skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Effects of adaptation accuracy and magnitude in affect-aware difficulty adaptation for the multi-attribute task battery
In affective computing, classification algorithms are used to recognize users’ psychological states and adapt tasks to optimize user experience. However, classification is never perfect, and the relationship between adaptation accuracy and user experience remains understudied. It is also unclear whether the adaptation magnitude (‘size’ of action taken to influence user states) influences effects of adaptation accuracy. To evaluate impacts of adaptation accuracy (appropriate vs. inappropriate actions) and magnitude on user experience, we conducted a ‘Wizard of Oz’ study where 112 participants interacted with the Multi-Attribute Task Battery over three 11-minute intervals. An adaptation accuracy (50 % to 80 %) was preassigned for the first 11-minute interval, and accuracy increased by 10 % in each subsequent interval. Task difficulty changed every minute, and participant preferences for difficulty changes were assessed at the same time. Adaptation accuracy was artificially induced by fixing the percentage of times the difficulty changes matched participant preferences. Participants were also randomized to two magnitude conditions, with difficulty modified by 1 (low) or 3 (high) levels each minute. User experience metrics were assessed after each interval. Analysis with latent growth models offered support for linear increases in user experience across increasing levels of adaptation accuracy. For each 10 % gain in accuracy, results indicate a 1.3 (95 % CI [.35, 2.20]) point increase in NASA Task Load Index scores (range 6–60), a 0.40 (95 % CI [.18, 0.57]) increase in effort/importance (range 2–14), and 0.48 (95 % CI [.24, 0.72]) increase in perceived competence (range 2–14). Furthermore, the effect of accuracy on Task Load Index scores was modulated by adaptation magnitude. No effects were observed for interest/enjoyment or pressure/tension. By providing quantitative estimates of effects of adaptation accuracy on user experience, the study provides guidelines for researchers and developers of affect-aware technologies. Furthermore, our methods could be adapted for use in other affective computing scenarios.  more » « less
Award ID(s):
2151464
PAR ID:
10547228
Author(s) / Creator(s):
; ; ; ;
Publisher / Repository:
Elsevier
Date Published:
Journal Name:
International Journal of Human-Computer Studies
Volume:
183
Issue:
C
ISSN:
1071-5819
Page Range / eLocation ID:
103180
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; (, IEEE)
    Physiological sensors are commonly applied for user state monitoring and consequent machine behavior adaptation in applications such as rehabilitation and intelligent cars. While more accurate user state monitoring is known to lead to better user experience, increased accuracy often requires more sensors or more complex sensors. The increased setup time and discomfort involved in the use of such sensors may itself worsen user experience. To examine this effect, we conducted a study where 72 participants interacted with a computer-based multitasking scenario whose difficulty was periodically adapted - ostensibly based on data from either a remote eye tracker or a lab-grade “wet” electroencephalography sensor. Deception was used to ensure consistent difficulty adaptation accuracies, and user experience was measured with the Intrinsic Motivation Inventory, NASA Task Load Index, and an ad-hoc scale. We found few user experience differences between the eye tracker and electroencephalography sensor - while one interaction effect was noted, it was small, and there were no other differences. This result is at first surprising and seems to indicate that comfort and setup time are not major factors for laboratory-based user experience evaluations of such technologies. However, the result is likely due to a suboptimal study protocol where each participant interacted with only one sensor. In future work, we will use an alternate protocol to further explore the effects of user comfort and setup time on user experience. 
    more » « less
  2. ; ; (, IEEE Transactions on Affective Computing)
    In affect-aware task adaptation, users' psychological states are recognized with diverse measurements and used to adapt computer-based tasks. User experience with such adaptation improves as the accuracy of psychological state recognition and task adaptation increases. However, it is unclear how user experience is influenced by algorithmic transparency: the degree to which users understand the computer's decision-making process. We thus created an affect-aware task adaptation system with 4 algorithmic transparency levels (none/low/medium/high) and conducted a study where 93 participants first experienced adaptation with no transparency for 16 minutes, then with one of the other 3 levels for 16 minutes. User experience questionnaires and physiological measurements (respiration, skin conductance, heart rate) were analyzed with mixed 2×3 analyses of variance (time × transparency group). Self-reported interest/enjoyment and competence were lower with low transparency than with medium/high transparency, but did not differ between medium and high transparency. The transparency level may also influence participants' respiratory responses to adaptation errors, but this finding is based on ad-hoc t-tests and should be considered preliminary. Overall, results show that the degree of algorithmic transparency does influence self-reported user experience. Since transparency information is relatively easy to provide, it may represent a worthwhile design element in affective computing. 
