CoChat
Sign Up
Cited by 38Technology Use by Older AdultsSpeech and dialogue systemsIntelligent Tutoring Systems and Adaptive Learning

OK Google, Let's Learn: Using Voice User Interfaces for Informal Self-Regulated Learning of Health Topics among Younger and Older Adults

Open access
Desai, Smit, Chin, Jessie·2023·Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
DOI 10.1145/3544548.3581507
Open this paper in CoChat — free accountOpen PDF

Abstract

In this paper, we present Health Buddy, a voice agent integrated into commercially available Voice User Interfaces (VUIs) to support informal self-regulated learning (SRL) of health-related topics through multiple learning strategies and examine the efficacy of Health Buddy on learning outcomes for younger and older adults. We conducted a mixed-factorial-design experiment with 26 younger and 25 older adults, assigned to three SRL strategies (within-subjects): monologue, dialogue-based scaffolding building, and conceptual diagramming. We found that while younger adults benefit more from scaffolding building and conceptual diagramming, both younger and older adults showed equivalent learning outcomes. Furthermore, interaction fluency (operationalized by the number of conversational breakdowns) was associated with learning outcomes regardless of age. While older adults did not experience less fluent conversations, interaction fluency affected their technology acceptance toward VUIs more than younger ones. Our study discusses age-related learning differences and has implications for designing VUI-based learning programs for older adults.

Integrity signals

Registry

Crossref · 10.1145/3544548.3581507

Cites

113 of 113 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

25 of 96 cited sources checked · 7 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Previous research has shown that younger and older adults tend to learn in different ways when using technology.

    “Previous research has shown that younger and older adults tend to learn in diferent ways when using technology [18,22,55,115].”
    Quote in the paperSubstantial claim1 of 4 sources checked
    • cites [18] 18. Li-Kuang Chen, Young Sek Kim, Paul Moon, and Sharan B. Merriam. 2008. A Review and Critique of the Portrayal of Older Adult Learners in Adult Education Journals, 1980-2006. Adult Education Quarterly 59, 1: 3-21. https://doi.org/10. 1177/0741713608325169 source
    • cites [22] 22. Ching Ju Chiu, Yi Han Hu, Dai Chan Lin, Fang Yu Chang, Cheng Sian Chang, and Cheng Fung Lai. 2016. The attitudes, impact, and learning needs of older adults using apps on touchscreen mobile devices: Results from a pilot study. Computers in Human Behavior 63: 189-197. https://doi.org/10.1016/j.chb.2016.05.020 source
    • cites [55] 55. Bora Jin, Junghwan Kim, and Lisa M. Baumgartner. 2019. Informal Learning of Older Adults in Using Mobile Devices: A Review of the Literature. Adult Education Quarterly 69, 2: 120-141. https://doi.org/10.1177/0741713619834726 source
    • cites [115] 115. Bo Xie, Ivan Watkins, Jen Golbeck, and Man Huang. 2012. Understanding and Changing Older Adults' Perceptions and Learning of Social Media. Educational Gerontology 38, 4: 282-296. https://doi.org/10.1080/03601277.2010.544580 source

      The Literature Review compares younger and older adults’ attitudes toward computers and discusses older adults’ learning of technology, but it does not compare how the two age groups learn. The Methods describe participants who were all older adults. Searches for age-related differences in technology learning found no passage establishing the claimed difference. (LLM, conf=0.75)

      Experimental — accuracy not yet measured
  • Premisecitation

    There is a comparatively significant body of work studying how VUIs could provide companionship to older adults.

    “However, there is a comparatively signifcant body of work (e.g., [57,87]) studying how VUIs could provide companionship to older adults.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [57] 57. Sunyoung Kim and Abhishek Choudhury. 2021. Exploring older adults' percep- tion and use of smart speaker-based voice assistants: A longitudinal study. Com- puters in Human Behavior 124: 106914. https://doi.org/10.1016/j.chb.2021.106914 source
    • cites [87] 87. Alisha Pradhan, Leah Findlater, and Amanda Lazar. 2019. "Phantom Friend" or "Just a Box with Information": Personifcation and Ontological Categorization of Smart Speaker-based Voice Assistants by Older Adults. Proceedings of the ACM on Human-Computer Interaction 3, CSCW: 214:1-214:21. https://doi.org/ 10.1145/3359316 source
  • Interpretationown data

    VUIs offer a low entry point by levelling the learning curve, as indicated in other VUI studies.

    “This result suggests that VUIs ofer a low entry point by levelling the learning curve, as indicated in other VUI studies [12,84,88].”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites [12] 12. T.W. Bickmore, L. Caruso, K. Clough-Gorr, and T. Heeren. 2005. "It's just like you talk to a friend" relational agents for older adults. Interacting with Computers 17, 6: 711-735. https://doi.org/10.1016/j.intcom.2005.09.002 source
    • cites [84] 84. F. Portet, M. Vacher, C. Golanski, C. Roux, and B. Meillon. 2013. Design and evaluation of a smart home voice interface for the elderly: Acceptability and objection aspects. Personal and Ubiquitous Computing 17, 1: 127-144. https: //doi.org/10.1007/s00779-011-0470-5 source
    • cites [88] 88. Alisha Pradhan, Amanda Lazar, and Leah Findlater. 2020. Use of Intelligent Voice Assistants by Older Adults with Low Technology Use. ACM Transactions on Computer-Human Interaction 27, 4: 31:1-31:27. https://doi.org/10.1145/3373759 source
  • Premisecitation

    VUIs are not conversational in nature.

    “As extensively noted in VUI literature, VUIs are not “conversational” in nature [24,25,31,36,45,83].”
    Quote in the paperNot a substantial claim3 of 6 sources checked
    • cites [24] 24. Leigh Clark, Phillip Doyle, Diego Garaialde, Emer Gilmartin, Stephan Schlögl, Jens Edlund, Matthew Aylett, João Cabral, Cosmin Munteanu, and Benjamin Cowan. 2019. The State of Speech in HCI: Trends, Themes and Challenges. Interacting with Computers 31, 4: 349-371. https://doi.org/10.1093/iwc/iwz016 source

      Section 2 (p. 2) says current system dialogues are stiff and questions how closely they simulate human conversation, supporting a qualified version of the claim. It does not say VUIs are categorically non-conversational; it also describes spoken dialogue systems as allowing two-way interaction. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [25] 25. Leigh Clark, Nadia Pantidi, Orla Cooney, Philip Doyle, Diego Garaialde, Justin Edwards, Brendan Spillane, Emer Gilmartin, Christine Murad, Cosmin Munteanu, Vincent Wade, and Benjamin R. Cowan. 2019. What Makes a Good Conversation?: Challenges in Designing Truly Conversational Agents. In Pro- ceedings of the 2019 CHI Conference on Human Factors in Computing Systems (CHI '19), 475:1-475:12. https://doi.org/10.1145/3290605.3300705 source
    • cites [31] 31. Benjamin R. Cowan, Nadia Pantidi, David Coyle, Kellie Morrissey, Peter Clarke, Sara Al-Shehri, David Earley, and Natasha Bandeira. 2017. "What Can I Help You with?": Infrequent Users' Experiences of Intelligent Personal Assistants. In Proceedings of the 19th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '17), 43:1-43:12. https: //doi.org/10.1145/3098279.3098539 source

      In “Personality and Voice” (page unavailable), the paper reports that users’ interactions with Siri were more about giving commands than having a conversation. This supports the point for Siri, but the citing sentence extends it to VUIs generally, which this passage does not establish. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [36] 36. Smit Desai and Michael Twidale. 2022. Is Alexa like a computer? A search engine? A friend? A silly child? Yes. In 4th Conference on Conversational User Interfaces (CUI 2022), 1-4. https://doi.org/10.1145/3543829.3544535 source
    • cites [45] 45. Joel E. Fischer, Stuart Reeves, Martin Porcheron, and Rein Ove Sikveland. 2019. Progressivity for voice interface design. In Proceedings of the 1st International Conference on Conversational User Interfaces (CUI '19), 1-8. https://doi.org/10. 1145/3342775.3342788 source
    • cites [83] 83. Martin Porcheron, Joel E. Fischer, Stuart Reeves, and Sarah Sharples. 2018. Voice Interfaces in Everyday Life. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI '18), 640:1-640:12. https://doi.org/ 10.1145/3173574.3174214 source
  • Premisecitation

    The majority of SRL literature with older adults was conducted using GUIs (text reading).

