Hikaru Sugimoto

Background: Early identification of older adults with lower cognitive screening performance may facilitate appropriate clinical evaluation and intervention to delay the progress of cognitive decline and onset of dementia. Because facial information can be acquired noninvasively and unobtrusively during everyday life, it has attracted increasing attention as a potential digital biomarker. However, it remains unclear which facial regions are associated with cognitive decline under a neutral facial condition. Furthermore, it is unknown whether these facial characteristics primarily reflect static facial appearance or facial movements. Objective: This study aimed to identify facial action unit (AU) patterns associated with lower cognitive screening performance during a fixation-cross viewing condition that is less susceptible to the influence of language and emotional elicitation. Methods: Older adults were classified into either a lower cognitive screening performance group (Montreal Cognitive Assessment Japanese version [MoCA-J] score ≤25; n=31) or a healthy control group (MoCA-J score ≥26; n=14). Facial videos were recorded while participants viewed a fixation cross for 1 minute. OpenFace 2.0 was used to extract the intensities of 17 AUs and the presence of 18 AUs. Video analyses were first performed to identify AU characteristics associated with cognitive screening performance. For AU intensity, frame-level mixed-effects models with participant-specific random intercepts and a first-order autoregressive (AR(1)) correlation structure were used, adjusting for age, Geriatric Depression Scale score, and sex, and participant-level sensitivity analyses were conducted to assess the consistency of the frame-level findings. For AU presence, participant-level analyses were performed because the frame-level models did not yield reliable parameter estimates. Subsequently, single-frame analyses were performed to examine whether the AU characteristics identified in the video analyses could be explained solely by static facial appearance. As this study was exploratory and hypothesis-generating, P values were not adjusted for multiple comparisons. Results: For AU intensity, frame-level video analyses showed that the lower cognitive screening performance group had a higher probability of zero intensity for AU10 (upper lip raiser), with participant-level analysis based on the proportion of zero-intensity frames yielding consistent results. In contrast, single-frame analyses of AU intensity showed no significant group difference for AU10, suggesting that the observed difference may reflect facial movements rather than static facial appearance. For AU presence, video analyses showed no significant group differences. Conclusions: Older adults with lower cognitive screening performance exhibited distinct AU patterns related to AU10 (upper lip raiser) during fixation-cross viewing, which may reflect subtle differences in facial movements around the mouth. These exploratory findings identify candidate facial biomarkers associated with lower cognitive screening performance and warrant validation in larger, balanced, independent samples.

Background Dementia prevention has the potential to bring considerable benefits in diverse contexts. Conversation requires various cognitive functions and is therefore considered a natural multi-component preventive measure that can be practiced in daily life. Although an emerging line of research supports the potential of conversation-based cognitive interventions, further research is needed to elucidate which elements are effective and through what mechanisms. In addition, given that previous studies have shown that older adults embedded in social networks with a high proportion of non-family members have higher cognitive function, fostering social interactions between older adults with those beyond their family would be beneficial, while assessing the ecological validity of delivering conversation-based program in naturalistic community settings is essential. Furthermore, it remains to be clarified whether participants can integrate lessons from conversation-based programs into their daily routines in a voluntary and sustained manner. This study aims to examine the feasibility of the long-term conversation-based program, with preliminary analyses of its key components for future rigorous effectiveness tests. Methods The study is a single-arm longitudinal intervention in a community setting in Japan. Participants will attend monthly sessions over five years, engaging in conversations based on the Coimagination method. The feasibility and adherence are assessed mainly by dropout rates and speech frequencies in sessions, respectively. The primary outcome is verbal fluency, repeatedly improved by previous conversational interventions. The key components of the intervention such as perspective taking and time orientation will also be measured. Preliminary intervention effects will be assessed by the interaction between adherence and time in a mixed-effects model. Discussion The study following this protocol is expected to provide insights into the feasibility and ecological validity of our conversation-based program and its barriers, as well as preliminary findings for conducting a future long-term randomized controlled trial to rigorously verify the intervention effects. Because the intervention will be conducted in parallel in two regions, suggestions for external validity are also expected.

