Personal Informatics Systems Research Articles

Lateralization Effects in Electrodermal Activity Data Collected Using Wearable Devices

Electrodermal activity (EDA) is a physiological signal that can be used to infer humans' affective states and stress levels. EDA can nowadays be monitored using unobtrusive wearable devices, such as smartwatches, and leveraged in personal informatics systems. A still largely uncharted issue concerning EDA is the impact on real applications of potential differences observable on signals measured concurrently on the left and right side of the human body. This phenomenon, called lateralization, originates from the distinct functions that the brain's left and right hemispheres exert on EDA. In this work, we address this issue by examining the impact of EDA lateralization in two classification tasks: a cognitive load recognition task executed in the lab and a sleep monitoring task in a real-world setting. We implement a machine learning pipeline to compare the performance obtained on both classification tasks using EDA data collected from the left and right sides of the body. Our results show that using EDA from the side that is not associated with the specific hemisphere activation leads to a significant decline in performance for the considered classification tasks. This finding highlights that researchers and practitioners relying on EDA data should consider possible EDA lateralization effects when deciding on sensor placement.

A Meta-Synthesis of the Barriers and Facilitators for Personal Informatics Systems

Personal informatics (PI) systems are designed for diverse users in the real world. Even when these systems are usable, people encounter barriers while engaging with them in ways designers cannot anticipate, which impacts the system's effectiveness. Although PI literature extensively reports such barriers, the volume of this information can be overwhelming. Researchers and practitioners often find themselves repeatedly addressing the same challenges since sifting through this enormous volume of knowledge looking for relevant insights is often infeasible. We contribute to alleviating this issue by conducting a meta-synthesis of the PI literature and categorizing people's barriers and facilitators to engagement with PI systems into eight themes. Based on the synthesized knowledge, we discuss specific generalizable barriers and paths for further investigations. This synthesis can serve as an index to identify barriers pertinent to each application domain and possibly to identify barriers from one domain that might apply to a different domain. Finally, to ensure the sustainability of the syntheses, we propose a Design Statements (DS) block for research articles.

Uncovering Bias in Personal Informatics

Personal informatics (PI) systems, powered by smartphones and wearables, enable people to lead healthier lifestyles by providing meaningful and actionable insights that break down barriers between users and their health information. Today, such systems are used by billions of users for monitoring not only physical activity and sleep but also vital signs and women's and heart health, among others. Despite their widespread usage, the processing of sensitive PI data may suffer from biases, which may entail practical and ethical implications. In this work, we present the first comprehensive empirical and analytical study of bias in PI systems, including biases in raw data and in the entire machine learning life cycle. We use the most detailed framework to date for exploring the different sources of bias and find that biases exist both in the data generation and the model learning and implementation streams. According to our results, the most affected minority groups are users with health issues, such as diabetes, joint issues, and hypertension, and female users, whose data biases are propagated or even amplified by learning models, while intersectional biases can also be observed.

The Role of Model Personalization for Sleep Stage and Sleep Quality Recognition Using Wearables

Personal informatics systems can help people promote their health and well-being. Recent studies have shown that such systems can be used to infer relevant health indicators such as, e.g., stress, anxiety, and sleeping habits. While automatic detection of sleep has been studied extensively, there is a lack of studies exploring how population and personalized models influence the performance of sleep detection. In this article, we address this challenge by investigating the recognition of sleep/wake stages and high/low sleep quality with a focus on the impact of personalized models. To evaluate our approach, we collect a dataset of physiological signals and self-reports about sleep/wake times and sleep quality score. The dataset contains 6557 hours of sensor data collected using wristbands from 16 participants over one month. Our results show that personalized models perform significantly better than population models for sleep quality recognition, and are comparably good for sleep stage detection. The balanced accuracy for sleep/wake and high/low sleep quality are 92.2% and 61.51%, which are significantly higher than baseline classifiers.

