Adapting Voice Assistant Technology for Older Adults: A Comprehensive Study on Usability, Learning Patterns, and Acceptance

Abstract

This study investigates the integration, usability, and learning patterns associated with voice assistant technology among older adults, focusing on the “Amazon Echo Show 10, 3rd generation” as a case study. Conducted with 32 participants aged 55 and above in senior and complementary households, this research employs a mixed-method approach, incorporating qualitative interviews and quantitative voice command logging over a twelve-week period. Our findings reveal a high level of learnability and usability of the voice assistant, with 90% of participants finding the device easy to learn and use. The study further explores the patterns of voice assistant use, highlighting a preference for listening to music and seeking information, predominantly on weekends. Despite initial reservations, participants reported a high satisfaction level, with most not feeling monitored by the device. Key recommendations for manufacturers include prioritizing the design and user experience to cater to older adults’ needs, aiming to enhance their digital inclusion and participation. This study contributes to the human–computer interaction (HCI) field by providing insights into older adults’ interactions with voice assistant technology, emphasizing the importance of designing accessible and user-friendly digital solutions for the aging population.

1. Introduction

Voice-based user interfaces provide an intuitive and accessible way to interact with technology using natural language. They enable hands-free operation, eliminating the need for fine motor skills or navigating complex menus [1,2,3]. In recent years, these interfaces have been increasingly integrated into various devices, ranging from simple input–output functions, such as automatic speech transcription or conversion of text to speech, to more advanced applications. Well-known examples include systems that interpret spoken user requests and execute corresponding actions, such as Siri, Google Assistant, Alexa, and Cortana. This combination of user interface and action execution is commonly referred to as a voice assistant (VA). This article investigates devices that rely on voice-based interaction with an integrated voice assistant—referred to as voice controlled devices (VCDs).

Despite the widespread adoption of VA globally, with an estimated 8.4 billion devices expected in use by 2024 [4], significant gaps remain in understanding how specific user groups, such as older adults, interact with these technologies. Older adults often face unique challenges due to age-related impairments, including reduced fine motor skills, cognitive limitations, and sensory deficits [5,6,7,8,9,10]. However, this demographic could benefit significantly from VCDs, as these devices offer more straightforward and intuitive operation than touchscreens or peripheral controls [11,12]. Unfortunately, many digital devices and applications fail to account for the specific needs of older users, often leading to frustration and feelings of being overwhelmed [13]. As noted by Sayago et al. [14], Kim [11], and Jakob [15], there is a pressing need for deeper insights into the learnability, usability, and accessibility of VAs among older adults. Moreover, most existing research relies on short-term studies. In contrast, longer-term field studies are essential to capturing the full spectrum of older adults’ perceptions, preferences, and usage behaviours in their home environments [8,16,17].

Key factors influencing the adoption of VCDs among older adults include usability [12,18,19,20,21] and acceptance [1,19,22,23]. This paper investigates these factors through a field study conducted in households of individuals aged 55+, using the “Amazon Echo Show 10, 3rd generation” equipped with “Alexa”.

The study seeks to answer the following research questions (RQ):

RQ 1:

To what extent can older adults 55+ learn to operate VCDs autonomously or with external support, and how do they evaluate their ease of use?

RQ 2:

What patterns and intensities of VCDs use are evident among older adults 55+? Which application areas are used, and how are the devices integrated into everyday life?

RQ 3:

How does the discrepancy between initial perceptions and reservations about VCDs and actual usage experiences evolve among older adults 55+?

RQ 4:

What specific problems and shortcomings in VCDs functionality do older adults 55+ identify, and how do these relate to objectively identifiable issues?

RQ1 aims to explore the learnability and usability of VCDs for older adults, a demographic that often faces unique challenges when adopting new technologies due to physical, cognitive, and sensory changes. By understanding whether participants can independently learn to operate VCDs and how they evaluate the ease of use, this question addresses the accessibility of VCDs. It also identifies whether external support, such as guidance from family members or instructional resources, plays a critical role in the learning process. The findings from this question contribute to improving the user-friendly VCDs design tailored to older adults’ needs.

With RQ2, we investigate how older adults use VCDs in their daily lives, focusing on their preferred application types (e.g., entertainment, information retrieval, or smart home integration) and frequency of use. Understanding these usage patterns helps to assess the practical utility and relevance of VCDs for older adults. It also shows how seamlessly these devices fit into their routines and provides insights into areas where VCDs can improve comfort and independence.

The development of user perceptions, particularly the discrepancy between initial skepticism or concerns and actual experiences after prolonged use, is the purpose of RQ3. We investigate this discrepancy to discover how practical experience and familiarity with VCDs can mitigate initial reservations and influence long-term acceptance. The results of this question provide insight into strategies for overcoming barriers to acceptance and emphasize the importance of experiential learning in forming positive attitudes in older adults.

Using RQ4, we aim to uncover the challenges and limitations that older adults encounter when using VCDs, including hardware, software, and privacy-related concerns. By correlating subjective complaints with objectively verifiable problems, such as operating errors or system limitations, areas where the functionality of VCDs can be improved are identified. The aim is to provide developers with actionable recommendations to address specific user needs, improve the usability of VCDs, and ensure that these devices meet the expectations and requirements of older adults.

To answer these questions, we conducted a field study over 3 months in 20 SH. Using periodic interviews at 4-week intervals and IT-supported usage logs, we collected data that gave us a deeper understanding of learnability, usability, usage patterns, and acceptance of the technology by older adults. Based on our findings, most participants learned the technology autonomously and did not require external help. The error rate when formulating voice commands was also very low, which indicates good usability. However, problems arose when using extended functions. The participants expressed the need for step-by-step instructions. The participants appreciated the visual feedback on the integrated touch display. Initial reservations about data protection and monitoring decreased significantly by the end of the study, indicating that prolonged use and positive experiences with the VCDs can alleviate concerns. This study contributes to the broader field of HCI by addressing identified research gaps. It expands upon the quantitative findings from our previous survey on older adults’ awareness, ownership, and usage patterns in the context of VCDs [24] with additional qualitative insights providing a nuanced understanding of the interplay between usability, acceptance, and long-term adoption of VCDs among older adults. The findings aim to inform the design of more inclusive and effective VCDs solutions, ultimately fostering digital inclusion and improving the quality of life for older populations.

This article provides an overview of the current state of research on this topic in Section 2. The methods and materials used for the field study are described in Section 3. The presentation of the results and their interpretation is presented in Section 4. Section 5 discusses and explains the findings and relates them to previous studies. In addition, the limitations of the work are discussed and conclusions and recommendations for further research are made at the end.

2. Related Work

In reviewing the relevant literature, we first discuss the current state of research on the usability and acceptance of VCDs by older adults in Section 2.1. This is followed by a presentation of the barriers and privacy concerns with Section 2.2. The section ends with the contribution of the study in Section 2.3.

