Abstract
This study employs multimodal conversation analysis to examine the precursors of resistance displays during Reminiscence Therapy (RT) sessions, the format of the resistance exhibited by Chinese older adults with mild cognitive impairment (MCI), and the ways it was managed by caregivers. It is found that older adults with MCI display resistance through behaviors like initiating repair, denying question premises, changing topics, refusing overtly, being silent, laughing, and giving minimal responses. Non-verbal cues such as gestures, head movements, averted gaze, and facial expressions are also employed to convey their resistance. In response, caregivers employ various facilitative practices to counter this resistance. Resistance occurs when the assumptions embedded in the caregiver’s question conflict with what the older adults have access to, or the question may infringe on the older adult’s epistemic rights. Our findings also highlight the dynamic and collaborative construction of power through negotiation of resistance and facilitation from epistemic perspective. These insights offer valuable guidance for enhancing the effectiveness of RT and fostering a more collaborative relationship between participants during RT sessions.
Keywords
Introduction
Given that dementia shows a high prevalence rate in the burgeoning aging population in China, the development of effective interventions for older adults at risk of developing dementia has become incredibly important. Mild Cognitive Impairment (MCI), a transitional stage between normal cognition and dementia, has the high possibility of reversing to a cognitively normal state, thereby providing a large time window to intervene (Jia et al., 2020). Reminiscence Therapy (RT) has been proposed as a non-pharmacological intervention to halt or ameliorate cognitive impairment, and improve psychological well-being in older adults with MCI (Dinius et al., 2023; Zhang et al., 2018). Until recently, numerous studies have paid attention to the effectiveness of RT in dementia care research, few have examined the interactional dynamics in the delivery of reminiscence interventions, especially in Chinese contexts.
RT predominantly involves participants sharing their life experiences, in which professional caregivers usually initiate most questions or topics (Slocombe et al., 2024a). In previous research, interactional practices of partners living with older adults that enable progressivity in conversations about shared memories were analyzed (Slocombe et al., 2024b). But caregivers’ interactional strategies when introducing and facilitating reminiscence activities have been little explored.
Furthermore, patient resistance has been observed to frequently occur within clinical settings and healthcare interactions (Bergen et al., 2017; Tate and Spencer, 2024; Tremblett et al., 2024). A key manifestation of resistance is the withholding of cooperation during joint activities, which slows or halts their progress (Huma et al., 2023). And older adults’ resistance to engaging in certain topics or questions may affect therapeutic effectiveness of RT. Therefore, it would be worth shedding more light on how resistance may negatively impact RT work. However, qualitative and systematic delineation of resistance manifested by Chinese older adults with MCI is exceedingly scant. Hence, this paper aims to focus on how and where displays of resistance to RT manifest and are handled in elderly patient-caregiver interactions in RT settings.
From epistemic perspective, resistance tends to be related to epistemic asymmetry between speakers and recipients (Heritage, 2012b). Clients or patients often made claims to their epistemic primacy to resist professionals epistemic authority (Tremblett et al., 2024). Additionally, when professionals emphasize clients’ obligation to answer questions, this can inadvertently provoke resistance (Muntigl, 2013). In healthcare contexts, client resistance displays are difficult to manage because orientation to epistemic primacy based on their own personal experience are relatively difficult to challenge (Bloch and Antaki, 2022). This resistance typically manifests through dispreferred responses, with its core lying in the rejection of unilaterally defined epistemic status by the other party, instead re-establishing epistemic symmetry through interactional practices.
Resistive behaviors are also associated with the negotiation of power in interactions (Haworth, 2006). According to Foucault (1978), resistance is another form of power often manifesting as a “tactical reversal” of domination by subjugated parties. Therefore, analyzing resistance may firstly reveal underlying power in RT intervention interactions. On the other hand, power is also inextricably linked with knowledge (Foucault, 1978: 27), yet few studies explore resistance through lens of power and epistemics. Furthermore, resistance often strains affiliation or alignment between participant and therapist (Muntigl, 2013), examining specific instances of resistance can reveal how it may negatively impact the collaborative relationship between caregivers and older adults with MCI.
This study aims to describe the precursors to resistance when caregivers introduce RT as a practice, and how Chinese older adults with MCI resist engaging in RT, and then how the resistance is oriented to by caregivers in RT interactions. Furthermore, it examines power dynamics and interpersonal effects that are realized in the interactional sequences from epistemic perspective.
Literature review
Power is constantly under negotiation, and is always open to resistance. Question-answer sequences serve as a revealing lens into dynamic power relations, resistance patterns, and the ongoing negotiation of epistemic rights and obligations.
Epistemics and questions
It is generally acknowledged that individuals possess epistemic primacy and obligation over knowledge pertaining to their own thoughts, emotions, or lived experiences (Pomerantz, 1980). However, these epistemic rights and obligations can be interactively negotiated or subtly contested through interaction (Heritage, 2012b; Heritage and Raymond, 2005; Mondada, 2013). The act of questioning plays a central role in the dynamic adjustment of epistemic gradients and the negotiation of epistemic stance (Mondada, 2013).
Heritage (2012b) posits that the fundamental premises for executing the action of asking a question are: (1) The questioner certainly doesn’t know the answer (K−); (2) The questioner believes the respondent do know the answer (K+); (3)The questioner has the social right to ask the question; (4) The questioner believes the respondent has obligation to answer. When these conditions are unmet, questions may be perceived as problematic.
Since epistemic access to knowledge is stratified between interactants, different interactant has different presupposition for epistemic status (Heritage, 2012a). Questioner may violate the principle of epistemic congruency in which epistemic stance at talk converges with the epistemic status of the recipient (Heritage, 2012a), and, in order to protect epistemic rights, the recipient may resist such attempts. Notably, epistemic access, primacy and responsibility would influence, and are influenced by, alignment and affiliation in interactions (Stivers et al., 2011: 8). Disregarding clients’ epistemic rights can thus lead to disaffiliation (Muntigl and Horvath, 2014). So epistemic stance may be associated with affiliation and alignment.
