Pablo E. Paredes is Clinical Assistant Professor at the Psychiatry and Behavioral Sciences Department, and by courtesy at the Epidemiology and Population Health Department at Stanford University School of Medicine. His research focuses on subtle interventions to reduce stress, such as guiding people to breathe slowly with subtle haptic cues from office or car furniture, and passive sensing of affective and physiological biomarkers derived from existing devices (such as computers, phones, etc.) Dr. Paredes leads the Pervasive Wellbeing Technology Lab, http://med.stanford.edu/pervasivewellbeingtech.html and recently was invited to give a TED Talk on his research, https://www.youtube.com/watch?v=f60UStZX0kA. He has advised several PhD, masters and undergrad students. Prior to joining the School of Medicine, he was a Postdoctoral Researcher in Computer Science at Stanford University for two years. Dr. Paredes earned his PhD in Computer Science from the University of California, Berkeley in 2015. During his PhD career, he held internships on behavior change and affective computing in Microsoft Research and Google. Before 2010 he was a senior strategic manager with Intel in Sao Paulo, Brazil, a lead product manager with Telefonica in Quito, Ecuador and an Entrepreneur in his natal Cuenca, Ecuador. Prior to that, he obtained a dual MBA and MS in Electrical Engineering from Georgia Tech with a Fulbright Scholarship. In these roles, he has had the opportunity to closely evaluate designers, engineers, business people and researchers in telecommunications, product development, and ubiquitous computing.
Clinical Assistant Professor, Psychiatry and Behavioral Sciences
Clinical Assistant Professor (By courtesy), Epidemiology and Population Health
Faculty Affiliate, Institute for Human-Centered Artificial Intelligence (HAI)
Honors & Awards
Fulbright Scholarship, Comision Fulbright del Ecuador (2001)
Engineer, Universidad Politecnica Salesiana, Electronics (1999)
MS, Georgia Institute of Technology, Electrical and Computer Engineering (2003)
MBA, Georgia Institute of Technology, Business Administration (2004)
PhD, University of California, Berkeley, Computer Science (2015)
Community and International Work
Laboratorio de Tecnología de Salud Mental, Cuenca - Ecuador
Mental Health Technology
Universidad de Cuenca
Ecuador / Latin America
Opportunities for Student Involvement
Pablo Paredes Castro. "United States Patent 17/500880 Popbots: a suite of chatbots to provide personalized support for stress management", Leland Stanford Junior University, Oct 13, 2021
Current Research and Scholarly Interests
In my lab we have two key areas of study: sensorless sensing, and subtle interventions. Our projects aim to design not only useful and efficient wellbeing interventions, but also passive behavioral and affective sensors.
Every couple of days, humanity generates more data than all the data that was generated until 2013. Sensorless sensing is a provocative term and approach focused on repurposing many existing behavioral data streams to passively measure and study affect, stress, behavior, wellbeing and mental health biomarkers.
TouchStress: Using touchpads as stress sensors.
Clinical Unobtrusive Stress Sensing: Use PC mice as unobtrusive stress sensors.
Skin Wearability: Exploring use cases for skin-like wearables.
Multimodal Stress Sensing: AI-enhanced multimodal sensing using wearables and unobtrusive passive sensing
We use human-centered design to develop affordable and ecologically valid interventions aimed at dramatically increasing adoption and engagement of wellbeing behaviors. We leverage cyber and physical personal spaces to deliver micro or slow interventions that do not distract or interrupt the user.
PopBots: an "army" of tiny chatbots for stress management.
Personal Digital Wellbeing: Repurposing popular media into coping strategies.
Haunted Desk: Non-volitional behavior change embedded in furniture
SubBreathe: Imperceptible modification of biorhythms.
PopTherapy - Coping with Stress through Popular Media, Microsoft Research
Repurposing popular media to create micro interventions for stress.
For More Information:
MouStress - Detecting Stress with Mouse Movement, Microsoft Research
Detecting stress with mouse motion using reverse dynamic modeling.