2.1. Usability and Acceptance of VCDs Among Older Adults

The usability of VCDs for older adults is influenced by factors such as learnability, user-friendliness, and interaction behavior. Existing studies generally report positive evaluations of learnability and minimal technological expertise requirements for operation [11,12,19,20,21]. However, despite these promising findings, there is a pressing need for the further exploration of specific usability factors, particularly from older users’ perspectives [11].

Research by Jaskulska et al. and Kowalski et al. highlights the importance of clear instructions and minimal training requirements to facilitate successful adoption by older adults [22,25].

Specific usability preferences among older adults, as identified by Gollasch and Weber [26], include step-by-step dialogues and simple sentence structures that reduce cognitive load. Additionally, integrating safety prompts can prevent unintentional actions, reflecting a desire for increased control and restrained, proactive suggestions [26].

The acceptance of VCDs is shaped by perceived usefulness, actual usefulness, and barriers to use. Perceived usefulness depends on older users recognizing the added value of VCDs for their daily tasks [19,22]. Social influences, transparent interaction flows, and ease of setup also play significant roles in acceptance [1,10,23]. Arnold et al. and Yu et al. recognize social influences and a transparent interaction flow as further factors that influence acceptance [1,23]. Nevertheless, actual usefulness often falls short of expectations due to challenges such as speech recognition inaccuracies, setup difficulties, and misunderstandings about device functionality [17,27,28,29,30,31]. Complementary screens, as noted by Blocker et al. and Kowalski et al., could mitigate these issues by offering visual feedback and additional interaction options [22,32].

2.2. Barriers to Use and Privacy Concerns

Privacy and security concerns represent substantial barriers to the adoption of VCDs among older adults. Fears of eavesdropping, unclear data usage, and constant surveillance are well-documented in the literature [27,29,33,34]. These concerns are amplified by the “always-on” functionality of VCDs, which raises apprehensions about unauthorized recordings and data breaches [11,19]. However, not all older adults prioritize these concerns, with some studies reporting that usability and functionality outweigh privacy considerations for many users [12,35].

2.3. Contribution of This Study

While existing research provides a foundation for understanding the usability and acceptance of VCDs, significant gaps remain regarding the long-term integration of these devices into older adults’ daily lives. Most studies rely on short-term experiments, overlooking the evolving perceptions and behaviors of older users over time [16,17]. Additionally, hybrid VCDs that combine voice control with alternative input methods, such as touch displays, remain underexplored despite their potential advantages for older adults [22,32].

This study addresses these gaps by investigating the usability and acceptance of VCDs through a longitudinal field study. It provides detailed insights into the learnability, interaction behavior, and usability challenges specific to older adults, focusing on hybrid devices like the “Amazon Echo Show 10, 3rd generation” which is only addressed by a few studies. The findings aim to inform the design of more inclusive and effective VCD solutions, fostering digital inclusion and enhancing quality of life for this demographic.

3. Material and Methods

Section 3.1 contains a detailed description of participant selection and recruitment. In addition, the role of complementary households (CHs) is presented and statements are made in compliance with data protection regulations. Building on this, we explain the criteria for device selection and the underlying research methodology in Section 3.2. Section 3.3 describes the entire research process and data collection in chronological form, while Section 3.4 explains the data analysis process.

3.1. Participants—Older Adults 55+, Selection, Recruitment, Role of CHs, and Ethics

The literature variably defines older adults with thresholds such as 55+, 60+, or 65+. We decided, also based on the use of these clusters in our previous studies, to investigate older adults aged 55 and over and to categorize this population into the following age groups based on their heterogeneity in terms of digital technologies:

• Age group “Pre-Retirement (55–64 years)”: This age group includes people close to the official retirement age. Many of them are still actively working, while others have already retired. Statistics show that employment is already declining significantly in this age segment. However, this group—also known as “baby boomers” [36]—makes up a significant proportion (15%) [37] of the total population of the Federal Republic of Germany. This group is essential because they have directly experienced the incipient digital transformation [38,39].
• Age group “Senior citizens (65–74 years)” is the age group of traditional pensioners. This age group already has fewer contact points with digital technologies [39].
• Age group “Ancient (75 years and older)” includes people who have already reached an advanced age. They can have different health challenges and are often dependent on support. Digital technologies have not played a role for this group of people for most of their lives [39].

3.1.1. Participant Selection

Before data collection, the socio-demographic and technical inclusion criteria relevant to selecting participants and the sample size were determined [40].

Table 1. Socio-demographic and technical inclusion criteria, justification, and planned sample size.

Criteria Category	Inclusion Criteria and Justification
Socio-demographic Criteria
Age	Participants must be 55+ years old, irrespective of marital status, gender, education level, income, or occupation. They should live alone to ensure an objective assessment of device usage without external influence from family members.
Residence	Participants must reside within 50 km of the research institution to minimize travel effort for interviews and enable quick technical support if needed.
Relatives and friends	Participants must have at least one relative or friend (not living in the same household) willing to participate as a complementary household (CH). Details are provided in Section 3.1.3.
Technical Criteria
Internet connection	Participants must have a WiFi router connected to the internet to ensure device connectivity.
Owning devices	Participants or their relatives/friends must own a smartphone or tablet capable of running the “Alexa App” to access additional digital services. Without a compatible device, usage is limited.
Amazon account	Necessary for device setup. If unavailable, the research team can assist in creating one.
Planned sample size	20 SHs and 20 corresponding CHs.

provides an overview of the inclusion criteria. For ethical reasons, vulnerable individuals unable to provide informed consent due to age or limited mental capacity or whose participation could pose undue strain or risk were excluded [41].

3.1.2. Participant Recruitment

Participants were recruited via “selective sampling” [40,42] from older adults aged 55+. Some participants were known from earlier studies, while others were recruited through recommendations and agreed to participate. The advantage of this approach was the guaranteed voluntary nature of participation, which generally leads to greater motivation. Participant selection aimed to recruit a deliberately heterogeneous [40], maximally contrasted, and therefore, an informative group of people for the study by the principle of variance maximization according to Patton [43]. Participants were provided a detailed explanation of the research process in a printed “Participant Information” document. This document included study details, contact information for the research institution, the responsible data protection officer, research team members, and a statement about voluntary participation and the right to withdraw at any time. Participants signed this document to confirm their understanding and consent. At the end of the study, the participants could choose whether to keep the VCD free of charge or return it.

A total of 32 participants (n = 32) from 20 SHs and 20 participating CHs were recruited to carry out the field study, considering the inclusion criteria. The total number of participants includes individual life partners, as one device was installed per household, which is used by all household members together, but in different ways and according to personal preferences. For this reason, the total number of participants differs from the number of SHs.

