Skip to main content Skip to navigation

Beyond the Winding Path of Learning: Exploring Affective, Cognitive, and Action-Oriented Prompts for Communication Skills

Project Overview

The document examines the integration of generative AI (GenAI) in education, focusing on its potential to enhance self-directed learning and improve communication skills among learners. It specifically analyzes the effectiveness of three distinct messaging styles—Affective, Cognitive, and Action-Oriented—created using a Large Language Model (LLM). The findings indicate that these styles significantly influence learner perceptions and engagement, which are crucial for improving retention and overall learning outcomes in online educational settings. Evaluator feedback emphasizes a preference for content that is not only evidence-based but also facilitates self-directed engagement, highlighting the importance of personalized learning experiences. Ultimately, the study suggests that leveraging GenAI can lead to more effective educational strategies, fostering an environment where students are empowered to take charge of their learning.

Key Applications

GenAI for instructional messaging

Context: Self-directed learning focusing on communication skills for online learners

Implementation: Utilized ChatGPT to generate 30 instructional items across three messaging styles and evaluated their desirability and appropriateness

Outcomes: Increased engagement, clearer instructional content, improved learner perceptions of skill-building experiences

Challenges: Initial negative reactions from learners due to lack of shared understanding and cognitive frameworks

Implementation Barriers

Cognitive Barrier

Learners may struggle with establishing a shared understanding of the content, leading to resistance and negative perceptions.

Proposed Solutions: Incorporate strategies to facilitate cognitive, affective, and action-oriented learning, and enhance the agent's social features to foster common ground.

Content Design Barrier

Overly affective content may lead to disengagement, and learners may have different expectations from AI versus human interactions.

Proposed Solutions: Balance affective elements in system interactions while focusing on cognitive and action-oriented instructional content.

Project Team

Naoko Hayashida

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Naoko Hayashida

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies