3983  In this study, we aim to acquire a better understanding of cultural differences when users take a cognitive test on a computer. You are being invited to participate in this study because you use a computer regularly at home and are free of cognitive and motor impairment to your hands. You can help us by using a computer application and providing feedback on your use experience. Your participation in this research study is entirely voluntary. This consent form, a copy of which has been given to you, is only part of the process of informed consent. It should give you the basic idea of what the research is about and what your participation will involve. If you would like more detail about something mentioned here, or information not included here, you should feel free to ask. Please take the time to read this carefully and to understand any accompanying information. If you wish to participate, you will be invited to sign this form but you should understand that you are free to withdraw your consent at any time and without giving any reasons for your decision. Who Is Conducting this Research? This research is conducted by a team of Investigators from UBC Computer Science and the Division of Neurology, Department of Medicine. Purpose: The purpose of this research is to acquire a more in-depth understanding of cultural differences when users take a cognitive test on a computer. We are investigating whether cultural variation regarding attitudes about uncertainty would affect performance on the test and satisfaction with its interface. 55  Procedure: This study should require up to one hour of your time. You will be asked to use a computer program to perform cognitive activities such as matching figures and making sentences. You will be asked to fill some questionnaires before using the program as well as answer some interview questions about your experience after. These questions are designed to collect data about your cultural profile, anxiety levels and satisfaction with the computer testing interface. Your test performance will be recorded as well. In all circumstances, you may refuse to answer questions that you are not comfortable with. Objective: The research objective is to inform and refine the design of a computerized program that is intended for cognitive health care purposes and is to be used by individuals with various cultural backgrounds. To achieve this, we need to first understand the effect of cultural differences -if any- on test performance and satisfaction with its computer interface. With this understanding we can continue to design effective and usable health care technologies which are culturally appropriate. Option for Photographing/Videotaping: For the purpose of data analysis, we would like to videotape and/or photograph your computer session and your interview. Please note that this is an optional procedure, which you are free to decline, and a refusal to videotape or photograph will in no way affect your eligibility for this study. Only the investigators of this study will have access to the recordings. The recordings will be stored in a secured departmental network of Computer Science for three years after the study, which will then be permanently erased. Participants’ identity will be protected by masking in publications and presentations. Please check and initial the ones you agree. · I agree that the researchers may videotape my computer session. __________ · I agree that the researchers may videotape my interview. __________ · I agree that the researchers may use the photographs taken during the study without modification, except for masking identities, for illustrative purposes in the dissemination of the study’s results, including but not limited to, presentations and publication of papers and/or theses. What are the Possible Harms and Side Effects of Participating? You may experience fatigue from performing the computer tasks and answering the questions. What are the Benefits of Participating in this Research? There may be no immediate, direct benefit to you as a result of participating in this study. However the findings from this study can help us improve future health care technologies for use in domestic environments that may benefit you, your family members and the community in the longer term.  56  What Happens If I Decide to Withdraw My Consent to Participate? Your participation in this research is entirely voluntary. You may withdraw from this study at any time, and are not required to provide any reason for withdrawing. If you choose to enter the study and then decide to withdraw at a later time, all data collected about you during your enrollment in the study will be retained for analysis. By law, these data cannot be destroyed. If you wish to withdraw your consent, we ask that you notify Dr. Claudia Jacova at 604-822-7540 or Shathel Haddad at 604-773-8269. What Happens If Something Goes Wrong? Signing this consent form in no way limits your legal rights against the sponsor, investigators, or anyone else, and you do not release the study doctors or participating institutions from their legal and professional responsibilities. Will My Taking Part in this Study be Kept Confidential? Your confidentiality will be respected. However, research records and health or other source records identifying you may be inspected in the presence of the Investigator or his or her designate by representatives of the UBC Clinical Research Ethics Board for the purpose of monitoring the research. No information