A total of 12 participants were recruited for the 55–64 age group (37%), 14 for the 65–74 age group (44%), and 6 for the 75 and older age group (19%). Of the participants, 8 were single, and 24 were married or cohabiting. The gender ratio was balanced, with 15 males and 17 females. The ratio of professional activity among the age group naturally deviates from a balance, as people of advanced age do not work or are only marginally employed. Accordingly, 4 participants were still working full-time, 4 were in part-time or marginal employment, and 24 had already retired and were no longer employed. All participants lived alone or with their partners in their households, were still mobile, and were not dependent on help in everyday life.

Table 2. Participant distribution according to attributes and characteristic values.

ID ∑(n = 32)	Age	Gender Male (M) Female (F)	Life Situation Marital Union / Cohabitation (M) Living Alone (A)	Technologies Used Smartphone (S) Tablet (T) PC/Notebook (P)
P1	61	F	A	S, T
P2	75	M	M	S
P3	71	F	M	—
P4	82	M	M	S, T, P
P5	82	F	M	S, T, P
P6	74	F	M	—
P8	66	F	A	S, P
P9	61	F	A	S, P
P10	61	M	A	S, T
P11	61	M	M	S, T, P
P12	57	F	M	S
P13	63	F	M	S, P
P14	70	M	M	S, P
P15	68	M	M	S, P
P16	69	F	M	S, T, P
P17	63	M	M	P
P18	62	F	M	S
P19	76	M	A	S, T, P
P20	63	F	A	S, T, P
P21	67	F	A	S, T, P
P22	68	M	M	S, T
P23	62	F	M	S
P24	74	M	M	P
P25	69	F	M	S, T, P
P26	70	M	M	S, T, P
P27	70	F	M	S, T, P
P28	57	F	M	S
P29	62	M	M	S
P30	77	M	A	P
P31	70	M	M	S
P32	70	F	M	S, P
	$\overline{x}$ = 68	∑ M: 15	∑ A: 8	∑ S: 27
	SD: 6.75	∑ F: 17	∑ M: 24	∑ T: 13
				∑ P: 19

shows the socio-demographic attributes of the participants. A balanced participant ratio could not be achieved.

3.1.3. Participants—The Role of CHs

In addition to the SHs, their relatives or acquaintances were included in the field study as CHs. The exclusion criterion was that they lived in their household, which was not located in the building and geographically outside the SH. This was intended to avoid multiple uses of the VCD as much as possible. In the event of problems with digital technologies (e.g., installing apps on smartphones, lack of affinity for technology and knowledge of how to use devices, operating problems), family members or friends are usually asked for help and support [44,45,46]. Within the family, for example, children and grandchildren are often an incentive to engage with new technologies in the first place. They can also act as knowledge brokers [47], as younger adults can be assumed to have a higher affinity for technology. Because relatives or acquaintances are seen as trusted persons, the involvement of CHs may make it easier for older adults to try out new functions in direct contact (e.g., through video calls). The verification of the statements made by the SHs in the interviews on the key topics mentioned should be mentioned as a further advantage.

3.1.4. Ethics

From a data protection perspective, the SHs and the CHs gave written consent to collect and process personal data (under Art. 6 para. 1 p. 1 lit. a, in conjunction with Art. 7 GDPR). In addition, they received printed participant information in general, easily understandable language, with a description of the entire study procedure and the data collection methods used.

3.2. Methodological Procedure

In Section 3.2.1, we describe the VCDs used and the reasons for choosing these devices. A description of the selected research methodology in Section 3.2.2 follows this. The procedure description is then explained in Section 3.3.

3.2.1. Selection of a Suitable VCD

The following criteria guided the selection of the most suitable VCD for the study:

• VCD with integrated VA: A Category 2 device (see Section 1).
• 10″ Touch Display: Provides visual feedback, which benefits older adults, particularly those with hearing impairments or age-related vision decline [22,32,48].
• Stationary Installation: Enables reliable operation, better sound quality for hearing-impaired users, continuous power supply, and integration into smart home systems, improving safety and quality of life [11,12].
• Integrated Camera: Facilitates home monitoring and video calls, supporting social interaction and security, especially for older adults living alone or distant from family [49].

The “Amazon Echo Show 10, 3rd generation” was selected based on these criteria. Amazon dominates the smart speaker market with its “Alexa” interface [50,51]. Alternatives like “Google Home Nest” lack integrated cameras and offer smaller displays, while the “Apple Home Pod” and other smart speakers do not include displays. The Echo Show 10 features a rotating screen that tracks the user’s movement, enhancing voice recognition and screen visibility. The use of this specific model has not been reported in prior research, making this study a unique exploration.

3.2.2. Research Method

This field study employs a descriptive research methodology with an exploratory focus, aiming to understand the investigated issues comprehensively. VCDs were provided to 20 SHs for 12 weeks, using a triangulated data collection approach [52]:

• Semi-structured personal guided interviews conducted at four-, eight-, and twelve-week intervals, combined with usage logs from the Amazon API in the SHs.
• Semi-structured telephone interviews with CHs after thirteen weeks.
• Analysis of all voice commands and responses recorded by the VCD using session cookies in usage logs.

This mixed-methods approach aligns with Schlomann et al. recommends combining traditional interviews with IT-supported tools (e.g., usage logs) over a minimum of four weeks, enhancing data validity and research credibility [16,52]. The study was conducted from 19 May 2021 to 9 September 2021.

3.3. Procedure

After a successful pre-test of the interview guide with three participants who could be included in the sample, personal, semi-structured guided interviews [53] were conducted. These included an initial interview at the beginning of the study and interim interviews after four, eight, and twelve weeks. The interviews were carried out over twelve weeks directly on-site in the SHs. Additionally, a final semi-structured telephone interview with the CHs was conducted after thirteen weeks. Two research team members conducted this interview [53]. Usage logs of the SHs over the entire study period supplemented the data collection [16,53].

Figure 1 provides a compact overview of the study process.

3.3.1. Initial Interview and Device Setup

In a personal semi-structured guided interview, SHs members were asked about the following topics: (1) their general attitudes towards digital technologies, (2) familiarity and opinions about VCDs, (3) willingness to participate in a VCDs study and meeting participation requirements, and (4) details about their personal life situation and daily living arrangements. Individuals were interviewed separately in households with multiple members (spouse/life partner) to prevent partner influence and ensure unbiased responses. The interview concluded with socio-demographic questions and the collection of contact details for the CH. The complete interview guides (initial and interim) are attached as supplementary material in the Appendix A (

Table A1. Preliminary interview guide—technology affinity.

Category: Technology Affinity
Guiding question:	What do you think of digital technology in general, such as smartphones, tablets, or PCs?
Maintaining the conversation:	What negative/positive experiences did you have with it? Why?
Contents	Questions
Attitude toward digital technologies	Which device(s) do you own?
	What do you like about the device(s)?
	Why are you not familiar with it?
	Have you tried the device before?
	Have there been any problems?
	Do you have any concerns/reservations?
Ownership of equipment	What devices do you own?
	Did you buy the device?
Use of the devices	For what purposes do you use the device?
Ability to use	Do you know how to operate the device?
Origin of knowledge in handling	Did you learn how to operate them yourself?
	Have you or did you already use digital devices during your professional career?
Support needs	Who can help you with any problems?

