Employing Consumer Wearables to Detect Office Workers' Cognitive Load for Interruption Management
The DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index.
Abstract
Office workers' productivity and well-being are reduced by interruptions, especially if they occur during an inconvenient moment. Interruptions in phases of high cognitive load are more disruptive than in phases of low cognitive load. Based on an explorative study, we suppose the presence of social codes that signal office workers' interruptibility. We propose a system that utilizes the cognitive load of an office worker to indicate situations suitable for interruptions. The cognitive load is inferred from office workers' physiological state measured by a consumer smartwatch. The system adapts an externally mounted smart device to indicate if the office worker is interruptible. To predict the cognitive load, we trained a classifier with ten office workers and achieved an accuracy between 66% and 86%. In order to validate the classifier's accurateness in an office setting, we performed a verification study with five office workers: We systematically triggered interruptions for each subject over an interval of half a day of office work. The classifier was able to infer the level of cognitive load for three office workers. This result supports our hypotheses that inferring cognitive load using a consumer smartwatch is a viable concept.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
(1994) · Cited by 21,711
The magical number seven, plus or minus two: Some limits on our capacity for processing information.
Miller, G. A. (1994) · Psychological Review · doi:10.1037/0033-295x.101.2.343 · Cited by 17,161
Development of NASA-TLX (Task Load Index): Results of Empirical and Theoretical Research
(1988) · Advances in Psychology · doi:10.1016/s0166-4115(08)62386-9 · Cited by 15,003 · cited 2×
“There are four different ways to quantify the cognitive load of a human [6]: • Subjective (self-reported) Measure: Quantifying the cognitive load by asking the subjects to rate their experienced cognitive load on a rating scale immediately after finishing a task [15].”
BackgroundMethodCognitive Load During Problem Solving: Effects on Learning
Sweller, J. (1988) · Cognitive Science · doi:10.1016/0364-0213(88)90023-7 · Cited by 11,048 · cited 1×
“The psychological concept of cognitive load refers to the total mental effort required to perform a task puts on the human working memory [33].”
BackgroundScikit-learn: Machine Learning in Python
(2011) · Journal of Machine Learning Research · doi:10.5555/1953048.2078195 · Cited by 8,539 · cited 1×
“The scikit-learn library was used for the classification of the discrete levels of cognitive load [31].”
Method(2006) · Springer eBooks · Cited by 8,123
Improving fluid intelligence with training on working memory
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0801268105 · Cited by 2,187 · cited 1×
“During these workload phases, the participant performs different variants of the Dual N-Back Test to induce low, medium, and high levels of cognitive load [17].”
MethodProceedings of the Human Factors and Ergonomics Society 2019 Annual Meeting
(2019) · Proceedings of the Human Factors and Ergonomics Society Annual Meeting · doi:10.1177/1071181319631171 · Cited by 1,432
Instructional control of cognitive load in the training of complex cognitive tasks
Paas, F. G. W. C., & Van Merriënboer, J. J. G. (1994) · Educational Psychology Review · doi:10.1007/bf02213420 · Cited by 968 · cited 1×
“The underlying assumption is that an overloaded working memory, which is an indicator of high cognitive load, will result in lower task performance [30].”
BackgroundEdge-centric Computing: Vision and Challenges
Garcia Lopez, P., Montresor, A., Epema, D., Datta, A., Higashino, T., Iamnitchi, A., Barcellos, M., Felber, P., & Riviere, E. (2015) · ACM SIGCOMM Computer Communication Review · doi:10.1145/2831347.2831354 · Cited by 934
The cost of interrupted work: more speed and stress
Mark, G., Gudith, D., & Klocke, U. (2008) · Proceedings of the SIGCHI Conference on Human Factors in Computing Systems · doi:10.1145/1357054.1357072 · Cited by 728 · cited 1×
“Other research argues that interruptions occurring at an inopportune moment have a negative effect on the well-being of humans since they cause stress [2, 24].”
