People with aphasia often do not use their Augmentative and Alternative Communication (AAC) devices in everyday conversations. Researchers have sought to understand why this is so in order to build more effective AAC devices. I take the approach of understanding the perspectives of caregivers, family members, and speech language therapists (SLTs) to inform design since they have ongoing roles in the use and adoption of AAC devices. By highlighting communication partners, this study emphasizes the bidirectional nature of communication often overlooked in the design of AAC devices. It highlights the impact of media substitution on AAC users’ experience and highlights dimensions along which AAC devices might support social connectedness. I draw on 43 hours of observation of people with aphasia (and their caregivers) during therapy sessions, interviews and technology tests with SLTs, and contextual interviews and technology testing with people with aphasia with their communication partners in naturalistic social settings.

Scientists have long sought to engage public audiences in research through citizen science projects such as biological surveys or distributed data collection. Recent online platforms have expanded the scope of what people-powered research can mean. Science museums are unique cultural institutions that translate scientific discovery for public audiences, while conducting research of their own. This makes museums compelling sites for engaging audiences directly in scientific research, but there are associated challenges as well. This project engages public audiences in contributing to real research as part of their visit to a museum. We present the design and evaluation of U!Scientist, an interactive multi-person tabletop exhibit based on the online Zooniverse project, Galaxy Zoo. We installed U!Scientist in a planetarium and collected video, computer logs, naturalistic observations, and surveys with visitors. Our findings demonstrate the potential of exhibits to engage new audiences in collaborative scientific discussions as part of people-powered research.

There is a growing need to design augmentative and alternative communication (AAC) devices that focus on supporting quality of life goals, such as increased social participation in leisurely activities. Yet, designing AAC applications that support leisurely activities is difficult, as the activity might require novel and specific language in a timely manner. Through observations and contextual interviews with people with aphasia, their social partners, and speech-language therapists, we characterize the important but challenging nature of supporting one specific leisure activity: meal ordering in restaurants. Based on our observational and interview data, we design and explore three prototype AAC systems to support people with aphasia in ordering meals in restaurants. Each prototype integrates a different AI technology, contributing insights into how AI may enhance AAC usage and design. The study opens up questions of designing accessible restaurant experiences for neurodivergent people and the role of AI in AAC devices more broadly.

Augmentative and Alternative Communication (AAC) devices help people with speech-language impairments due to aphasia express themselves more independently and fluently. However, current AACs are still limited in that they must be pre-programmed with common words or phrases prior to use. When people with aphasia encounter unexpected and unplanned situations or contexts, they may have difficulty using their AAC device to generate the desired speech. We present the design of an AAC application that uses image recognition technology to give people with aphasia just in time access to words. Our design automatically carries out an extensive retrieval of words associated with a captured image. The relevance, accuracy and usefulness of each word can then be verified independently by the user through an intensive retrieval of images associated with each word.

Electrical circuits are difficult to understand. In this paper, we introduce Spark, a museum exhibit that enables learners to interact with multiple representations of electrical circuits at the same time. The goal is to familiarize children with fundamental concepts of electricity such as current and resistance. We tested our design with 6 parent-child dyads at a major science museum. Our study investigates how families make sense of representations at the level of circuits and at the level of electrons and ions. Our findings show a wide variety of visitor interactions with the exhibit, in particular when connecting the two representations of a circuit.