,

Table A2. Preliminary interview guide—voice assistants/Alexa.

Category: Voice Assistants/Alexa
Guiding question:	Have you ever heard of “Alexa”? What is your attitude towards it?
Maintaining the conversation:	Have you ever had any experience with it? Why?
Contents	Questions
Attitude to “Alexa”	Do you know “Alexa”?
	What have you heard about it?
Functionality of “Alexa”	Do you know what you can use “Alexa”?
	Do you think “Alexa” could be fun for them?
Interest in the technology	Would you like to try “Alexa”?
	What prevents you from trying “Alexa”?
	Suppose you were given “Alexa” as a gift. Would you still not want to try “Alexa”?

and

Table A3. Preliminary interview guide—living conditions/daily life.

Category: Living Conditions/Daily Life
Guiding question:	Please tell us something about yourself: How do you live? Are you still working? How is your everyday life?
Maintaining the conversation:	Is there anything else? And further? What happened next? And what else?
Contents	Questions
Family status	Do you live alone or in a cohabitation?
	Are there any family members living in your house/apartment?
	Do you have children?
	What is your children’s profession?
	Do you have grandchildren?
Professional practice	Are you still employed?
	What is your profession?
	When did you retire?
	Do you still have a marginal job?
Mobility	Do you own a car?
	How do you do your shopping or get to the doctor?
Living Situation/Daily Life	What does your daily routine look like?
	Does anything change on the weekends?
	Are your everyday tasks shared with your life partner?
	Do you do things together with your life partner?
	Do you prepare your meals?
	Who helps you with this?
	Do you depend on external help, e.g., cleaning help, relatives?
	Do you sometimes feel bored or perhaps lonely?
Hobbies	What hobbies do you practice?

).

The research team installed the VCD at the location chosen by participants and guided them through the step-by-step setup process. This included entering the SSID identifier, Wi-Fi password, location information (address for local weather), and Amazon credentials. If participants lacked an Amazon account, one was created on-site with their consent. The device played an introductory video tutorial after establishing the internet connection and linking the Amazon account. The team demonstrated an example command (“Alexa, can you sing?”) and encouraged participants to try commands themselves, omitting further operational guidance to investigate self-directed learnability.

The “Alexa-App” was installed on a household member’s smartphone where possible. Amazon voice purchases and newsletters were disabled and replaced by a weekly identical Deggendorf Institute of Technology (DIT) newsletter. This newsletter, sent via post to SHs and CHs, included information about new functions, voice commands, and warnings about paid services, such as audiobooks from Audible. The VCD was mailed to CHs after an advance telephone briefing and their agreement to participate. Being younger and digitally adept, these participants received no additional instructions or pre-configurations. No usage logs were saved for CHs.

3.3.2. Interim Interviews (After Four and Eight Weeks of Use)

Four and eight weeks after installing the VCDs in the SHs, semi-structured qualitative guided interviews were conducted to reassess participants’ experiences, identify changes, and explore additional expectations or support needs. Operational assistance was minimal, with participants only inquiring about music subscriptions, compatible smart devices, and calendar integration with Google. During these sessions, personal instruction on video telephony was provided. Contact details for the corresponding CHs were stored on the devices, enabling participants to make direct video calls to their CHs. The complete interview guide (after four and eight weeks) is attached as Supplementary Materials in the Appendix A (

Table A4. Interview guide—after four weeks.

Guiding question:	What do you think of “Alexa”, and how do you get along with “Her”?
Maintaining the conversation:	Why do you have this impression? What problems of a technical nature have occurred?
	Are there any other problems? Is the operation of “Alexa” simple?
Contents	Questions
Device operation	Do you know the meaning of the different colored light strips (orange)?
	Do you know how to retrieve and delete received messages?
	Do you know how to recognize that there is an Internet connection?
	Do “Alexa” understand you?
	How did you react to it?
Usability	What do you like about “Alexa” and what don’t you like?
	Have you been angry about “Alexa” or perhaps particularly pleased?
	Is “Alexa” polite to you?
	What do you think of the rotating display?
	Does anything about “Alexa” bother you?
	Has “Alexa” responded without being addressed by you?
Usage/non-usage	Why have you not yet used/tried “Alexa”?
	Do you know how to use “Alexa”?
	Would it be helpful if we explained “Alexa” to you again?
Features	Do you like the possible applications of “Alexa”?
	Do you know what else “Alexa” can do?
	Can you think of anything you want to do with “Alexa”?
	Is the newsletter helpful?
	Have you tried out the tips?
	Have you told friends or acquaintances that you have an “Alexa”?
	Have you talked to friends or acquaintances about “Alexa”?
Reservations	Do you have any reservations about using “Alexa”?
	Have you also unplugged “Alexa” once?
	Have you muted the microphone?
	Do you feel you are being monitored by “Alexa”?
	Has the camera been switched off by you?

and

Table A5. Interview guide—after eight weeks.

Guiding question:	Has anything changed regarding how you deal with “Alexa” since our last visit?
Maintaining the conversation:	Can you think of anything else to say about this? What else?
	Is there anything else you can tell us?
Contents	Questions
Device operation	What did you do as a result?
	Are there still problems with “Alexa”?
	Can we help you?
Usage / non-usage	Did you notice something in the use of “Alexa”?
	Do you now know how to use “Alexa”?
	Do you now use “Alexa” with our newsletters more often than before?
Features	Have you tried out new features with “Alexa”?
	Have you tried video telephony?
Reservations	Do you still feel you are being monitored or followed?

).

3.3.3. Final Interviews with SHs (After Twelve Weeks of Use)

At the end of the 12-week study, a final set of semi-structured guided interviews was conducted to evaluate participants’ overall experience with the VCDs. The interviews aimed to address RQ1–RQ4 by covering the following areas: (1) intention and attitude toward the device, (2) purposes and frequency of use, (3) ease of use, (4) information and system satisfaction, (5) system reliability and flexibility, (6) integration into daily life, (7) reservations and privacy, and (8) comparisons with other digital technologies. Participants were informed that the study had concluded, data collection had ceased, and the DIT newsletter would no longer be sent. The complete interview guides (after twelve weeks) are attached as Supplementary Materials in Appendix A (

Table A6. Interview guide—after twelve weeks.