or records that disclose your identity will be published without your consent, nor will any information or records that disclose your identity be removed or released without your consent unless required by law. You will be assigned a unique study number as a subject in this study. Only this number will be used on any research-related information collected about you during the course of this study, so that your identity [i.e. your name or any other information that could identify you] as a subject in this study will be kept confidential. Information that contains your identity will remain only with the Principal Investigator and/or designate. The list that matches your name to the unique study number that is used on your research-related information will not be removed or released without your consent unless required by law. Your rights to privacy are legally protected by federal and provincial laws that require safeguards to insure that your privacy is respected and also give you the right of access to the information about you that has been provided to the sponsor and, if need be, an opportunity to correct any errors in this information. Further details about these laws are available on request to your study doctor. Who do I Contact if I have any Questions or Concerns about the Study? If you have any questions or desire further information with respect to this research, you should contact Dr. Claudia Jacova at 604-822-7540 or Shathel Haddad at 604-7738269. If you have any concerns about your rights as a research subject and/or your experiences while participating in this study, you should contact the Research Subject Information Line at the University of British Columbia’s Office of Research Services at 604-822-8598, toll free 1-877-822-8598, or email RSIL@ors.ubc.ca 57  Subject Consent to Participate: • I have read and understood the subject information and consent form. • I have had sufficient time to consider the information provided and to ask for advice if necessary. • I have had the opportunity to ask questions and have had satisfactory responses to my questions. • I understand that all of the information collected will be kept confidential and that the results will only be used for scientific objectives such as research and publications. • I understand that I can refuse to answer any questions that I do not feel comfortable answering from this study. • I understand that my participation in this study is voluntary and that I am completely free to refuse to participate or to withdraw from this study at any time. • I understand that I am not waiving any of my legal rights as a result of signing this consent form. • I understand that there is no guarantee that this study will provide any benefits to me. • I have read this form and I freely consent to participate in this study. • I have been told that I will receive a dated and signed copy of this form.  Signatures  Printed Name of Participant  Signature and Date  ______________________________________________________________________ Principal Investigator or designated representative Signature and Date  58  Appendix B Study Questionnaires: •  B.1 - Usability Questionnaire.  •  B.2 - Uncertainty Avoidance Profile Questionnaire.  •  B.3 - Behavioral Inhibition Questionnaire.  •  B.4 - State Trait Anxiety Inventory.  59  B.1  Usability Questionnaire  Note: partially adapted from the Unified Theory of Acceptance and Use of Technology (UTAUT) questionnaire [29]. Questions (1) through (12) are asked after exposure to both the Rich and Minimal interfaces. Questions (13) through (17) are only administered after exposure to the Rich interface.  Instructions: Please answer the following questions regarding the matching test interface you just used. 1. My interaction with the test interface was clear and understandable. Strongly disagree  Disagree  Agree  Strongly Agree  Agree  Strongly Agree  2. I found the test interface easy to use. Strongly disagree  Disagree  3. I could complete the task using the interface if there was no one around to tell me what to do as I go Strongly disagree  Disagree  Agree  Strongly Agree  Agree  Strongly Agree  4. I felt anxious while doing the task. Strongly disagree  Disagree  5. It scared me to think that I could make a mistake by hitting the wrong key or button. Strongly disagree  Disagree  Agree  Strongly Agree  6. I hesitated while doing the task for fear of making mistakes that I could not correct. Strongly disagree  Disagree  7. The interface was somewhat intimidating to me. 60  Agree  Strongly Agree  Strongly disagree  Disagree  Agree  Strongly Agree  8. I clearly understood what I was required to do in the task. Strongly disagree  Disagree  Agree  Strongly Agree  9. The test interface was too cluttered (that is there was too many things on the interface) Strongly disagree  Disagree  Agree  Strongly Agree  Agree  Strongly Agree  10. I felt frustrated while using the test interface. Strongly disagree  Disagree  11. I have confidence in the legitimacy of the research team which designed the interface. Strongly disagree  Disagree  Agree  Strongly Agree  12. The information provided in this test interface was ______________ to do the task. Not enough  Just right  Too much  13. I found the "video demo" button useful and it helped me understand what I had to do Strongly disagree Agree did not use  Disagree  Agree  Strongly  Agree  Strongly  did not notice  14. I found the "test background information" button useful Strongly disagree Agree did not use  Disagree did not notice  15. I found the tooltips (that is the additional information that pops up when I place the mouse on certain elements) useful Strongly disagree Agree  Disagree  Agree  did not notice 61  Strongly  16. I noticed the logo of the University of British Columbia "UBC" on the webpage Yes  No  Not sure  17. I noticed a "padlock" icon on the webpage Yes  No  Not sure  62  B.2  Uncertainty Avoidance Profile Questionnaire  Note: partially adapted from the Hofstede 2008 Values Survey Module.  