Background(2010) · Cambridge University Press eBooks · doi:10.1017/cbo9780511844744 · Cited by 575
(2001) · Cited by 269
Psychological Issues in Support of Multiple Activities
MIYATA, Y., & NORMAN, D. A. (1986) · User Centered System Design · doi:10.1201/b15703-13 · Cited by 245 · cited 1×
“External interruptions are caused by changes in the environment, e.g., a ringing phone [27].”
BackgroundMonitoring of mental workload levels during an everyday life office-work scenario
Cinaz, B., Arnrich, B., La Marca, R., & Tröster, G. (2013) · Personal and Ubiquitous Computing · doi:10.1007/s00779-011-0466-1 · Cited by 235 · cited 14×
“We achieve an accuracy for the general classifier between 31% and 35% compared to an accuracy between 35% and 40% using a GSR sensor [29].”
BackgroundMethodInfluentialKrause, A., Smailagic, A., & Siewiorek, D. P. (2006) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2006.18 · Cited by 208 · cited 1×
“present a system that applies unsupervised machine learning to detect the context-dependent personal preferences of the user using the state, context, and user interactions with the system [22].”
BackgroundIqbal, S. T., & Bailey, B. P. (2005) · CHI '05 Extended Abstracts on Human Factors in Computing Systems · doi:10.1145/1056808.1056948 · Cited by 203 · cited 1×
“Research reveals that interruptions during phases of high cognitive load are perceived as more disruptive than in phases of low cognitive load [13, 16, 20].”
BackgroundRobust Multimodal Cognitive Load Measurement
Chen, F., Zhou, J., Wang, Y., Yu, K., Arshad, S. Z., Khawaji, A., & Conway, D. (2016) · Human–Computer Interaction Series · doi:10.1007/978-3-319-31700-7 · Cited by 192 · cited 12×
“We achieve an accuracy for the general classifier between 31% and 35% compared to an accuracy between 35% and 40% using a GSR sensor [29].”
InfluentialCognitive Heat: Exploring the Usage of Thermal Imaging to Unobtrusively Estimate Cognitive Load
Abdelrahman, Y., Velloso, E., Dingler, T., Schmidt, A., & Vetere, F. (2017) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3130898 · Cited by 154
PrefMiner: mining user's preferences for intelligent mobile notification management
Mehrotra, A., Hendley, R., & Musolesi, M. (2016) · Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing · doi:10.1145/2971648.2971747 · Cited by 137 · cited 1×
“The project PrefMiner [25] tries to reduce the number of notifications by classifying them into important and unimportant and consequently only presents the important notifications.”
BackgroundTask Interruption and its Effects on Memory
B. Edwards, M., & Gronlund, S. D. (1998) · Memory · doi:10.1080/741943375 · Cited by 132 · cited 1×
“Multiple studies reveal that interruptions make humans slower and more error-prone in performing tasks [8, 13].”
BackgroundEstimating heart rate using wrist-type Photoplethysmography and acceleration sensor while running
Fukushima, H., Kawanaka, H., Bhuiyan, M. S., & Oguri, K. (2012) · 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society · doi:10.1109/embc.2012.6346570 · Cited by 126 · cited 5×
“The ECG combined with the multiple PPG sensors offers a more accurate measurement of the HRV as well as a measurement of the blood pressure.”
BackgroundInfluentialUsing mental load for managing interruptions in physiologically attentive user interfaces
Chen, D., & Vertegaal, R. (2004) · CHI '04 Extended Abstracts on Human Factors in Computing Systems · doi:10.1145/985921.986103 · Cited by 120 · cited 1×
“…precedent systems for physiological measurement and response to cognitive loading are important background for this research project: Chen and Vertegaal developed a system that automatically manages mobile phone notifications depending on the movement state and the cognitive load of the user [5].”
BackgroundGSR and Blink Features for Cognitive Load Classification
Nourbakhsh, N., Wang, Y., & Chen, F. (2013) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-40483-2_11 · Cited by 98 · cited 12×
“This is comparable accuracy to similar research [29] in the classification of the level of cognitive load.”