Guiding question:	Please tell us about your experience with “Alexa”?
Maintaining the conversation:	What do you enjoy about “Alexa”?
	What don’t you enjoy about “Alexa”?
	Can you think of anything else?
	And what else?
	Is there anything else you can tell us?
	What do members of the complementary household have to say?
	Do you find “Alexa” intrusive?
Contents	Questions
Intention and attitude	Would you buy an “Alexa” for yourself?
	Would “Alexa” be good in other rooms of your home?
	Would you recommend “Alexa” to relatives/acquaintances/friends, and what would you say to them?
	Do you continue using “Alexa” after the test period?
	Do you want to keep the device?
Usage	For what purposes do you use “Alexa”?
	How often do you use “Alexa”?
Simplicity	How do you find the use of “Alexa”?
	Was it easy to learn to use “Alexa”?
	Did you research “Alexa features” beyond the newsletter?
	Did you need to seek help from complementary households or others?
Information and system satisfaction/quality	Are you satisfied with the information you receive from “Alexa”?
	Which functions do you particularly like, and which do you dislike?
Reliability	Do you think “Alexa” works reliably?
Flexibility	Would you like “Alexa” to have certain additional functions?
Integration	How has “Alexa” integrated into your daily life?
	Has your contact behavior changed due to video telephony with “Alexa”?
	Are there certain circumstances in which you use “Alexa”?
	Do you use “Alexa” to pass the time out of boredom or loneliness?
Reservations/Privacy	Do you have reservations in connection with “Alexa”?
	Do you feel that “Alexa” restricts your privacy?
Personification	Has “Alexa” annoyed you?
	Does “Alexa” seem like a technical device to you?
Comparison to other digital technologies	How do you find “Alexa” compared to the smartphone?
	For which purposes would you prefer “Alexa” to the smartphone?
	For which purposes would you prefer to use the smartphone/tablet or the PC?

).

3.3.4. Telephone Interviews with CHs (After 13 Weeks of Use)

To validate the SHs participants’ statements, semi-structured telephone interviews were conducted with the CHs. These interviews explored (1) whether the SHs managed well with the device, (2) whether they sought help for operational issues, (3) the use of video telephony for communication, and (4) who initiated video calls. Interviews were recorded with the CHs’s consent. The complete interview guide (after 13 weeks) is attached as Supplementary Materials in the Appendix A (

Table A7. Interview guide—after thirteen weeks—complementary households.

Guiding question:	Please share your thoughts on your relative/friend/acquaintance’s experience with “Alexa”?
Maintaining the conversation:	What else?
	Is there anything else you can tell us?
	Do you think your relative/acquaintance/friend got along well with “Alexa”?
	Have you been asked for help by your relative/acquaintance/friend?
	Did you use video telephony with your relative/acquaintance/friend?
	Who took the initiative to use the video telephony function?

).

3.3.5. Usage Logs

The usage logs of all voice commands and device responses were extracted from the DIT database with participants’ explicit consent. Session cookies were deleted at the study’s conclusion, ensuring no further data collection occurred.

3.4. Data Analysis

The data analysis included interviews, transcripts, and usage logs. Individual interviews were conducted face-to-face in the SHs (n = 32), while telephone interviews were held with the main contact person in each of the 20 CHs (n = 20). Interviews were recorded, and the audio files were transcribed from dialect into standard language using the software “f4-Transkript” based on predefined transcription rules. Each transcript was reviewed by a research team member and checked for quality by a second researcher. The transcriptions formed the basis for qualitative analysis, which was performed using “MAXQDA 2022”. The usage logs of the 20 VCDs were imported as CSV files into Microsoft™ Excel 365 and subsequently analyzed.

A thematic coding approach combining deductive and inductive methods was applied to analyze the interviews [54]. Deductive top-level codes were derived from research questions and a systematic literature review by Jakob [15], which also informed the development of the interview guidelines. Inductive sub-codes were generated by one team member who reviewed the transcripts in full and indexed them subsumptively [40]. Another team member reviewed the coded transcripts, and discrepancies were resolved in discussions to achieve consensus. In total, 148 interviews were conducted (128 with SHs, 20 with CHs), resulting in 2169 min of data (SHs: 1878 min; CHs: 291 min). The final codebook comprised 156 top-level codes and 1021 sub-codes, which can be accessedat https://doi.org/10.5281/zenodo.14326495 (Last accessed: 9 December 2024).

The analysis of the usage logs included coding 12,556 data records into 20 top-level categories [40], focusing on commands and responses. Only the “commands sent by the user” (CT) and “responses from the VCD” (AR) were analyzed, as parameters like “automatic speech recognition” (ASR) and “text-to-speech” (TTS) were not relevant to the study’s objectives. To ensure consistency, coding was verified through random sampling and team discussions.

The analysis yielded insights into user interactions with the VCD. Categories such as “System commands,” “Command incorrect”, and “Command understood” highlighted usability and participants’ learning progress (RQ1), while “Wrong answer/answer not known” identified functional deficits, contributing to RQ4. A detailed summary of the top-level categories, their explanations, and examples is shown in

Table 3. Top-codes usage logs with explanations and examples.

Top-Code	Explanation and Example
Weather outlook/weather report	Questions about weather forecasts. Example: “Alexa, what will the weather be like in XX?”
Radio stations and music tracks	Commands to play radio stations or specific songs. Example: “Alexa, play XX.”
Knowledge in reference works	General knowledge inquiries. Example: “Alexa, what is the highest mountain in the world?”
Other information	Queries about varied topics such as time, COVID-19 statistics, or translations. Example: “Alexa, what are the incidence values in XX?”
System commands	Commands for volume control, canceling interactions, and settings.
Latest news	Requests for current news. Example: “Alexa, show me the news.”
Alarms and timers	Commands to set alarms or timers. Example: “Alexa, set an alarm for 7 o’clock.”
Cuisine recipes	Requests for recipes. Example: “Alexa, give me a recipe for currywurst.”
Reminders, appointments, notes	Commands to manage reminders or notes. Example: “Alexa, remind me of my doctor’s appointment.”
Communication	Phone or video calls. Example: “Alexa, call my daughter.”
Smart home device control	Commands for controlling smart home devices. Example: “Alexa, switch on the socket.”
Skills	Third-party information retrieval. Example: “Alexa, what is the animal of the day?”
Command incorrect	Commands not recognized by the VCD. Example response: “I’m not sure.”
Command understood	Correctly recognized and executed commands. Example response: “OK!”
Wrong answer/answer not known	Incorrect or unknown responses. Example: “Unfortunately, I don’t know that.”

. This categorization provided a sufficient framework to evaluate the usability, learnability, and functionality of the VCD devices objectively.

4. Results

This section presents the findings on the participants’ ability to autonomously learn to use the VCD and their evaluation of usability (RQ1). We also discuss usage patterns, purposes of use, and integration into daily life (RQ2), the evolution of initial perceptions versus actual experiences (RQ3), and identified problems alongside objectively verifiable deficiencies (RQ4).

4.1. Autonomous Learnability and Evaluation of Usability

The following interview quotes from the participants have been translated from German into English.