Power
Power is fundamentally tied to language and discourse (van Dijk, 2008), often defined as control over others (Fairclough, 1989) or an external factor shaping communication (Brown and Levinson, 1987). However, earlier research overlooked the endogenous, dynamic construction of power (Wu and Yang, 2022). In particular, a paucity of studies has delved into the dynamic nature of power construction and the pragmatic motivations driving its transformation.
Chiang (2009) provides a fresh perspective, suggesting that power can be interpreted as a set of resources utilized by participants to achieve communicative goals. Meanwhile, power is dynamically achieved moment-to-moment and demonstrably exercised during social encounters, with all participants potentially acting as either claimers or ratifiers of power (Schegloff, 1995; Stevanovic, 2018). Consequently, power can be acknowledged, challenged, or denied by the recipients (Wu and Yang, 2022). Clearly, individuals with language and speech impairments are also subject to a much more covert and insidious form of power relations in interactions. The exercise of power in therapist-client interactions is not only consistent with collaborative dialogic practice, but may be necessary to facilitate dialogic communication and ultimately affect therapeutic outcomes (Ong et al., 2023). Question-answer sequences, in particular, can reveal power through the rights and obligations they assign to participants (Heritage, 2005; Wang, 2006).
Resistance
Foucault (1978) claims that where there is power, there is resistance. Clients or patients usually show resistance, affecting the outcome of therapy and adjusting power relations between them (Muntigl, 2013; Tremblett et al., 2024). Resistance, conceptualized as an interactional accomplishment, is manifested by delaying or blocking the trajectory of a certain course of interactional action, or by misaligning with the epistemic, deontic, or affective stances proposed by initiating actions (Dionne et al., 2024). A vast body of Conversation Analytic research has found that individuals use the linguistic, prosodic, sequential, and embodied resources to accomplish resistance (Huma et al., 2023).
Disalignment is central to resistance, as it invariably constitutes a form of non-cooperation with the objectives of a prior action, such as disagreeing with a previous assessment, declining a request, or failing to answer a question (Heritage, 1984), thereby impeding the ongoing interactional project. Thus resistance often hinders the progress of therapy and even results in disaffiliation in health interactions (Muntigl, 2013). However, the ways in which social bonds between caregivers and older adults with MCI become reinforced or weakened, resulting from the ebb and flow of resistance of engaging in RT, have not been interactively displayed.
Moreover, since pragmatic communication involves the interaction of various elements across symbolic, cognitive, motor, sensory and other systems (Huang et al., 2024), power can also be constructed through multimodal resources, including both verbal and nonverbal resources. What is also lacking in previous studies is scrutiny of power and resistance from a multimodal perspective in RT interactions.
Participants and data
Participants
Participants are individuals aged over 60 with Mild Cognitive Impairment (MCI), selected according to Petersen’s criteria. They have completed tests for MMSE, MOCA-B, working memory, semantic memory, and verbal fluency in the authors’ university’s affiliated hospital. Demographic and clinical information were also recorded at the same time (see Table 1). The caregivers, with an average age of 25, had all received training and held professional qualifications in reminiscence therapy. The Research Ethics Committee of [removed for review] approved this study, and all participants provided written informed consent.
Participant characteristics.
Data
The data for this study were collected from videotaped sessions of RT interventions. All names are pseudonyms, and participant images have been anonymized. The interactions during Reminiscence Therapy (RT) sessions were transcribed using Jefferson’s transcription system, which provides detailed insights into verbal changes. Additionally, Mondada’s multimodal transcription conventions were employed to capture nuances of nonverbal interactions. The transcription symbols used in the data of this paper are presented in Appendix 1. The inclusion of nonverbal behaviors is meant to portray the interplay of various multimodal resources: gestures, facial expression, and head movement, gaze. We specifically focus on multimodal elements of interactions especially when they are emphasized by older adults with MCI during RT interactions.
Method
Resistance was identified when older adults with MCI showed their unwillingness in conducting reminiscence work. As such, they were not cooperating with their conversational positioning in the caregivers’ turns at talk. Conversation analysis was then conducted on all instances of resistance. Our analysis had a specific focus on: (a) caregivers’ interactional strategies when they introduce RT as a practice, (b) then “resistance” as a conversational strategy and how it is manifested, and then (c) how that resistance is responded to.
Drawing on Heritage’s (2010) framework, we examined how questions impose interactional constraints, including topic control, presuppositions, epistemic stance, and response preferences and how older adults either conformed to or resisted these constraints. Epistemic dynamics (Heritage, 2012a) were analyzed to understand how participants signaled knowledge ownership during questioning sequences.
We then documented how caregivers managed resistance. Special attention was given to both alignment/disalignment and (dis)affiliation with caregivers’ initiatives. Finally, adopting a multimodal CA approach (Bryan et al., 1998), we expanded the analysis beyond verbal exchanges to include nonverbal resources (e.g. gaze, gesture, posture). This enabled examination of resistance and power dynamics through integrated verbal and embodied practices.
Result
Like dementia quizzes, RT can also be seen as a “social practice” according to social practice theories (Webb et al, 2024). We found older adults with MCI displayed resistance to reminiscence practices framed as questions. We first examine caregivers’ interactional strategies for introducing RT, and then analyze how older adults’ resistance manifests and is responded to. The analysis demonstrates the role of epistemic access, epistemic primacy, epistemic responsibility and epistemic asymmetry in this resistance. Our analysis highlights asymmetrical knowledge regarding older adults’ life experience, particularly private experiences they have the right to withhold. Finally, we show how caregivers managed resistance.
