Persistence of implicit recognition in neurocognitive diseases

Abstract

Background

Recognition abilities are often considered severely impaired in the advanced stages of Alzheimer's disease and related neurocognitive diseases. However, implicit forms of recognition, supported by emotional and relational processes, may remain preserved despite profound cognitive decline.

Objective

This study aimed to explore implicit recognition in older adults with advanced neurocognitive diseases through repeated interactions with an unfamiliar experimenter in emotionally modulated contexts.

Methods

Forty residents of long-term care facilities were randomly assigned to one of two experimental conditions: a positive emotional condition (warm interaction and emotional expressions) or a neutral emotional condition (detached and factual interaction). An 11-session protocol was designed to capture the progressive emergence of recognition behaviors—verbal and nonverbal, spontaneous or elicited—toward the experimenter. Behavioral responses were analyzed across sessions using statistical analyses examining recognition type, emotional condition, and temporal progression.

Results

A main effect of recognition type was observed, with verbal responses, particularly those elicited by prompts, being predominant. The positive emotional condition significantly enhanced the expression of implicit recognition behaviors. A significant temporal progression was also identified, peaking at session 11, suggesting implicit learning across repeated exposures.

Conclusions

Despite advanced neurocognitive diseases, older adults appear to retain the capacity for new learning mediated by implicit memory. Repeated exposure and emotionally meaningful interactions foster familiarity with the experimenter and highlight the persistence of relational and affective competencies. These findings emphasize the role of affective memory in sustaining interpersonal connections and support emotionally attuned human interaction as a valuable therapeutic approach in institutional care.

Keywords

Alzheimer's disease emotions implicit recognition new encoding older

Introduction

In recent decades, the progressive aging of the global population has led to a significant rise in age-related dependency and the prevalence of neurocognitive diseases (NCDs). Among these conditions, Alzheimer's disease is the most common, accounting for 60% to 80% of cases.¹ NCDs, previously referred to as neurodegenerative diseases, are characterized by irreversible neuronal damage within the central nervous system, leading to a progressive deterioration of cognitive, motor, and behavioral functions.² As the disease advances, individuals with NCDs experience increasing difficulties in performing daily activities, often requiring continuous support. These functional and cognitive impairments, such as prosopagnosia, a deficit in facial recognition frequently observed in advanced stages, are among the primary reasons for admission into long-term care facilities.

Beyond cognitive decline, NCDs progressively impair verbal communication. As language abilities deteriorate, nonverbal modalities gradually become the main channels of interaction. Despite severe disease progression, older adults remain capable of engaging with caregivers and relatives through nonverbal cues.³ Facial expressions, gestures, posture, and eye contact continue to play a critical role in sustaining social and emotional exchanges, even in the later stages of the disease.

While declarative memory is typically affected early in NCDs, implicit memory tends to remain relatively preserved, even in advanced stages.^4–5 Implicit memory encompasses both procedural memory, which supports unconscious skills and habits, and emotional memory. As a core component of long-term memory,⁶ implicit memory is defined by the unconscious retrieval of past experiences, operating independently of conscious awareness.⁷ It relies on automatic processes⁸ and manifests through the influence of prior experiences on current behavior in new situations. Throughout the progression of NCDs, subcortical structures involved in non-declarative memory tend to remain relatively preserved, allowing emotional experiences to be implicitly reactivated.⁹ Supporting this view, Jelicic and collaborators¹⁰ showed that many implicit abilities persist despite disease progression, particularly in emotional, motor, and social domains. These findings underscore the importance of considering implicit memory as a therapeutic resource, one that may contribute to preserving autonomy, communication, and human connection. This perspective challenges the assumption that severe cognitive deterioration impairs the potential for forming new relational bonds or recognition.

Nevertheless, individuals with NCDs are still too often perceived solely in terms of their impairments, leading to social exclusion and misinterpretation of their behavior.¹¹ A common misconception holds that interaction with such individuals is needless, they are presumed unable to remember these encounters due to their advanced condition. Contrary to this belief, several studies^12–13 have shown that, even in severe stages, individuals with NCDs can continue to exhibit emotional responsiveness, largely owing to the preservation of emotional capacities supported by implicit memory. These abilities may underlie the implicit recognition of familiar individuals through unconscious emotional and behavioral responses. Moreover, research on emotional enhancement of memory has demonstrated that emotionally salient information, whether positive or negative, is more effectively encoded and retrieved than neutral content.^14–15 Implicit memory often emerges through the automatic reactivation of emotionally meaningful experiences.⁹

Within this framework, we focus on the concept of implicit recognition, an emerging construct in the field of NCDs that currently lacks a standardized operational definition. We define it as the process by which individuals with NCDs display familiarity toward others (in this case, an experimenter) through verbal or nonverbal behaviors, despite being unable to express such recognition explicitly. In light of the absence of consensus in the literature, we developed an ad hoc observational grid to systematically capture these manifestations. This tool was designed to encompass a range of verbal and nonverbal indicators (pointing gestures, posture, smiling, nodding), thereby allowing for the quantification of subtle behavioral expressions of familiarity across repeated encounters.

Our research aims to investigate the influence of emotional context on implicit recognition, testing whether emotionally positive environments facilitate implicit recognition in older adults with advanced stages of NCDs. Building on the work of Pria Veillon,¹⁶ we hypothesize that repeated and consistent interactions foster the development of implicit recognition toward a previously unknown person (the experimenter). In other words, we propose that individuals with severe cognitive impairment retain the capacity to form new implicit associations. Based on prior research on emotional memory,^17–19 we hypothesize that exposure to a positive emotional condition will significantly enhance implicit recognition compared to a neutral condition. Furthermore, drawing on the findings of Delamarre,³ we expect nonverbal indicators of recognition to be more prevalent than verbal ones. Moreover, we expect that nonverbal recognition will be predominantly expressed through pointing gestures, which represent a salient and intentional form of attention direction toward the familiar other.^20–21 Finally, grounded in the theoretical perspectives of Delage and Lejeune,⁹ who emphasize the fundamental role of implicit memory in relational dynamics, we hypothesize that elicited responses, those triggered by the experimenter's solicitations, will be significantly more frequent than spontaneous ones, reflecting the relational anchoring of implicit recognition.

Methods

Participants

Initially, we contacted 73 families whose relatives met our inclusion criteria; 57 consented to participate, 5 explicitly declined and the remaining families did not respond despite repeated contact attempts. Among the 57 enrolled, 5 passed away during the study period, 6 were withdrawn due to substantial deterioration in behavioral symptoms and 6 were excluded from analyses owing to non-attendance in more than three sessions.