The evaluation of learnability and usability of the VCD was based on participant feedback, observational data, and usage logs over 12 weeks. Below, the findings are summarized.

After four weeks, most of the participants reported no difficulty learning to use the VCD independently, without assistance from CHs or external individuals. One participant required support from a CH, while four did not provide direct feedback. By the end of the study (12 weeks), nearly all participants confirmed that they found the VCD easy to learn, with only one participant encountering challenges with advanced features, such as video telephony, calendar linking, or Bluetooth pairing.

These observations were supported by CHs, who noted that participants successfully learned to operate the VCD autonomously. Participant P3 reflected on their experience: “Yes, you didn’t have to learn much. It’s different from a computer where I had to rethink how to interact”.

Usage log analysis showed an average command error rate of 4.9%, with 13 out of 20 SHs achieving error rates below 5%. The correctness ratio was calculated as follows:

$$r_{correctness}={\displaystyle \frac{c_u}{c_u+c_w}}$$

Detailed results are shown in

Table 4. Ratio of incorrectly sent commands to those understood by the VCD and error rates.

SH-ID	Incorrect Commands	Understood Commands	Sum	Error Rate (%)
∑ (n = 20)	589	11,066	12,556	4.9
$\overline{x}$	29.4	553.3	628	4.4

.

These findings demonstrate that participants successfully learned to use the VCD independently, with minimal errors, confirming the interview data and highlighting strong learnability.

Regarding usability, the overwhelming majority of participants reported a good understanding of the VCD operation after four weeks. By 12 weeks, the participants rated its usability positively. Positive feedback emphasized the device’s features, such as the rotating screen, always-on functionality, and immediate access to information without needing a PC. Participant P7’s comments reflected this: “For me, it is perfect. (…) I can give commands from my seat, and it works most of the time”. However, three participants experienced difficulties with advanced functions such as smart home integration and calendar setup. P31 noted the following: “No, I don’t find that easy. If you’re not familiar with technology, it’s not easy, at least for me.”

While some participants required no assistance during the study, some sought occasional help. According to the CHs, participants managed operations well, requiring minimal external support. Positive feedback highlighted the rotating screen, always-on functionality, and immediate access to information without needing a PC. Overall, a major part of SHs affirmed the VCD’s ease of use.

These findings demonstrate that VCDs are accessible and learnable for older adults, with high usability ratings from most participants. Despite minor challenges with advanced features, the overall results highlight the device’s potential for supporting independent use among older individuals.

4.2. Purposes and Intensity of Use

The analysis of encoded usage data allowed the categorization of VCD functions by frequency. The most common purpose was listening to radio stations and music (23%), followed by queries for “other information” (14%), weather forecasts (7%), general knowledge (5%), and smart home controls (3%). The complete frequency distribution is shown in Figure 2.

These results indicate that entertainment purposes dominate VCD usage among participants, with other functionalities playing a secondary role.

The intensity of VCD usage was analyzed based on days of the week, times of day, and frequency of daily use. Sundays showed the highest activity (17 %), followed closely by Saturdays (16 %), with weekday usage ranging from 12% to 14 %. Usage by time of day revealed peak activity in the early morning (23%), noon (22%), and afternoon (20%), with reduced activity in the evening (13%) and at night (2%). Detailed data are presented in

Table 5. Times of day for VCD usage (all participants).

Time of Day	Early	Morning	Noon	Afternoon	Evening	Night
Commands (n = 12,556)	2907	2401	2777	2517	1653	301
Percentage (%)	23.2	19.1	22.1	20.0	13.2	2.4

.

Regarding daily frequency, participants used the VCD several times a day (71%), with fewer days showing once-a-day use (4%). However, on 25% of days, the VCD was not used, indicating either absence from the household or lack of interest.

These findings demonstrate that VCDs were well-integrated into participants’ daily routines, mainly for entertainment and retrieving information, with consistent use across most days and peak activity during morning and midday.

4.3. Discrepancy Between Initial Perceptions and Actual Usage

Data from qualitative interviews and usage logs were analyzed to addressRQ3, focusing on differences between initial perceptions and actual usage.

All participants chose to keep the VCD after the study. However, the participants indicated that they would not purchase the device independently, citing high costs, lack of prior knowledge, or aversion to technology. Despite this, most of the participants said they would recommend the VCD to others.

Regarding satisfaction, participants expressed satisfaction with the system and the quality of the information but also indicated limitations. Common complaints included unanswered or insufficiently detailed responses. Usage logs corroborated this, showing that the VCD did not know the answer in 7% of commands (“answer not known”) and provided incorrect responses in 5% due to command wording.

4.4. Reservations and Barriers Hindering Usage

Participants were asked whether they felt surveilled during the study. After four weeks, a large majority of the participants reported no feelings of surveillance. By the end of the study, few participants no longer felt surveilled, and some noted improvement. However, a predominant part of the participants maintained no such feelings throughout the study. Feedback from CHs confirmed these assessments. The evolution of surveillance concerns is shown in Figure 3.

Participants were also asked if they felt observed or followed by the VCD, mainly due to the rotating display. After 12 weeks, the participants stated they did not feel followed or observed, while two expressed uncertainty and two felt followed. P7 noted, “When it moved towards me, I turned it off because I didn’t want to feel observed”. Despite the low concern, some few participants deactivated the integrated camera using the built-in slider.

4.5. Problems and Functional Deficits in the Use of VCD

To addressRQ4, participants were asked to describe problems and functional deficits encountered while using the VCDs.

Several participants criticized the requirement for a paid subscription (Amazon Music Unlimited) to play specific music titles (P4, P23) and the need for user intervention to reject the subscription offer (P9). Technical issues were common, including a lack of interaction due to missing Internet connections (P13) and difficulties configuring the VCD via the “Alexa-App”. Challenges with advanced functions, such as linking to Google Calendar, Bluetooth integration, and smart home components, were also reported (P11, P16, P26).

Incorrect or incomplete responses from the VCD were highlighted by participants (P6, P7, P12). For example, P17 suggested improving the device as a reference tool, noting that general knowledge queries were often unanswered. Regarding speech recognition, most participants reported no difficulties in issuing commands. However, participants P2, P3, P8, P11, P12, P18, P19 experienced occasional problems, resolved through repeated or reformulated commands. P7 humorously remarked, “She understands me better than my wife”.

P19 noted deficits in regional information, such as private telephone directory searches. Others expressed dissatisfaction with the physical arrangement of control buttons on the VCD.

Video telephony presented mixed results. Some participants successfully used the video telephony function, while others faced setup challenges or lacked support for configuration. Common obstacles included CHs being unavailable for setup (P4, P5, P6, P7, P17, P18, P24, P31, P32) and participants’ limited knowledge or need for the feature (P11, P19, P22, P23, P25). Despite these challenges, the overwhelming majority of the participants rated the video telephony function positively, whereasa few found it unfavorable (P7, P17, P22, P23, P24, P25, P26, P27, P29).