Caregivers’ strategies for introducing RT
Resistance often arose from how caregivers initiated RT sessions. In Extract 1, a caregiver’s attempt to structure Reminiscence Therapy (RT) through question reformulation, based on inappropriate presupposition about the older adult’s epistemic territory, not only controlled the topical development of conversations but also served as a significant precursor to resistance (Figure 1).
(Line 22) P1 averts gaze and displays minimal engagement while verbally disattending the topic of memorable life events (“There was nothing”).
下面我们来聊一聊,您一生当中,您觉得最难忘的事情或者最重要的时刻
01
C1:
Let’s talk about the most unforgettable moments or the most important events in your life
02
(7s)
最难忘的时刻啊?
03
P1:
The most unforgettable moment?
嗯,或者最难忘的事情
04
C1:
Yeah:
05
Or the most unforgettable events
啊?::
06
P1:
Ah?::
嗯,可以仔细回忆一下
07
C1:
Yeah, take a moment to reminisce carefully
回忆您的一生,像小时候啊,后来长大,工作,结婚到现在为止
08
C1:
Reflect on your life, like your childhood, growing up, working, getting married,
and up to now
((低头看向桌子))
09
P1:
((looks down to the table))
10
(12s)
我年纪大了,做事也做不来了
11
P1:
I’m in poor health because of advanced age
嗯嗯::
12
C1:
Mm-hmm::
我现在不能自己做好任何事情
13
P1:
and I can’t do anything well by myself now
嗯,我理解,但是现在您回忆您的童年,后来长大后,当你工作的时候,结婚呢
以后 您觉得你一生当中,最让你难忘的事情或重要的时刻是什么?
14
C1:
Mm: I understand, but if you think back your childhood, after growing up,
when you were working and getting married now:
15
C1:
What do you think has been the most unforgettable event or important moment in your life?
P1:
什么重要的时刻?
((用手扶着下巴,注视着 C1))
16
What is important moment?
((Cleaning on chin with hand, gazing at C1))
重要的时刻,比如说,您孩子出生的时候,这应该对那您难忘的了吧?
17
C1:
Important moments, like when you give birth to your child, that must have been quite unforgettable for you, right?
或者您孙子出生的时候,您可能记得很清楚吧。
18
C1:
Or the birth of your grandchild, these are things that you probably remember quite vividly
怎么啦? 出生的时候怎么了?
19
P1:
What’s the matter? What’s the matter with children’s birth?
就是您孩子出生的时候,你有没有觉得很难忘,让你现在都觉得,非常美好
20
C1:
When your child was born, did you find it very unforgettable, that moment, and even
now you still feel it was very beautiful?
In Extract 1, one can see how questions strongly constrained a recipient’s response by its mode and its presuppositions (Schegloff, 1968). Reformulating a question involves paraphrasing or framing topics in a particular way, influencing discussion direction and focus (Sue, 2025). C1 initiates with an open, ostensibly patient-centered question (line 01). However, P1’s difficulty comprehending the open question (clarification requests in lines 03 and 06, embodied through looking down and prolonged silence in line 10) and subsequent shift to discussing current physical limitations (line 11) were treated as non-conforming responses. Topics are sequentially implicated (Schegloff and Sacks, 1973), and topic control is often regarded as an important means to exercise power (Wang, 2006). So C1’s reformulation (lines 16) asserts his power by explicitly rejecting P1’s offered topic (aging/health) and persisting with the original topic. Meanwhile, this reformulation, compounded by exemplification (“like when you give birth. . .,” line 17), presupposed both the existence of such positive memories and P1’s willingness to share them, which isn’t consistent with P1’s epistemic territory and epistemic primacy regarding her personal experience. C1’s persistence in offering institutionally-preferred topics (childbirth, grandchild birth 17, 20) despite P1’s embodied disengagement (line 09) and minimal responses, framed RT work as a constrained practice rather than an open invitation. So this series of moves transforms C1’s initial open question into a precursor for resistance.
The caregiver’s (C2) interactional approach in Extract 2 illustrates how institutional power triggers resistance by challenging older adult’s epistemic rights and responsibilities (Figure 2).
(Line 31) P2 challenges institutional framing: sustained gaze during direct confrontation (“Why are you asking this question?”).
那您:第一任先生是什么时候没了的?有印象吗?
01
C2:
Do you remember when did your first husband pass away?
((摇头))
02
P2:
((Shakes head))
也没有啊:
03
P2:
You don’t, huh. . .
那您第二次结婚是几几年?还有印象吗?
04
C2:
Do you remember which year you got married for the second time?
((摇头))
05
P2:
((Shakes head))
06–18
(13 lines Omitted)
P2:
Mm:
19
(Nods)
C2:
有的:有没有50岁啊?
20
C2:
You were:
21
Were you over 50?
P2:
((摇头))
22
P2:
(Shakes head)
C2:
那有没有 45 岁啊?
23
C2:
So, were you around 45 years old?
P2:
((注视 C2))
24
P2:
((gazes at C2))
25
(1.8s)
C2:
是不是您 45 岁左右的时候,您这个第一任老伴儿没了?
26
C2:
Was it around your 45th year when your first spouse passed away?
27
(2.5s)
P2:
这个忘了 ((目光转向前方))
28
P2:
I forgot that:
29
((gazes away))
C2:
嗯:忘掉了?
30
C2:
Oh, forgot?
C2:
嗯,没有做什么
就是:就是有的时候呢:
我们通过回忆,我们刺激我们的大脑,能够帮助我们的大脑更有活力
33
C2:
Oh::nothing.