Forty older adults (36 women and 4 men), aged between 78 and 99 years (M = 89.57, SD = 4.98), participated in this study. All had been diagnosed with advanced stages of NCDs and were residing in long-term care facilities in the French departments of Rhône and Haute-Garonne. The main clinical and demographic characteristics of the participants are summarized in Table 1, including age, sex distribution, diagnostic status, cognitive severity as assessed by the Mini-Mental State Examination (MMSE), and educational level. Overall, the sample reflects a broad diversity of social backgrounds and former occupational trajectories, consistent with the heterogeneity typically observed among individuals living in French long-term care facilities (Établissements d’Hébergement pour Personnes Âgées Dépendantes, EHPADs). Inclusion criteria required a score of ≤10 on the Mini-Mental State Examination,²² indicating severe cognitive impairment (M = 4.84, SD = 2.71). Cognitive status was regularly monitored by the care teams. Participants presenting any associated neurological condition other than a major neurocognitive disease (movement disorders, pseudobulbar syndrome, essential tremor, or other intercurrent neurological pathologies) were excluded from the study, based on medical information provided by the physicians responsible for the residents’ clinical follow-up. In addition, the video recordings were reviewed under the supervision of an experienced neuropsychologist to ensure that the behavioral indicators coded as implicit recognition were not attributable to involuntary motor or neurological symptoms.

Table 1.

Participants information.

Sex	Age	Educational level	Diagnostic	MMSE score	Experimental condition
Female	92	Primary education	Alzheimer's disease	9	Positive
Female	87	Secondary education	Mixed dementia	4	Positive
Female	90	Higher education	Mixed dementia	9	Positive
Female	97	Higher education	Alzheimer's disease	5	Neutral
Female	84	Primary education	Alzheimer's disease	3	Neutral
Female	91	Higher education	Alzheimer's disease	1	Positive
Female	93	Higher education	Alzheimer's disease	1	Positive
Female	91	Secondary education	Vascular dementia	10	Neutral
Female	96	Secondary education	Alzheimer's disease	10	Positive
Female	97	Secondary education	Alzheimer's disease	2	Positive
Female	94	Secondary education	Alzheimer's disease	10	Positive
Male	85	Secondary education	Mixed dementia	6	Neutral
Female	91	Higher education	Alzheimer's disease	3	Neutral
Male	88	Primary education	Alzheimer's disease	9	Neutral
Female	77	Primary education	Alzheimer's disease	9	Neutral
Female	99	Primary education	Alzheimer's disease	4	Positive
Female	84	Secondary education	Alzheimer's disease	2	Neutral
Female	96	Primary education	Vascular dementia	2	Positive
Female	89	Secondary education	Alzheimer's disease	4	Positive
Female	97	Higher education	Mixed dementia	9	Positive
Female	88	Secondary education	Alzheimer's disease	3	Neutral
Female	93	Secondary education	Alzheimer's disease	1	Neutral
Male	84	Higher education	Alzheimer's disease	9	Positive
Female	85	Secondary education	Alzheimer's disease	8	Neutral
Female	94	Secondary education	Alzheimer's disease	7	Neutral
Female	92	Secondary education	Alzheimer's disease	8	Positive
Female	96	Primary education	Alzheimer's disease	8	Neutral
Female	78	Secondary education	Mixed dementia	0	Neutral
Female	80	Primary education	Alzheimer's disease	5	Positive
Female	82	Secondary education	Alzheimer's disease	5	Neutral
Male	90	Higher education	Alzheimer's disease	3	Neutral
Female	84	Primary education	Alzheimer's disease	6	Positive
Female	84	Secondary education	Alzheimer's disease	1	Neutral
Female	78	Secondary education	Alzheimer's disease	4	Neutral
Female	93	Primary education	Alzheimer's disease	5	Neutral
Female	93	Secondary education	Alzheimer's disease	1	Positive
Female	97	Higher education	Alzheimer's disease	0	Positive
Female	96	Primary education	Mixed dementia	5	Positive
Female	86	Secondary education	Mixed dementia	1	Neutral
Female	92	Primary education	Alzheimer's disease	4	Positive

The study was conducted in close collaboration with the multidisciplinary teams of the participating EHPADs. To minimize bias in recognition-related responses, participants with major uncorrected visual or auditory impairments were excluded.

Ethical approval for the study was obtained from the Research Ethics Committee of Toulouse (reference: 2022–533). Written informed consent was obtained from the legal guardians or family members of all participants, in accordance with the Declaration of Helsinki. Consent also included authorization for the use of video recordings and personal data, which were anonymized to ensure confidentiality.

Materials

This study assessed implicit recognition through an experimental protocol comprising 11 structured sessions with an experimenter previously unfamiliar to the participants. The first 10 sessions were conducted daily from Monday to Friday over two consecutive weeks (five sessions per week). A final (11th) session, held 6 days after the tenth, aimed to assess whether implicit recognition of the experimenter had developed over time (Figure 1).

Figure 1.

Temporal structure of the experimental protocol.

The protocol involved a relational task using a 40-piece wooden construction game varying in color and shape (Figure 2).

Figure 2.

Construction game.

During each session, the same selection of pieces was randomly arranged on a table prior to the participant's entry into the experimental room. Experimental conditions were standardized across all sessions (Figure 3). All sessions were conducted in the same room, unfamiliar to the participants, to limit the potential influence of previously acquired contextual cues. The experimenter was also unknown to the participants, enabling an assessment of whether novel learning could emerge through repeated interaction.

Figure 3.

Experimental setup of the interaction context.

Two experimental conditions were defined, differing in the emotional condition induced during the interaction between the participant and the experimenter. The forty participants were randomly assigned to one of two experimental conditions: a positive emotional condition (n = 20) and a neutral emotional condition (n = 20). In the positive emotional condition, the experimenter adopted a warm and encouraging demeanor, characterized by frequent smiling, explicit verbal encouragement, and sustained engagement throughout the activity. Conversely, the neutral emotional condition involved a more distant and reserved attitude, in which the experimenter refrained from emotional expression and limited their role to pointing out errors or awkward attempts made by the participant.

Data analysis

Statistical analyses were performed to examine the effects of recognition type (verbal and nonverbal), recognition origin (spontaneous and elicited), emotional condition (positive and neutral), and session on implicit recognition behaviors. Mixed-design ANOVAs and repeated-measures ANOVAs were used to analyze behavioral frequencies across sessions and conditions, with appropriate post hoc tests applied when necessary.

Given the absence of standardized tools for assessing implicit recognition in older individuals with advanced stages of NCDs, we developed a customized behavioral coding grid tailored to the specificities of our protocol. A review of the literature revealed a significant gap in validated indicators of verbal and nonverbal recognition within this population. Consequently, we drew upon frameworks from developmental psychology and integrated insights from foundational works,^3,23 theoretical perspectives,^12,24 and empirical studies.^25–27

All experimental sessions were video recorded to allow for repeated viewing and detailed analysis of interactions. This procedure enabled the identification of both verbal and nonverbal responses indicating recognition of the experimenter, the environment, or the ongoing activity. Verbal recognition was defined by spontaneous responses, contextually grounded references (“like last time”), and fluent speech when participants referred to previous encounters. Nonverbal recognition was indicated by sustained eye contact, expressive facial cues (smiling, attentive gaze), proximity-seeking responses (voluntary touch, leaning toward the interlocutor), and active participation in the interaction.

