I recently asked ChatGPT to list the five most influential instructional designers.
ChatGPT, of course, is an artificial intelligence chatbot that has grabbed the attention of the world. Locally, it was put to the test by University of Minnesota Law School. It answered questions from law school exams in several legal subject areas and when judged blindly, it earned a C+. On every front, businesses and institutions are assessing opportunity – and risk. (New York City has blocked ChatGPT on school devices and networks*.) Every news outlet is talking about it. Artificial intelligence has stepped off the silver screen and into our lives in a new way.
ChatGPT Lists 5 Influential Instructional Designers
I’ve asked ChatGPT several questions over the past two months. But last evening, I turned its attention on instructional design and learning experience design.
So, I ask it: who are the five most influential instructional designers.
Here was ChatGPT’s reply:
Influential Instructional Designers:
- Richard Mayer – Known for his work on multimedia learning and cognitive theory of multimedia learning, Mayer has made significant contributions to the field of instructional design.
- Clark Aldrich – A pioneer in the field of simulation-based learning, Aldrich has designed numerous training programs and simulations that are used widely in corporate and military settings.
- Michael Allen – The founder of Allen Interactions, Allen has designed over 1,000 e-learning courses and is known for his work in instructional design and performance support.
- Julie Dirksen – Dirksen is a leader in the field of user experience (UX) design for learning, and has written several influential books on instructional design and UX.
- Will Thalheimer – A researcher and consultant, Thalheimer is known for his work on evidence-based instructional design and his writing on instructional design best practices.
The five designers listed are certainly influential. In our Learner Experience Design graduate courses at Metropolitan State University, we’ve included the writings and/or texts of Clark Aldrich, Michael Allen and Julie Dirksen. In working with faculty, I’ve used research from both Richard Mayer and Will Thalheimer.
And so, I liked its answer and my curiosity was piqued. To explore further the role of AI in Learning Experience Design, I wondered if ChatGPT could provide content for a quick lesson on that very topic — Learning Experience Design. (It also gave me the opportunity to take the latest nightly build of LodeStar 10 on a little road test.)
The query produced interesting results.
Before I show that content, a few admissions and disclaimers. First, I’m not advocating for the use of AI in the instructional design process. In the near future, I will certainly evaluate its utility. Secondly, in my example I offered ChatGPT no directives. I didn’t specify grade level, target audience, prerequisite knowledge, cultural or situational context. I simply asked it to spit out information.
A further admission is that I’m very uneasy about the topic of artificial intelligence. AI obviously presents both opportunity and challenge. I have a dystopian dread over AI and yet I recognize the many opportunities that it presents. I’ve long held out hope for truly adaptive learning systems that could individualize the scope and sequence of instruction for the benefit of learners. Many systems have appeared on the market with ‘semantic’ engines that don’t quite measure up to a system that can learn from a students’ successes, missteps, preferences and confidence level, and make appropriate decisions. Adaptive Learning is one area that benefits tremendously from AI.
The other thing I recognize is that the information that ChatGPT prints out is, at best, raw material for an instructional designer. The design must include learning experiences that help students engage with the material, think critically, make decisions, make meaning, and make memory.
David Wiley blogs about this more eloquently in AI, Instructional Design, and OER – improving learning (opencontent.org). Here is an excerpt from his blog:
“What distinguishes an educational resource from an informational resource is that the latter shows some intentional application of knowledge about how people learn. I have previously argued that the minimum amount of effort you could invest to convert an informational resource into an educational resource was to add practice with feedback. That simple sounding task quickly explodes in complexity as you consider the research on what form that practice should take, how long it should last, when it should be scheduled, what kind of feedback should be provided, whether the feedback should come immediately or after some delay, etc.”
That leads to my biggest disclaimer. Just for fun, I wanted to see what ChatGPT would produce and how I could use it in a project with minimum effort.
David Wiley writes:
“LLMs will dramatically increase the speed of creating the informational resources that comprise the content infrastructure. Of course the drafts of these informational resources will need to be reviewed and improvements will need to be made – just as is the case with all first drafts – to insure accuracy and timeliness. But it appears that LLMs can get us 80% or so of the way to reasonable first drafts orders of magnitude faster, eliminating the majority of the expense involved in this part of the process.”
I took the ChatGPT output and sprinkled in some activities. The disclaimer is that these are not activities that any of the above authors would endorse. The design is just a simple present-and-check. But I would hazard to say its not unlike the content that gets pumped out by course mills. It’s not unlike the content that one consumes to meet a compliance check or earn continuing professional education credit. It’s not unlike most of the stuff out there that our 5 influential instructional designers decry.
In doing this short exercise, I was reminded of David Thornburg’s often quoted “Any teacher that can be replaced by a computer, deserves to be.” Perhaps, any instructional designer who can be replaced by an artificial intelligence chatbot deserves to be. Or perhaps the real line of enquiry should be: can a chatbot be trained to make decisions that imitate an instructional designer? Can a chatbot be trained to pass the LXD equivalent of a Turing test? (See appendix A for what happened when I gave ChatGPT more information about the audience.)
But disclaimers aside, it is intriguing to me to view what ChatGPT produced.