34
Just because sometimes::you know::when we recall past things:: we stimulate
our brains::which can help our brains be more vibrant.
In Extract 2, C2 uses documented information and initiates a line of questioning about P2’s deceased spouse (lines 01, 20–21), pursuing a deeply personal, culturally sensitive topical talk. C2’s institutional position grants her interactional authority to probe into P2’s intimate life details. This stands in stark contrast to P2’s epistemic primacy—her inherent right as primary knower of such sensitive experiences. C2’s persistence, employing increasingly specific yes/no interrogatives (YNIs) (lines 23, 26) designed to elicit confirmation (“type-conforming ‘yes’”), demonstrates an exercise of institutional power (Wang, 2006). She attempts to guide P2 by imposing a framework of possible answers, effectively ignoring P2’s asserted epistemic stance (“I forgot”). When P2 ultimately resists not the topic but the practice itself with the meta-question “Why are you asking this question?” (line 31), she directly challenges RT’s validity and benefit. C2’s response (lines 33–34), framing the probe as necessary brain stimulation, serves to justify the institutional intrusion into private knowledge by appealing to the therapeutic rationale of RT. This demonstrates how question constraints can lead to resistance and alter the course of agenda (Stivers and Hayashi, 2010). The interaction sequence vividly shows how caregiver’s institutional role enables he/she to define relevant topics and pursue answers, ultimately fostering resistance.
Older adults’ resisting practices and caregivers’ corresponding responses
Behaviors that demonstrated their resistance were categorized into six types: (1) implicit resistance through laughter and minimal responses, (2) denying question premises; (3) withholding from taking turns and maintaining long silences, (4) repeat-formatted other-repair initiation, (5) changing the topic, (6) overt refusal due to sorrow. Resistance was often displayed through claims to physical factors (“I’m in poor health because of advanced age” “I forgot it”), emotional factors (“I don’t really want to talk about it”), or discomfort with constraints imposed by questions (“duiba?” “I didn’t really do much work”). These resistive moves and corresponding responses (such as follow-up guidances, topical renewal, implicit prompts, repair and candidate answers, pick up another topic, acceptance) (in Figure 7), dynamically reflect power relations co-constructed by interlocutors, and shape interlocutors’ relationships.
Physical factors
Other repair-initiation not only identifies and initiates the correction of the source of trouble but also delivers composite actions, thus turning the initiation of repair into a composite behavior. For instance, “open” repair initiations (Drew, 1997) primarily indicate the initiator’s difficulty in comprehending the trouble source’s turn at talk, while another behavior conveys to the speaker of the trouble source that the turn performed was inappropriate. Such other-initiated repairs simultaneously perform three actions: (1) questioning the prior turn; (2) treating the turn containing the question as the repair source and initiating a repair; (3) implicitly resisting the practice of RT raised by caregivers. In this context, “questioning” serves as the means to achieve “repair initiation” and “repair initiation” serves as the means to achieve “implicit resistance.”
In Extract 1, when the caregiver (C1) explicitly initiates a question (line 01), the older adult with MCI (P1) subsequently initiates question-intoned repeat-formatted other-initiated repairs, “The most unforgettable thing?” (line 03) and “Ah?” (line 06). C1 initially treats this initiation as a “query” and confirms it with a “Yeah” response (line 04). Subsequently, C1 further treats P1’s repair initiation as a comprehension issue, providing an explanation (“reflect on your life, like your childhood, growing up, working, get married, up to now”). If P1’s repair initiation are to perform only the two main actions of questioning and initiating repair, then the sequence should come to an end at this point. However, P1 looks down at the table and remains silent (line 09–10). And then she claims inability (“I’m in poor health because of advanced age” in line 11; line 13), revealing that P1 is not yet willing to engage in the topic. P1 initiates question-intoned repeat-formatted other repairs again (line 16; line19), and shows resistance to engaging in RT work (“There was nothing” in line 22). Older respondents exhibiting age-related cognitive decline appear to be at a heightened risk of either failing to comprehend complex questions or terminating the question-answering process, which is attributable to the necessary cognitive processes are too burdensome for them (Ziniel, 2008).
Epistemically, knowledge distribution creates gradients where participants hold K+ (greater access) or K− (lesser access) positions, which is termed epistemic status (Heritage, 2012a). Type-1 knowledge (personal experiences) typically grants epistemic rights to the experiencer (Pomerantz, 1980). On the other hand, the experiencer bears greater explanatory responsibility for Type 1 knowledge (Stivers et al., 2011). So older adult attributes her resistance to physical factors (“I’m in poor health because of advanced age” in line 11). By claiming insufficient access to life experience knowledge (“and I can’t do anything well by myself now” in line 13), she intentionally repositions herself from expected K+ (knowledgeable) to K− (unknowing) on the topics. Since accessing to personal experiences entails epistemic obligations (Pomerantz, 1980), avoiding RT topics constitutes resistance to caregiver’ institutional authority/social power and fosters disalignment.
Extract 2 shows explicit resistance to the caregivers’ question similar to that seen in Extract 1. Older adults with MCI sometimes refrain from responding to caregivers’ questions and instead focus on their prompting practices. Resistance functions at the level of epistemic responsibility (Pomerantz, 1980; Stivers et al., 2011). As subject-actor, P2 holds primary rights and obligations over Type 1 knowables (personal experiences such as bereavement). Yet C2’s institutional role imposes asymmetrical responsibility: P2 is compelled to disclose traumatic memories despite physical factor (cognitive limitations) (line 28–29), while C2 assumes the privileged position of evaluating her epistemic stance. This forced allocation of responsibility constitutes a fundamental power imbalance.