    “Since the majority of SRL literature with older adults was conducted using GUIs (i.e., text reading, e.g., [26,91]), our fndings have unique contributions to understanding age diferences in self-regulated learning with VUIs.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [26] 26. Rachel Clark, Michael Freedberg, Eliot Hazeltine, and Michelle W. Voss. 2015. Are There Age-Related Diferences in the Ability to Learn Confgural Responses? PLoS ONE 10, 8: e0137260. https://doi.org/10.1371/journal.pone.0137260 source

      The Materials and Methods (Participants; Configural learning task, page unavailable) describe older adults learning from faces displayed on a computer screen. But the task involved face–finger mappings and keypresses, not text reading or self-regulated learning. The paper also does not establish that most older-adult SRL studies used GUIs. (LLM, conf=0.78)

      Experimental — accuracy not yet measured
    • cites [91] 91. Jodi Price, Christopher Hertzog, and John Dunlosky. 2010. Self-Regulated Learn- ing in Younger and Older Adults: Does Aging Afect Metacognitive Control? Neuropsychology, development, and cognition. Section B, Aging, neuropsychol- ogy and cognition 17, 3: 329-359. https://doi.org/10.1080/13825580903287941 source

      Experiment 2’s Method and Procedure describe a computer-screen interface where older adults selected and studied Spanish–English vocabulary pairs, supporting this paper as an example of screen-based self-regulated learning. Experiment 1’s Method instead describes studying vocabulary pairs on note cards. Neither experiment involved reading prose texts, and this paper does not establish that a majority of the older-adult SRL literature used GUIs. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • May combine two claims. 55
  • Classified as probably not a substantial claim. 29
  • The quote only partly states this claim. 1

  • May combine two claims.

    Premisecitation

    Compared to GUIs, VUIs offer a more accessible, easily learnable, and usable interaction modality and are positively perceived by older adults.

    “Compared to GUIs, VUIs ofer a more accessible, easily learnable, and usable interaction modality [12,84,88] and, therefore, are positively perceived by older adults [102,117].”
    Quote in the paperSubstantial claim1 of 5 sources checked
    • cites [12] 12. T.W. Bickmore, L. Caruso, K. Clough-Gorr, and T. Heeren. 2005. "It's just like you talk to a friend" relational agents for older adults. Interacting with Computers 17, 6: 711-735. https://doi.org/10.1016/j.intcom.2005.09.002 source
    • cites [84] 84. F. Portet, M. Vacher, C. Golanski, C. Roux, and B. Meillon. 2013. Design and evaluation of a smart home voice interface for the elderly: Acceptability and objection aspects. Personal and Ubiquitous Computing 17, 1: 127-144. https: //doi.org/10.1007/s00779-011-0470-5 source
    • cites [88] 88. Alisha Pradhan, Amanda Lazar, and Leah Findlater. 2020. Use of Intelligent Voice Assistants by Older Adults with Low Technology Use. ACM Transactions on Computer-Human Interaction 27, 4: 31:1-31:27. https://doi.org/10.1145/3373759 source
    • cites [102] 102. Brodrick Stigall, Jenny Waycott, Steven Baker, and Kelly Caine. 2019. Older Adults' Perception and Use of Voice User Interfaces: A Preliminary Review of the Computing Literature. In Proceedings of the 31st Australian Conference on Human-Computer-Interaction (OZCHI'19), 423-427. https://doi.org/10.1145/ 3369457.3369506 source
    • cites [117] 117. Randall Ziman and Greg Walsh. 2018. Factors Afecting Seniors' Perceptions of Voice-enabled User Interfaces. In Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems (CHI EA '18), 1-6. https://doi.org/10. 1145/3170427.3188575 source

  • May combine two claims.

    Premisecitation

    VUIs are comparatively more accessible to older adults and people with disabilities due to their speech-based nature.

    “Due to the speech-based nature of VUIs, they are comparatively more accessible to older adults [87,88,107] and people with disabilities [102].”
    Quote in the paperSubstantial claim2 of 4 sources checked
    • cites [87] 87. Alisha Pradhan, Leah Findlater, and Amanda Lazar. 2019. "Phantom Friend" or "Just a Box with Information": Personifcation and Ontological Categorization of Smart Speaker-based Voice Assistants by Older Adults. Proceedings of the ACM on Human-Computer Interaction 3, CSCW: 214:1-214:21. https://doi.org/ 10.1145/3359316 source

      Sections 2.3 and 3.1 describe a study of older adults using voice assistants, but do not say that speech-based interfaces are more accessible to them than other interfaces. A search for accessibility, ease of use, and speech found no passage establishing that comparison. The separate claim about people with disabilities is cited to [102], not this source. (LLM, conf=0.74)

      Experimental — accuracy not yet measured
    • cites [88] 88. Alisha Pradhan, Amanda Lazar, and Leah Findlater. 2020. Use of Intelligent Voice Assistants by Older Adults with Low Technology Use. ACM Transactions on Computer-Human Interaction 27, 4: 31:1-31:27. https://doi.org/10.1145/3373759 source

      Section 2.2 says VUIs have the potential to lower older adults’ barriers to using digital technology compared with traditional interfaces, and Section 6.4 reports increased confidence associated with the ease of voice-based interaction. The paper therefore supports the accessibility rationale, but the citing sentence states a general comparative advantage as fact where the paper presents it as a possibility and studies older adults with low technology use. In the citing sentence, the separate disability claim is attached to a different citation. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [107] 107. Michel Vacher, Sybille Cafau, François Portet, Brigitte Meillon, Camille Roux, Elena Elias, Benjamin Lecouteux, and Pedro Chahuara. 2015. Evaluation of a Context-Aware Voice Interface for Ambient Assisted Living: Qualitative User Study vs. Quantitative System Evaluation. ACM Transactions on Accessible Computing 7, 2: 5:1-5:36. https://doi.org/10.1145/2738047 source
    • cites [102] 102. Brodrick Stigall, Jenny Waycott, Steven Baker, and Kelly Caine. 2019. Older Adults' Perception and Use of Voice User Interfaces: A Preliminary Review of the Computing Literature. In Proceedings of the 31st Australian Conference on Human-Computer-Interaction (OZCHI'19), 423-427. https://doi.org/10.1145/ 3369457.3369506 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Older adults' use of technology has been linked to improved physical and mental health, increased quality of life and autonomy, and decreased loneliness.