To ensure objective coding, the video recordings were analyzed in a double-blind procedure by two independent raters from psychology and nursing backgrounds. All video recordings were analyzed independently by each rater. The assessments were conducted individually, without any joint viewing, discussion, or consultation between raters during the coding process. Both raters were unaware of the study's hypotheses and experimental conditions. The behavioral coding grid facilitated the classification of recognition responses along three key dimensions: (1) the type of recognition (verbal and nonverbal), (2) its origin (spontaneous and elicited by the experimenter's questions), and (3) the emotional condition of the interaction context (positive or neutral). In order to capture the embodied dimensions of implicit recognition, we focused on four nonverbal behavioral indicators that may reflect familiarity and interpersonal attunement: pointing gestures, used to reference shared objects or spaces; smiling, as a spontaneous or responsive expression of emotional resonance; postural orientation, such as leaning toward the experimenter as a sign of relational engagement; and nodding, which may indicate acknowledgment or recognition. These responses were selected based on their communicative salience in the context of advanced NCDs, where verbal expression is often compromised. Their occurrence was systematically coded in terms of frequency across all sessions to assess the temporal dynamics and emotional modulation of nonverbal recognition.

All occurrences were coded based on the video data collected across 11 sessions. To ensure the reliability of the coding process, multiple independent observers were trained and conducted individual evaluations. Inter-rater reliability was assessed using the intraclass correlation coefficient (ICC), with values ranging from 0.782 to 1. These scores indicated a high level of agreement, considered good (ICC > 0.75) to excellent (ICC > 0.90) according to current standards. These findings confirm the methodological soundness of the coding system and reinforce the objectivity of the behavioral analysis.

Results

Systematic analysis of the video recordings revealed a clear trend of participants recognizing the experimenter as sessions progressed. We first tallied the number of individuals who exhibited at least one behavioral indicator of recognition toward the experimenter in each session. By session 2, 21 participants demonstrated such behavior; this figure rose steadily to 34 by session 11, indicating a progressive familiarization with the experimenter. Importantly, all participants exhibited at least one recognition behavior toward the experimenter over the course of the 11 sessions. The evolution in the number of participants showing at least one recognition indicator across all 11 sessions is depicted in Figure 4.

Figure 4.

Evolution of the number of participants exhibiting at least one indicator of recognition toward the experimenter across sessions.

Implicit recognition of the experimenter: influence of recognition type, elicitation origin, and emotional condition

Table 2 displays the descriptive statistics for each recognition modality, verbal / nonverbal and spontaneous/elicited, across the two emotional conditions.

Table 2.

Descriptive statistics for implicit recognition modalities of the experimenter by emotional valence.

Emotional valence	N	Spontaneous verbal M (SD)	Spontaneous nonverbal M (SD)	Elicited verbal M (SD)	Elicited nonverbal M (SD)
Positive	20	3.95 (3.69)	0.60 (0.99)	6.00 (3.32)	1.15 (2.76)
Neutral	20	0.75 (1.58)	0.90 (2.10)	3.90 (1.80)	0.05 (0.22)
Total	40	2.35 (3.24)	0.75 (1.63)	4.95 (2.85)	0.60 (2.04)

Table 3 displays the descriptive statistics for the four nonverbal recognition modalities—pointing gesture, smiling, postural orientation, and nodding—across the two emotional conditions (positive and neutral).

Table 3.

Descriptive statistics for the four nonverbal recognition modalities by emotional valence.

Emotional valence	N	Pointing gesture M (SD)	Smiling M (SD)	Postural orientation M (SD)	Nodding M (SD)
Positive	20	8.45 (2.50)	5.55 (1.90)	4.65 (3.03)	3.90 (1.86)
Neutral	20	3.60 (1.54)	3.25 (1.86)	1.55 (1.54)	1.10 (0.85)
Total	40	6.03 (3.20)	4.40 (2.19)	3.10 (2.84)	2.50 (2.01)

A mixed-design ANOVA was conducted with recognition type (verbal and nonverbal) and recognition origin (spontaneous and elicited) as within-subjects factors, and emotional condition (positive or neutral) as a between-subjects factor.

Main effects

A significant main effect of recognition type was observed: F_{(1, 398)} = 238.89, p < 0.0001, $η_{p}^{2}$ = 0.375, indicating a large effect size. Verbal recognition responses were significantly more frequent than nonverbal ones across all conditions. This finding confirmed the dominance of language-based expressions in implicit familiarity.

There was also a strong effect of recognition origin, F_{(1, 398)} = 27.58, p < 0.001, $η_{p}^{2}$ = 0.065, corresponding to a moderate effect size, with elicited recognitions occurring more frequently than spontaneous ones.

The emotional condition of the interaction had a significant impact as well, F_{(1, 398)} = 23.40, p < 0.001, $η_{p}^{2}$ = 0.056, also indicative of a moderate effect, suggested that participants exposed to a positive emotional context exhibited more implicit recognition responses toward the experimenter than those in the neutral condition.

Interaction effects and post hoc analyses

Recognition type and emotional condition

A significant interaction between recognition type and emotional condition was observed, F_{(1, 398)} = 32.74, p < 0.001, $η_{p}^{2}$ = 0.076, representing a moderate effect size, indicating that the difference between verbal and nonverbal recognition was more pronounced in the positive condition.

Post hoc comparisons, adjusted using Holm's method, clarified the significant interaction between recognition type and emotional condition. Verbal recognition was significantly higher in the positive condition compared to the neutral condition (mean difference = 0.26 ± 0.043, p < 0.001). Furthermore, verbal recognition under positive condition significantly outperformed all nonverbal modalities, both in the neutral (mean difference = 0.40 ± 0.027, p < 0.001) and positive conditions (mean difference = 0.44 ± 0.036, p < 0.001), suggesting a combined influence of language and emotional condition on the strength of implicit recognition.

Nonverbal recognition was also modulated by emotional condition, with higher scores observed in the positive condition than in the neutral one (mean difference = 0.15 ± 0.036, p < 0.001). However, this effect was less pronounced than that observed for verbal recognition. No significant difference was found between nonverbal recognition in the two emotional conditions (mean difference = 0.04 ± 0.028, p = 0.156), indicating a relative stability of nonverbal responses across emotional contexts.

Temporal evolution of nonverbal recognition across sessions and emotional condition

As shown in Table 2, pointing gestures were the most prevalent nonverbal indicator of implicit recognition, followed by smiling, postural orientation, and nodding. Participants in the positive condition exhibited consistently higher mean scores across all nonverbal modalities compared to those in the neutral condition. The largest difference was observed for pointing gestures (M = 8.45, SD = 2.50 in the positive condition vs. M = 3.60, SD = 1.54 in the neutral condition), followed by smiling, postural orientation, and nodding. These results suggest that emotionally enriched contexts are associated with a greater frequency of nonverbal behaviors indicative of implicit recognition.

A repeated-measures ANOVA was conducted on the four nonverbal behavioral modalities, pointing, smiling, posture and nodding, across the ten experimental sessions. These indicators had been selected to capture nonverbal expressions of implicit recognition observed throughout the protocol.

