SócrateS: personality-based teaching strategies for web tutoring systems
In the Call Notes project we studied people's needs to recall information exchanged during mobile phone calls. A study was conducted over 2 months in which participants had their calls semiautomatically transcribed for their later annotation. Analysis of the calls and the subjects’ annotation behavior revealed that phone recall is indeed a relevant user need. We found that patterns in phone calls (numbers, names, interrogative adverbs) as well as some contextual parameters are better indicators of annotation needs than variables related to the callers' profile or call quality. We also observed that annotation needs can change over time: while some annotations
might not be considered relevant after a couple of weeks, other pieces of information originally regarded as irrelevant might become important items for archival. Analyses of these findings helped us identifying important implications for the design of mobile phone annotation tools.
In the Online Privacy project we studied how users value their personally identifiable information (PII) while web browsing, as well as their perceptions of the economic usage of their PII by online service providers. In order to collect PII valuations in context (time and place when it was generated), we used the refined Experience Sampling method. Furthermore, we used a reverse second price auction to guarantee that users would provide honest valuations to their PII. A sample of 168 subjects living in Spain installed and used a web browser plugin during 2 weeks to extract valuations of their PII. We found that users value items of their online browsing history for about 7€ (≈10 USD), and they give higher valuations to their offline PII, such as age and address (about 25€ or ≈36 USD). When it comes to PII shared in specific online services, users value information pertaining to financial transactions and social network interactions more than activities like search and shopping. No significant distinction was found between valuations of different quantities of PII (e.g. one vs. 10 search keywords), but deviation was
found between types of PII (e.g. photos vs. keywords). Finally, the users' preferred goods for exchanging their PII included money and improvements in service, followed bygetting more free services and targeted advertisements.
In the first phase of the Psychographics project, we proposed a set of user models capable of predicting customers' personality profile based on mobile phone usage as captured by telecommunication operators, i.e. using Call Detail Records (e.g., number of calls, duration of calls, time of calls, SMS sent/received) and variables derived from social network analysis of the call graph (e.g., number of peers, strength of peers). Models are based on ground truth personality data collected from customers of a mobile operator in Mexico, as well as their mobile phone usage logged in the operator's data warehouse. Initial findings corroborate the efficacy of predicting personality profile from mobile phone usage.
In the second phase, we empirically validated a conceptual model that measures the influence of people's personality on their satisfaction with basic mobile phone services. Our findings reveal that: (1) extroversion, conscientiousness, and intellect have a significant impact on customer satisfaction—positively for the first two traits and negatively for intellect; (2) extroversion positively influences mobile phone usage; and (3) extroversion and conscientiousness positively influence the users' perceived usability of mobile services. Interestingly, usability has the strongest positive impact on satisfaction, whereas mobile phone usage has a negative impact on satisfaction. Relevant implications for the design of mobile phone services were proposed based on these findings.
my roles: project leader survey/statistical specialist
MoviPill is a mobile phone-based application that motivates patients to be more adherent to their medication prescription by means of social competition with simple rules: more points are given to patients that take their medication closer to the prescribed time and less or negative points otherwise. In a 6-week user study conducted with 18 elders, the use of MoviPill increased compliance by 60% and the accuracy in drug intake time by: (1) 43% when applied to the entire user base and (2) 56% when the participants that were not interested in games were filtered out. This reveals the importance of applying personalized persuasive techniques.
my roles: project leader software developer UX researcher/designer
In the Contextual Services initiative, we conducted ethnographic observations, diary studies and a large-scale quantitative questionnaire (n=395) to study the reasons for adoption and refusal of context-aware mobile applications. Through a qualitative study we identify 24 user needs that these applications fulfill and 9 barriers for adoption (i.e. information distrust, lack of privacy, low service popularity, hardness to use, embarrassment, information overload, lack of usefulness, lack of personalization, and danger of use). We found that for many of the identified needs the end-goal is not that of receiving information, thus complementing work on mobile information needs. Also, this work offers an actionable list of obstacles that prevent contextual services to reach a larger audience. For instance, a project manager in charge of increasing the user base of an online mobile social network service could leverage on the findings of our quantitative study to better understand why implementing novel privacy-awareness features could yield better results than conducting marketing campaigns.