    “Older adults' use of technology has been linked to improved physical and mental health, increased quality of life and autonomy, and decreased loneliness [27,32,72,112].”
    Quote in the paperNot a substantial claim1 of 4 sources checked
    • cites [27] 27. Michael J. Cody, Deborah Dunn, Shari Hoppin, and Pamela Wendt. 1999. Silver surfers: Training and evaluating internet use among older adult learners. Communication Education 48, 4: 269-286. https://doi.org/10.1080/ 03634529909379178 source
    • cites [32] 32. Sara J. Czaja and Chin Chin Lee. 2006. The impact of aging on access to tech- nology. Universal Access in the Information Society 5, 4: 341. https://doi.org/10. 1007/s10209-006-0060-x source
    • cites [72] 72. Alex Mihailidis, Brent Carmichael, and Jennifer Boger. 2004. The use of computer vision in an intelligent environment to support aging-in-place, safety, and independence in the home. IEEE transactions on information technology in biomedicine: a publication of the IEEE Engineering in Medicine and Biology Society 8, 3: 238-247. https://doi.org/10.1109/titb.2004.834386 source
    • cites [112] 112. Shengzhi Wang, Khalisa Bolling, Wenlin Mao, Jennifer Reichstadt, Dilip Jeste, Ho-Cheol Kim, and Camille Nebeker. 2019. Technology to Support Aging in Place: Older Adults' Perspectives. Healthcare 7, 2: 60. https://doi.org/10.3390/ healthcare7020060 source

      The Introduction (page not provided) says these technologies have potential to support independent living, improve physical and mental health, and increase quality of life. That supports much of the claim, including autonomy in the sense of independent living, but our copy does not link technology use to decreased loneliness. (LLM, conf=0.87)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Older adults are often underrepresented in learning studies, possibly due to stereotypes about their attitudes toward new technology and learning.

    “Furthermore, older adults are often underrepresented in learning studies, possibly due to stereotypes about their attitudes toward new technology and learning [18,32,50,74,75,110].”
    Quote in the paperSubstantial claim1 of 6 sources checked
    • cites [18] 18. Li-Kuang Chen, Young Sek Kim, Paul Moon, and Sharan B. Merriam. 2008. A Review and Critique of the Portrayal of Older Adult Learners in Adult Education Journals, 1980-2006. Adult Education Quarterly 59, 1: 3-21. https://doi.org/10. 1177/0741713608325169 source
    • cites [32] 32. Sara J. Czaja and Chin Chin Lee. 2006. The impact of aging on access to tech- nology. Universal Access in the Information Society 5, 4: 341. https://doi.org/10. 1007/s10209-006-0060-x source
    • cites [50] 50. Melinda Heinz, Peter Martin, Jennifer A. Margrett, Mary Yearns, Warren Franke, Hen-I. Yang, Johnny Wong, and Carl K. Chang. 2013. Perceptions of technology among older adults. Journal of Gerontological Nursing 39, 1: 42-51. https://doi. org/10.3928/00989134-20121204-04 source
    • cites [74] 74. J. Morgan Morris. 1994. User interface design for older adults. Interacting with Computers 6, 4: 373-393. https://doi.org/10.1016/0953-5438(94)90009-4 source
    • cites [75] 75. M.E. Morris, B. Adair, K. Miller, E. Ozanne, R. Hansen, A.J. Pearce, N. Santamaria, L. Viega, M. Long, and C.M. Said. 2013. Smart-home technologies to assist older people to live well at home. Journal of Aging Science 1, 1: 1-9. source
    • cites [110] 110. Hartmut Wandke, Michael Sengpiel, and Malte Sönksen. 2012. Myths about older people's use of information and communication technology. Gerontology 58, 6: 564-570. https://doi.org/10.1159/000339104 source

      The Introduction (p. 2) discusses stereotypes about older adults’ technology use and learning, and the section “Older People Lack the Physical Capabilities to Use ICT” (p. 5) describes a teacher whose assumption changed how he taught older users. The paper does not say that older adults are underrepresented in learning studies or that stereotypes cause such underrepresentation. (LLM, conf=0.78)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Older adults are believed to be a targeted market for smart speakers because they provide a lower entry barrier than other input modalities and can be easily incorporated into daily lives.

    “In particular, older adults are believed to be a targeted market for smart speakers [101] as they provide a lower entry barrier than other input modalities and can be easily incorporated into daily lives [58,117].”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites [101] 101. Jaisie Sin, Cosmin Munteanu, Numrita Ramanand, and Yi Rong Tan. 2021. VUI Infuencers: How the Media Portrays Voice User Interfaces for Older Adults. In CUI 2021 -3rd Conference on Conversational User Interfaces (CUI '21), 1-13. https://doi.org/10.1145/3469595.3469603 source
    • cites [58] 58. Jarosław Kowalski, Anna Jaskulska, Kinga Skorupska, Katarzyna Abramczuk, Cezary Biele, Wiesław Kopeć, and Krzysztof Marasek. 2019. Older Adults and Voice Interaction: A Pilot Study with Google Home. Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems: 1-6. https: //doi.org/10.1145/3290607.3312973 source
    • cites [117] 117. Randall Ziman and Greg Walsh. 2018. Factors Afecting Seniors' Perceptions of Voice-enabled User Interfaces. In Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems (CHI EA '18), 1-6. https://doi.org/10. 1145/3170427.3188575 source

  • May combine two claims.

    Premisecitation

    Prior research indicates that older adults and infrequent VUI users tend to have difficulties knowing what to say and how to operate the device, resulting in frustration and discontinued use.

    “This design decision was especially useful for older adults and infrequent users of VUIs, as prior research indicates that these users tend to have difculties knowing what to say and how to operate the device [31,57], resulting in frustration and discontinued use.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [31] 31. Benjamin R. Cowan, Nadia Pantidi, David Coyle, Kellie Morrissey, Peter Clarke, Sara Al-Shehri, David Earley, and Natasha Bandeira. 2017. "What Can I Help You with?": Infrequent Users' Experiences of Intelligent Personal Assistants. In Proceedings of the 19th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '17), 43:1-43:12. https: //doi.org/10.1145/3098279.3098539 source

      The User Accent and Speech Recognition section reports that infrequent users had to think carefully about how to phrase requests, and Interruptability and Integration reports frustration with having to use the screen during hands-free operation. The paper mentions an older-adult study in People’s Experiences of IPAs, but does not establish that older adults generally have these difficulties or that uncertainty about what to say and how to operate the device causes discontinued use. Page numbers are unavailable in our copy. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [57] 57. Sunyoung Kim and Abhishek Choudhury. 2021. Exploring older adults' percep- tion and use of smart speaker-based voice assistants: A longitudinal study. Com- puters in Human Behavior 124: 106914. https://doi.org/10.1016/j.chb.2021.106914 source

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Older adults tend to perform significantly worse in recall-based tasks and take more time to learn the same amount of information compared to younger adults.