The analysis revealed a significant main effect of modality, F₍₃₁₁₄₎ = 23.95, p < 0.001, $η_{p}^{2}$ = 0.387 (large effect size), indicating that these nonverbal signals were not equally expressed. Pointing and smiling were overall more frequent than postural or nodding cues. Post hoc comparisons revealed that pointing behavior significantly increased between session 2 and session 11 (p < 0.001), whereas no other pairwise comparisons reached significance.

A robust effect of session was also found, F₍₉₃₄₂₎ = 7.75, p < 0.001, $η_{p}^{2}$ = 0.169 (large effect size), suggesting that nonverbal expressions of recognition evolved over time. Post hoc comparisons further indicated that nonverbal recognition responses significantly increased by session 11 when compared to several earlier sessions, including session 2 (p < 0.001), session 3 (p = 0.006), session 4 (p = 0.008), session 5 (p < 0.001), session 6 (p = 0.004), session 7 (p < 0.001) and session 9 (p = 0.028).

Additionally, a significant interaction between modality and session, F_(27,1026) = 2.10, p < 0.001, $η_{p}^{2}$ = 0.052 (small to moderate effect), indicated that some nonverbal responses were more sensitive to temporal change than others. Post hoc comparisons provided more granular insights into the temporal evolution of nonverbal recognition behaviors. For pointing gestures, significant increases were observed between session 11 and several earlier sessions: session 2 (p < 0.001), session 3 (p = 0.003), session 4 (p = 0.049) and session 8 (p = 0.006), suggesting a marked reinforcement of this referential behavior over time. For smiling, a significant increase was found between session 7 and session 11 (p = 0.031), indicating a delayed but notable enhancement in emotional expressivity.

Most notably, a strong effect of emotional condition was observed, F_(1,38) = 122.00, p < 0.001, $η_{p}^{2}$ = 0.762 (very large effect size), with significantly more nonverbal recognition responses in the positive condition. Post hoc analyses clarified this effect. All nonverbal modalities were significantly more expressed in the positive condition compared to the neutral one. Pointing gestures (p < 0.001), smiles (p = 0.002), postural engagement (p < 0.001), and nodding (p < 0.001) were all markedly more frequent when the experimenter interacted using emotionally positive cues.

However, interactions involving emotional condition and time or modality did not reach significance, suggesting parallel trajectories across emotional contexts. These findings provide further evidence that implicit interpersonal recognition is not only emotionally modulated but also behaviorally differentiated and temporally dynamic.

Recognition type and recognition origin

Another significant interaction emerged between recognition type and recognition origin, F_{(1, 398)} = 38.04, p < 0.001, $η_{p}^{2}$ = 0.087 (moderate to large effect size), suggesting that the relative frequency of verbal versus nonverbal recognition depended on whether the behavior was spontaneous or elicited. Specifically, elicited verbal recognition was the most prominent modality across conditions.

Post hoc analyses, adjusted using Holm's method, clarified the significant interaction between recognition type (verbal and nonverbal) and recognition origin (spontaneous and elicited) in the implicit recognition of the experimenter. Multiple significant contrasts were identified across conditions.

Spontaneous verbal recognition yielded significantly higher scores than verbal elicited recognition (mean difference = −0.26 ± 0.039, p < 0.001), nonverbal spontaneous recognition (mean difference = 0.16 ± 0.029, p < 0.001,) and nonverbal elicited recognition (mean difference = 0.17 ± 0.029, p < 0.001). These findings indicate a distinctive predominance of spontaneous verbal recognition in the implicit interpersonal recognition processes observed in individuals with NCDs.

Conversely, verbal elicited recognition was significantly less frequent than both nonverbal spontaneous recognition (mean difference = 0.41 ± 0.031, p < 0.001) and nonverbal elicited recognition (mean difference = 0.43 ± 0.029, p < 0.001), highlighting a pronounced contrast between these modalities.

Finally, no significant difference emerged between nonverbal spontaneous and nonverbal elicited recognition (mean difference = 0.02 ± 0.023, p = 0.514), suggesting a relative consistency in nonverbal recognition patterns, regardless of origin.

Recognition type, origin, and emotional condition

Finally, a three-way interaction was found among recognition type, recognition origin, and emotional condition: F_{(1, 398)} = 8.33, p = 0.004, $η_{p}^{2}$ = 0.020 (small effect size). This complex interaction suggests that the combined effects of recognition type and origin were further modulated by emotional condition. Notably, the positive emotional context selectively enhanced elicited verbal recognition, highlighted the modulatory role of emotion in implicit familiarity processes.

Post hoc comparisons, adjusted using Holm's method, of the three-way interaction between recognition type, origin (spontaneous and elicited), and emotional condition revealed several significant effects.

In the spontaneous verbal condition, scores were significantly higher under positive condition than under neutral condition (mean difference = 0.315 ± 0.058, p < 0.001), and also higher than elicited verbal recognition within the same positive context (mean difference = −0.200 ± 0.055, p = 0.004). Moreover, spontaneous verbal recognition under positive condition significantly exceeded all nonverbal modalities, both spontaneous and elicited, in both emotional conditions (positive: mean differences = 0.336 ± 0.041 and 0.300 ± 0.048; neutral: 0.275 ± 0.041 and 0.380 ± 0.049; all p < 0.001).

Conversely, under neutral condition, spontaneous verbal recognition was significantly lower than all other forms of recognition, both verbal (elicited: mean difference = −0.515 ± 0.057, p < 0.001) and nonverbal (mean differences ranging from −0.275 ± 0.055 to −0.380 ± 0.041, all p < 0.001), suggesting an overall reduction in implicit recognition in emotionally neutral contexts.

Elicited verbal recognition was significantly higher under positive condition compared to neutral condition (mean difference = 0.200 ± 0.057, p = 0.006), and also significantly greater than all nonverbal forms, regardless of emotional condition (mean differences = 0.475 ± 0.049, p < 0.001). This pattern supports the notion that positive emotional condition preferentially enhances verbal responses to external prompts.

Regarding nonverbal recognition, differences were generally less marked. A few significant effects were observed, for instance, elicited nonverbal recognition in the positive condition was higher than its spontaneous counterpart within the same context (mean difference = 0.385 ± 0.041, p < 0.001), and also higher than the same elicited modality under neutral condition (mean difference = 0.330 ± 0.047, p < 0.001). However, several comparisons among nonverbal modalities did not reach significance (mean difference = 0.015 ± 0.048, p = 1.00), highlighting the relative consistency of nonverbal responses across emotional contexts.

Temporal evolution of implicit recognition across sessions

Table 4 illustrates the session-by-session trajectory of mean implicit recognition scores toward the experimenter, stratified by emotional condition (positive or neutral).

Table 4.

Mean implicit recognition scores (sd) by session and emotional valence condition.

