Emotions and Dementia

Background: Dementia is a group of neurodegenerative diseases that affect cognitive processing. Nowadays, more than 270.000 persons have dementia in the Netherlands, a figure that will double over the next 25 years. Besides problems in cognitive functioning, 80-90% of the patients also suffer from problems in emotional functioning. It is important to gain more insight in the emotional functioning of persons with dementia as it is highly important to maintain quality of life in person-centered care. Current research uses observational instruments to assess emotional functioning in persons with dementia. However, they do not provide the fine-grained insights that are necessary to understand the factors influencing emotional responses and expressions of persons with dementia. Emotional expression is a continuous process that involves many features of behavioural, facial, vocal, and verbal modalities. Given this complexity, few psychological studies have addressed emotion recognition in an everyday context of persons with dementia.
Recent technological innovations in the field of affective computing aim to take the complexities of emotional expression into account. Automatic emotion detection makes it possible to study latent features that are difficult to observe and track by human beings between different modalities. It also allows us to investigate larger sets of video data in a smaller amount of time and for unobtrusive analysis and monitoring of everyday emotions. However, little is known about how these lab-based technologies generalize to real world problems. Rather than a one-size-fits-all-solution, existing tools need to be adapted to specific user groups in more natural settings. They also need to take large individual differences into account.

Objective: The primary goal of this study is to gain a better understanding of how dementia affects the multimodal expression of emotions in face, gestures, verbal and non-verbal expressions when discussing emotional laden autobiographical memories. Therefore, we carry out an observational comparative study between patients with dementia and matched healthy older adults, as well as a longitudinal study on the development of emotion expression in patients with early dementia across time to gain insight how dementia affects emotions versus normal aging as well as the effect is when the disease progressed over time.
The secondary goal of this project is to advance technologies that allow for (automatic) recognition of emotions in persons with dementia. We therefore explore machine learning techniques to advance technologies for multimodal emotion recognition. Furthermore, we intend to create a corpus of the processed video, audio and qualitative data for other researchers to use.

Study design: This is an observational study that consists of 1) a comparative study between people with early dementia and healthy elderly; and 2) a longitudinal study within people with early dementia. The comparative study consists of two sessions. In the Session 1, participants have to fill in questionnaires, and assess emotional laden memories with a word association task. In Session 2, they discuss these memories in detail via a life story book based on the first session and react to affective pictures. For the comparison study, the data of healthy elderly was already collected in a previous study with ethical approval of the committee at the University of Twente. The longitudinal study has 3 follow-up measurement points: after 3 months (Session 3), 6 months (Session 4) and 12 months (Session 5).

The primary goal of this study is to gain a better understanding of how dementia affects the multimodal expression of emotions in face, gestures, verbal and non-verbal expressions when discussing emotional laden autobiographical memories. Therefore, we carry out an observational comparative study between patients with dementia and matched healthy older adults, as well as a longitudinal study on the development of emotion expression in patients with early dementia across time to gain insight how dementia affects emotions versus normal aging as well as the effect is when the disease progressed over time.
The secondary goal of this project is to advance technologies that allow for (automatic) recognition of emotions in persons with dementia. We therefore explore machine learning techniques to advance technologies for multimodal emotion recognition. Furthermore, we intend to create a corpus of the processed video, audio and qualitative data for other researchers to use.

The person with dementia will have in total 5 sessions of about 120 minutes each, so a total time investment of 10 hours over a period of one year.
Participants are free to collaborate in the study. They may quit without reason and without consequences. There are no known negative effects of reminiscence. Possible side effect might be the recollection of negative memories. The principal investigator has experience with dealing with negative memories and privacy issues. Furthermore, persons with psychotrauma are excluded. Last, participants are free to decide whether and under which conditions their data may be included in the corpus.

None, explorative and observation study