Background: Asking questions is common in conversations, and while asking questions, we need to listen carefully to what others say and consider the perspective our questions adopt. However, difficulties persist in verifying the effect of asking questions on older adults’ cognitive function due to the lack of a standardized system for conducting experiments at participants’ homes. Objective: This study examined the intervention effect of cognitive training moderated by robots on healthy older adults. A focus on the feasibility of the intervention at participants’ homes was also maintained. Feasibility was evaluated by considering both the dropout rate during the intervention and the number of questions posed to each participant during the experiment. Methods: We conducted a randomized controlled trial with 81 adults older than 65 years. Participants were recruited through postal invitations and then randomized into 2 groups. The intervention group (n=40) received sessions where participants listened to photo-integrated stories and posed questions to the robots. The control group (n=41) received sessions where participants listened to photo-integrated stories and only thanked the robots for confirming participation. The participants participated in 12 dialogue sessions for 2-3 weeks. Scores of global cognitive functioning tests, recall tests, and verbal fluency tasks measured before and after the intervention were compared between the 2 groups. Results: There was no significant intervention effect on the Telephone Interview for Cognitive Status-Japanese scores, recall tests, and verbal fluency tasks. Additionally, our study successfully concluded with no participant dropouts at follow-up, confirming the feasibility of our approach. Conclusions: There was no statistically significant evidence indicating intervention benefits for cognitive functioning. Although the feasibility of home-based interventions was demonstrated, we identified areas for improvement in the future, such as setting up more efficient session themes. Further research is required to identify the effectiveness of an improved cognitive intervention involving the act of asking questions.

BACKGROUND Following empirical evidence suggesting a strong connection between social activities and longevity/well-being, this topic has now entered an intervention phase, which evaluates the effectiveness of social activity interventions and accumulates practical knowledge. OBJECTIVE This study examined the effects of "Photo-Integrated Conversation Moderated by Application" (PICMOA), an application-based remote conversational intervention, on the cognitive function and psychological well-being of community-dwelling older adults with subjective cognitive concerns in Japan. METHODS The PICMOA trial is a randomized controlled trial introducing an open-label, two parallel group trial design with a 1:1 allocation. In this trial, the effects of the PICMOA intervention, a weekly application-based group conversation, were compared with weekly health education videos. The participants included community dwellers aged 65 years and older with subjective cognitive concerns. In total, 81 participants were enrolled in the trial and divided into groups (intervention: n = 41; control: n = 40). The main outcomes included cognitive function assessed through a telephone Interview for Cognitive Status in Japanese, verbal fluency test, and Digit Span Forward and Backward tests. The secondary outcomes included psychological and social well-being. The effects of the PICMOA intervention were examined using a mixed linear model. RESULTS In total, 75 participants (intervention: n = 35; control: n = 40) who completed the intervention and evaluations were analyzed. Overall, there were no significant improvements associated with cognitive function and psychological metrics in the intervention group. However, bilateral patterns emerged in the intervention effects on category fluency when individuals were classified by their smartphone-use proficiency at baseline: significant negative effects for those who were unfamiliar with smartphones at baseline (β = -5.47, SE = 2.04, P = .009) and significant positive effects for those who were familiar with smartphones at baseline (β = 6.43, SE = 2.47, P = .01) were observed. CONCLUSIONS This trial showed no significant improvements in cognitive and psychological outcomes after the PICMOA intervention. Further research on digital health practices is required to accumulate more evidence. CLINICALTRIAL UMIN Clinical Trials Registry UMIN000047247. INTERNATIONAL REGISTERED REPORT RR2-10.3389/fpsyg.2023.1114790

This study investigated the time orientation of older adults in group conversations and its association with their memory functioning. We hypothesized that those with a greater ability to memorize recently accessed information would talk about more recent topics in natural conversations. To test this hypothesis, we used dictation data from older adults’ group conversations and categorized their utterances in terms of the time to which the utterances refer. While inheriting the conventional coding categories for the temporal focus of the utterances, Past, Present, and Future, this study assigned a new category, Recent, to an utterance that referred to an event within the past month to distinguish between recent and remote memories. We further classified utterances concerning whether they were experience-based or knowledge-based, based on the distinction between two types of declarative memories: episodic and semantic memories. We found that the frequency of knowledge-based utterances had significant positive associations with memory functioning if they were Recent-tagged. However, the frequency of experience-based utterances had no significant association with memory functions, irrespective of whether the sentence was Past- or Recent-tagged. Based on the results, we discussed what type of intervention program would be effective in preserving or improving memory function in older adults.

BACKGROUND Asking questions is common in conversation, and while asking questions, we need to listen carefully to what others say and consider the perspective our questions adopt. However, difficulties remain in verifying its effect on older adults’ cognitive function owing to the lack of a standardized system to conduct experiments in daily life. OBJECTIVE This study examined the intervention effect of cognitive training moderated by robots on healthy older adults. In the intervention program, the participants listened to a story related to a photo displayed on the tablet screen and subsequently asked questions about it to the robots within a predetermined time. A focus on the feasibility of the intervention at the participant's home was also maintained. METHODS We conducted a randomized controlled trial with 81 adults over 65 years of age. Participants were randomized into two groups: the intervention group received sessions where participants listened to photo-integrated stories and asked questions to the robots (n=40), and the control group received sessions where participants listened to photo-integrated stories and only thanked the robots for confirming participation (n=41). The participants participated in 12 dialogue sessions for 2-3 weeks. Cognitive functioning scores measured before and after the intervention were compared between the two groups. RESULTS The improvement in category fluency in the intervention group was higher than that in the control group (P = .09), which is partially consistent with previous studies. There was no significant intervention effect on the Telephone Interview for Cognitive Status-Japanese score (P = .31). Additionally, our study was completed with no loss of participants at follow-up. CONCLUSIONS We confirmed that the intervention program has a potentially reliable cognitive training effect on category fluency in healthy older adults. The feasibility of home-based interventions was also demonstrated, and we identified issues that need to be improved in the future, such as setting up more efficient session themes for cognitive training to develop more effective intervention programs. CLINICALTRIAL The studies are registered at Clinical Trials (UMIN000039489)

