The AI Fusionist Playbook for Learning Theories

Part 3 of the AI Fusionist Series for Learning & Development

Learning is a complex thing.?

We don’t just read something and become instant experts. Instead, we engage with it. We absorb it. We make connections with the materials, our past experiences, and interactions with peers. We adapt based on feedback and reinforcements. And we reflect.?

All of these things happening behind the scenes in our minds help determine whether we retain what we learn and can effectively apply it in our work.

But how do we know what’s happening in our minds or our learners’ minds? It comes down to our evidence-based learning theories.

What are Learning Theories?

Learning theories are frameworks that describe how we learn - including both internal influencers (like our experiences and motives) and external (like our environment).

Each theory presents its own perspective, shedding light on how we process, engage with, and apply new knowledge. These learning theories are what inform our instructional design models and approaches.

Fusing AI with our Learning Theories

As Learning & Development professionals, especially those adopting the AI Fusionist mindset, we need to understand how we can apply these theories as we integrate AI into our processes and the learning experiences we develop.

This AI Fusionist Playbook for ChatGPT and Learning Theories covers:

• Behaviorism
• Cognitivism
• Constructivism?
• Social Learning
• Dual Coding Theory (DCT)

If it seems like some important ones are missing, you’re right.

Some theories or theory categories are so impactful that they are worthy of their own playbook. This includes the AI Fusionist Playbook for Knowles’ Adult Learning Theory that I’ve previously shared and ones in the works like a playbook specific to motivation-related learning theories, among others.?

Behaviorism & AI

Behaviorism , with its origins in the early 20th century through the work of psychologists like John B. Watson and B.F. Skinner, is grounded on the idea that all behaviors are acquired through conditioning.?

Repetition and reinforcement. Within this framework, learning occurs through repeated interactions with the environment, which influences behavior. Through these experiences, we learn to associate specific behaviors with particular consequences, either in the form of rewards or punishments.

For example, if a student performs well on a test (the stimulus), they may receive an ‘A’ grade (a positive reinforcement or reward), encouraging them to continue studying effectively in the future. Conversely, if they perform poorly, receiving a low grade serves as a punishment, discouraging them from neglecting their studies in the future.

In a workplace context, a learner might watch an educational video and answer a quiz correctly, receiving immediate positive feedback and earning a digital badge (positive reinforcement). This recognition and reward encourage the learner to engage with more learning materials. If they answer incorrectly, the system might direct them to review the material again before retaking the quiz (a form of mild punishment), guiding them to invest more effort in understanding the content.

Fusing Behaviorism with AI

In Behaviorism, the emphasis is on repetition and reinforcement.?With AI, we can leverage tools like ChatGPT to effectively recognize and reinforce desirable behaviors while providing corrective feedback as necessary.?

The goal is to create an environment where learners can practice and refine their behaviors through continuous interaction with the AI, receiving feedback that is timely, relevant, and constructive. You can also use AI like ChatGPT in your own workflow to develop incentive strategies, identify opportunities to provide feedback, and more.

Example Prompt to Create an Incentive Plan

Use a prompt like the one below to get ChatGPT to develop a performance-based incentive plan for you.

The Prompt

I need to develop an incentive plan to encourage positive behaviors and high performance within our customer service team. 

The plan should motivate the team to improve customer satisfaction, enhance collaboration among team members, and contribute positively to our organizational culture. 

Help me brainstorm and design this incentive system. Start by asking me questions one at a time to gather relevant information. Wait for my response to each question. Use my information to craft an incentive plan tailored to our specific needs and context.        

The Results

Cognitivism & AI

Introduced in the 1960s as a reaction to Behaviorism, Cognitivism , championed by psychologists like Jean Piaget, focuses on the inner workings of the mind. It explores how information is received, processed, stored, and retrieved by the brain, emphasizing the role of internal cognitive processes in understanding how learning occurs.?

Cognitivism pivots away from the external, observable behaviors emphasized by Behaviorism, instead highlighting the internal cognitive mechanisms that underpin learning and the essential roles played by memory, perception, and thought in the learning process that drive observable behaviors.

For example, in a workplace training situation, an employee might be learning to use new software. The Cognitivism perspective would focus not just on the employee’s observable behaviors (like typing and clicking) but also on their internal processes, such as how they understand, remember, and apply the instructions or guidelines provided during the training. The training program might incorporate various aids like visual diagrams, written instructions, and interactive simulations to facilitate the employee’s cognitive processing and understanding of the new information.

Fusing Cognitivism & AI

From the Cognitivism perspective, we can use generative AI to support and enhance cognitive processes involved in learning. For example, ChatGPT can help structure information in ways that facilitate easier processing and understanding by our learners, using techniques like chunking, categorization, and sequencing.?

AI tools can also offer adaptive learning experiences that respond to the learner’s cognitive needs, providing support, scaffolding, and challenge as required in turn promoting deeper understanding and mastery of the content.

As we apply Cognitivism with AI, our focus should be aligning with and supporting the cognitive processes of learners. This involves crafting learning experiences that not only deliver content but also help with mental processing, organization, and retrieval of information, actively engaging learners in cognitive acts of learning and understanding.

Constructivism & AI

Constructivism emerged in the mid-20th century and emphasizes the active role of learners in building their own understanding. As learners, this theory suggests we actively construct our own knowledge and understanding by synthesizing new information with our existing knowledge.?

Within the constructivist framework, learning is seen as a dynamic process where learners actively engage with the material, participating in problem-solving, critical thinking exercises, and the development of deep, conceptual understanding.