Emotional valence	N	Session 2 M (SD)	Session 3 M (SD)	Session 4 M (SD)	Session 5 M (SD)	Session 6 M (SD)	Session 7 M (SD)	Session 8 M (SD)	Session 9 M (SD)	Session 10 M (SD)	Session 11 M (SD)
Positive	20	0.50 (0.61)	0.75 (1.25)	1.40 (1.39)	1.35 (1.39)	0.70 (0.57)	1.20 (1.77)	1.00 (1.77)	1.70 (2.79)	1.05 (0.69)	1.90 (1.45)
Neutral	20	0.20 (0.41)	0.50 (1.40)	0.45 (0.51)	0.60 (0.59)	0.45 (0.51)	0.75 (1.16)	0.55 (0.89)	0.80 (1.58)	0.45 (0.51)	0.85 (0.75)
Total	40	0.35 (0.53)	0.63 (1.31)	0.93 (1.14)	0.98 (1.27)	0.58 (0.55)	0.98 (1.49)	0.78 (0.86)	1.25 (2.28)	0.75 (0.67)	1.38 (1.25)

A repeated measures ANOVA was conducted to assess changes in implicit recognition scores across the ten-session protocol.

A significant main effect of Session emerged: F_{(9, 342)} = 2.607, p = 0.006, $η_{p}^{2}$ = 0.064 (moderate effect size), indicating that implicit recognition responses varied significantly over time. Post hoc comparisons using Tukey's test (p < 0.05) revealed that Session 11 exhibited significantly higher scores than several earlier sessions, including Session 2 (p < 0.001), Session 6 (p = 0.014), Session 7 (p = 0.019), and Session 9 (p = 0.009). Taken together, these findings indicate a progressive and cumulative increase in implicit recognition over time, culminating in a marked peak during the final session.

No significant interaction was found between Session and Emotional Condition: F_{(9, 342)} = 0.725, p = 0.686, $η_{p}^{2}$ = 0.019 (small effect size), suggesting that the temporal pattern of recognition was consistent across both emotional conditions.

However, a strong effect of Emotional Condition was observed: F_{(1, 38)} = 6.70, p = 0.014, $η_{p}^{2}$ = 0.150 (large effect size), indicating that participants exposed to the positive condition consistently demonstrated higher levels of implicit recognition than those in the neutral condition, regardless of session.

Discussion

The present study, which aimed to investigate implicit recognition in older adults with advanced stages of NCDs, employed both quantitative and behavioral approaches to better understand the impact of repeated emotionally modulated interactions. Specifically, we examined how recognition type (verbal and nonverbal), recognition origin (spontaneous and elicited) and the emotional condition of the experimental context influenced participants’ implicit recognition of the experimenter, while also tracking changes across sessions.

Several significant main effects emerged. First, implicit recognition expressed through verbal language occurred significantly more often than nonverbal recognition. Second, elicited recognition was more frequent than spontaneous recognition. A robust effect of emotional condition was also observed: participants exposed to a positive emotional context exhibited higher implicit recognition scores than those in the neutral condition. Longitudinal analysis further revealed a significant increase in recognition scores over time, culminating in a pronounced peak at session 11.

Scientific literature on implicit recognition remains extremely limited. This scarcity led us to explore and bridge several theoretical domains. Each of these fields, psychoanalysis, neuropsychology, emotion theory, gerontological clinical practice, and developmental psychology, offers partial insights into the dynamics of recognition and interpersonal connection, yet none directly addresses the specific phenomenon of implicit recognition in NCDs.

To address this gap, in this study we adopted an integrative approach, not as a mere juxtaposition of perspectives but as a dialogue between complementary theoretical frameworks. This articulation allows for a multidimensional understanding of recognition processes, encompassing cognitive, emotional, relational and embodied dimensions.

From a neurocognitive perspective, implicit recognition can be linked to preserved nondeclarative memory systems^6,23,28 and to affective memory mechanisms that remain functional even in advanced dementia.^29,30 From a clinical and relational perspective, interactionist and systemic theories^12,24,31 highlight the persistence of communication and emotional resonance beyond the loss of explicit cognition. Developmental psychology further contributes by illuminating the continuity of embodied communicative forms, such as gaze, gesture, and affective attunement, throughout the lifespan.

By integrating these complementary approaches, the present study advances a theoretical model in which implicit recognition is conceptualized as a process of affective and relational continuity rather than a mere cognitive residue. This model situates recognition at the intersection of emotion, memory, and interaction, suggesting that the capacity to “recognize” the other may persist as a form of embodied relational knowledge, even when declarative awareness has faded. Such an integrative framework provides a coherent foundation for interpreting the complex interplay of verbal, nonverbal, and emotional indicators of familiarity observed in individuals with advanced stages of NCDs.

Verbal and nonverbal implicit recognition

Verbal communication is often considered to be severely impaired in advanced stages of NCDs. Common difficulties include word-finding problems, paraphasia, and mutism.²³ According to this view, verbal expression is thought to be gradually supplanted by nonverbal modes of communication.^3,12

However, the present findings challenge this perspective. Contrary to expectations, verbal implicit recognition occurred significantly more frequently than nonverbal recognition, especially when elicited by the experimenter. This reversal is not a trivial detail but a central finding, suggesting that lexical expression may persist and even be reactivated under emotionally salient conditions, despite advanced stages of NCDs.

This predominance of verbal responses is consistent with prior research by Rousseaux and collaborators,²⁶ who emphasized that emotional engagement during social exchanges can facilitate verbal expression, even in individuals with advanced stages of NCDs. Hence, the emotional context may play a pivotal role in reactivating linguistic automatisms, allowing participants to implicitly express familiarity through verbal cues.

Post hoc analyses further confirmed that positive emotional condition amplified this effect: verbal recognition in the positive condition was significantly higher than in the neutral condition and exceeded all other nonverbal recognition modalities. These findings reinforce the hypothesis that a warm emotional climate, combined with a repetitive interactional framework, fosters the activation of implicit affective memory and its verbal manifestations.

In terms of nonverbal recognition, our results revealed a remarkable stability across conditions, regardless of origin (spontaneous and elicited) or emotional condition. No significant differences were observed between spontaneous and elicited forms, nor between positive and neutral contexts, suggesting that nonverbal communication functions as a relatively autonomous expressive channel, less sensitive to contextual fluctuations. This interpretation aligns with the work of Watzlawick and collaborators²⁴ and Delamarre,³ who emphasize the persistence of analogical expressions (gaze, gestures, posture) even when higher cognitive functions are compromised.

Nevertheless, qualitative analysis revealed a progressive enrichment of nonverbal responses across sessions in the positive condition: increased eye contact, spontaneous smiling, proximity-seeking gestures, self-initiated actions, and emotionally congruent responses. These findings indicate that although quantitatively stable, nonverbal responses may evolve qualitatively, reflecting growing affective engagement.

Thus, while nonverbal communication appears less modulated by experimental conditions in terms of frequency, it plays a fundamental role in establishing and sustaining implicit social bonds, particularly when verbal abilities are diminished. As highlighted by Delamarre,³ nonverbal behavior remains a genuine “language of the body,” conveying emotion, memory, and implicit recognition.

Taken together, these results demonstrate that both verbal and nonverbal recognition contribute complementarily to the expression of implicit recognition in older adults with NCDs. Verbal expression appears more sensitive to emotional context, whereas nonverbal communication ensures expressive continuity, reflecting progressive familiarity and relational safety. These findings advocate for the rehabilitation of preserved expressive capacities in individuals with neurocognitive conditions and underscore the importance of interactional strategies that promote both verbal and nonverbal engagement.

