Chelsea Chandler

The technological basis of automated essay evaluation is used in other domains for assessing language-based responses. In general, these technologies assess knowledge and cognition, mental state, affective and cognitive functioning, and can even be harnessed to identify students at risk. As such, language can be viewed as a window into cognitive, social, and affective functioning. Underlying these different domains are many of the same techniques: spoken discourse is transcribed, the text is reduced to a set of computational features, machine learning models are built and trained to predict an indicator of performance (e.g., essay score, clinical diagnosis, and team cohesion), and actionable inferences are made with the predictions. This chapter discusses this pipeline as a common denominator in assessing variants of higher-level cognitive functioning, details other domains in which this toolbox of technologies has been applied, and ends with a discussion around ethical considerations in the context of measuring cognitive processing.

The use of Artificial Intelligence (AI) in K-12 education is showing considerable promise to enhance student learning, yet existing tools continue to situate AI tutoring firmly within the context of one-on-one instruction and personalized learning. As HCI, learning science, and team science researchers we envision AI to help students become better collaborators—a highly valued skill for their lives after school. In this demonstration we present “CoBi”—a multi-party AI partner that focuses on the relationship dimension of collaboration. CoBi helps students to co-negotiate classroom agreements along four dimensions: respect, equity, community, and thinking. CoBi then uses state-of-the-art speech and language technologies to look for and visualize evidence of these agreements as they occur during small group student talk. Through these feedback visualizations, students can hone collaboration skills, collaboratively reflect about and identify areas for improvement, and develop critical AI literacy skills.

INTRODUCTION: Traditional Alzheimer's disease (AD) and mild cognitive impairment (MCI) screening lacks the sensitivity and timeliness required to detect subtle indicators of cognitive decline. Multimodal artificial intelligence technologies using only speech data promise improved detection of neurodegenerative disorders.METHODS: Speech collected over the telephone from 91 older participants who were cognitively healthy (n = 29) or had diagnoses of AD (n = 30) or amnestic MCI (aMCI; n = 32) was analyzed with multimodal natural language and speech processing methods. An explainable ensemble decision tree classifier for the multiclass prediction of cognitive decline was created.RESULTS: This approach was 75% accurate overall—an improvement over traditional speech‐based screening tools and a unimodal language‐based model. We include a dashboard for the examination of the results, allowing for novel ways of interpreting such data.DISCUSSION: This work provides a foundation for a meaningful change in medicine as clinical translation, scalability, and user friendliness were core to the methodologies.Highlights: Remote assessments and artificial intelligence (AI) models allow greater access to cognitive decline screening. Speech impairments differ significantly between mild AD, amnestic mild cognitive impairment (aMCI), and healthy controls. AI predictions of cognitive decline are more accurate than experts and standard tools. The AI model was 75% accurate in classifying mild AD, aMCI, and healthy controls.