Take, for example, a workplace situation where a team is introduced to a new project management tool. Instead of simply following a step-by-step tutorial, employees are encouraged to explore the tool's features and functionalities themselves. They might collaborate with colleagues, sharing insights and discussing ways to integrate the tool into their existing workflow.

Through this process, each employee constructs their understanding of the tool, relating its features to their specific roles and responsibilities and the challenges they face in their work. This learning approach not only facilitates a deeper understanding of the tool but also empowers employees to apply their knowledge creatively and efficiently, enhancing their problem-solving and adaptiveness in the workplace.

Fusing Constructivism & AI

Generative AI offers us a lot of potential for integrating constructivism in learning. For example, AI can provide realistic and complex problem-solving scenarios, facilitating learners as they navigate through challenges, experiment with solutions, and construct their own understanding. It can also provide scaffolding, giving guidance and support as learners engage with tasks and gradually building their confidence and competence.

When we adopt a constructivist approach, we use AI in a way that provides not just information, but also opportunities for exploration, experimentation, and reflection. It should be designed to prompt critical thinking, support problem-solving, and facilitate the learner's active construction of knowledge, aligning with the principles of Constructivism.

Social Learning Theory & AI

Developed in the early 1960s by Albert Bandura, the Social Learning Theory offers insight into how individuals acquire knowledge, emphasizing that learning predominantly occurs through observation, imitation, and modeling of others within their communities. This theory underscores the importance of social interactions, observation, and processing of information. It suggests that learning is a social endeavor intrinsically linked to the surrounding environment.

Bandura’s Social Learning Theory suggests that individuals don't learn in isolation. Instead, they learn by watching and imitating others. Observational learning, as outlined in the theory, implies that individuals can learn new behaviors and acquire new knowledge by observing others' actions and the consequences of those actions. This approach to learning highlights the importance of social context and acknowledges the significant influence of environmental factors on the learning process.

For example, consider a workplace training scenario where employees are learning customer service skills. Instead of relying solely on manuals or written guidelines, the training program might incorporate videos demonstrating exemplary customer service interactions.?

Additionally, experienced employees might model effective communication and problem-solving skills during role-play exercises. New employees observe these behaviors and the positive outcomes associated with them. Through observation and imitation, trainees learn not only the technical skills necessary for customer service but also the nuances and interpersonal skills that are crucial for success in their roles.

Fusing Social Learning & AI

Generative AI cannot replace human interaction. But it can simulate and enhance important social interactions for learning. It can guide learners through the necessary critical thinking processes for effective social learning by providing prompts, feedback, and additional resources to deepen understanding and reflection.

AI avatars tools like Synthesia can also demonstrate diverse skills and behaviors via AI avatars, allowing learners access to consistent and repeatable demonstrations. This helps learners observe and practice at their own pace, facilitating a self-driven learning experience that’s still rooted in social observation and interaction.

Dual Coding Theory (DCT) & AI

Dual Coding Theory emphasizes that the human mind processes verbal and non-verbal information in separate, related systems. Learners can benefit from receiving information both in visual and verbal forms, as it facilitates better understanding and retention.

For example, if employees are being introduced to a new software too, the training facilitator doesn't rely solely on verbal explanation. They use a well-integrated approach combining verbal instructions with visual aids, such as diagrams, screenshots, and live demonstrations of the software. Employees might also engage with interactive tutorials where they can simultaneously listen to instructions and navigate through the software's interface.?

This dual-coding approach in training allows employees to understand the new tool more effectively, as they can relate the verbal instructions to visual representations and vice versa, reinforcing their learning and retention of the new skill.

Fusing DCT & AI

One of the common misconceptions about AI is that it is text-based only. There are many tools available that offer the ability to provide information in different forms. For example, Dante AI recently announced they now support image recall. With this functionality, you can create a chatbot that retrieves relevant images from your uploaded documents or images. Learners can also have a verbal conversation with the chatbot rather than just text.

AI also provides the ability to generate content in a variety of formats. For example, Eleven Labs can instantly generate realistic AI audio narration. When combined with other tools like ChatGPT generated scripts or storyboards and human generated reference materials, it can improve learning retention.

Example of Theory in Action

In a recent LinkedIn post, Isabella Bedoya shared Dante AI's new image recall feature. She also shared to potential use cases: Pulling images from a manual into the chatbot conversation and finding images to include in a report..

AI & Learning Theories in Action - DuoLingo

Duolingo is the poster-child company for fusing AI with learning.

Even before ChatGPT became mainstream, Duolingo had been using machine learning and other technologies to provide quality digital language tutors at scale. But the technology alone wasn’t what made the learning experiences so successful. It was the fusion of learning theory and AI.

When we think about Behaviorism and learning and development, e-Learning gamification is an easy connection. Learners perform an action and are rewarded accordingly. Using their AI system Birdbrain , Duolingo reinforces performance with rewards and levels.

But Duolingo goes beyond gamification. One of the limitations of Behaviorism is it doesn’t account for our internal influencers and how we process information. This is where other learning theories come into play.

Duolingo chunks information making it easier to learn and dynamically adapts materials based on the learner’s need. Duolingo Max now also leverages GPT-4 for more authentic role-play practice as side-quests along the main path which earn XP.

Fusing Learning Theories with AI

As Duolingo showed, using AI for L&D is most effective when we are grounded in the foundations of learning. Our professional knowledge and expertise are to a successful AI collaboration. AI isn’t a substitute but a complement to our expertise, enhancing the learning experience.

Want to know more about AI in workplace L&D?

?? Visit the AI Innovation Lounge website for resources, prompts, and more

??? Listen to the AI Learner Lounge podcast