AAC research has traditionally focused on input speed, leaving higher-level communication goals such as relational maintenance under-explored. Through semi-structured interviews with AAC users with motor and speech impairments and their primary family caregivers, we offer a nuanced understanding of AAC’s roles in maintaining close relationships. Our inductive analysis reveals emerging themes including how AAC users and their partners share the physical and mental workload to overcome communication barriers in complex situations. Our deductive application of the Relational Maintenance Strategies framework exposes the efforts made and the challenges encountered in managing social engagements, providing mutual support, and decoding implicit expressions. From these insights, we propose novel research directions for better supporting maintenance strategies and social purposes of communication, including notably mediating relational tensions, leveraging empowerment and identity, and supporting interactions for social closeness and etiquette, which we hope will motivate discussion in HCI communities on expanding AAC research space.

The concurrent think-aloud protocol—in which participants verbalize their thoughts when performing tasks—is a widely employed approach in usability testing. Despite its value, analyzing think-aloud sessions can be onerous because it often entails assessing all of a user's verbalizations. This has motivated previous research on developing categories to segment verbalizations into manageable units of analysis. However, the way in which a category might relate to usability problems is currently unclear. In this research, we sought to address this gap in our understanding. We also studied how speech features might relate to usability problems. Through two studies, this research demonstrates that certain patterns of verbalizations are more telling of usability problems than others and that these patterns are robust to different types of test products (i.e., physical devices and digital systems), access to different types of information (i.e., video and audio modality), and the presence or absence of a visualization of verbalizations. The implication is that the verbalization and speech patterns can potentially reduce the time and effort required for analysis by enabling evaluators to focus more on the important aspects of a user's verbalizations. The patterns could also potentially be used to inform the design of systems to automatically detect when in the recorded think-aloud sessions users experience problems.