    “It is well-established in the learning literature that older adults tend to perform signifcantly poorly in recall-based tasks and take more time to learn the same amounts of information compared to younger adults [49,94].”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [49] 49. Joellen T. Hartley, Carrie C. Stojack, Thomas J. Mushaney, T. A. Kiku Annon, and Diane W. Lee. 1994. Reading speed and prose memory in older and younger adults. Psychology and Aging 9: 216-223. https://doi.org/10.1037/0882-7974.9.2. CHI '23, April 23-28, 2023, Hamburg, Germany 216 source
    • cites [94] 94. Stephen Rhodes, Nathaniel R. Greene, and Moshe Naveh-Benjamin. 2019. Age- related diferences in recall and recognition: a meta-analysis. Psychonomic Bulletin & Review 26, 5: 1529-1547. https://doi.org/10.3758/s13423-019-01649-y source

      The Abstract reports that older adults perform more poorly than younger adults on free-recall tests, with a substantial age difference in the meta-analysis. The paper does not establish that older adults take more time to learn the same amount of information; its Description of the included studies discusses learning conditions, not time needed to reach equal learning. Page numbers are unavailable in our copy. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    A cued recall protocol is common in studies involving text information learning with older adults in short learning sessions and is used as a proxy for learning.

    “Such a recall-based protocol is common in studies involving text information learning [19,40,91] with older adults in short learning sessions.”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [19] 19. Jessie Chin, Darcie D Moeller, Jessica Johnson, Elise A G Duwe, James F Graum- lich, Michael D Murray, and Daniel G Morrow. 2018. A Multi-faceted Approach to Promote Comprehension of Online Health Information Among Older Adults. The Gerontologist 58, 4: 686-695. https://doi.org/10.1093/geront/gnw254 source

      The Methods (Procedure and Passage Memory Measures; page unavailable) describe older adults reading passages for no more than nine minutes each, then summarizing them and answering multiple-choice questions. This supports the use of recall-based measures in a short text-learning task, but the paper does not call its protocol cued recall or establish that such a protocol is common across studies. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [40] 40. John Dunlosky, Julie M. C. Baker, Katherine A. Rawson, and Christopher Hertzog. 2006. Does aging infuence people's metacomprehension? Efects of processing ease on judgments of text learning. Psychology and Aging 21, 2: 390-400. https: //doi.org/10.1037/0882-7974.21.2.390 source
    • cites [91] 91. Jodi Price, Christopher Hertzog, and John Dunlosky. 2010. Self-Regulated Learn- ing in Younger and Older Adults: Does Aging Afect Metacognitive Control? Neuropsychology, development, and cognition. Section B, Aging, neuropsychol- ogy and cognition 17, 3: 329-359. https://doi.org/10.1080/13825580903287941 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Previous VUI literature suggests key differences in how native and non-native speakers perceive and interact with VUIs.

    “Previous VUI literature suggests that there are key diferences in how native and non-native speakers perceive and interact with VUIs [31].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [31] 31. Benjamin R. Cowan, Nadia Pantidi, David Coyle, Kellie Morrissey, Peter Clarke, Sara Al-Shehri, David Earley, and Natasha Bandeira. 2017. "What Can I Help You with?": Infrequent Users' Experiences of Intelligent Personal Assistants. In Proceedings of the 19th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '17), 43:1-43:12. https: //doi.org/10.1145/3098279.3098539 source

      The ‘User Accent and Speech Recognition’ section reports participants’ perceptions that Siri struggles with local or non-native accents, supporting a difference in interaction associated with accent. It does not compare native and non-native speakers’ perceptions or interactions as groups, so the broader claim is not established. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Harrington et al., in a study with 30 Black older adults, concluded that cultural factors influence human-VUI conversations, and not accounting for them could result in low technology acceptance.

    “Harrington et al. [48], in a study with 30 Black older adults, concluded that cultural factors infuence human-VUI conversations, and not accounting for them could result in low technology acceptance.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [48] 48. Christina N. Harrington, Radhika Garg, Amanda Woodward, and Dimitri Williams. 2022. "It's Kind of Like Code-Switching": Black Older Adults' Ex- periences with a Voice Assistant for Health Information Seeking. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (CHI '22), 1-15. https://doi.org/10.1145/3491102.3501995 source

  • May combine two claims.

    Findingown data

    There was no interaction effect of age and interaction fluency on learning outcome (N=0.03, SE=0.48, t=0.06, p>.05), suggesting interaction fluency influenced the learning outcome of younger and older adults equally.

    “Results showed no interaction efect of age and interaction fuency on learning outcome (N=0.03,SE=0.48, t=0.06, p>.05), suggesting interaction fuency infuenced the learning outcome of younger and older adults equally.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    About 70% of participants indicated that they never or rarely used VUIs and about 90% reported having no or little familiarity with VUIs, yet this did not affect their learning performance.

    “In our study, about 70% of participants indicated that they never or rarely used VUIs and about 90% of them reported having no or little familiarity with VUIs-nevertheless, it did not afect their learning performance.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There were no main effects of age or strategy on learning outcome, but there was an interaction effect of age and learning strategy (B=2.10, SE=0.74, t=2.84, p<.05).

    “Results suggested no main efects of age or strategy on learning outcome.However, there was an interaction efect of age and learning strategy, suggesting that the diferences in the learning outcome in monologue and the other two strategies vary with age groups (B=2.10,SE=0.74, t=2.84, p<.05).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Learning outcome was not associated with technology acceptance, while age and the interaction term of age and interaction fluency were associated with technology acceptance (B=-0.33, SE=0.55, t=-2.51, p<.05; B=-0.3, SE=0.55, t=-2.29, p<.05).

    “As shown in Model 10 of Table 6, while learning outcome was not associated with technology acceptance, age and the interaction term of age and interaction fuency were associated with the technology acceptance, respectively (B=-0.33,SE=0.55, t=-2.51,p<.05; B=-0.3,SE=0.55, t=-2.29,p<.05).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    Participants experienced an average of 1.26 conversation breakdowns or failures in the study, and the error rate was 11.45%.

    “Overall, our participants experienced an average of 1.26 times conversation breakdowns or failures in the study.The error rate was 11.45%.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    In this study, older and younger adults learned equally well with VUIs, with no age differences in learning performance (total ideas recalled and proportion of important ideas recalled).

    “However, we did not observe any such diferences in our study.”
    • Also stated in: § 6.3 Design implications for older adults · plain
      “Older adults learn as well as younger adults with VUIs; however, older adults were more sensitive to interaction fuency while developing their technology acceptance compared to younger adults, impacting long-term adoption of VUIs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There was no effect of learning strategy on learning outcome for older adults (F(2,48)=0.94, p=0.40, partial eta squared=0.04), suggesting older adults learned an equivalent amount from the three strategies.

    “Results suggested there was no efect of the learning strategy on learning outcome for older adults (F(2,48)=0.94,p=0.40, partial eta 2 =0.04), suggesting older adults learned an equivalent amount of information from the three strategies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Although still low, younger adults used VUIs significantly more frequently and were more familiar with VUIs compared to older adults.

    “However, although still low, younger adults used VUIs signifcantly more frequently and were more familiar with VUIs compared to older adults.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There were no main effects of age and working memory or their interaction on interaction fluency (Model 5), and no main effects of age and verbal ability or their interaction on interaction fluency (Model 6).

    “Results suggested no main efects of age and working memory or its interaction efects on interaction fuency (Model 5).There were also no main efects of age and verbal ability or its interaction efects on interaction fuency (Model 6).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    There were no main effects of age and technology experience or their interaction on interaction fluency (Model 7), and no main effects of age and familiarity with VUIs or their interaction on interaction fluency (Model 8).