A Model of Socially Sustained Self-Tracking for Food and Diet

Studies of personal informatics systems primarily examine people's use or non-use, but people often leverage other technology towards their long-term behavior change processes such as social platforms. We explore how tracking technologies and social platforms together help people build healthy eating behaviors by interviewing 18 people who use Chinese food journaling apps. We contribute a Model of Socially Sustained Self-Tracking in personal informatics, building on the past model of Personal Informatics and the learning components of Social Cognitive Theory. The model illustrates how people get advice from social platforms on when and how to track, transfer data to and apply knowledge from social platforms, evolve to use social platforms after tracking, and occasionally resume using tracking tools. Observational learning and enactive learning are central to these processes, with social technologies helping people to gain deeper and more reliable domain knowledge. We discuss how lapsing and abandoning of tracking can be viewed as evolving to social platforms, offering recommendations for how technology can better facilitate this evolution.

Psychophysiological Therapy from a Distance: The Art of Sharing

Psychophysiological Therapy from a Distance: The Art of Sharing

A Multi-Sensor Approach to Automatically Recognize Breaks and Work Activities of Knowledge Workers in Academia

Personal informatics systems for the work environment can help improving workers' well-being and productivity. Using both self-reported data logged manually by the users and information automatically inferred from sensor measurements, such systems may track users' activities at work and help them reflect on their work habits through insightful data visualizations. They can further support interventions like, e.g., blocking distractions during work activities or suggest the user to take a break. The ability to automatically recognize when the user is engaged in a work activity or taking a break is thus a fundamental primitive such systems need to implement. In this paper, we explore the use of data collected from personal devices -- smartwatches, laptops, and smartphones -- to automatically recognize when users are working or taking breaks. We collect a data set of of continuous streams of sensor data captured from personal devices along with labels indicating whether a user is working or taking a break. We use multiple instruments to facilitate the collection of users' self-reported labels and discuss our experience with this approach. We analyse the available data -- 449 labelled activities of nine knowledge workers collected during a typical work week -- using machine learning techniques and show that user-independent models can achieve a (F1 score) of 94% for the identification of work activities and of 69% for breaks, outperforming baseline methods by 5-10 and 12-54 percentage points, respectively.

Being (In)Visible: Privacy, Transparency, and Disclosure in the Self-Management of Bipolar Disorder.

Research in personal informatics (PI) calls for systems to support social forms of tracking, raising questions about how privacy can and should support intentionally sharing sensitive health information. We focus on the case of personal data related to the self-tracking of bipolar disorder (BD) in order to explore the ways in which disclosure activities intersect with other privacy experiences. While research in HCI often discusses privacy as a disclosure activity, this does not reflect the ways in which privacy can be passively experienced. In this paper we broaden conceptions of privacy by defining transparency experiences and contributing factors in contrast to disclosure activities and preferences. Next, we ground this theoretical move in empirical analysis of personal narratives shared by people managing BD. We discuss the resulting emergent model of transparency in terms of implications for the design of socially-enabled PI systems. CAUTION: This paper contains references to experiences of mental illness, including self-harm, depression, suicidal ideation, etc.

Practical Challenges for the Design of Personal Informatics Systems

Practical Challenges for the Design of Personal Informatics Systems

Personal Informatics in Interpersonal Contexts

Personal informatics systems for supporting health largely grew out of a "self"-centric orientation: self-tracking, self-reflection, self-knowledge, self-experimentation, self-improvement. Health management, however, even when self-driven, is inherently social and depends on a person's direct relationships and broader sociocultural contexts, as an emerging line of research is coming to recognize, study, and support. This is particularly true in the case of mental health. In this paper, we engage with individuals managing the serious mental illness bipolar disorder and members of their support circles to (a) identify key social relations and the roles they play in condition management, (b) characterize patients' complex interactions with these relations (e.g., positive or negative, direct or peripheral, steady or unstable), and (c) understand how personal informatics mediates these recovery relations. Based on these insights, we offer a model of this social ecology, along with design implications for personal informatics systems that are sensitive to these interpersonal contexts.