Postdoctoral Research Mentor
Calm Commute: Guided Slow Breathing for Daily Stress Management in Drivers
Interactive, Mobile, Wireless, Ubiquitous Technologies
View details for DOI 10.1145/3380998
- Just Breath - Guided Breathing Interventions for Automobile Commuters (in press) Journal of Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT) 2018: 10
MouStress – Detecting Stress with Mouse Motions
ACM Conference on Human Factors in Computing Systems (CHI 2014)
View details for DOI 10.1145/2556288.2557243
- Fast & Furious: Detecting Stress with a Car Steering Wheel ACM Conference on Human Factors in Computing Systems (CHI 2018) 2018: 10
Inquire: Large-Scale Early Insight Discovery for Qualitative Research
ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2017)
View details for DOI 10.1145/2998181.2998363
PopTherapy: Coping with Stress through Pop Culture
8th EAI International Conference on Pervasive Computing for Healthcare (PervasiveHealth 2014)
View details for DOI 10.4108/icst.pervasivehealth.2014.255070
Evaluating In-Car Movements in the Design of Mindful Commute Interventions
Journal of Medical Internet Research (JMIR)
The daily commute could be a right moment to teach drivers to use movement or breath towards improving their mental health. Long commutes, the relevance of transitioning from home to work, and vice versa and the privacy of commuting by car make the commute an ideal scenario and time to perform mindful exercises safely. Whereas driving safety is paramount, mindful exercises might help commuters decrease their daily stress while staying alert. Increasing vehicle automation may present new opportunities but also new challenges.This study aimed to explore the design space for movement-based mindful interventions for commuters. We used qualitative analysis of simulated driving experiences in combination with simple movements to obtain key design insights.We performed a semistructured viability assessment in 2 parts. First, a think-aloud technique was used to obtain information about a driving task. Drivers (N=12) were given simple instructions to complete movements (configural or breath-based) while engaged in either simple (highway) or complex (city) simulated urban driving tasks using autonomous and manual driving modes. Then, we performed a matching exercise where participants could experience vibrotactile patterns from the back of the car seat and map them to the prior movements.We report a summary of individual perceptions concerning different movements and vibrotactile patterns. Beside describing situations within a drive when it may be more likely to perform movement-based interventions, we also describe movements that may interfere with driving and those that may complement it well. Furthermore, we identify movements that could be conducive to a more relaxing commute and describe vibrotactile patterns that could guide such movements and exercises. We discuss implications for design such as the influence of driving modality on the adoption of movement, need for personal customization, the influence that social perception has on participants, and the potential role of prior awareness of mindful techniques in the adoption of new movement-based interventions.This exploratory study provides insights into which types of movements could be better suited to design mindful interventions to reduce stress for commuters, when to encourage such movements, and how best to guide them using noninvasive haptic stimuli embedded in the car seat.
View details for DOI 10.2196/jmir.6983
View details for PubMedCentralID PMC5735252
INQUIRE Tool: Early Insight Discovery for Qualitative Research
Proceeding CSCW '17 Companion Companion of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing.
This demo presents an instance of Inquire, a tool designed to support qualitative researchers in the early stages of research. The tool enables the search over millions of users' records to extract early insights to aid in the formulation of research strategies. The tool presents the work described in the Inquire paper by Paredes, et. al.  in this demo we show how queries become a part of the inductive process, enabling researchers to try multiple ideas while gaining intuition and discovering less-obvious insights. We leverage LiveJournal (LJ) as a rich source of public posts, and Google News (GN) embeddings which link formal language (e.g. "reminiscence triggers") with colloquial expressions (e.g. "music brings back memories"). Using our tool, we elicit the interplay between tool and user to enhance qualitative and social research opportunities.
Design Principles for the Conceptualization of Games for Health Behavior Change.
Gamification Workshop @ CHI 2013.
This paper presents a list of principles that can be used to conceptualize games for health behavior change. These principles are derived from lessons learned after teaching two design-centered courses on Gaming and Narrative Technologies for Health Behavior Change. Course sessions were designed to create many rapid prototypes on specific topics coupling behavior change theory with iterative human-centered and game design techniques. The design task had two broad goals: 1) designing efficacious technologies, with an emphasis on short-term behavior change and 2) using narratives and game dynamics as vehicles for increased engagement and long-term sustained change. Example prototypes resulting from this design approach are presented.
Sensor-less Sensing for Affective Computing and Stress Management Technology.
This paper describes our vision on what should be the research around sensing and adaptive interventions to make affective computing and stress management technology pervasive and unobtrusive. With the use of common computer peripherals and mobile computing devices as affect sensors, personalized and adaptive intervention technologies can be developed. Furthermore, physiological sensing can be performed without the introduction of extraneous factors such as wearable devices or focused software. Different methods for sensing and complementary adaptable interventions and interactions are described and proposed. We show initial lab evidence of the use of a computer mouse in the detection of stress.
CalmMeNow: exploratory research and design of stress mitigating mobile interventions
Proceeding CHI EA '11 CHI '11 Extended Abstracts on Human Factors in Computing Systems.
This paper describes design explorations for stress mitigation on mobile devices based on three types of interventions: haptic feedback, games and social networks. The paper offers a qualitative assessment of the usability of these three types of interventions together with an initial analysis of their potential efficacy. Social networking and games show great potential for stress relief. Lastly, the paper discusses key findings and considerations for long-term studies of stress mitigation in HCI, as well as a list of aspects to be considered when designing calming interventions.
Synestouch: Haptic + Audio Affective Design for Wearable Devices
6th International Conference on Affective Computing and Intelligent Interfaces (ACII2015)
View details for DOI 10.1109/ACII.2015.7344630
Under Pressure: Sensing Stress of Computer Users
ACM Conference on Human Factors in Computing Systems (CHI2014)
View details for DOI 10.1145/2556288.2557165
Fiat Lux: Efficient Wellbeing Interactive Urban Lights
ACM Conference on Design of Interactive Systems (DIS 2016)
View details for DOI 10.1145/2901790.2901832