2027 DT&L Conference | General Sessions

 

February 3 | 11:05 AM ET

Behavioral Science for Better AI Use

AI adoption is creating a new set of behavior challenges. People may know that they should validate AI outputs, think critically about what they delegate to the tool, and use appropriate human oversight—but knowing those things does not mean they will do them consistently. AI is particularly tricky because its outputs can be fluent and convincing, and because the tool makes it unusually easy to delegate thinking and judgment.

Behavioral science provides practical ways to understand why these behaviors break down and what to do about it. In this session, we’ll use common AI-use challenges—including validating outputs, cognitive offloading, calibrating AI supervision, and evaluating AI-assisted work—as case studies for applying behavioral science to workplace performance.

You’ll learn how to define the behaviors you actually need, diagnose the capability, motivation, habit, feedback, social, and environmental factors that influence them, and determine when training is the right response—or when better results require workflow changes, decision aids, feedback loops, social expectations, practice, or system constraints.

In this session, you will learn:


  • How cognitive biases, habits, and prior technology use influence high-risk AI behaviors

  • How to use behavioral science frameworks to diagnose why effective AI behaviors do or do not happen

  • How to think about validation, cognitive offloading, and human supervision based on task risk and expertise

  • How to select training, environmental support, workflow, and governance strategies that make effective AI use more likely

 

  • Julie Dirksen
    Author + Learning Strategy Consultant

    Julie Dirksen

Julie Dirksen is the author of the books Design For How People Learn and Talk to the Elephant: Design Learning for Behavior Change. She is a learning strategy consultant with a focus on incorporating behavioral science into learning interventions. Her MS degree is in Instructional Systems Technology from Indiana University. She’s been an adjunct faculty member at the Minneapolis College of Art and Design and is a LearningGuild Guildmaster.

She is happiest when she gets to learning something new, and you can find her at usablelearning.com.

 

 

 

 

 

 



February 4 | 11:05 AM ET

AI Reviews and the Cost of “Good Enough”

AI tools are increasingly being used to review courses and programs, identify evidence, assess alignment, and provide feedback on instructional quality. The appeal is obvious: faster reviews, immediate feedback, and scalable support for faculty and instructional designers. But identifying relationships among learning outcomes, activities, materials, and assessments is not the same as determining whether those elements meaningfully support the intended learning.

This session explores what AI can and cannot reliably establish about instructional alignment and quality, drawing on work with AI-supported course review and program mapping. Through practical examples, we'll examine the challenges of evaluating cognitive demand, interpreting instructional context, and determining whether students are given meaningful opportunities to develop and demonstrate the intended knowledge and skills. We'll also consider what happens when AI-generated evaluations appear convincing but are based on incomplete evidence or unsupported assumptions.

Participants will be introduced to a practical framework for distinguishing what AI can establish, what it can suggest, and what requires deeper professional judgment. Attendees will leave better equipped to evaluate AI-generated conclusions about instructional quality, recognize the limitations of automated review, and make informed decisions about where AI can support the process and where human expertise remains essential.

  • Dr. Bethany Simunich
    Vice President of Innovation and Research, Quality Matters

    Bethany Simunich

Dr. Bethany Simunich is Vice President of Innovation and Research at Quality Matters, where she leads applied research and innovation initiatives focused on online learning quality, institutional strategy, and evidence-based improvement. With more than 20 years of experience in higher education, she has worked as faculty, instructional designer, faculty developer, online learning administrator, and consultant, helping institutions strengthen instructional design practices and quality assurance processes. She is Co-Director of the Changing Landscape of Online Education (CHLOE) Project, a national research initiative examining online education strategy, leadership, and institutional practices through the perspectives of chief online learning officers and other senior online learning leaders. She is also co-author of High Impact Design for Online Courses (HIDOC), an instructional design model developed specifically for online learning. Her work focuses on translating research into practical strategies that strengthen institutional decision-making, instructional quality, and the learner experience.