People Like Me

Increases in data complexity in personal informatics systems require new ways of contextualizing personal data to facilitate meaningful reflection. An emerging approach for providing such context includes augmenting one's personal data with the data of others "like them" to help individuals make sense of their data. However, we do not yet understand how an individual's self-reflection process is affected when the data of others is made available. In this paper, we investigate how people reflect on three types of personal data when presented alongside a large set of aggregated data of multiple cohorts. We conducted personal and cohort data reviews using a subset of participants from a mobile-sensing study that collected physical activity, digital social activity, and perceived stress, from 47 students over three weeks. Participants preferred to use characteristics of the data (e.g., maxima, minima) and graphical presentation (e.g., appearance of trends) along with demographic identities (e.g., age, gender) when relating to cohorts. We further characterize how participants incorporated cohort data into their self-reflection process, and conclude with discussion of the implications for personal informatics systems that leverage the data of "people like me" to enable meaningful reflection.

On Being Told How We Feel

Algorithms and sensors are increasingly deployed for highly personal aspects of our everyday lives. Recent work suggests people have imperfect understanding of system outputs, often assuming sophisticated capabilities and deferring to feedback. We explore how people construe algorithmic interpretations of emotional data in personal informatics systems. A survey (n=188) showed strong interest in automatic stress and emotion tracking, but many respondents expected these systems to provide objective measurements for their emotional experiences. A second study examined how algorithmic sensor feedback influences emotional self-judgments, by comparing three system framings of physiological ElectroDermal Activity data (EDA): Positive ("alert and engaged"), Negative ("stressed"), and Control (no frame) in a mixed-methods study with 64 participants. Despite users reporting strategies to test system outputs, users still deferred to feedback and their perceived emotions were significantly influenced by feedback frames. Some users overrode personal judgments, believing the system had access to privileged information about their emotions. Based on these findings, we explore design implications for personal informatics including risks of users trusting systems that seemingly "unlock" hidden aspects of the self. We propose design approaches that provide opportunities for future emotion-monitoring systems to exploit these framing effects, and for users to more actively construe emotional states.

Personal informatics systems for supporting self‐regulation in online learning environments

AbstractPersonal informatics (PI) systems have emerged as powerful tools for helping people to be aware of their behaviors in diverse domains. Nevertheless, little attention has been devoted to educational scenarios. For this reason, this paper presents an educational PI system called Glance and, moreover, evaluates its usefulness in two fully online undergraduate courses.

Application of Genetic Algorithm and Personal Informatics in Stock Market

The financial market is extremely attractive since it moves trillion dollars per year. Many investors have been exploring ways to predict future prices by using different types of algorithms that use fundamental analysis and technical analysis. Many professional speculators or amateurs had been analysing the price movement of some financial assets using these algorithms. The use of genetic algorithms, neural networks, genetic programming combined with these tools in an attempt to find a profitable solution is very common. This study presents a prototype that utilizes genetic algorithms (GAs) and personal informatics system (PI) for short-term stock index forecast. The prototype works according to the following steps. Firstly, a collection of input variables is defined through technical data analysis. Secondly, GA is applied to determine an optimal set of input variables for a one-day forecast. Â The data is gathered from the Saudi Stock Exchange as being the target market. Thirdly, PI is utilised to create a smart environment, which enables visualisation of stock prices. The outcome indicates that this approach of forecasting the stock price is positive. The highest accuracy obtained is 64.67% and the lowest one is 48.06%.

Designing a personal informatics system for users without experience in self-tracking: a case study

ABSTRACT Thanks to the advancements in ubiquitous and wearable technologies, Personal Informatics (PI) systems can now reach a larger audience of users. However, it is not still clear whether this kind of tool can fit the needs of their daily lives. Our research aims at identifying specific barriers that may prevent the widespread adoption of PI and finding solutions to overcome them. We requested users without competence in self-tracking to use different PI instruments during their daily practices, identifying five user requirements by which to design novel PI tools. On such requirements, we developed a new system that can stimulate the use of these technologies, by enhancing the perceived benefits of collecting personal data. Then, we explored how naïve and experienced users differently explore their personal data in our system through a user trial. Results showed that the system was successful at helping individuals manage and interpret their own data, validated the usefulness of the requirements found and inspired three further design opportunities that could orient the design of future PI systems.