Overall, participants reported a range of issues, including subscription requirements, technical challenges, speech recognition limitations, and functionality gaps. However, the most positively perceived the VCD’s usability and functionality, including video telephony, indicating that its benefits often outweighed its drawbacks.

4.6. Distinguishing Perceived Problems and Functional Deficits from Actual Issues

To addressRQ4, participants were asked to describe the issues and functional deficits encountered while using the VCD.

Participants criticized poor accessibility and the lack of color-coded control buttons for microphone mute and volume, which are small and located on the device’s top, making them difficult for older adults to use. These are valid issues but can be addressed with user manuals. Conversely, the inability to use the VCDs without an Internet connection is not a hardware defect but a design feature, as the device relies entirely on online content. Connectivity issues can often be resolved by restarting the router, indicated by the VCD’s orange light strip.

Participants raised concerns about the inability to retrieve specific music without a subscription (e.g., Amazon Music Unlimited). This is a standard distribution model, similar to other streaming services, and not a functionality deficit. However, if they subscribe, older adults may require support to setup payment methods.

Speech recognition issues were linked to both user errors and software limitations. Problems arose from non-standard command syntax, unclear pronunciation, or being too far from the device. Repeating or rephrasing commands often resolved these issues.

Complaints about incorrect or insufficient answers, particularly regarding regional information (e.g., local events, doctor office hours), represent software-related deficiencies. The VCD lacks the necessary data sources to address such queries effectively, creating unmet user expectations.

Participants struggled with advanced functions such as linking Google Calendar, entering telephone contacts, and configuring telephony or video telephony via the “Alexa-App” or web interface. Due to their complexity and lack of intuitive guidance, these tasks require external assistance. Some participants found the need to use a wake word (e.g., “Alexa”) to initiate interactions unsatisfactory. However, this is a standard feature of voice-controlled devices and not a functionality flaw. Users must adapt to this requirement, which does not necessitate external support.

The perceived problems largely stemmed from unmet expectations, a lack of familiarity, or system limitations rather than inherent flaws. Actual issues included hardware accessibility challenges, insufficient regional information capabilities, and the complexity of advanced functions. Solutions range from user adaptation and manuals to potential software enhancements.

5. Discussion

This study provides novel insights into the usability and learnability of VCDs among older adults through a longitudinal field study, addressing significant research gaps identified in the previous literature [11,14,24]. This research expands on prior findings by exploring participants’ learnability, usability challenges, and perceptions over 12 weeks. It contributes to a deeper understanding of older adults’ interaction with voice-controlled technologies in real-world contexts. However, this discussion also highlights nuanced findings related to users’ perceptions of privacy and surveillance, which are critical for understanding barriers to adoption and long-term engagement with these technologies.

This section begins with a comparison of the findings with the relevant literature in Section 5.1. Subsequently, identified new insights and conclusions to be drawn from them are made in Section 5.2. The identified research gaps and implications of the study are discussed in Section 5.3. At the end of the section, the limitations of the study are discussed in detail in Section 5.4.

5.1. Comparison with Related Work

Existing research has highlighted that VCDs offers significant potential for enhancing digital inclusion among older adults [11,14,24]. However, most studies have focused on short-term interactions or controlled environments, overlooking these devices’ long-term usability and integration into daily life. This study extends the scope by examining older adults’ evolving perceptions and interaction behaviors in their home environments over three months.

The findings corroborate earlier reports on the intuitive nature of VCDs [11,12], with most of the participants successfully learning to operate the devices autonomously. The reported average error rate of 4.9% aligns with the literature, indicating the effectiveness of VCDs in recognizing and executing commands. However, challenges with advanced functionalities such as video telephony, calendar integration, and Bluetooth pairing highlight a significant usability gap [16,24]. Unusually, most participants wanted to use the video telephony function to establish visual contact with their relatives. Due to problems with the setup and a lack of support from the CHs, this plan failed. The astonishing thing is that the CHs could not assist with the setup due to claimed time constraints. This phenomenon has not been observed in any previous study.

Compared to studies emphasizing the importance of transparent interaction flows [1,23], our results underscore the critical role of step-by-step guidance and support for older adults when using complex functions. This supports prior findings that enhanced onboarding processes and simplified instructions can significantly improve the user experience [26].

5.2. New Insights and Contributions

In contrast to previous studies [16,22,25], except one participant, none of the participants required training material or further training to use the basic functions.

A unique contribution of this study is the focus on hybrid VCDs, such as the “Amazon Echo Show 10, 3rd generation”, which combines voice control with a touch display. Participants highlighted the advantages of visual feedback, particularly for those with hearing impairments, affirming prior suggestions that hybrid systems could mitigate some limitations of voice-only devices [22,32]. Despite these advantages, some participants’ lack of familiarity with touch interfaces suggests that hybrid systems require further optimization for older users.

Another key insight is integrating VCDs into participants’ daily routines. The usage logs reveal that entertainment functions, such as listening to music and radio, dominate usage patterns, which aligns with prior findings [24]. However, low utilization of smart homes and advanced features highlights a need for more accessible and user-friendly interfaces tailored to this demographic. Somewhat surprising was the fact that the integration of smart home devices, in particular, only appeared to be beneficial for a small proportion of participants, even though these assistive technologies can support older adults in their everyday lives.

Moreover, this study sheds light on the evolution of participants’ perceptions over time. Initial reservations about privacy and surveillance diminished by the end of the study, suggesting that prolonged exposure and positive experiences with VCDs can alleviate concerns. This aligns with research emphasizing the importance of trust and familiarity in adopting new technologies [10,33].

5.3. Research Gaps and Implications

The findings of this study highlight several research gaps and provide valuable implications for future investigations and the design of VCDs. A critical limitation of the existing literature is the lack of attention to the long-term usage experiences of older adults with voice-controlled technologies. Most prior studies focus on short-term usage, failing to capture this user group’s evolving interaction patterns and perceptions. This study demonstrates that integrating voice control into the daily lives of older adults requires time and ongoing support, especially for more complex functions such as calendar integration, Bluetooth pairing, or smart home control.

Another significant research area is the discrepancy between initial expectations and actual usage of such devices. While older adults often approach voice assistants with reservations rooted in privacy concerns or a lack of technical confidence, the findings reveal that many participants reported positive experiences after using the devices, even expressing appreciation for them despite initial skepticism. This discrepancy underscores the need for tailored training materials and onboarding processes that help older users understand and leverage the benefits of such technologies.

The findings also emphasize the necessity of further exploring hybrid devices, such as the “Amazon Echo Show 10, 3rd generation”. Combining voice control with visual feedback via a screen offers unique potential for older users but is often overlooked in existing research. These devices enable intuitive interaction but present challenges for specific features like video calling or app-based operation. Addressing these barriers requires improvements in user interface design and targeted support for older adults.