Background: There is a difference in the neuronal state of individuals. However, this has not been taken into consideration in most intervention studies. Recent advances in analytical technologies in hematological examination enabled us to evaluate neuronal states in a relatively convenient manner. Objective: Using these advanced technologies, we aimed to investigate whether cognitive intervention effects vary as a function of levels of blood-based biomarkers, such as neurofilament light chain (NfL), since plasma NfL could be a biomarker of neurodegenerative diseases including Alzheimer’s disease. Methods: In this study, we employed a group conversation-based intervention methodology named Photo-Integrated Conversation Moderated by Robots (PICMOR), which has been shown to improve verbal fluency in older adults. To examine the possibility of varying effects of this intervention method according to the neuronal state of each individual, we conducted a randomized controlled trial (UMIN Clinical Trials Registry number: UMIN000036599) and investigated how longitudinal changes in cognitive performance, such as verbal fluency, vary with the NfL level measured at the baseline. Results: As the main result, positive intervention effects of PICMOR on verbal fluency were observed in individuals with lower level of NfL, which indicate a relatively intact neuronal state, whereas negative intervention effects were identified in individuals with higher NfL level. Conclusion: Our findings suggest that cognitive intervention effects vary depending on level of Nfl in the plasma. Thus, future intervention studies should take into account the neuronal status of the participants to examine intervention effects.

Social comparison orientation (SCO) refers to the tendency to compare oneself with others and has two distinct dimensions: one about opinions and the other about abilities. Although dissociable neural mechanisms underlying the two dimensions of social comparison can be assumed, little is known about how each dimension of SCO is associated with cognitive and brain health among older adults. To investigate this, we analyzed the SCO scale questionnaire data, neuropsychological assessment data, and multimodal MRI data collected from 90 community-dwelling older adults. We found that global cognitive performance was positively correlated with the score of the opinion subscale but not with the score of the ability subscale and the total score. Similarly, hippocampal volume was positively correlated with opinion score alone. Additionally, the resting-state functional connectivity between the hippocampal seed and the default mode network showed a positive correlation only with the opinion score. Moreover, fractional anisotropy in the hippocampal cingulum was positively correlated with opinion score only. These findings suggest that global cognition and hippocampal properties in older age are associated with the SCO of opinion, which could reflect a regular habit of performing the types of cognitively demanding activities involved in evaluation of self and other opinions.

Diffusion tensor imaging (DTI) enables the investigation of white matter properties in vivo by applying a tensor model to the diffusion of water molecules in the brain. Using DTI metrics including fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD), an attempt has been made to detect age-related alterations in the white matter microstructure in aging research. However, the use of comprehensive DTI measures to examine the effects of cognitive intervention/training on white matter fiber health in older adults remains limited. Recently, we developed a cognitive intervention program called Photo-Integrated Conversation Moderated by Robots (PICMOR), which utilizes one of the most intellectual activities of daily life, conversations. To examine the effects of PICMOR on cognitive function in older adults, we conducted a randomized controlled trial and found that verbal fluency task scores were improved by this intervention. Based on these behavioral findings, we collected in this pilot study diffusion-weighted images from the participants to identify candidate structures for white matter microstructural changes induced by this intervention. The results from tract-based spatial statistics analyses showed that the intervention group, who participated in PICMOR-based conversations, had significantly higher FA values or lower MD, AD, or RD values across various fiber tracts, including the left anterior corona radiata, external capsule, and anterior limb of the internal capsule, compared to the control group, who participated in unstructured free conversations. Furthermore, a larger improvement in verbal fluency task scores throughout the intervention was associated with smaller AD values in clusters, including the left side of these frontal regions. The present findings suggest that left frontal white matter structures are candidates for the neural underpinnings responsible for the enhancement of verbal fluency. Although our findings are limited by the lack of comparable data at baseline, we successfully confirmed the hypothesized pattern of group differences in DTI indices after the intervention, which fits well with the results of other cognitive intervention studies. To confirm whether this pattern reflects intervention-induced white matter alterations, longitudinal data acquisition is needed in future research.