In traditional Chinese culture, the topic of bereavement is often regarded as a taboo (Li et al., 2024) and is rarely discussed in casual conversations. Direct inquiries about the death of someone’s relatives may be perceived as impolite or presumptuous. Moreover, cultural norms exempt individuals from publicly accounting for such sensitive and private Type 1 knowables, granting P2 moral legitimacy to resist. When C2 prompts reminiscence, P2 makes full use of nonverbal resources (nods and shakes hands) to respond to C2 (line 02+05). Based on P2’s responses in line 19 and line 22, the caregiver produces a yes/no interrogative (YNI) designed to elicit a type-conforming “yes” response (Raymond, 2003) in line 23 (“So, were you around 45 years old?”) and line 26 (“Was it around your 45th year when your first spouse passed away?”). However, P2 gazes away and explicitly rejects the imposed epistemic responsibility by claiming ignorance in line 28 (“I forgot that:”). This dispreferred design of P2’s turn not only displays a direct refusal to answer the question, but also challenges C2’s institutional authority, which can be identified as a key mode of resistance to power and control (Harris, 1989: 140–146). P2’s gaze (line 32) and meta-question (“Why are you asking this question?” in line 31) in response to the original question naturally transitions the responder’s role from a recipient of information to an interrogator. This is a form of “resistance” by the respondent. P2 seizes the conversational floor temporarily and claims her epistemic stance, exposing how power relations are constructed through the coercive assignment of epistemic obligations. The extended sequence thus reveals how disalignment emerges through P2’s face-threatening act that challenges institutionalized authority.
Emotional factors
Older adults with MCI also displays resistance through claims to emotional factors, as shown in Extract 3 (Figure 3).
01–06
(lines omitted)
C3:
学习成绩一直很好! 是吗?
07
C3:
Her academic performance has always been excellent, hasn’t it?
P3:
((抬头看向 C3+点头))
(7.7)
((低头+用手擦拭眼泪))
08
P3:
((raises head, looks at C3, and then nods))
09
(7.7)
10
((Lowers head and wipes tears with hand))
C3:
那不是挺好的嘛:您孩子那么优秀:又在日本
11
C3:
It’s very good, isn’t it? Your daughter is so outstanding, and now lives in Japan
P3:
((用手擦拭眼泪))
12
P3:
((Wipes tears with hand))
13–19
(lines omitted)
P3:
((用手抹眼泪))
20
P3:
((Wipes tears with hand))
C3:
那她那么优秀,在日本也是生活的很好。所以您也应该安心才对啊
(2.8)
21
C3:
Since she’s so outstanding and living well in Japan, you shouldn’t have also
been worried about her, should you?
22
(2.8)
23
C3:
骄傲嘛:这种 [
C3:
嘿,她一个人啊。
27
C3:
Oh, she’s alone. . .
(Line 25) P3’s display of distress: lowering head and wiping tears with both hands while rejecting C3’s positive framing (“She’s not married . . . living alone”).
Both resistance and facilitation coexist in Extract 3. The duration of pauses can indicate the amount of time a speaker spends in formulating their subsequent response (Chafe, 1979), suggesting resistance to the question. Topic extension is achieved through implicit prompts, allowing P3 to assert knowledge about why she becomes distressed and cries when thinking of her daughter. P3’s implicit resistance is evident in responses like silence (Heritage and Sefi, 1992). C3’s implicit facilitation was generally also accomplished through silences or confirmation-seeking queries (Schegloff, 1982).
Although the design of P3’s turn, which mainly uses nonverbal resources (lower head and wipe tears in line 25), displays a resistance and hesitation to further reminiscence. Topic initiations can be facilitated by a high tolerance for silence in the relevant sequential environment (Emborg, 2024), following prolonged silence, P3 explains the real reason for her sadness (“She. . . she’s not married. . . living alone” in line 25–26), a dispreferred response marked by nonverbal distress (tears, lowered head), successfully extending the topic and continuing reminiscence. Ibrahim (1999) found that interlocutors who manage to get topics developed mostly control the conversation by ignoring the co-participants’ negative face. In this Extract, it is C3 who extends the topic and controls the conversation, simultaneously harming P3’s negative face and then building up disaffiliation between them. According to Pomerantz (1984), interlocutors can demonstrate their epistemic access to the referent through assessment, and the initiation of the first assessment itself seizes the epistemic primacy. Tag questions and assessment (“It’s very good, isn’t it? Your daughter is so outstanding, and now lives in Japan” in line 11) constrain the C3’s conversational practice and seize epistemic primacy through evaluating P3’s emotional state (Hepburn and Potter, 2011). Here, C3 assumes the right to define P3’s feelings (epistemic primacy), pressuring P3 to accept this interpretation or disclose further. C3’s facilitative practices and repeated tag questions successfully sustain the RT work. Power negotiation thus emerges through contested epistemic primacy manifested by C3’s first assessment and P3’s second assessment). C3’s topic extension superficially facilitates reminiscence but damages affiliation by disregarding P3’s epistemic and emotional autonomy. In China, parents often feel humiliated and would be under tremendous social pressure if a child remains unmarried due to tradition. P3, a 73-year-old mother, has a 50-year-old daughter who hasn’t married yet and immigrated to Japan. P3’s practices indicate her unwillingness, humiliation, and resistance to self-disclosure.
Extracts 4 and 5 also show explicit resistance to the caregivers’ question similar to that seen in Extract 3. And the sequences of them eventually led into a rupture in the conversation (Figure 4).
01–03
(3 lines omitted)
C4:
嗯,对
嗯: : : 我听他们工作人员说 : 您那个: : 太太已经::已经走了: 是吧?
04
C4:
Mm-hmm: okay
05
Mm: I heard from staff here that your: your wife has: has passed away,
is that correct?