Beyond the mode of expression, the origin of implicit recognition emerges as another critical factor in the analysis of relational dynamics, warranting further investigation.

Temporal evolution of nonverbal recognition across sessions and emotional condition

Our findings highlight the persistence and specificity of nonverbal expressions of implicit recognition in individuals with advanced stage of NCDs. Despite limited verbal abilities, participants displayed consistent and differentiated behavioral cues, particularly pointing and smiling, suggesting preserved capacities for relational engagement through embodied modalities. This aligns with clinical literature emphasizing the communicative value of the body even in advanced stages of NCDs,^3,12 and with the foundational premise that “one cannot not communicate”.²⁴

Pointing gestures were the most sensitive to temporal progression, with significant increases across sessions, particularly between session 2 and session 11. These gestures likely served a referential function, indexing the recognition of the experimenter. While largely studied in early development,^20,21 pointing may reflect a preserved form of intentional communication grounded in implicit memory mechanisms, such as familiarity or procedural learning.³²

Smiling also increased significantly but with a delayed onset, suggesting a progressive affective attunement fostered by repeated interactions. These smiles, interpreted as expressions of satisfaction rather than social convention, may indicate the emotional resonance of the experimental setting.^27,33 Nodding, though less frequent, appeared as a subtle cue of assent or alignment, while postural engagement reflected a broader motivational stance rather than direct recognition.²⁷

Together, these results provide compelling evidence that nonverbal implicit recognition is not only emotionally modulated but also modality-specific and temporally dynamic. The progressive reinforcement of certain gestures supports the hypothesis that repeated emotionally enriched encounters can foster embodied recognition, even in the absence of explicit memory. These insights open promising avenues for rethinking communication and relational care in advanced stages of NCDs.

Spontaneous and elicited implicit recognition

Our findings revealed a significant main effect of the origin of recognition: elicited recognition, triggered by the experimenter's prompts, occurred more frequently than spontaneous recognition. This result highlights the proactive role of interaction in activating memory capacities in individuals with advanced stages of NCDs.

These findings align with models that conceptualize implicit memory as a reactive system, one that is more readily activated by contextual or emotional cues than by autonomous retrieval.^6,8 Such models emphasize the importance of environmental and emotional stimuli in reactivating implicit traces. In our study, both verbal and nonverbal cues provided by the experimenter appeared to serve as external triggers, facilitating recognition responses even when participants’ spontaneous communicative initiative was compromised by their condition.

These challenges a widely held assumption in the literature, that the most authentic and meaningful forms of recognition are necessarily spontaneous. On the contrary, our results suggest that elicited responses in interactive settings may constitute a privileged access point to implicit memory traces. Rather than being dismissed as experimental artifacts, these responses should be viewed as meaningful and legitimate indicators of implicit recognition.

Furthermore, our results underscored the central role of the experimenter in activating abilities in older individuals with advanced stages of NCDs. They emphasized the importance of interaction and more specifically, relational attunement in fostering implicit recognition, even in the absence of spontaneous expression. This dynamic supports the notion of interaction as a therapeutic or relational lever for reactivating preserved memory systems, particularly in individuals with diminished communicative initiative. Such a view is consistent with interactionist models of care, which stress the need for attentive and adaptive relational engagement, especially from professionals in institutional settings, to support the expression of presence, connection, and even recognition.¹²

Implicit recognition and the role of experimental emotional condition

Our findings revealed a significant effect of emotional condition on participants’ implicit recognition of the experimenter. Participants exposed to the positive condition demonstrated higher levels of implicit recognition compared to those in the neutral condition. This effect was observable from the early sessions and tended to increase over time, which suggested that emotionally enriched interactions progressively reinforce mechanisms of social familiarity. These results support a growing body of evidence indicating that emotion, particularly positive emotion, acts as a facilitator of memory processes, including implicit memory. As highlighted by Damasio^34–36 and Eustache and collaborators,³⁷ positive emotions serve as powerful catalysts for the consolidation and retrieval of memory traces. In individuals with advanced stages of NCDs, this emotional dynamic appears to play a key role in reactivating familiarity, even when explicit recognition is severely impaired.^6,28

The main effect of emotional condition confirms that the relational climate in which interactions take place is crucial for implicit encoding. Participants exposed to the positive condition (characterized by warm interactions, joyful facial expressions, and verbal encouragement) showed significantly higher levels of implicit recognition than those in the neutral condition. In other words, implicit recognition scores were significantly higher in emotionally positive interactions, supporting the hypothesis that affective engagement facilitates implicit familiarity. These findings align with the literature on the emotional enhancement of memory, which demonstrates that emotional condition enhances both encoding and implicit retrieval.^14,15,38

This influence of emotion also echoes the work of Ekman,^39–41 who demonstrated that emotional expressions, particularly micro-expressions, are essential components of nonverbal communication, remaining active even in the presence of severe cognitive decline. In our protocol, the experimenter's emotional expression, conveyed through both verbal and nonverbal positive cues, likely contributed to eliciting affective responses and fostering a sense of implicit familiarity.

Furthermore, our results can be interpreted in light of face recognition models.^17,42 Repeated exposure to the experimenter's face in an emotionally supportive context seemed to foster a sense of implicit familiarity. Although participants were unable to explicitly name or identify the experimenter, they consistently exhibited behavioral cues, such as sustained eye contact, smiling, and contextual references, that signaled implicit recognition. These findings are consistent with studies on prosopagnosia by Bauer,^43,44 which show that emotional recognition can persist despite the loss of conscious identification.

This phenomenon further supported the hypothesis of a preserved dorsal route for recognition, more implicit and affective, distinct from the ventral stream associated with explicit identification. Even when declarative memory is profoundly compromised, emotionally enriched relational dynamics and repeated exposure may activate these preserved neural pathways. Our observations align with those of Jelicic and collaborators,¹⁰ who demonstrated that patients with advanced stages of NCDs can exhibit meaningful implicit responses even when explicit memory is severely impaired. These findings reinforce the notion that implicit recognition can be supported through consistent affective and social stimulation, offering therapeutic value by leveraging preserved abilities. They also resonate with the Preserved Implicit Memory (PIM) model proposed by Harrison and collaborators,⁴⁵ which emphasizes the potential for individuals with NCDs to mobilize implicit skills within adaptive, emotionally responsive interactions, despite extensive declarative memory loss.

Longitudinal dynamics of implicit recognition

A longitudinal analysis of the ten experimental sessions revealed a significant, though non-linear, trajectory in implicit recognition scores. Notably, a peak emerged during session 11, conducted six days post-intervention, suggesting a delayed consolidation effect of implicit memory. This behavioral stabilization points to the emergence of familiarity-based recognition. Strikingly, this phenomenon occurred in individuals with advanced stages of NCDs, underscoring the resilience of implicit learning mechanisms despite severe cognitive decline.

