Agata Rozga

Designing Conversational AI systems to support older adults with MCI necessitates robustness in handling complex and personalized interactions. In this study, we investigate how older adults with MCI experience, navigate and attempt to repair breakdowns while interacting with a voice-based AI system in their homes for medication management through the lens of interactional ”seams”. Through a 20-week in-home deployment with 7 participant dyads, we analyzed 844 interactions to identify interactional and conversational breakdowns and user repair strategies. Drawing on post-deployment interviews, we reflect on the nature of these breakdowns and older adults’ experiences of mitigating them. We further illustrate how older adults draw on their situated knowledge and environment to make sense of and recover from these disruptions, highlighting the cognitive effort required in doing so. Our findings emphasize the collaborative nature of interaction in human-AI contexts, and point to the need for voice assistants to better align with users’ expectations for memory, their routines and external resources in their environment. We conclude by discussing opportunities for voice assistants to integrate contextual knowledge from older adults’ sociotechnical environment and to facilitate more meaningful and user-centered interactions.

Older adults aging in place often have informal support systems to help them maintain independence and quality of life. As they age, many older adults deal with the onset of Mild Cognitive Impairment (MCI), which introduces a new set of functional and cognitive changes that affect their ability to manage daily routines . The approach to arranging and coordinating support for everyday activities for older adults with MCI varies across informal care networks, but typically involves a primary care partner and a network of family, friends and others. In this paper, we present a thematic analysis of in-depth interviews with older adults with MCI and their primary care partners to gain a holistic picture of their day-to-day lived experience . Our analysis uses a multi-dimensional lens of people (“ who ”), activities (“ what ”), and tools (“ how ”) to reveal insights about the nature of informal care coordination in MCI. Our results characterize informal care for MCI as a set of complex orchestration tasks by a primary care partner that support the practical, cognitive and emotional needs of the diagnosed individual and mediate the involvement of the broader care network. We uncover that coordination is not solely a matter of logistical organization, but also a deeply relational process shaped by negotiation with technological tools and evolving roles. Through this work, we reframe care coordination for MCI as a distinct and underexplored design space, one that demands systems capable of scaffolding autonomy, adapting to shifting capacities, responding to socio-emotional needs, and fostering collaborative caregiving.

Designing Conversational AI systems to support older adults requires more than usability and reliability, it also necessitates robustness in handling conversational breakdowns. In this study, we investigate how older adults navigate and repair such breakdowns while interacting with a voice-based AI system deployed in their homes for medication management. Through a 20-week in-home deployment with 7 older adult participant dyads, we analyzed 844 recoded interactions to identify conversational breakdowns and user-initiated repair strategies. Through findings gleaned from post-deployment interviews, we reflect on the nature of these breakdowns and older adults' experiences of mitigating them. We identify four types of conversational breakdowns and demonstrate how older adults draw on their situated knowledge and environment to make sense of and recover from these disruptions, highlighting the cognitive effort required in doing so. Our findings emphasize the collaborative nature of interactions in human-AI contexts, and point to the need for AI systems to better align with users' expectations for memory, their routines, and external resources in their environment. We conclude by discussing opportunities for AI systems to integrate contextual knowledge from older adults' sociotechnical environment and to facilitate more meaningful and user-centered interactions.

Over the past 20+ years, researchers have worked toward identifying early behavioral predictors of autism spectrum disorder (ASD) and developing observation-based screeners to supplement existing parent-report methods. This study is a follow-up, 3 to 8 years later, with parents/caregivers of 57 children previously enrolled in a U.S. university-based study evaluating early ASD-risk. The original study evaluated infants’ (ages 15–35 months) ASD-risk through both observation-based and parent-report screeners. At follow-up, caregivers completed a phone interview inquiring about their child’s developmental progress and diagnostic outcomes. Results indicated screener at-risk status agreement in infancy predicted only one of the four parent-reported ASD diagnoses at follow-up. Single instrument at-risk status aligned with two additional ASD diagnoses (one per screener), and both screeners missed one ASD diagnosis at follow-up. Results did not indicate significant added utility for the observation-based screener over the commonly used parent-report screener, suggesting that ASD behavioral markers may be hard to observe at early ages.

While commercial conversational agents (CA) (i.e. Google assistant, Siri, Alexa) are widely used, these systems have limitations in error-handling, flexibility, personalization and overall dialogue management that are amplified in care coordination settings. In this paper, we synthesize and articulate these limitations through quantitative and qualitative analysis of 56 older adults interacting with a commercial CA deployed in their home for a 10 week period. We look at the CA as a compensatory technology in an older adult's care network. We argue that the CA limitations are rooted in the rigid cue-and-response style of task-oriented interactions common in CAs. We then propose a redesign for CA conversation flow to favor flexibility and personalization that is nonetheless viable within the limitations of current AI and machine learning technologies. We explore design tradeoffs to better support the usability needs of older adults compared to current design optimizations driven by efficiency and privacy goals.

Eye contact is among the most primary means of social communication used by humans. Quantification of eye contact is valuable as a part of the analysis of social roles and communication skills, and for clinical screening. Estimating a subject’s looking direction is a challenging task, but eye contact can be effectively captured by a wearable point-of-view camera which provides a unique viewpoint. While moments of eye contact from this viewpoint can be hand-coded, such a process tends to be laborious and subjective. In this work, we develop a deep neural network model to automatically detect eye contact in egocentric video. It is the first to achieve accuracy equivalent to that of human experts. We train a deep convolutional network using a dataset of 4,339,879 annotated images, consisting of 103 subjects with diverse demographic backgrounds. 57 subjects have a diagnosis of Autism Spectrum Disorder. The network achieves overall precision of 0.936 and recall of 0.943 on 18 validation subjects, and its performance is on par with 10 trained human coders with a mean precision 0.918 and recall 0.946. Our method will be instrumental in gaze behavior analysis by serving as a scalable, objective, and accessible tool for clinicians and researchers.

In this short-term longitudinal study, 30 preschool-aged children with autism were first observed in Ainsworth's Strange Situation Procedure and, separately, interacting with the primary caregiver in the home. One year later, each child completed both a developmental assessment and an observational assessment of empathic responding. Behaviors typical for children with autism were distinguished from behaviors suggestive of relationally based attachment disorganization. Forty-five percent of the children were classified as securely attached. The secure group demonstrated language skills superior to those of the insecurely attached group, concurrently and during the follow-up. Compared to parents of children who were insecurely attached, parents of securely attached children were rated as more sensitive. Compared to both organized insecure and disorganized children, secure children were rated as more responsive to an examiner's apparent distress during the follow-up relative to their ratings at intake, whereas empathy ratings of children with insecure classifications did not increase. Importantly, attachment security was associated with empathy above and beyond the contribution of children's language level. These results indicate that the sequelae of attachment security in autism may be similar to those documented for typically developing children.