BackgroundMethodResultInfluentialInterruptibility of Software Developers and its Prediction Using Psycho-Physiological Sensors
Züger, M., & Fritz, T. (2015) · Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems · doi:10.1145/2702123.2702593 · Cited by 96 · cited 1×
“Züger and Fritz present a system that predicts the interruptibility of professional software developers using the data of wearable physiological sensors [36].”
BackgroundMultitasking in the Digital Age
Mark, G. J. (2015) · Synthesis lectures on human-centered informatics · doi:10.2200/s00635ed1v01y201503hci029 · Cited by 88 · cited 1×
“Humans need on average about twenty-three minutes to resume their task after they got interrupted [23].”
BackgroundA review on stress inducement stimuli for assessing human stress using physiological signals
Karthikeyan, P., Murugappan, M., & Yaacob, S. (2011) · 2011 IEEE 7th International Colloquium on Signal Processing and its Applications · doi:10.1109/cspa.2011.5759914 · Cited by 86 · cited 1×
“Stress influences the HRV [21], GSR, and the heart rate [18].”
BackgroundReducing Interruptions at Work: A Large-Scale Field Study of FlowLight
Züger, M., Corley, C., Meyer, A. N., Li, B., Fritz, T., Shepherd, D., Augustine, V., Francis, P., Kraft, N., & Snipes, W. (2017) · Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3025453.3025662 · Cited by 80 · cited 2×
“do not report on such correlation in their long-term study with 449 participants [35].”
BackgroundWhat happens when we switch tasks: Pupil dilation in multitasking.
Katidioti, I., Borst, J. P., & Taatgen, N. A. (2014) · Journal of Experimental Psychology: Applied · doi:10.1037/xap0000031 · Cited by 59 · cited 1×
“Research reveals that interruptions during phases of high cognitive load are perceived as more disruptive than in phases of low cognitive load [13, 16, 20].”
BackgroundFerreira, E., Ferreira, D., Kim, S., Siirtola, P., Roning, J., Forlizzi, J. F., & Dey, A. K. (2014) · 2014 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain (CCMB) · doi:10.1109/ccmb.2014.7020692 · Cited by 51 · cited 1×
“measure the cognitive load using an EEG, an ECG, and a GSR sensor [9].”
MethodMultitasking and Interruptions during Mobile Web Tasks
Nagata, S. F. (2003) · Proceedings of the Human Factors and Ergonomics Society Annual Meeting · doi:10.1177/154193120304701105 · Cited by 49 · cited 1×
“, someone entering the office without an appointment [28].”
BackgroundInterrupted by Your Pupil: An Interruption Management System Based on Pupil Dilation
Katidioti, I., Borst, J. P., de Haan, D. J. B., Pepping, T., van Vugt, M. K., & Taatgen, N. (2016) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2016.1198525 · Cited by 47 · cited 2×
“By detecting changes in these patterns the cognitive load can be inferred [19].”
BackgroundGoyal, N., & Fussell, S. R. (2017) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3130917 · Cited by 46 · cited 1×
“Goyal and Fussell identify suitable moments of interruptions to reduce their negative effects on the task performance [14].”
BackgroundGould, S. J. J., Brumby, D. P., & Cox, A. L. (2013) · Proceedings of the Human Factors and Ergonomics Society Annual Meeting · doi:10.1177/1541931213571034 · Cited by 40 · cited 2×
“Multiple studies reveal that interruptions make humans slower and more error-prone in performing tasks [8, 13].”
BackgroundInterruptibility Estimation Based on Head Motion and PC Operation
Tanaka, T., Abe, R., Aoki, K., & Fujita, K. (2015) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2014.986635 · Cited by 20 · cited 1×
“al develop an interruption management system that uses head movements and the interaction with the computer to identify phases of low cognitive load [34].”