Additionally, the study underscores the need for better access to local and regional information via voice assistants, as many participants expressed dissatisfaction with the limited availability of relevant content. This indicates that the algorithms and databases powering these devices must be more aligned with the needs of regional user groups.

The implications of this study suggest that developers of voice-controlled systems must consider not only technical but also social and psychological factors to foster broader acceptance. This includes designing user-friendly interfaces, enhancing regional content availability, and providing practical training resources to empower older adults to use these devices confidently and effectively.

5.4. Limitations

The study has several limitations that must be acknowledged to contextualize the findings and guide future research. First, the sample size was relatively small, with only 32 participants, which limits the generalizability of the results to a broader population of older adults. While efforts were made to recruit a heterogeneous group regarding age, gender, and digital literacy, the findings may not fully represent all older users’ diverse experiences and needs. Moreover, the study was conducted in a specific geographic region, potentially introducing cultural and socio-economic biases that may not reflect other contexts. As the study was only conducted within a specific region in Germany, the results cannot be generalized to the entire Federal Republic of Germany or in an international context.

Another limitation pertains to the duration of the study. Although the 12-week period provided insights into long-term usage patterns, this timeframe may still be insufficient to fully understand the evolving interactions and potential shifts in perceptions over even longer periods. For instance, participants’ initial enthusiasm or frustrations may not accurately predict their sustained engagement or eventual abandonment of the device after several months or years.

The study also relied on a single type of VCDs, the “Amazon Echo Show 10, 3rd generation”, which incorporates both voice and visual interface features. While this hybrid model offers unique insights into usability for older adults, the findings cannot be easily extrapolated to other VCDs that rely solely on voice commands or have different hardware configurations. Similarly, the Alexa voice assistant may have influenced user experiences, as differences in natural language processing capabilities and response accuracy across platforms were not explored.

Additionally, the study design included a significant degree of external support during the setup phase and periodic interactions with researchers, which may have mitigated potential frustrations and barriers that participants would face in real-world, unsupported scenarios. This could result in overestimating the device’s ease of learning and usability. Furthermore, the involvement of CHs as part of the study design may have inadvertently influenced participants’ behaviors and perceptions, as they could rely on these individuals for assistance, thus reducing the challenges associated with independent device use.

Another limitation lies in the reliance on self-reported data through interviews, which may introduce bias due to participants’ tendency to provide socially desirable answers or underreport challenges. Although usage logs were employed to triangulate findings, they primarily captured command-level interactions and lacked contextual information about why certain features were used or ignored.

Lastly, privacy concerns, a critical factor in adopting VCDs, were assessed through participant feedback but not rigorously tested under real-world conditions. For example, participants were aware that their data were monitored for the study, which might have influenced their comfort levels with the device’s always-on functionality and data usage.

These limitations highlight the need for further research with more extensive and more diverse samples, extended study durations, and comparisons across different device types and voice assistants. Future studies should also aim to minimize researcher intervention to simulate real-world use better and explore the broader contextual factors influencing adoption, such as privacy concerns, regional differences, and evolving user needs.

5.5. Conclusion and Future Directions

The integration, usability, and learning processes of voice assistant technology among older adults were explored using the Amazon Echo Show 10 (3rd Generation) as a case study. Conducted with 32 participants aged 55 and above, the research combined qualitative interviews and quantitative voice command analysis over twelve weeks in SHs and CHs. This study contributes to the growing body of research on VCDs by providing detailed insights into their usability, learnability, and integration into older adults’ daily lives. Furthermore, the study highlights the potential of VCDs to support older adults to maintain independence and engage in technology while identifying critical barriers and opportunities for improvement. Privacy-related perceptions, including the sense of surveillance and observation, play a significant role in shaping user acceptance and must be carefully addressed through device design, user education, and support. Although the findings underscore the potential of VCDs to promote digital inclusion, they also emphasize the need for targeted design improvements and support mechanisms. In summary, we answer the research questions based on our study findings.

RQ1: Older adults aged 55+ demonstrated a high capacity for learning to operate VCDs autonomously. After four weeks, most participants reported no difficulty using the VCD independently. By the conclusion of the 12-week study, the participants affirmed that learning the VCD was straightforward. While minor challenges were encountered with advanced features, such as video telephony and Bluetooth pairing, these did not impede basic functionality.

Regarding usability, the participants positively evaluated the operation of the VCD. Key features such as the rotating screen, hands-free interaction, and immediate access to information were highlighted as strengths. Overall, the findings demonstrate that VCDs are accessible, with minimal external assistance required, and are generally perceived as easy to use by older adults.

RQ2: The study identified distinct usage patterns, with participants primarily employing VCDs for entertainment and information retrieval. The most frequent applications included listening to music and radio, seeking general information, and checking weather forecasts. These preferences highlight the practicality of VCDs in providing convenient access to everyday services. Usage intensity varied across time and days, with peak activity observed on Sundays and Saturdays. Most interactions occurred during the early morning, noon, and afternoon. Daily usage was consistent. Most of the participants interacted with the device multiple times per day. These findings underscore the integration of VCDs into participants’ daily routines, particularly for entertainment and practical tasks.

RQ3: A discrepancy was observed between initial skepticism and eventual acceptance of VCDs. Initially, the participants expressed reservations about purchasing the device, citing concerns over cost, perceived complexity, and unfamiliarity with the technology. However, after using the VCD for twelve weeks, all participants chose to keep the device and indicated they would recommend it to others. This positive shift in perception was attributed to the device’s ease of use, practical utility, and features such as hands-free operation and access to real-time information. Privacy concerns also diminished over time, with most users becoming more comfortable as they gained familiarity with the VCD’s functionality. The findings suggest that experiential learning can mitigate initial reservations and foster positive attitudes toward VCDs.

RQ4: Participants identified several limitations. Some participants found the physical controls, such as small buttons for volume and microphone mute, difficult to use. These accessibility challenges are particularly relevant for individuals with limited fine motor skills or vision impairments. Inaccurate or incomplete responses were reported. Additionally, due to non-intuitive configuration processes, advanced features like video calling, calendar integration, and Bluetooth pairing were challenging for many participants. A minority of participants expressed initial concerns about surveillance and being observed by the device’s rotating screen. However, most concerns diminished over time as participants gained familiarity with the device and utilized privacy features such as the camera cover. Despite these issues, participants overall found the VCD functional and effective for daily use. The findings indicate that while some usability challenges persist, they are outweighed by the benefits of convenience and ease of interaction provided by VCDs.

Future research should explore adaptive and personalized interfaces, investigate the long-term impact of VCDs on older adults’ quality of life, and expand the scope to include diverse cultural and geographic contexts. By addressing these areas, researchers and designers can work toward developing inclusive technologies that improve the lives of older adults.