    more » « less
  3. ; ; ; ; ; (, GeoHealth)
    Abstract Greenspace in schools might enhance students' academic performance. However, the literature—dominated by ecological studies at the school level in countries from the Northern Hemisphere—presents mixed evidence of a beneficial association. We evaluated the association between school greenness and student‐level academic performance in Santiago, Chile, a capital city of the Global South. This cross‐sectional study included 281,695 fourth‐grade students attending 1,498 public, charter, and private schools in Santiago city between 2014 and 2018. Student‐level academic performance was assessed using standardized test scores and indicators of attainment of learning standards in mathematics and reading. School greenness was estimated using Normalized Difference Vegetation Index (NDVI). Linear and generalized linear mixed‐effects models were fit to evaluate associations, adjusting for individual‐ and school‐level sociodemographic factors. Analyses were stratified by school type. In fully adjusted models, a 0.1 increase in school greenness was associated with higher test scores in mathematics (36.9 points, 95% CI: 2.49; 4.88) and in reading (1.84 points, 95% CI: 0.73; 2.95); as well as with higher odds of attaining learning standards in mathematics (OR: 1.20, 95% CI: 1.12; 1.28) and reading (OR: 1.07, 95% CI: 1.02; 1.13). Stratified analysis showed differences by school type, with associations of greater magnitude and strength for students attending public schools. No significant associations were detected for students in private schools. Higher school greenness was associated with improved individual‐level academic outcomes among elementary‐aged students in a capital city in South America. Our results highlight the potential of greenness in the school environment to moderate educational and environmental inequalities in urban areas. 
    more » « less
  4. ; ; ; ; ; ; ; ; ; ; et al (, British Journal of Cancer)
    Abstract BackgroundPrevious social determinants of health (SDoH) studies on laryngeal cancer (LC) have assessed individual factors of socioeconomic status and race/ethnicity but seldom investigate a wider breadth of SDoH-factors for their effects in the real-world. This study aims to delineate how a wider array of SDoH-vulnerabilities interactively associates with LC-disparities. MethodsThis retrospective cohort study assessed 74,495 LC-patients between 1975 and 2017 from the Surveillance-Epidemiology-End Results (SEER) database using the Social Vulnerability Index (SVI) from the CDC, total SDoH-vulnerability from 15 SDoH variables across specific vulnerabilities of socioeconomic status, minority-language status, household composition, and infrastructure/housing and transportation, which were measured across US counties. Univariate linear and logistic regressions were performed on length of care/follow-up and survival, staging, and treatment across SVI scores. ResultsSurvival time dropped significantly by 34.37% (from 72.83 to 47.80 months), and surveillance time decreased by 28.09% (from 80.99 to 58.24 months) with increasing overall social vulnerability, alongside advanced staging (OR 1.15; 95%CI 1.13–1.16), increased chemotherapy (OR 1.13; 95%CI 1.11–1.14), decreased surgical resection (OR 0.91; 95%CI 0.90–0.92), and decreased radiotherapy (OR 0.97; 95%CI 0.96–0.99). DiscussionIn this SDoH-study of LCs, detrimental care and prognostic trends were observed with increasing overall SDoH-vulnerability. 
    more » « less
  5. ; ; (, Academic Medicine)
    Abstract PurposeTo examine graduating medical student reports of burnout by sex, race and ethnicity, and sexual orientation and explore trends within intersectional demographic groups from 2019–2021 in a national sample. MethodThe authors obtained medical student responses to the 2019–2021 Association of American Medical Colleges (AAMC) Graduation Questionnaires (GQs) linked to data from other AAMC sources. The dataset included year of GQ completion, responses to a modified Oldenburg Burnout Inventory (exhaustion subscale range: 0–24; disengagement subscale range: 0–15), and demographics previously shown to relate to the risk of burnout in medical students, residents, or physicians. Multivariable linear regression analysis was performed to evaluate independent associations between demographics and burnout. ResultsOverall response rate was 80.7%. After controlling for other factors, mean exhaustion scores were higher among Asian (parameter estimate [PE] 0.38, 95% confidence interval [CI] 0.21, 0.54), bisexual (PE 0.97, 95% CI 0.76, 1.17), and gay or lesbian (PE 0.55, 95% CI 0.35, 0.75) students than those who did not identify with each of those respective groups. Mean disengagement scores were lower among female (PE −0.47, 95% CI −0.52, −0.42), Hispanic (PE −0.11, 95% CI −0.22, −0.01), and White (PE −0.10, 95% CI −0.19, 0.00) students and higher among Asian (PE 0.17, 95% CI 0.07, 0.27), Black or African American (PE 0.31, 95% CI 0.18, 0.44), bisexual (PE 0.54, 95% CI 0.41, 0.66), and gay or lesbian (PE 0.23, 95% CI 0.11, 0.35) students than those who did not identify with each of those respective groups. From 2019–2021, mean exhaustion and disengagement scores were relatively stable or improved across nearly all intersectional groups. ConclusionsMale, Asian, Black or African American, and sexual minority students had a higher risk of burnout, while female, Hispanic, White, and heterosexual or straight students had a lower risk of burnout. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email