BackgroundKim, D., Seo, Y., & Salahuddin, L. B. (2008) · International Conference on Pervasive Computing Technologies for Healthcare · doi:10.1109/pcthealth.2008.4571093 · Cited by 20
Binsch, O., Wabeke, T. R., & Valk, P. J. L. (2016) · IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN) · doi:10.1109/bsn.2016.7516272 · Cited by 20 · cited 1×
“Nevertheless, research shows that consumer wearables (Microsoft Band 2, Fitbit Surge, Apple Watch) are reliable for measuring the heart rate in resting condition (sitting) [3].”
BackgroundMultitasking and interruptions during mobile web tasks
Nagata, S. F. (2003) · PsycEXTRA Dataset · doi:10.1037/e576962012-005 · Cited by 12
Gloria Mark, Multitasking in the digital age , Morgan & Claypool, 2015
Kavanagh, É., & Roberge, J. (2018) · Interfaces numériques · Cited by 1
Research works printed but not linked: 3
- 10.5555/1162264 · doi:10.5555/1162264
- 10.5555/186897 · doi:10.5555/186897
- 10.1207/s15516709cog1202_4 · doi:10.1207/s15516709cog1202_4 · Paper page
Cited by
A Survey on Measuring Cognitive Workload in Human-Computer Interaction
Kosch, T., Karolus, J., Zagermann, J., Reiterer, H., Schmidt, A., & Woźniak, P. W. (2023) · ACM Computing Surveys · doi:10.1145/3582272 · Cited by 304
“In a darker environment, pupils dilate to acquire more light – pupils contract to reduce the amount of light in a brighter environment.”
Datasets for Cognitive Load Inference Using Wearable Sensors and Psychological Traits
Gjoreski, M., Kolenik, T., Knez, T., Luštrek, M., Gams, M., Gjoreski, H., & Pejović, V. (2020) · Applied Sciences · doi:10.3390/app10113843 · Cited by 120
Physiological correlates of discomfort in automated driving
Beggiato, M., Hartwich, F., & Krems, J. F. (2019) · Transportation Research Part F Traffic Psychology and Behaviour · doi:10.1016/j.trf.2019.09.018 · Cited by 92
Differentiating Higher and Lower Job Performers in the Workplace Using Mobile Sensing
Mirjafari, S., Masaba, K., Grover, T., Wang, W., Audia, P., Campbell, A. T., Chawla, N. V., Swain, V. D., Choudhury, M. D., Dey, A. K., et al. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3328908 · Cited by 83
“They demonstrate that physiological data relates to mental state and can determine when a person is busy and intellectually invested in their work [61].”
BackgroundInfluentialKawakami, A., Chowdhary, S., Iqbal, S. T., Liao, Q. V., Olteanu, A., Suh, J., & Saha, K. (2023) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3610207 · Cited by 100
“In addition, prior research has used multimodal sensing through smartphone, wearable, bluetooth, and wireless sensors to try to infer job performance [111], mood and cognition [17, 18, 104, 133, 145], social interactions [24, 107], organizational personas [42], organizational fit [45].”
BackgroundMethod
Metadata
- Authors
- Schaule, F., Johanssen, J. O., Bruegge, B., & Loftness, V.
- Affiliations
- Technical University of Munich (DE)
- Carnegie Mellon University (US)
- Venue
- Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Published
- 2018-03-26
- Volume
- 2
- Issue
- 1
- Pages
- 1-20
- ISSN
- 2474-9567 (ISSN-L)
- OpenAlex
- W2794857355
- Cited by
- 77
- References
- 40
- Citations by year
- 2026: 10 · 2025: 8 · 2024: 10 · 2023: 15 · 2022: 5 · 2021: 10 · 2020: 10 · 2019: 5 · 2018: 4
- FWCI
- 10.29
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies (publishedVersion)
- Funders
- Deutscher Akademischer Austauschdienst
- Topics
- Social Sciences › Decision Sciences › Information Systems and Management › Personal Information Management and User Behavior
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Social Sciences › Human Factors and Ergonomics › Human-Automation Interaction and Safety
- Keywords
- cognitive load, interruptions, office workers, physiological monitoring, interruptibility, social codes, wearable devices, productivity, well-being, classification accuracy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.