P4:
这这这:::::::::::::::::::::
((开始哭泣))
06
P4:
This: this: this:::::::::::::::::
07
((begins to cry))
P4:
fig.:
#图 4
08
P4:
09
fig.:
#figure 4
C4:
[好,我不讲了这个好吧]
10
C4:
[Alright, I won’t talk about it then, Okay?
(Line 08 and 09) P4’s overt refusal: gazing away and gesturing dismissal while pleading cessation (“Don’t talk about this”).
In preceding conversation sequences in Extract 4, P4 recounts his daily life with his wife as if she were still alive. The elderly gentleman is somewhat depressed, and the death of his wife may have been a significant blow. The caregiver (C4) initially draws on interactional resources to elicit an answer from P4. Implicit elicitation is generally accomplished through indirect speech and confirmation-seeking queries in line 05 (“I heard from staff here that your: your wife has: has passed away, is that correct?”) (Schegloff, 1982). P4 repeats “this. . .” (line 06), representing hesitation and grief when his wife’s death is mentioned. His voice trembles for an extended period, and his facial expression appears extremely distressed, culminating in tears (line 07). Moreover, the shift in gaze has a topic termination-implicature (Jefferson, 1984), projecting potential transitions in the topic and indicating a change in the coherence of the discourse. In line 09, P4 gazes away and waves his right hand, indicating his unwillingness to reminisce. At the same time, P4’s design of turn explicitly shows resistance to reminiscence (“Don’t: :don’t: :don’t: ::don’t: don’t talk about this” in line 08). P4’s directive action threatens C4’s face. Consequently, the conversation on the topic of his wife’s death is blocked.
Similarly, another older adult (P5), who has divorced with his wife, overtly refuses to discuss his marriage (“I don’t really want to talk about it” in line 02) and shows deep empathy and doesn’t insist on the topic. Thus, there is a display of disalignment through their resistance, leading to a momentary lapse in talk.
C5:
那您啊:这是工作的事情(.)那您可以讲讲您的什么:结婚的时候的事情;
或者说:后来的:就是::结婚啊家庭啊之类的这些
01
So, what you talk is all about your work (.) Can you tell me about your. . .like
when you got married, or. . . later: you know: things like marriage, family and so on?
P5:
嗯:结婚呢就是:结婚呢反正就是我不想讲这个结婚的事情
02
Well, about marriage:: I don’t really want to talk about it
嗯: : :
03
Oh::
P5:
因为我这个:这个呢:我这个结婚是:婚姻不幸福的
04
Because (.) my marriage is:: my marriage is unhappy
嗯:您婚姻不幸啊?=
05
Oh:: Your marriage is unhappy?=
=不幸福的: :
06
=It’s unhappy.
07–11
C5:
(4 lines omitted)
P5:
别人问我婚姻的事情(.)我说我是一个人的
12
When people ask me about my marriage (.) I said I’m living alone
啊:嗯嗯::我知道我知道:您还是不太愿意讲:是吧?
13
Ah: I see:: You’re still not keen on talking about it, right?
嗯,那行,那我们下次再聊,我们聊别的话题。
15
Alright, let’s talk about something else next time.
Extracts 4 and 5 demonstrate explicit resistance was realized by refusing to undertake epistemic responsibility. In Extract 4, C4 invokes knowledge (K+) about P4’s wife’s death (line 05). P4 resists reminiscence by asserting ultimate epistemic access (K+) and rejecting epistemic responsibility through emotional breakdown and commands (lines 06–08). His embodied actions (gaze aversion, hand-waving) defend epistemic right, forcing C4 not to claim epistemic primacy (line 10). This halts knowledge transfer, causing temporary disalignment. Similarly, P5 (Extract 5) claims epistemic primacy over his marital narrative (line 02), admitting minimal knowledge (“unhappy marriage,” line 04) while refusing elaboration. P5 neutralizes C5’s conversational control by epistemic stance. C5’s compliance (line 15) avoids disaffiliation by honoring P5’s epistemic right. Power negotiation occurs when older adults utilize epistemic right to override caregivers’ institutional authority. Resistance thus constitutes epistemic boundary work, where disalignmention arises from violated epistemic rights and resolves through deference to elders’ epistemic authority.
Discomfort with constraints imposed by questions
In the Extracts 6 and 7, the caregivers start with an opener that imposes constraint for the older adult to pick a topic and develop it themselves, but older adults with MCI show their implicit resistance to reminiscence and challenge caregivers’ conversational dominance, reflecting dyadic tension (Figure 5).
C6:
那您:觉得自己来说()比如说()小的时候()然后:就是十几岁的时候()
二十多岁()三十多岁()再到五六十岁()对吧?这一个一小时时间段:
你觉得每个时间段()比较重要的。事件或者说*转折点是什么样的?
(*转头,眼睛注视 Z *)
01
So, as for yourself ( ) Like ( ) when you were a child ( ) and then ( ) you know in your
teenage ( ) your twenties thirties ( ) and then in your fifties and sixties ( ) right? what,
do you think, are the more significant events
02
*or turning points in each of these periods ?*
03
((* Moves head and directs gaze towards Z *))
04
(14 lines omitted)
C6:
有的时候()要回忆一点过去了()做点总结,对吧?
18
Sometimes (.) you need to reminisce (.) and reflect about the past, Duiba?
C6:
[不想回忆?对,¥ 就是被我* 逼着† *在回忆 ¥
((*转过头,眼睛看向 Z*)
21
[¥ Don’t want to reminisce? Yeah, it’s like
22
*I’m forcing! you to reminisce*¥
((*Moves head and gazes towards Z*))
C6:
其实反过来说()是不是您觉得自己()嗯()年轻的时候()其中有一个转折
点()其实就是:身体下降()身体不好()您觉得这是一个转折点?