    “Results suggested no main efects of age and technology experience or its interaction efects on interaction fuency (Model 7).Similarly, there were no main efects of age and familiarity with VUIs or its interaction efects on interaction fuency (Model 8).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    The interaction effect of age and strategy remained after entering the interaction term of age and interaction fluency, suggesting that interaction fluency could not explain why older and younger adults performed differently across the strategies.

    “In addition, the interaction efect of age and strategy remained after entering the interaction term of age and interaction fuency, suggesting that interaction fuency could not explain why older and younger adults performed diferently across the strategies (see Model 4).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    Interaction fluency significantly explained learning outcome (B=-0.59, SE=0.27, t=-2.43, p<.05), with participants recalling more information if they experienced higher interaction fluency.

    “As shown in Model 3 (Table 4), while we still found the age x learning strategy interaction efect on learning outcome, interaction fuency signifcantly explained the learning outcome (B=-0.59,SE=0.27, t=-2.43,p<.05), which participants recalled more information if they experienced higher interaction fuency in the study.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    Younger adults showed higher technology acceptance of Health Buddy than older adults, while older adults were more sensitive to interaction fluency in determining their technology acceptance of Health Buddy.

    “While younger adults showed higher technology acceptance of Health Buddy than older adults, older adults were more sensitive to the interaction fuency to determine their technology acceptance of Health Buddy.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    The average total number of ideas recalled was 21.3 (SD=6.48), and the average proportion of important ideas recalled was 41.92% (SD=12.77%).

    “The average total number of ideas recalled is 21.3 (SD=6.48),and the average proportion of important ideas recalled is 41.92% (SD=12.77%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Technology familiarity and technology experience were highly correlated (r=0.71, p<.001), so their standardized scores were averaged into a technology experience composite score.

    “Due to high correlations between technology familiarity and technology experience (r=0.71,p<.001), we average the standardized scores of technology experience and familiarity as the technology experience composite score (TE).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    After controlling for prior knowledge, there were no differences in learning outcome between monologue and the two other strategies, or between scaffolding building and conceptual diagramming.

    “Results suggested that, after controlling the efects of prior knowledge, there were no diferences in learning outcome between the monologue (M) and the two other strategies (B=-0.44,SE=0.37, t=-1.19), or between scafolding building (SB) and conceptual diagraming (CD) (B=0.29,SE=0.42, p=0.67).”
    • Also stated in: § 5.2.1 Efects of Learning Strategies on Learning Performance. · plain
      “Participants learned an equivalent amount of information from the three learning strategies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Younger adults experienced an average of 1.17 (SE=0.19) and older adults 1.36 (SE=0.19) conversational breakdowns and failures in the study.

    “Younger adults experienced an average of 1.17 times (SE=0.19),and older adults experienced 1.36 times (SE=0.19) of conversational breakdowns and failures in the study.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • May combine two claims.

    Findingown data

    Approximately 70% of participants indicated that they hardly or never used VUIs, and about 90% reported no or less familiarity with VUIs.

    “Approximately 70% of the participants indicated that they hardly or never used VUIs, and about 90% of the participants reported no or less familiarity with VUIs.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    For younger adults, there was an effect of learning strategy on learning outcome (F(2,50)=5.28, p<.01, partial eta squared=0.17); younger adults learned more in scaffolding building and conceptual diagramming than in monologue (d=-1.52, SE=0.52, p<.01; d=-1.79, SE=0.64, p<.01).

    “For younger adults, there was an efect of learning strategy on learning outcome (F(2,50)=5.28,p<.01, partial eta 2 =0.17), that is younger adults learned more information in the scafolding building mode and conceptual diagramming mode than the monologue mode, respectively (d=-1.52,SE=0.52, p<.01; d=-1.79,SE=0.64, p<.01 ) .”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    After TE was entered into the model, age was not associated with technology acceptance, suggesting the associations between age and technology acceptance could be explained by the variance of TE.

    “After TE was entered into the model, age was not associated with technology acceptance, suggesting the associations between age and technology acceptance could be explained by the variance of TE.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Participants were often unaware of what went wrong with the interaction; the VUI's response 'Sorry, I could not understand' is insufficient to move the conversation forward.

    “Often, participants were simply unaware of what went wrong with the interaction-VUI's chimeric system response: “Sorry, I could not understand, ” is insufcient in moving the conversation forward.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Rarely, the VUI stopped responding due to unknown reasons (system error) and provided improper feedback (feedback error).

    “Rarely, the VUI would stop responding due to unknown reasons (system error) and provide improper feedback to the participants' requests (feedback error).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Most participants were able to recover from conversational breakdowns, with experimenters offering suggestions on what might have gone wrong and what to say to recover.

    “In the case of our study, while most of the participants were able to recover from conversational breakdowns, the experimenters would ofer suggestions on what might have gone wrong, and what the participants could say to recover the conversation if the frustrations persisted.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Health Buddy failed to recognize participants' speech and asked them to repeat their responses.

    “In these instances, Health Buddy (HB) failed to recognize the speech from the participants (P) and asked participants to repeat their responses.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The suggestion to provide an account of what went wrong was particularly helpful in this study.

    “We found this suggestion to be particularly helpful.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    During the study, older adults took too long to respond and were interrupted while still speaking.

    “During our study, we found instances of older adults taking too long to respond and being interrupted when they were still speaking.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Health information seeking can often lead to learning and can serve as a starting point for health information learning.

    “Although health information seeking and health information learning are two distinct activities that require diferent strategies, needs, and levels of engagement, health information seeking can often lead to learning and can serve as a starting point for health information learning [61,66].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [61] 61. Louise Limberg. 1998. Experiencing information seeking and learning A study of the interaction between two phenomena: summary. Svensk biblioteksforskning 1: 5-23.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [66] 66. Gary Marchionini. 1995. Information seeking in electronic environments. Cam- bridge University Press, USA. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Falkner et al. found conceptual diagramming to be the most successful self-regulated learning strategy to support computer science students learning autonomously.

    “Falkner et al. [44] found conceptual diagramming to be the most successful SRL strategy to support computer science students learning autonomously.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [44] 44. Katrina Falkner, Rebecca Vivian, and Nickolas J.G. Falkner. 2014. Identifying computer science self-regulated learning strategies. In Proceedings of the 2014 conference on Innovation & technology in computer science education (ITiCSE '14), 291-296. https://doi.org/10.1145/2591708.2591715 source

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Premisereasoning

    Building SRL agents in commercially available VUIs has the potential to tackle these weaknesses, given their accessible and scalable dissemination to diverse users.

    “Building SRL agents in commercially available VUIs has the potential to tackle these weaknesses, given their accessible and scalable dissemination to diverse users.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Scaffolding-building strategies have been deployed in artificial-intelligence-based pedagogical agents via natural language, but their applicability to commercial VUIs is currently limited to specific contexts.

    “These strategies have been deployed in artifcial intelligence-based pedagogical agents via natural language before [46], but their applicability to commercial VUIs is currently limited to specifc contexts.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [46] 46. Arthur Graesser and Danielle McNamara. 2010. Self-Regulated Learning in Learning Environments With Pedagogical Agents That Interact in Natural Language. Educational Psychologist 45, 4: 234-244. https://doi.org/10.1080/ 00461520.2010.515933 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Xu and Warschauer successfully built and deployed a storytelling conversational agent for children based on scaffolding-building strategies.