Rules of Engagement
Here were my rules when applying the content:
- Spend very little time (under two hours)
- Retain all of ChatGPT’s text – even if there were errors (example, Bloom’s Taxonomy was produced in 1956, not the sixties.)
- Format the content in different ways, including placing it in a time line
- Add videos or audio from external sources to supplement ChatGPT and provide some human connection.
- Return ChatGPT’s content to the engine and ask ChatGPT to create a final quiz from that content
You can see this for yourself by clicking on the following link and/or joining me on a screen-by-screen tour (not all screens shown).
Link to content: Learner Experience Design ChatGPT (lodestarlearning.github.io)
Screen by Screen Commentary
First, in my title screen, I give full credit to ChatGPT.
I asked ChatGPT to explain instructional design versus learning experience design. I simply presented the information as a choice between two explanations: one that matched Learning Experience Design and the other that matched Instructional Design. ChatGPT provided the content; I added this little warm-up learner engagement.
I added ChatGPT’s definition of What is Learning Experience Design.
I then added a video from YouTube so that we have an LXD person describe the characteristics of LXD. This reinforces the information and humanizes it.
I asked ChatGPT to list the important events in the history of Instructional Design. It put out a list of events. Interestingly, the earliest event was the publishing of Didactica Magna in 1600s. I took that information and presented it in an interactive timeline.
ChatGPT output the importance of Learner’s Needs and Goals. I added a video where the host really challenges his guests on how LXD designers actually translate goals and user research into meaningful experiences.
ChatGPT output an explanation of instructional methodology. I did a strike-through on key words and asked LodeStar to convert them into fill-in-blanks. I then provided a randomized word list. So rather than just reading, the learner has to think about the context and type in the appropriate word.
To supplement ChatGPT’s explanation of Technology and Tools, I added a podcast of instructors talking about the use of technology and tools in response to COVID.
In the Assessment and Feedback session I added a word problem related to calculating the volume of a cylinder (as an example). The dimensions of the cylinder (radius and height) are randomly generated from a defined range. Students can work on these problems all day and always get a new set when they refresh the lesson.
For User Experience and Interaction Design, ChatGPT spat out content and I added a slideshow viewer of three interesting designs.
Finally, I gave back ChatGPT ‘s content and asked it to produce an assessment. I then took the question stem and answer options and fed them into a Multiple Choice/Multiple Select Question Widget.
Most of the activities were assigned a point value. I then added a report page to send this off to either a Learning Management System or a Learner Record Store (CMI5/xAPI).
I submitted the content to Open AI’s text classifier so that it could determine whether or not the content was possibly AI-generated.
Here is the result:
*OpenAI announced the AI Text Classifier to help determine whether or not content was AI-generated
OpenAI already offers specific application programming interfaces (API) to its language processing engine. For example it offers auto completion in its Text Completion API. OpenAI is now inviting developers to its waiting list to use ChatGPT in new and existing applications. One can easily imagine the day when toolmakers will invoke the API commands in order to, in David Wiley’s words, “dramatically increase the speed of creating the informational resources that comprise the content infrastructure”.
We’ve just completed LodeStar 10 and continue to test it. We have no current plans to interface with AI but have been added to the waiting list by OpenAI to explore ChatGPT API capabilities. Who knows where that will lead? In the meantime, our focus is to help designers start with simple projects and then offer them more capability as their designs grow from simple present-and-checks to short simulations, story-telling, decision-making scenarios and CCAF (Context Challenge Activity Feedback) projects that the five influential instructional designers would endorse. As far as AI is concerned, we have a lot to learn about the role of AI in learning experience design. I would love to read your comments.
Doubtless, the course mills will go into overdrive on this technology. Caveat emptor.
In a second attempt I gave ChatGPT a little more information about the target audience.
Here is my prompt:
Write about Learning Experience Design for an audience of high school students who know very little about instructional design or cognitive science. Write at an eighth grade level.
Here is ChatGPT’s response:
Learning Experience Design is all about creating educational experiences that are engaging, effective, and enjoyable. It’s a field that combines ideas from instructional design and cognitive science to help people learn new things.
Think about it like this: have you ever taken a boring class where you couldn’t stay focused or retain information? That’s where Learning Experience Design comes in! The goal is to create lessons that grab your attention and make you remember what you’ve learned.
The designers use research from cognitive science to understand how people learn best. For example, they know that people remember information better when they’re actively engaged in the learning process, instead of just passively listening. They also understand that people remember information better when it’s presented in a way that makes sense to them and connects to their own experiences.
So, Learning Experience Design combines these ideas to create learning experiences that are tailored to the way people learn. This might include interactive activities, simulations, games, videos, and more. The goal is to make learning a fun and memorable experience that sticks with you, so you can use what you’ve learned in the future.
Overall, Learning Experience Design is a field that’s all about making learning better. If you’re interested in making education more exciting and effective, this might be a great field for you to explore!
When I asked ChatGPT to comment on the reading level of this example, it answered:
This content is appropriate for high school students or early college students.
When I applied the Flesch-Kincaid Reading Grade Level test, it scored 10.4. Just saying.