The lack of a significant interaction between session progression and emotional condition indicates that these dimensions may function independently. Repetition appears to facilitate a gradual process of relational habituation, whereas emotional condition enhances the affective resonance of the interaction. These findings are consistent with prior research,^12,42 which emphasized the importance of relational continuity and emotional quality in maintaining affective engagement through implicit processes.

Qualitative observations corroborated this trajectory, revealing a progressive increase in behavioral engagement. Participants exhibited more sustained eye contact, spontaneous smiling, and proximity-seeking behaviors, such as gently touching the experimenter's arm. Contextual verbal references (“like last time”) also emerged, suggesting implicit reminiscence. Although often subtle, these manifestations reflect enhanced emotional security and an evolving sense of relational familiarity.

Taken together, these findings suggest that, even in the context of advanced stages of NCDs, older adults retain the capacity to engage in relational processes when situated within stable, repetitive, and emotionally supportive environments. Implicit recognition thereby emerges as a meaningful proxy for preserved affective memory, offering promising avenues for therapeutic intervention in institutional care settings. Despite the absence of explicit verbal recall, participants’ responses consistently revealed an implicit recognition of the experimenter.

Limits

While the present study provides novel insights into implicit recognition processes among older adults with advanced NCDs, several limitations should be acknowledged to contextualize and qualify the findings.

Although the eleven-session protocol allowed us to capture a progressive emergence of implicit recognition, the observation period remains relatively short. It does not provide sufficient insight into the long-term stability or consolidation of implicit recognition behaviors. Future studies should consider extending the observation window over several weeks or months to determine whether implicit familiarity can be maintained, fade, or be reinforced over time, especially in individuals with advanced stages of NCDs whose cognitive functioning fluctuates considerably.

The coding grid used to assess implicit recognition was developed specifically for this study, due to the lack of standardized tools in this population. While inter-rater reliability was good to excellent, future studies should aim to validate this grid or compare it with other behavioral or physiological markers of implicit familiarity.

In addition, the sample was predominantly composed of female participants, reflecting the demographic characteristics of very old populations in institutional settings. While this distribution limits the generalizability of the findings, especially with regard to potential sex differences in implicit recognition, current evidence on sex-related differences in implicit memory remains limited. Future studies including more balanced samples would be necessary to explore this issue more thoroughly.

Although participants with associated neurological disorders were excluded based on medical records, a formal neurological assessment of the video recordings was not conducted. Future studies would benefit from a multidisciplinary approach, including a neurologist's evaluation, to further differentiate relational nonverbal behaviors from involuntary motor phenomena.

Conclusion

This study underscores the persistence of implicit recognition in older adults with advanced stages of NCDs. Our findings suggest that, although often not verbalized, implicit recognition can be triggered through consistent interpersonal interactions and is reinforced by emotionally positive contexts. Verbal recognition elicited by prompts emerged as particularly salient, highlighting the critical role of the interlocutor in reactivating memory traces. Furthermore, the positive emotional condition of the experimental setting acted as a facilitating factor, emphasizing the central role of emotion in mechanisms of implicit familiarity.

In line with the work of Blessing and collaborators,⁴² our results indicate that emotionally driven implicit learning can endure in individuals with advanced stages of NCDs, provided that it is supported by a stable and emotionally engaging relational environment. These findings reinforce the hypothesis that affective memory continues to structure interpersonal bonds over time.^4,5 The experimenter, emotionally salient due to repeated interactions, was implicitly recognized by most participants, not through explicit naming, but via behavioral cues indicative of familiarity.

Taken together, these results call for renewed attention to the role of affective exchange and support the integration of human interaction as a key therapeutic lever in the care of individuals with NCDs.

The present findings carry significant clinical implications for long-term care practices. In particular, they suggest that implicit recognition can serve as a therapeutic lever within the caregiving relationship, offering a valuable resource for professionals working with individuals in advanced stages of NCDs. Far from being limited to residual automatisms, implicit recognition reflects an ongoing capacity for relational engagement, one that persists beneath the surface of verbal or explicit communication. Recognizing and supporting this capacity requires a shift in caregiving practices from a deficit-based model to a relational framework grounded in emotional continuity and intersubjective presence.

For healthcare professionals, this implies the need for a sensitive clinical posture that actively acknowledges and responds to subtle behavioral cues (gaze, smiles, gestures) as meaningful indicators of preserved affective connection. Regularity in interactions, emotional attunement, and the establishment of familiar relational patterns appear to be key conditions for sustaining implicit recognition. In this regard, caregiver training programs should be expanded to include modules on embodied communication, affective memory and the recognition of low-intensity or ambiguous signs of familiarity.

These insights also opened new avenues for research. Further studies could explore the role of implicit recognition in buffering caregiver burden, improving institutional adjustment, and supporting dyadic resilience in the context of progressive cognitive decline. Investigating how implicit recognition affects the subjective well-being and ethical positioning of caregivers, both professional and familial, may provide a more comprehensive understanding of its role in care settings. Future research may also benefit from longitudinal and interdisciplinary designs that integrate behavioral, emotional, and neurophysiological indicators of implicit relational processing, thereby refining our understanding of how recognition operates beyond explicit recall.

Ultimately, by foregrounding implicit recognition as a marker of relational vitality, this study invites a redefinition of care in advanced stages of NCDs not as the management of loss, but as the co-construction of meaning and presence in the face of cognitive dissolution.

Footnotes

Acknowledgements

We would like to thank all the staff at the Groupe ACPPA and the ORPEA Group, who supported this research by hosting one of the authors as part of their experimental initiative on the theme of implicit recognition.

ORCID iD

Ingrid Hennebert

Ethical considerations

The research protocol was approved by the Ethics Committee of the University of Toulouse Jean Jaures (approval number: 2022–533) on October 14, 2022. Written informed consent was obtained from participants’ legal guardians prior to the beginning of the experimental sessions. Consent also included authorization for video recordings and the use of anonymized personal data, in accordance with ethical standards and the Declaration of Helsinki.

Consent to participate

All participants’ legal guardians provided written informed consent for their participation in the study prior to the start of the sessions. The requirement for written informed consent was reviewed and approved by the Research Ethics Committee of the University of Toulouse Jean Jaures (approval number: 2022–533, dated October 14, 2022).

Consent for publication

Consent for publication of anonymized personal data and video recordings was also obtained in writing from the participants’ legal guardians. Identifiable information was excluded from the manuscript in accordance with ethical standards and the journal's policies on research integrity.

Author contribution(s)

Ingrid Hennebert: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing – review & editing.

Thierry Darnaud: Writing – review & editing.

Christine Vanessa Cuervo-Lombard: Writing – review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

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 and analyzed during the current study are not publicly available due to ethical and confidentiality restrictions related to the protection of participants living in long-term care facilities. Anonymized data may be available from the corresponding author upon reasonable request and subject to approval by the relevant ethics committee.

References

Alzheimer’s Association. 2022 Alzheimer’s disease facts and figures. Alzheimers Dement 2022; 18: 700–789.