Instructions: Please indicate how much you agree/disagree with the following statements based on your personal beliefs. 1. One can be a good manager without having a precise answer to every question that a subordinate may raise about his or her work Strongly disagree  Disagree  Agree  Strongly Agree  2. A company's or organization's rules should not be broken - not even when the employee thinks breaking the rule would be in the organization's best interest Strongly disagree  Disagree  Agree  Strongly Agree  3. Children must be taught to be organized and avoid ambiguity Strongly disagree  Disagree  Agree  Strongly Agree  4. People who are resilient to change and can adapt to different environments are appreciated in society Strongly disagree  Disagree  Agree  Strongly Agree  Agree  Strongly Agree  Agree  Strongly Agree  5. People should always carry an ID on them Strongly disagree  Disagree  6. It is improper to express feelings in public Strongly disagree  Disagree  7. There are some customs and rules that all people must respect Strongly disagree  Disagree  Agree 63  Strongly Agree  8. Making online purchases is safe and secure today Strongly disagree  Disagree  Agree  Strongly Agree  Please answer the following question: 9. How often do you feel nervous or tense? Never  Not Often  Sometimes  64  Usually  Always  B.3  Behavioral Inhibition Questionnaire  Note: adapted from the behavioral inhibition portion of the BIS/BAS scale [4]. Questions (1), (3), (5) and (8) are filler questions.  Instructions: Each item of this questionnaire is a statement that a person may either agree with or disagree with. For each item, indicate how much you agree or disagree with what the item says. Please respond to all the items; do not leave any blank. Choose only one response to each statement. Please be as accurate and honest as you can be. Respond to each item as if it were the only item. That is, don't worry about being "consistent" in your responses. 1. A person's family is the most important thing in life. very false for me  somewhat false for me  somewhat true for me  very  true for me 2. Even if something bad is about to happen to me, I rarely experience fear or nervousness. very false for me  somewhat false for me  somewhat true for me  very  somewhat true for me  very  somewhat true for me  very  true for me 3. How I dress is important to me. very false for me  somewhat false for me  true for me 4. Criticism or scolding hurts me quite a bit. very false for me  somewhat false for me  true for me  65  5. It's hard for me to find the time to do things such as get a haircut. very false for me  somewhat false for me  somewhat true for me  very  true for me 6. I feel pretty worried or upset when I think or know somebody is angry at me. very false for me  somewhat false for me  somewhat true for me  very  true for me 7. If I think something unpleasant is going to happen I usually get pretty "worked up." very false for me  somewhat false for me  somewhat true for me  very  somewhat true for me  very  true for me 8. I often wonder why people act the way they do. very false for me  somewhat false for me  true for me 9. I feel worried when I think I have done poorly at something important. very false for me  somewhat false for me  somewhat true for me  very  somewhat true for me  very  somewhat true for me  very  true for me 10. I have very few fears compared to my friends. very false for me  somewhat false for me  true for me 11. I worry about making mistakes. very false for me  somewhat false for me  true for me 66  B.4  State Trait Anxiety Inventory  Note: adapted from [27].  Instructions: Read each statement and select the appropriate response to indicate how you feel right now, that is, at this very moment. There are no right or wrong answers. Do not spend too much time on any one statement but give the answer which seems to describe your present feelings best. 1. I feel calm Not at all  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  2. I feel secure Not at all 3. I feel tense Not at all 4. I feel strained Not at all 5. I feel at ease Not at all 6. I feel upset Not at all  7. I am presently worrying over possible misfortunes Not at all  A little  Somewhat  8. I feel satisfied 67  Very Much So  Not at all  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  9. I feel frightened Not at all 10. I feel uncomfortable Not at all 11. I feel self confident Not at all 12. I feel nervous Not at all 13. I feel jittery Not at all 14. I feel indecisive Not at all 15. I am relaxed Not at all 16. I feel content Not at all 17. I am worried Not at all 18. I feel confused Not at all  68  19. I feel steady Not at all  A little  Somewhat  Very Much So  A little  Somewhat  Very Much So  20. I feel pleasant Not at all  69  Appendix C Low fidelity paper prototypes of the Rich and Minimal interfaces. C.1  Low Fidelity Minimal Interface  70  C.2  Low Fidelity Rich Interface  71  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0052193/manifest

Comment

Related Items