Editorial of the Special Issue on Quantified Self and Personal Informatics

In recent years, we witnessed the spreading of a plethora of wearable and mobile technologies allowing for a continuous and “transparent” gathering of personal data [...]

NEAT-Lamp and Talking Tree: Beyond Personal Informatics towards Active Workplaces

A growing number of personal informatics (PI) systems have been designed to break the habit of prolonged sitting and to encourage physical activity during workdays and leisure hours. Few studies, however, have investigated the nature of local movement and mobility in workspaces. Relatively little is known about how such movement patterns are shaped and in what ways micro-mobility in workplaces could be increased. By undertaking a concept-driven design approach, and on the basis of our ethnographic prestudy, we introduce a conceptual framework. In this conceptual framework, we indicate the five main agencies that shape local movement and mobility among office workers. On the basis of this empirical and conceptual work, two prototypes, the non-exercise activity thermogenesis (NEAT)-Lamp and Talking Tree, have been designed, implemented and observed in an office environment. This paper describes this design project and articulates the role of discussions in socially established settings in work environments in order to increase daily movement. The paper concludes by highlighting not only technology, but also collective reflections to spark behavioral change in office environments as social settings.

Dealing With Information Overload in Multifaceted Personal Informatics Systems

Personal informatics systems are tools that capture, aggregate, and analyze data from distinct facets of their users’ lives. This article adopts a mixed-methods approach to understand the problem of information overload in personal informatics systems. We report findings from a 3-month study in which 20 participants collected multifaceted personal tracking data and used a system called Exist to reveal statistical correlations within their data. We explore the challenges that participants faced in reviewing the information presented by Exist, and we identify characteristics that exemplify “interesting” correlations. Based on these findings, we develop automated filtering mechanisms that aim to prevent information overload and support users in extracting interesting insights. Our approach deals with information overload by reducing the number of correlations shown to users by about 55% on average and increases the percentage of displayed correlations rated as interesting to about 81%, representing a 34 percentage point improvement over filters that only consider statistical significance at p < .05. We demonstrate how this curation can be achieved using objective data harvested by the system, including the use of Google Trends data as a proxy for subjective user interest.

Personal Informatics, Self-Insight, and Behavior Change: A Critical Review of Current Literature

Personal informatics (PI) systems allow users to collect and review personally relevant information. The purpose commonly envisioned for these systems is that they provide users with actionable, data-driven self-insight to help them change their behavioral patterns for the better. Here, we review relevant theory as well as empirical evidence for this self-improvement hypothesis. From a corpus of 6,568, only 24 studies met the selection criteria of being a peer-reviewed empirical study reporting on actionable, data-driven insights from PI data, using a “clean” PI system with no other intervention techniques (e.g., additional coaching) on a nonclinical population. First results are promising—many of the selected articles report users gaining actionable insights—but we do note a number of methodological issues that make these results difficult to interpret. We conclude that more work is needed to investigate the self-improvement hypothesis and provide a set of recommendations for future work.

ESelf or Computerized Self Network

Empowering individuals is carried out through improving self-knowledge and self-awareness and maximization of their autonomous self-regulatory potential. In order to improve individual self-knowledge the author needs to provide evidence-based and need-oriented information to individuals and help them to process and self-reflect this information This paper attempts to describe a conceptual system, called Electronic-Self or Computerized Self Network, encompassing Medical Informatics Systems, Personal Informatics System, and Optimal Health Care Providers and explain how such a system can be considered as a practical solution. eSelf a complex adaptive system and based complexity sciences rules. eSelf strategy for individual empowerment based on each individual should be in personal health context play two role philosopher and scientist in personal health information processing and embodied self awareness.