Know Thyself: Monitoring andReflecting on Facets of One's Life

Abstract

People strive to gain better knowledge of themselves bycollecting information about their behaviors, habits, andthoughts. Personal informatics systems can help byfacilitating the collection of personal information andthe reflection on that information. These systemssatisfy people's innate curiosity about themselves andencourage holistic engagement with one's life.Development of such systems poses new challenges inhuman-computer interaction and opens opportunitiesfor new applications and collaborations between diversedisciplines, such as design, life-logging, ubiquitouscomputing, persuasive technologies, and informationvisualization.

Keywords

Personal informatics, reflection, awareness, behavior,life logging, visualizations, study methods

ACM Classification Keywords

H5.m. Information interfaces and presentation (e.g.,HCI): Miscellaneous.

General Terms

Design, Experimentation, Human Factors

CHI 2010

, April 10 – 15, 2010, Atlanta, GA, USAACM 978-1-60558-930-5/10/04.

Ian Li

Human Computer InteractionInstituteCarnegie Mellon UniversityPittsburgh, PA 15213 USAianli@cmu.edu

Jodi Forlizzi

Human Computer InteractionInstituteCarnegie Mellon UniversityPittsburgh, PA 15213 USAforlizzi@cs.cmu.edu

Anind Dey

Human Computer InteractionInstituteCarnegie Mellon UniversityPittsburgh, PA 15213 USAanind@cs.cmu.edu

Web Page

More information about this workshop is available athttp://personalinformatics.org/chi2010/.

Introduction

Knowing oneself has many benefits, such as fosteringself-insight [2], increasing self-control [4], andpromoting positive behaviors such as energyconservation [5]. However, knowing oneself is difficultbecause we often have incomplete knowledge of ourselves [6], we cannot monitor our behaviors all thetime, and we cannot easily find patterns in ourbehaviors.Computers can help; they can store large amounts of data, analyze the data for patterns, visualize the data,and provide feedback to users at opportune times. Anew class of applications and web sites called

personal informatics

is appearing that collects and providesaccess to personal behavioral information (

e.g.

,Mint.com and Google Web History).Personal informatics poses new challenges and opensopportunities for research in human-computerinteraction. What behaviors would benefit frompersonal informatics? How do we make it easier for theuser to monitor his/her behaviors? How do we presentthe information to the user? Lastly, how do weintegrate collection of data and reflection on the datainto a seamless interaction? This workshop will combinea diverse set of participants to tackle the many issuesof this growing field.

Approach

We will organize the themes of the workshop along thefive stages of personal informatics systems [3]:

Preparation

Collection

Integration

Reflection

, and

Action

. These stages will serve as starting points fordiscussions about need-finding, behavioral studies,technological solutions, and interaction design issues.

Preparation: What behaviors would benefit from personal informatics?

If we introduce personalinformatics to help users learn about their behaviors,we need to understand what users want to learn. Whatkinds of information are pertinent to a particularbehavior? Additionally, not everyone needstechnological assistance in learning about behaviors.Finding the appropriate users and information is criticalin building effective personal informatics systems.

Collection: How do we collect behavioral data withminimal burden on the user?

Personal informaticssystems often require that users record behavioralinformation for several days, even weeks or months.Various technologies can automate sensing of information about the user,

e.g.

, GPS for the user’slocation, financial infrastructure for the user’spurchasing habits. For some information, systems mustrequire users to input information manually becauseeither the sensing technology does not exist yet or thetechnology is not robust enough for everyday use. Inthese cases, interaction with the application should bepainless or the application must motivate the user tokeep recording.

Integration: How do we integrate the monitoring and providing information to the user to provide a seamlessinteraction to users?

Personal informatics tools requireusers to interact with the system during collection andreflection,

e.g.

, carrying a sensing device daily,recording information on a web form, exploring

visualizations for patterns, and listening to advice fromsuggestions. Engaging the user only during collectionmay not allow the user to reflect about their behaviorsover the long-term. On the other hand, engaging theuser only during reflection may lead to lack of datarecording.

Reflection: How do we present the information to theuser?

Providing the wealth of information aboutdifferent behaviors to users is a challenge. Onechallenge is to provide the information for easy andquick understanding. Another challenge is determiningthe level of detail at which information is presented tousers.

Action: What are the effects of personal informatics ondaily life?

Once the information has been recorded andpresented to the user, how will the interaction affectthe user? This is a challenge because this will requireusers studies over long periods of time; the study hasto allow users to collect enough information foreffective feedback. There are several effects to study indetail: trust in the system, motivation, better decision-making, loss of control, etc.

Workshop Goals and Themes

One goal of this workshop is to define opportunities forexploration of human-computer interaction in personalinformatics. A unique aspect of personal informatics isthat the user is both source of the input and thereceiver of the output. This has several interestingimplications in the technology and design of systemsand applications. Different areas such as, ubiquitouscomputing, life logging, and visualizations, have tackleddifferent aspects of the monitoring and feedback partsof personal informatics. However, we argue thatsuccessful personal informatics systems must integratemonitoring and feedback seamlessly.Another goal is to share expertise between differentdisciplines to better tackle the many challenges thatpersonal informatics poses on interaction withcomputers. Researchers need to study how personalinformatics can benefit people’s daily lives as well asdevelop the technologies that will make personalinformatics available in daily life.Lastly, we want to get more researchers andpractitioners interested in this burgeoning field. Designguidelines and infrastructures need to be created tohelp more people build personal informatics systemsand applications.

Topics of Interest

We invite contributions from various disciplines ontopics including but not limited to:

•

New and current personal informatics applicationsand systems on the desktop and online

•

Sensor and life-logging technologies that monitorvarious personal behavioral information

•

Effective feedback techniques, such asvisualizations, virtual agents, and persuasivetechnologies, that help users become more awareof their own behaviors

•

Interaction techniques that alleviate the burdenthat personal informatics impose on engagement

•

Effects of self-knowledge and self-awareness onbehaviors and daily life

•

Methods of conducting long-term studies todetermine effects of information on user behavior