23
Actually, in other words (.) don’t you think that (.) when you were young (.)one of the
24
turning points was actually that: your health declined (.) you weren’t well: (.)
25
and you think this was a turning point?
P6:
((点头))
26
((Nods))
C6:
因为您不是一直身体不好 (.) 就像上一次 (.)您说了(.)
27
C6:
since you’ve always been in poor health (.) as you mentioned last time
(Lines 19–20) P6 covers mouth while laughing and gazing at C6 during refusal (“I don’t want to reminisce”).
In Extract 6, caregivers, as institutional representatives, inherently occupy a position of epistemic authority (K+) concerning the institutional agenda and record-keeping, while older adults hold epistemic primacy (K+) over their personal life experiences. The Chinese particle “对吧” (Am I right?) (line 18) seeks the interlocutor’s (P6) alignment with C6’s opinion in Chinese, with the aim of establishing an alliance for the beginning of RT in Extract 6. This question also implicitly positions C6 as the director of the interaction, wielding the institutional power inherent in initiating sequences and defining the agenda. But rather than responding with “yes,” P6 explicitly shows reluctance (“I don’t want to reminsce” in line 19) to enter into reminiscence session. This rejection constitutes a powerful assertion of her epistemic right not to disclose personal experiences. At the same time, P6 laughs loudly while covering her mouth with her hands, indicating she is aware of not aligning with the caregiver’s agenda. P6’s dispreferred response embodies much resistance to reminiscing important turning points of her life. And she tries to mitigate embarrassment and potential face-threatening acts with laughter (Jefferson, 1984), which is also regarded as passive resistance when encountering difficulty (Tremblett et al., 2024; Lihe Huang et al., 2023). In terms of sequence organization, that resistance tends to block RT progressivity. Thereafter, the caregiver affiliates with P6 by acknowledging her unwillingness humorously with laughter (line 21–22), enhancing affiliation by maintaining P6’s positive face. C6 continues RT work by proceeding to ask follow-up specific questions (line 23–25), asserting control over the interaction. C6’s reference to prior knowledge (“as you mentioned last time” in line 27), follow-up confirmation questions and yes/no interrogatives (line 25) are geared to encourage P6 to reminisce and share more personal experiences. P6’s resistance is overturned by C6’s facilitative practices. And the topic is driven forward successfully because of the addition of new K+/K− contributions (Heritage, 2012a), demonstrating C6’s superior topic-control power.
Questions set an agenda in a variety of ways (Heritage, 2010), usually steering the topic of conversation through interrogative pronouns (for example, What, Where, When, How), embodying an inherent epistemic asymmetry where the questioner claims a K− (unknowing) position while projecting the recipient as K+ (knowing) regarding the solicited information. In Extract 7, the older adults with MCI show their implicit resistance also because of misaligned questions (Figure 6).
(Line 13) P7 shifts the topic: turning toward the caregiver with a resigned smile while dismissing work history (“I didn’t really do much work”).
C7:
那就是您,呃,之前刚刚提到就是有好几份工作对吧?
您现在印象最深的就是哪一份工作?
01
So, you’ve mentioned that you’ve held several positions, right?
Which is the most fresh one in your memory?
(1.3)
P7:
我我在邮局里做啊?
(1.3)
02
When I was working at the post office?
C7:
嗯,就是之前做的所有工作里面,你有没有印象比较深的
然后可以讲给我下
03
C7:
Yeah, is there one position that you remember more clearly, and you could tell me
about that?
P7:
我嘛. . . . . .不做就回来了嘛
(2.5)
改改改:少了我就回来了吗?
我 45 岁就回来了。
05
Well, I didn’t really do much work. I just came back home after retirement
(2.5)
06
My company underwent reform and reorganization several times,
employees had been continuously laid off. And later on, I came back home after
retirement by the age of 45!
07–10
(4 lines omitted)
C7:
嗯,当时就是都是在那个企业里。都具体做过一些什么呢?
11
Well, what exactly did you do in your company at that time?
15
C7:
(1.4) 就让您退休啦!
(1.4) They asked you to retire, and you just [ah
When the caregiver (C7) asks a question that aims to invite the older adult with MCI (P7) to talk about her positions in company, P7 displays resistance to the topic by denying the premise that she held specific roles in the company (line 05), transforming the question’s agenda from recalling specific work experiences to explaining her early retirement (“I came back home after retirement “ in line 06). This shift is not merely a topical evasion but an assertion of epistemic primacy: P7 counters the institutional agenda (seeking detailed work history) by asserting her own epistemic territory (the circumstances and consequences of her retirement), thereby challenging C7’s interactional power to define the relevant topic. Despite C7’s persistence in reintroducing the same topic with more specific questions (“what exactly did you do” in line 11), P7 negates the presupposition again and redirects the conversation to another topic, sustaining eye contact and smiling (line 13). This response can be characterized as agenda transformation, as it tries to alter the focus and presuppositions within the question (Stivers and Hayashi, 2010). C7 fails to battle resistance and defers to her, letting her introduce her own topic (line 15), showing alignment through topic concession. The practices of C7 and P7 also demonstrate that topic maintenance and development are accomplished by the co-interactants’ cooperation (Janusz et al., 2021).
Abrupt topic shifts signal conversational power because it helps the topic shifter dominate the conversation (Sacks, 1992). When a topic introduced by a question isn’t picked up, the questioner loses the positions. In Extract 7, faced with P7’s persistent resistance and agenda transformation, C7 ultimately aligns with P7 by relinquishing her attempt to control the topic. This marks a breach of the balance of power in favor of C7. However, it aligns with the overall goal of RT interaction to get older adults to talk more, even to talk “out of turn.” P7 competes with C7 for topic control in conversation, indicating that the status of interlocutors in this extract is fluctuates and power is negotiated. Ultimately, P7 attains status and asserts power over C7 during the conversation. C7’s claim of power in conversation does not mean that P7 must accept it, and P7 may reject or adjust this power. This process reflects the dynamic negotiation of power in collaborative interaction. Thus, power performativity highlights the dynamic and non-summative feature (Figure 7).