    “Most notably, Xu and Warschauer [116] successfully built and deployed a storytelling conversational agent for children based on scafoldingbuilding strategies.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [116] 116. Ying Xu and Mark Warschauer. 2019. Young Children's Reading and Learning with Conversational Agents. In Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems (CHI EA '19), 1-8. https://doi.org/10. 1145/3290607.3299035 source

  • May combine two claims.

    Premisecitation

    Computer-Based Learning Environments suffer from disadvantages like lack of affordability and inaccessibility, due to which their adoption is limited.

    “However, CBLEs sufer from disadvantages like lack of afordability and inaccessibility [30], due to which their adoption is limited.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [30] 30. David A Cook. 2007. Web-based learning: pros, cons and controversies. Clinical Medicine 7, 1: 37-42. https://doi.org/10.7861/clinmedicine.7-1-37 source

  • May combine two claims.

    Study design and conductown data

    The six health topics were selected to have similar difficulty and readability level, with a mean Flesch ease score of 54.16 (SD=5.94).

    “We selected health topics of similar difculty and readability level (calculated using textstat 0.7.3 Python package using the full text as input for each health topic in all modes; Mean Flesch ease score=54.16,SD=5.94)”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The lengths of the conversations were equivalent across the six health topics.

    “The lengths of the conversations were equivalent across six topics.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    The content of the health topics remained the same across all three modes of Health Buddy, and the conversation length was approximately the same.

    “Notably, the content of the health topics remained the same across all three modes of HB, and the same information was provided using the three diferent learning strategies.As a result, the conversation length was approximately the same.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    The three learning strategies were selected based on their feasibility for implementation in VUIs and their effectiveness in previous HCI literature on learning.

    “These strategies were selected based on their feasibility for implementation in VUIs [34] and their efectiveness in previous HCI literature on learning [44].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [34] 34. Smit Desai and Jessie Chin. 2020. An Explorative Analysis of the Feasibility of Implementing Metacognitive Strategies in Self-Regulated Learning with the Conversational Agents. Proceedings of the Human Factors and Ergonomics So- ciety Annual Meeting 64, 1: 495-499. https://doi.org/10.1177/1071181320641112 source
    • cites [44] 44. Katrina Falkner, Rebecca Vivian, and Nickolas J.G. Falkner. 2014. Identifying computer science self-regulated learning strategies. In Proceedings of the 2014 conference on Innovation & technology in computer science education (ITiCSE '14), 291-296. https://doi.org/10.1145/2591708.2591715 source

  • May combine two claims.

    Premisecitation

    In the United States, 75% of younger adults and 43.9% of older adults sought health information on the internet in 2018.

    “In the United States, 75% of younger adults [47], and 43.9% of older adults [53] sought health information on the internet in 2018.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [47] 47. Nicola J. Gray, Jonathan D. Klein, Peter R. Noyce, Tracy S. Sesselberg, and Judith A. Cantrill. 2005. Health information-seeking behaviour in adolescence: the place of the internet. Social Science & Medicine 60, 7: 1467-1478. https: //doi.org/10.1016/j.socscimed.2004.08.010 source
    • cites [53] 53. Li-Yu Hung, Jennifer G. Lyons, and Chung-Hsuen Wu. 2020. Health information technology use among older adults in the United States, 2009-2018. Current Medical Research and Opinion 36, 5: 789-797. https://doi.org/10.1080/03007995. 2020.1734782 source

  • May combine two claims.

    Premisecitation

    The monologue learning strategy minimizes distractions and exerts less cognitive load on the user because it requires minimum back-and-forth interaction.

    “Since the monologue learning strategy requires minimum back-and-forth interaction with HB, it minimizes distractions and exerts a less cognitive load on the user [52,67,73].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [52] 52. Bruce D. Homer, Jan L. Plass, and Linda Blake. 2008. The efects of video on cognitive load and social presence in multimedia-learning. Computers in Human Behavior 24, 3: 786-797. https://doi.org/10.1016/j.chb.2007.02.009 source
    • cites [67] 67. Richard E. Mayer and Roxana Moreno. 1998. A split-attention efect in mul- timedia learning: Evidence for dual processing systems in working memory. Journal of Educational Psychology 90: 312-320. https://doi.org/10.1037/0022- 0663.90.2.312 source
    • cites [73] 73. Roxana Moreno, Richard E. Mayer, Hiller A. Spires, and James C. Lester. 2001. The case for social agency in computer-based teaching: Do students learn more deeply when they interact with animated pedagogical agents? Cognition and Instruction 19: 177-213. https://doi.org/10.1207/S1532690XCI1902_02 source

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Unless there was a major conversational error and the participant could not continue, the experimenter did not interfere with the learning session.

    “Unless there was a major conversational error and the participant could not continue the interaction, the experimenter did not interfere with the learning session.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Two independent raters scored the recall, with Cohen's kappa greater than 0.85.

    “Two independent raters scored the recall (Cohen's kappa > 0.85).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Both younger and older adults performed equivalently well and effectively learned information using Health Buddy.

    “We found that both younger and older adults performed equivalently well and efectively learned information using Health Buddy.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductreasoning

    Three versions of Health Buddy (alpha, beta, and gamma) were designed to be identical in content but differ in the topics assigned to each learning strategy and the order of the strategies, to ensure the validity of the experiments.

    “To ensure the validity of our experiments, we designed three versions of HB (alpha, beta, and gamma) that were identical in content but difered in the topics assigned to each learning strategy and the order of the strategies.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    To keep learning performance limited to information given by Health Buddy, no points were given if participants mentioned relevant information about a topic that was not part of their interaction with Health Buddy.

    “To keep the learning performance limited to just the information given by Health Buddy, no points were given to the participants if they mentioned relevant information about a topic that was not part of their interaction with Health Buddy.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Google Assistant, along with Amazon Alexa, is the most popular and commercially adopted VUI.

    “We chose Google Assistant because along with Amazon Alexa, it is the most popular and commercially adopted VUI [13].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [13] 13. Bratten. As Smart Speakers Evolve, So Do Consumers. Comscore, Inc. Retrieved December 13, 2022 from https://www.comscore.com/Insights/Blog/As-Smart- Speakers-Evolve-So-Do-Consumers

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationown data

    Although there were no age differences in overall learning outcomes, younger adults took more advantage of scaffolding building and conceptual diagramming than older adults.

    “Although there were no age diferences in overall learning outcomes, younger adults took more advantage of scafolding building and conceptual diagramming than older adults.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The cognitive measures used are commonly used in learning studies, especially with older adults, as they are predictors of cognitive abilities important for knowledge acquisition for learning.

    “These measures are commonly used in learning studies (especially with older adults), as they are predictors of cognitive abilities, which is important for knowledge acquisition for learning [7].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [7] 7. Margaret E. Beier and Phillip L. Ackerman. 2005. Age, ability, and the role of prior knowledge on the acquisition of new domain knowledge: promising results in a real-world learning environment. Psychology and Aging 20, 2: 341-355. https://doi.org/10.1037/0882-7974.20.2.341 source

  • May combine two claims.

    Interpretationown data

    There was no age difference in interaction fluency with Health Buddy; older adults experienced the same number of minimum errors during interactions with Health Buddy as younger adults.

    “There was no age diference in interaction fuency with Health Buddy; older adults experienced the same number of minimum errors during the interactions with Health Buddy as younger adults.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Using mixed-effects modeling controls the variance of participants and topics.