Hirsch

Brugeron

. Avant-propos. Les maladies neuro-évolutives. Vivre avec, vivre ensemble. In: Hirsch

(eds) Vivre avec une maladie neuro-évolutive: Enjeux éthiques et sociétaux. Toulouse: Érès, 2019, pp.15–18.

Delamarre

. Communication verbale et non verbale. In: Delamarre

(eds) Alzheimer et communication non verbale: Maladie d’Alzheimer et maladies apparentées. Paris: Dunod, 2014, pp.29–38.

Son

Therrien

Whall

. Implicit memory and familiarity among elders with dementia. J Nurs Scholarsh 2002; 34: 263–267.

Quoniam

Ergis

Fossati

, et al. Implicit and explicit emotional memory for melodies in Alzheimer’s disease and depression. Ann N Y Acad Sci 2003; 999: 381–384.

Graf

Schacter

. Implicit and explicit memory for new associations in normal and amnesic subjects. J Exp Psychol Learn Mem Cogn 1985; 11: 501–518.

Schachter

Singer

. Cognitive, social, and physiological determinants of emotional state. Psychol Rev 1962; 69: 379–399.

Hasher

Zacks

. Automatic and effortful processes in memory. J Exp Psychol Gen 1979; 108: 356–388.

Delage

Lejeune

. La mémoire sans souvenir. Paris: Odile Jacob, 2017.

10.

Jelicic

Bonebakker

Bonke

. Implicit memory performance of patients with Alzheimer’s disease: a brief review. Int Psychogeriatr 1995; 7: 385–392.

11.

Van der Linden

A-C

. Changer les représentations de la démence: Un long et difficile chemin!. Mythe Alzheimer 2017.

12.

Ploton

. Ce que nous enseignent les malades d’Alzheimer. In: Arfeux-Vaucher

(eds) Les démences au croisement des non-savoirs: Chemins de la complexité. Rennes: Presses de l’EHESP, 2012, pp.17–34.

13.

Gil

Arroyo-Anllo

. Émotions et maladie d’Alzheimer: neuropsychologie et enjeux éthiques. Neurol Psychiatr Gériatr 2019; 19: 233–240.

14.

Ochsner

. Are affective events richly recollected or simply familiar? The experience and process of recognizing feelings past. J Exp Psychol Gen 2000; 129: 242–261.

15.

Hamann

. Cognitive and neural mechanisms of emotional memory. Trends Cogn Sci 2001; 5: 394–400.

16.

Pria Veillon

. Le lien conjugal à l’épreuve de la maladie d’Alzheimer: une étude exploratoire auprès de quatre couples, lorsqu’un des conjoints atteint de démence est placé en institution. Thérap Familiale 2017; 38: 71–87.

17.

Bruce

Young

. Understanding face recognition. Br J Psychol 1986; 77: 305–327.

18.

Blanc

Syssau

Brouillet

. Émotion et cognition. Quand l’émotion parle à la cognition. Paris: IN PRESS Édition, 2006.

19.

Imbeault

Gagnon

Pigot

, et al. Impact of AP@LZ in the daily life of three persons with Alzheimer's disease: long-term use and further exploration of its effectiveness. Neuropsychol Rehabil 2018; 28: 755–778.

20.

Camaioni

Perucchini

Bellagamba

, et al. The role of declarative pointing in developing a theory of mind. Infancy 2004; 5: 291–308.

21.

Colonnesi

Stams

GJJM

Koster

, et al. The relation between pointing and language development: a meta-analysis. Dev Rev 2010; 30: 352–366.

22.

Folstein

McHugh

. Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198.

23.

Eustache

Faure

Desgranges

. Démences et syndromes démentiels. In: Eustache

Faure

Desgranges

(eds) Manuel de neuropsychologie. Paris: Dunod, 2018, pp.339–376.

24.

Watzlawick

. De la parole à la communication. Quand interpréter c’est changer. In: Quand interpréter c’est changer. Paris: Seuil, 1995, pp.139–151.

25.

Covelet

. La familiarité du personnel gérontologique envers les résidents. Gérontol Soc 2003; 26: 115–124.

26.

Rousseaux

Sève

Vallet

, et al. An analysis of communication in conversation in patients with dementia. Neuropsychologia 2010; 48: 3884–3890.

27.

Abric

. La dimension non verbale de la communication. In: Psychologie de la communication: théories et méthodes. Paris: Dunod, 2019, pp.59–69.

28.

Tulving

. Memory systems: Quo vadis? In: Gazzaniga

(eds) The cognitive neurosciences. Cambridge, MA: MIT Press, 1995, pp.1–31.

29.

Péruchon

. Le déclin de la vie psychique. Paris: Dunod, 1994.

30.

Damasio

. Emotion in the perspective of an integrated nervous system. Brain Res Rev 1998; 26: 83–86.

31.

Maisondieu

Le crépuscule de la raison: en finir avec Alzheimer sans frontières. 6e éd. Paris: Bayard, 2018.

32.

Mundy

Block

Delgado

, et al. Individual differences and the development of joint attention in infancy. Child Dev 2007; 78: 938–954.

33.

Dayan

. Inné et social, le sourire du bébé. Spirale 2023; 106: 175–179.

34.

Damasio

. Emotion in the perspective of an integrated nervous system. Brain Res Rev 1998; 26: 83–86.

35.

Damasio

. The feeling of what happens: Body and emotion in the making of consciousness. New York: Harcourt College Publishers, 1999.

36.

Damasio

. Spinoza avait raison: Joie et tristesse, le cerveau des émotions. Paris: Odile Jacob, 2003.

37.

Eustache

Guillery-Girard

Dayan

. Les liens ténus et complexes entre mémoire et émotions. In Analysis 2017; 1: 32–38.

38.

Gil

Arroyo-Anllo

. La relation au visage : du sujet normal au malade Alzheimer, du visage nu au visage masqué, d'une neuropsychologie humaniste à une éthique incarnée. Gériatr Psychol Neuropsychiatr Vieillissement 2022; 20: 219–233.

39.

Ekman

Friesen

. Hand movements. J Commun 1972; 22: 353–374.

40.

Ekman

. What the face reveals: Basic and applied studies of spontaneous expression using the Facial Action Coding System (FACS). Oxford: Oxford University Press, 1997.

41.

Ekman

. Basic emotions. In: Dalgleish

Power

(eds) Handbook of cognition and emotion. Chichester: John Wiley & Sons, 1999, pp.45–60.

42.

Blessing

Zöllig

Dammann

, et al. Implicit learning of affective responses in dementia patients: a face-emotion-association paradigm. Aging Neuropsychol Cogn 2010; 17: 633–647.

43.

Bauer

. Autonomic recognition of names and faces in prosopagnosia: a neuropsychological application of the Guilty Knowledge Test. Neuropsychologia 1984; 22: 457–469.

44.

Bauer

Verfaellie

. Electrodermal discrimination of familiar but not unfamiliar faces in prosopagnosia. Brain Cogn 1988; 8: 240–252.

45.

Harrison

Son

G-R

Kim

, et al. Preserved implicit memory in dementia: a potential model for care. Am J Alzheimers Dis Other Demen 2007; 22: 286–293.