Resistance behaviors. The first column denotes the conversational behaviors that reflect older adults’ resistance to engaging in RT work. The second column indicates the facilitative behaviors used by caregiver in response. This framework helps to understand how resistance and facilitation interact to shape the RT flow during Reminiscence Therapy sessions.
Discussion
This paper has examined the interactional precursors to resistance during reminiscence therapy (RT) sessions conducted by caregivers for older adults with mild cognitive impairment (MCI). We details specific questioning strategies used by caregivers to introduce RT sessions. It reveals that caregivers’ interactive practices when questioning—such as transforming open invitations into constrained questions or dismissing the elderly-initiated topics—often trigger explicit and straightforward resistance from older adults. In response, caregivers tend to either accept the resistance or prompt topic development implicitly, though institutional power (or social right) is sometimes reasserted through persistent and culturally insensitive questioning. Furthermore, the analysis demonstrates that resistance is systematically tied to issues of epistemic positioning and primacy. Specifically, resistance frequently arises when caregivers make presuppositions that misalign with the older adults’ actual epistemic access. While older adults retain epistemic authority over their personal life and experiences, caregivers often exercise institutional authority in selecting and constraining topics. We show that resistive interactions emerge when caregivers’ questions conflict with the older adult’s epistemic access or infringe upon their epistemic rights.
Secondly, we have explored the varied ways in which older adults manifest resistance to RT. Their resistive practices include other-initiation of repair, denial of question premises, topic shifts, overt refusals, extended silences, laughter, minimal responses, and a range of nonverbal cues (e.g. gestures, gaze aversion, head movements). This multimodal conversation analysis extends prior work on dispreferred responses (Schegloff, 2007a, 2007b), showing how questions function as instruments of power (Wang, 2006) and especially how resistance directly challenges institutional authority (Pomerantz, 1984). Older adults’ resistance displays relatively explicitly and straightforwardly, which may be related to Chinese older adults’ higher power because of their advanced age in Chinese context (Ma and Huang, 2024). Furthermore, the paper illustrates how RT practice is interactionally co-constructed: caregivers initiate sessions, older adults resist, and caregivers subsequently manage that resistance, leading either to continuation or termination of the reminiscence talk. Through the lens of the Conjoint Co-constituting Model of Communication (Arundale, 2010), we argue that power is dynamically negotiated through turn-taking and sequence organization. The power relations between caregivers and older adults with MCI is like two ends of a jumping board, which can be co-constructed through the dynamic negotiation of institutional authority and epistemic right through both resistive and facilitative behaviors (Figures 8 and 9). This further demonstrates that power is collaboratively constructed, constantly open to challenge (Haworth, 2006). Thus, we recommend that caregivers adopt affiliative practices—such as accepting resistance through laughter—to foster more balanced interactions.
Power contrast between caregivers and older adults with MCI.
Power contrast between caregivers and older adults with MCI.
Thirdly, we have analyzed how caregivers respond to resistance and identified practices that lead to more productive interactional trajectories. Some older adults with MCI display resistance to engaging in RT sessions by claiming their declined memory, which can be mitigated through forced-choice questions that prompt declarative memory (Hall et al., 2018). Other resistance reflects cultural aversions to sensitive topics (e.g. bereavement), particularly within Chinese communicative norms (Li et al., 2024). Critically, we argue that caregivers share responsibility for interactional breakdowns when they pursue epistemically or culturally insensitive lines of questioning. Actually, sensitive topics can be oriented using delicacy features, which can minimize resistance from recipients (Pecanac et al., 2023). Since resistance tends to follow various kinds of imposition from the speakers (Antaki, 2023), we suggest avoiding overly constraining questions after an open invitation so as to minimize resistance, as these often provoke non-cooperation and dispreferred responses. Instead, caregivers should foster affiliative environments that acknowledge residents’ agency, as accepting resistance could support smooth interaction and prevent escalation of resistance (Albury et al., 2022). In summary, this study offers fine-grained insights into the interactional mechanics of resistance in RT sessions, highlighting the need for epistemically sensitive and culturally attuned caregiving practices.
Footnotes
Appendix
The transcription symbols used in the data.
| [ ] | Encloses talk in overlap, that is, when more than one speaker is speaking |
| Fig | The exact moment at which a screen shot has been taken |
| # | Is indicated with a specific symbol showing its position within the turn at talk |
| s::::: | Sustained or stretched sound; the more colons, the longer the sound |
| (0.6) | Silence in seconds |
| (.) | Silence of less than two tenths of a second |
| ((action)) | Double parentheses indicate a description of the actions |
| °quiet° | Encloses talk which is produced quietly |
| = | Links talk produced in close temporal proximity (latched talk) |
| ↑ | Indicates a sudden rise in pitch |
| ¥ | Indicates speaking while laughing |
| ¿ | Indicates a rising intonation |
| * | The caregiver C’s synchronized actions |
| ^ | The older adults’s synchronized actions |
Ethical considerations
The Research Ethics Committee of School of Foreign Studies at Tongji University approved our research (Approval Number: tjsflrec202501) on Month 4, 2025.
Informed consent statements
Participants gave written consent for data collection and signature before starting video recordings.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Project of National Social Science Foundation of China (NSSFC) “A study of linguistic features of cognitive disorders and the machine-learning-based automatic screening” (No. 24BYY120).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
The datasets generated during and/or analyzed during the current study are not publicly available due to confidentiality, but are available from the corresponding author on reasonable request.