    “By using the mixed-efects modeling, we can control the variance of participants (due to their diferences in capacities or background knowledge) and topics (due to the diferences in difculties or familiarities with the individual participant).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Younger adults took more advantage of scaffolding building and conceptual diagramming, while older adults performed roughly the same across the three strategies.

    “Hence, younger adults took more advantage of the scafolding building and conceptual diagraming, while older adults performed roughly the same across the three strategies.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Participants learned six topics assigned to three modes in a counterbalanced order, with two topics in each mode.

    “Participants learned six topics assigned to three modes (also operationalized as learning strategies) in a counterbalanced order (2 topics in each mode).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Under each topic, five most important ideas were previously identified by two researchers independently and reconciled through discussions.

    “Under each topic, there are fve most important ideas which were previously identifed by two researchers independently and reconciled through discussions.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    To maintain consistency, the number of quizzes in scaffolding-building mode and the number of diagrams in conceptual-diagramming mode remained the same, at 3 or 4 per health topic.

    “To maintain consistency in the interactions, the number of quizzes in SB mode and the diagrams in CD mode remained the same (3 or 4 for each health topic).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    During the cued recall session, the experimenter provided no additional hints or support to help the participant recall more or any information about the health topics.

    “During the cued recall session, the experimenter provided no additional hints or support to help the participant recall more or any information about the health topics.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    All three learning strategies resulted in good enough learning performances among older and younger adults, with proportions of recall of important ideas comparable with previous studies in reading comprehension.

    “All three learning strategies resulted in good enough learning performances among older and younger adults (i.e., their proportions of recall of important ideas were comparable with previous studies in reading comprehension (e.g., [41,114])).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [41] 41. Christopher P. Dwyer, Michael J. Hogan, and Ian Stewart. 2013. An examination of the efects of argument mapping on students' memory and comprehension performance. Thinking Skills and Creativity 8: 11-24. https://doi.org/10.1016/j. tsc.2012.12.002 source
    • cites [114] 114. Robert M. Wilson, Linda B. Gambrell, and Warren R. Pfeifer. 1985. The Efects of Retelling Upon Reading Comprehension and Recall of Text Information. The Journal of Educational Research 78, 4: 216-220. https://doi.org/10.1080/00220671. 1985.10885604 source

  • Classified as probably not a substantial claim.

    Premisecitation

    According to Merriam & Kee, formal and informal learning activities of older adults promote an active and engaged lifestyle that helps create and preserve community.

    “According to Merriam & Kee [70], “formal and informal learning activities of older adults promote an active and engaged lifestyle that helps create and preserve community.””
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [70] 70. Sharan B. Merriam and Youngwha Kee. 2014. Promoting Community Wellbeing: The Case for Lifelong Learning for Older Adults. Adult Education Quarterly 64, 2: 128-144. https://doi.org/10.1177/0741713613513633 source

  • May combine two claims.

    Study design and conductown data

    Although the sample included a relatively diverse group of younger adults, the older participants were mostly Caucasian.

    “Further, although our sample included a relatively diverse group of younger adults, the older participants were mostly Caucasian.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    One suggestion from Fischer et al. was to provide an 'account of what went wrong.'

    “One of the suggestions was to provide an “account of what went wrong.””
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [45] 45. Joel E. Fischer, Stuart Reeves, Martin Porcheron, and Rein Ove Sikveland. 2019. Progressivity for voice interface design. In Proceedings of the 1st International Conference on Conversational User Interfaces (CUI '19), 1-8. https://doi.org/10. 1145/3342775.3342788 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Myers et al. identified four categories of obstacles prevailing in all interactions with VUIs: unfamiliar intent, NLP error, feedback error, and system error.

    “As previously stated, Myers et al. [79] identifed four categories of obstacles prevailing in all interactions with VUIs: unfamiliar intent, NLP error, feedback error, and system error.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [79] 79. Chelsea Myers, Anushay Furqan, Jessica Nebolsky, Karina Caro, and Jichen Zhu. 2018. Patterns for How Users Overcome Obstacles in Voice User Interfaces. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI '18), 6:1-6:7. https://doi.org/10.1145/3173574.3173580 source

  • May combine two claims.

    Premisecitation

    Older adults take longer to respond and are more verbose than younger adults in discourse.

    “Older adults take longer to respond and are more verbose than younger adults in discourse [57].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [57] 57. Sunyoung Kim and Abhishek Choudhury. 2021. Exploring older adults' percep- tion and use of smart speaker-based voice assistants: A longitudinal study. Com- puters in Human Behavior 124: 106914. https://doi.org/10.1016/j.chb.2021.106914 source

  • May combine two claims.

    Study design and conductreasoning

    Due to the online nature of the study, participants required access to a laptop or tablet, so participants with comparatively higher technological literacy were enrolled.

    “due to the online nature of this study; participants were required to have access to a laptop or a tablet-consequently, participants with comparatively higher levels of technological literacy were enrolled.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Myers et al. found that NLP errors were the most frequent out of the four categories.

    “They found that NLP errors were the most frequent out of the four categories.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [79] 79. Chelsea Myers, Anushay Furqan, Jessica Nebolsky, Karina Caro, and Jichen Zhu. 2018. Patterns for How Users Overcome Obstacles in Voice User Interfaces. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI '18), 6:1-6:7. https://doi.org/10.1145/3173574.3173580 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    Data from study sessions that suffered from technical problems (internet connectivity, background noise) were excluded.

    “Eforts were made to counteract this limitation by not including data from study sessions that sufered from technical problems (e.g., internet connectivity issues, excessive background noise, etc.)”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Interpretationreasoning

    Greater sensitivity to interaction fluency impacts long-term adoption of VUIs for older adults.

    “impacting long-term adoption of VUIs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Previous work in older adults' use of VUIs found older adults were critically affected by functional errors and inclined to blame themselves rather than the device.

    “This fnding is consistent with previous work in older adults' use of VUIs [57], wherein older adults were critically afected by functional errors of the device and were inclined to (unfairly) blame themselves rather than the device for the poor performance.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [57] 57. Sunyoung Kim and Abhishek Choudhury. 2021. Exploring older adults' percep- tion and use of smart speaker-based voice assistants: A longitudinal study. Com- puters in Human Behavior 124: 106914. https://doi.org/10.1016/j.chb.2021.106914 source

  • May combine two claims.

    Premisecitation

    VUIs adopt a request-response mechanism and fail to grasp the subtlety of the logical model of human-human conversations.

    “They adopt a request-response mechanism and fail to grasp the subtlety of the logical model of human-human conversations [83].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [83] 83. Martin Porcheron, Joel E. Fischer, Stuart Reeves, and Sarah Sharples. 2018. Voice Interfaces in Everyday Life. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI '18), 640:1-640:12. https://doi.org/ 10.1145/3173574.3174214 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Fischer et al. devised suggestions helpful to designers to ascertain that VUI responses to errors could be fruitful for 'progressivity.'

    “Using conversation analysis, Fischer et al. [45] devised suggestions helpful to designers to ascertain that the VUI's responses to errors could be fruitful for 'progressivity.'”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [45] 45. Joel E. Fischer, Stuart Reeves, Martin Porcheron, and Rein Ove Sikveland. 2019. Progressivity for voice interface design. In Proceedings of the 1st International Conference on Conversational User Interfaces (CUI '19), 1-8. https://doi.org/10. 1145/3342775.3342788 source
76 of 76

  • The paper relies on this without support.

