Artificial Emotional Intelligence in Socially Assistive Robots for Older Adults: A Pilot Study (2201.11167v1)

Abstract: This paper presents our recent research on integrating artificial emotional intelligence in a social robot (Ryan) and studies the robot's effectiveness in engaging older adults. Ryan is a socially assistive robot designed to provide companionship for older adults with depression and dementia through conversation. We used two versions of Ryan for our study, empathic and non-empathic. The empathic Ryan utilizes a multimodal emotion recognition algorithm and a multimodal emotion expression system. Using different input modalities for emotion, i.e. facial expression and speech sentiment, the empathic Ryan detects users' emotional state and utilizes an affective dialogue manager to generate a response. On the other hand, the non-empathic Ryan lacks facial expression and uses scripted dialogues that do not factor in the users' emotional state. We studied these two versions of Ryan with 10 older adults living in a senior care facility. The statistically significant improvement in the users' reported face-scale mood measurement indicates an overall positive effect from the interaction with both the empathic and non-empathic versions of Ryan. However, the number of spoken words measurement and the exit survey analysis suggest that the users perceive the empathic Ryan as more engaging and likable.

The paper "Artificial Emotional Intelligence in Socially Assistive Robots for Older Adults: A Pilot Study" (Abdollahi et al., 2022 ) presents a pilot paper evaluating the effectiveness of integrating Artificial Emotional Intelligence (AEI) into a social robot named Ryan, designed to offer companionship and social interaction for older adults experiencing depression and dementia.

The paper involved two versions of the robot: an empathic version and a non-empathic version. The empathic Ryan used a multimodal emotion recognition algorithm analyzing facial expressions and speech sentiment, coupled with a multimodal emotion expression system to detect and respond to user emotions with affective dialogues. Conversely, the non-empathic Ryan lacked facial expressions and used scripted dialogues irrespective of the user's emotional state. Ten older adults from a senior care facility participated, interacting with both versions of Ryan twice weekly over three weeks, totaling six sessions. Participants were divided into two groups, alternating between the empathic and non-empathic versions after the initial three sessions. User engagement was gauged by the average number of words spoken by the user. Emotional state was evaluated using a face-scale mood assessment before and after each session. An exit survey was conducted to assess the robot's likeability and perceived empathy. Participants' depression levels were evaluated using the Patient Health Questionnaire-9 (PHQ-9) and the Geriatric Depression Scale (GDS) assessments pre- and post-paper.

Key findings include statistically significant improvements in participants' self-reported mood after interacting with both versions of Ryan. The empathic Ryan was found to be more engaging and likable based on the number of spoken words and exit survey responses. The inclusion of emotion (ON/OFF) demonstrated a significant effect on word count, with participants reacting more positively when Ryan exhibited empathy. Depression scores improved in 7 out of 10 participants based on GDS and PHQ-9 assessments.

The paper concludes that both empathic and non-empathic social robots can positively influence the mood of older adults. The integration of AEI and empathic capabilities may enhance user engagement and foster a more positive perception of the robot. The results suggest that social robots with AEI have the potential to improve the quality of interaction and engagement with older adults. The authors recommend further research to determine the long-term impact of such robots on depression in older adults.