In the Multimedia Storytelling project, we developed and evaluated a novel photo collection summarization system that learns some of the users' social context by analyzing their online photo albums, and includes storytelling principles together with face and image aesthetic ranking in order to assist users in creating new photo albums to be shared online. An in-depth user study conducted with 12 subjects validated the proposed system as a first step in the photo album creation process, helping users reduce workload to accomplish such a task. Furthermore, our findings confirm that the proposed system performs as well as a human audio/video professional with cinematographic skills. Findings of this work supported a few guidelines that should help designers and multimedia technology experts to build social storytelling solutions
Most scholars define Near-Duplicate Video Clips (NDVC) by means of non-semantic features (e.g., different image/audio quality), while a few also include semantic features (i.e., different videos of similar content). Findings of four large scale online surveys (n = 1335) confirmed the relevance of both types of features. However, similarity was diminished by simultaneous changes in multiple features, and by increased informative value. Participants preferred to see only one NDVC in the search results and they were more tolerant to changes in the audio than in the video tracks. Finally, a human-centric NDVC definition was proposed along with implications for designing video sharing websites.
People share pictures online to increase their social presence. However, recent studies have shown that most of the content shared in social networks is not looked at by peers. uTag is the project in which we studied different persuasive mechanisms to support peers in providing metadata for multimedia content that can be used for a person’s self-promotion in online social networks. Through an iterative design and experimentation process, we demonstrated how this methodology can be used effectively to increase one’s social presence thus building more enjoyable, rich, and creative content that is shared in the social network. As a result, we identified several implications that inform the design of social games with a purpose.
The External Recruitment Program (ERP) is a non-research project that aims to ease the recruiting of subjects to test prototypes and participate in research experiments conducted in Telefonica. The approach is very simple: a recruitment program is periodically advertised through Telefonica shops, news portals, social networks, and corporate web sites. Volunteer data is then collected via a web form (ERP frontend) and stored in a secure and centralized database. Whenever product/research teams from Telefonica need to carry out a user study, they can use the system's researcher interface (ERP backend) in order to filter potential participants according to their study constraints (e.g. mobile phone users between 25 and 35 years old). Finally, eligible subjects can be invited to participate in the study directly from the ERP web-based interface. After 1.5 month of deployment, the ERP system received more than 4,000 subscribers from all over Spain.
TripleBeat is a mobile phone system that encourages users to achieve their exercise goals by: (1) letting them establish cardiovascular goals from high-level desires –e.g. lose fat–, (2) providing a real-time musical feedback to guide their pace, (3) motivating exercise by means of a virtual competition, and (4) displaying recommendations in a glanceable interface. TripleBeat’s competition encourages users to achieve their exercise goals in a healthy manner. It was built on top of its predecessor MPTrain from Microsoft Research. In a study conducted with 10 runners, all subjects preferred TripleBeat instead of MPTrain and they all exercised for a longer period of time inside their target heart rate zone when using it.
my roles: project leader software developer UX researcher/designer
Consistency Priorities is a multi-device design approach for contexts of interchange (where tasks are accomplished using different devices) and task migration (when one starts a task using device A and switches to device B to accomplish it). The model preserves specific design principles and a 3-tier consistency hierarchy. The first two levels give support to the user’s expectation in order to guarantee easiness of learning/ remembering and safety in contexts of interchange and task migration. Finally, the third level provides task personalization according to the user’s interests towards higher efficiency and satisfaction of use with a specific device. The approach was evaluated by means of a user study in which participants accessed an e-learning desktop application via 3 mobile interfaces generated by different approaches: (1) Direct Migration, (2) Linear, and (3) Automatic -applying only the first two levels of the Consistency Priorities hierarchy. Our findings confirm that he 2-tier automatic approach was the best to maintain the user’s mental model by preserving easiness, efficiency and safety of use for inter-device interaction. Additionally, efficacy (accuracy to execute task) and efficiency (average time to execute task) were either similar or better with this approach. Conversely, users preferred the task personalization of the Linear approach. This result supports our proposal suggesting that the efficacy generated by the first two levels of the Consistency Priorities hierarchy (task perception and execution) should be combined with the third level of personalization.
my roles: project leader software developer UX researcher/designer
SócrateS is an e-learning system capable of suggesting and adapting study guides dynamically using teaching strategies adequate to the student's personality traits (i.e., thinking/feeling, extraversion/introversion, etc.). In order to validate our proposal, a user study was conducted with 47 undergraduate students that accessed SócrateS for two weeks to learn basic concepts of recursive programming. Results revealed that those who were presented the study guide did not significantly change their behavior, as opposed to the control group. By analyzing the logs of the students interaction with the system, we could observe that, for the sample that did not receive the study guide, the SócrateS dynamic adaptation would have achieved a performance 2.5 times better than the static study guide (based only on the personality traits). This result could indicate a mismatch between innate and on-demand preferences, confirming the importance of designing systems that automatically adapt to the user’s current needs.
my roles: project leader software developer UX researcher/designer