    There is no existing research on using commercially available VUIs for self-regulated learning of health topics in informal settings.

    “While there is no existing research on using commercially available VUIs for self-regulated learning of health topics in informal settings”

  • The paper relies on this without support.

    There is no prior work on utilizing smart speakers for self-regulated learning among older adults.

    “Furthermore, there is no prior work on utilizing smart speakers for self-regulated learning among older adults.”

  • The paper relies on this without support.

    This is the first study investigating self-regulated informal learning of health topics using commercially available VUIs.

    “To the best of our knowledge, this is the frst study investigating self-regulated informal learning of health topics using commercially available VUIs.”

  • The paper relies on this without support.

    Early evidence in the literature suggests that VUIs have the potential to alleviate the lives of older adults and support aging in place.

    “Early evidence in the literature suggests that VUIs have the potential to alleviate the lives of older adults and support aging in place.”

  • The paper relies on this without support.

    The health information from MedlinePlus, Mayo Clinic, and American Cancer Society is credible and accurate for the study's informal learning content.

    “We introduced participants to learn six health topics using conversation scripts based on the materials from credible health information websites, including MedlinePlus, Mayo Clinic, and American Cancer Society.”

  • The paper relies on this without support.

    The interaction fluency score, operationalized as the maximum number of conversational breakdowns minus the actual number of breakdowns, is a valid measure of interaction fluency.

    “Interaction fuency is operationalized by the reverse grading of the number of breakdowns or failures in the conversations between participants and Health Buddy.We used the max number of conversational breakdowns or failures (N=6) minus the actual number of conversational breakdowns or failures as the interaction fuency score.”

  • The paper relies on this without support.

    No research has investigated how voice user interfaces could be deployed to provide learning solutions among older adults.

    “Currently, there is some research investigating how voice user interfaces could be deployed to provide learning solutions to children and younger adults-as previously mentioned in the related work section of our paper-but no research investigating the same among older adults.”

  • The paper relies on this without support.

    There is a major population gap in learning literature, which has traditionally prioritized younger adults and sidelined developing pedagogical agents for older adults due to myths surrounding older adults' attitudes towards technology and learning.

    “exposes a major population gap in learning literature which has traditionally prioritized younger adults and sidelined developing pedagogical agents for older adults due to myths surrounding their attitudes towards technology and learning.”

  • The paper relies on this without support.

    VUIs' inability to recover from errors is a major obstacle inhibiting natural conversation.

    “One major obstacle inhibiting natural conversation is VUIs' inability to recover from errors.”
9 of 9

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • ConsistentF(2, 48) = 0.94reported p = .4 · recomputed p = .398

    “F(2,48)=0.94,p = .40”

  • ConsistentF(2, 50) = 5.28reported p < .01 · recomputed p = .008

    “F(2,50)=5.28,p < .01”

  • Consistentt(49) = -0.7reported p = .49 · recomputed p = .487

    “t(49)=-0.70,p = .49”

  • ConsistentF(5, 45) = 3.81reported p < .01 · recomputed p = .006

    “F(5,45)=3.81,p < .01”

  • Consistentt(49) = 0.75reported p = .46 · recomputed p = .457

    “t(49)=0.75, p = .46”

5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Fluency coefficient points opposite to the claimed learning effect

    Major

    Higher interaction-fluency scores mean fewer breakdowns, but the reported negative coefficient associates higher scores with lower recall. The stated conclusion says higher fluency is associated with recalling more information.

  • Age effect on technology acceptance has an incompatible t statistic

    Major

    In Model 10, the age coefficient divided by its reported standard error is −0.60, not −2.51. The reported significance of this age effect therefore is not supported by the displayed coefficient and standard error.

    Reported -2.51 · recomputed -0.6

  • The non-native English speaker is assigned to the wrong age group

    Minor

    The participant description identifies the sole non-native English speaker as a younger adult. The demographic appendix instead lists an older adult whose native language is Chinese.

3 of 3

Cites

Flagged first, then by citations

Research works printed but not linked: 18

  1. 10.1093/geronb · DOI
  2. 10.1093/geront · DOI
  3. 10.1093/geront · DOI
  4. 10.1093/geront · DOI
  5. 10.1093/iwc · DOI
  6. Ruth B Ekstrom, John W French, Harry H Harman, and Diran Dermen. 1976. Manual for kit of factor-referenced cognitive tests: 1976. Education Testing Service, Princeton N.J.
  7. 10.1155/2014 · DOI
  8. 10.1093/jamia · DOI
  9. Michael Pressley and Christine B. McCormick. 1997. Advanced Educational Psychology for Educators, Researchers, and Policymakers. Allyn & Bacon, New York, NY.
  10. 10.1093/geronb · DOI
  11. David Wechsler and Psychological Corporation. 1997. WAIS-III: administration and scoring manual: Wechsler Adult Intelligence Scale. Psychological Corporation, San Antonio, Tex.
  12. Barry J. Zimmerman and Dale H. Schunk. 2008. Motivation: An essential dimension of self-regulated learning. In Motivation and self-regulated learning: Theory, research, and applications. Lawrence Erlbaum Associates Publishers, Mahwah, NJ, US, 1–30.
  13. 2021. Sleep: The healthy habit that promotes weight loss - Mayo Clinic. Retrieved December 13, 2022 from https://web.archive.org/web/20210617222710/https://www.mayoclinic.org/healthy-lifestyle/adult-health/in-depth/sleep-promotes-weight-loss/art-20270122
  14. Get the facts about what causes cancer. Mayo Clinic. Retrieved December 13, 2022 from https://www.mayoclinic.org/diseases-conditions/cancer/in-depth/cancer-causes/art-20044714
  15. LDL Cholesterol: What It Is & How to Lower It. Cleveland Clinic. Retrieved December 13, 2022 from https://my.clevelandclinic.org/health/articles/24391-ldl-cholesterol
  16. Vitamin B-12. Mayo Clinic. Retrieved December 13, 2022 from https://www.mayoclinic.org/drugs-supplements-vitamin-b12/art-20363663
  17. What new Alzheimer's treatments are on the horizon? Mayo Clinic. Retrieved December 13, 2022 from https://www.mayoclinic.org/diseases-conditions/alzheimers-disease/in-depth/alzheimers-treatments/art-20047780
  18. Alexa in Education. Amazon Web Services, Inc. Retrieved August 27, 2021 from https://aws.amazon.com/education/alexa-edu/

Cited by

Sorted by citations

Related

Authors
Desai, S., & Chin, J.
Affiliations
University of Illinois Urbana-Champaign (US)
ORCID
Venue
Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
Publisher
ACM
Published
2023-04-19
Pages
1-21
DOI
10.1145/3544548.3581507
OpenAlex
W4366590507
Semantic Scholar
0a6403e799985e3c5c838dbd73d61b946f75090b
Cited by
38
References
99
Citations by year
2026: 5 · 2025: 19 · 2024: 9 · 2023: 5
FWCI
34.70
Type
conference paper
Language
en
Open access
Licence
www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations
CHI Conference on Human Factors in Computing Systems (publishedVersion)
Topics
  • Social Sciences › Demography › Technology Use by Older Adults
  • Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
  • Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
Keywords
voice user interface, voice assistants, health education, older adults, age differences in learning, scaffolding, technology acceptance, chatbots

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.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Cited sources are checked by gpt-6-sol.

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