Appendix 1: Accessible text versions of the Learning through analogies prompts and outputs
Prompt:
You are an academic in Education, I am a first year university student. Please explain Vygotsky’s theory of the zone of proximal development, using an analogy.
Output:
Sure, I’d be happy to explain Vygotsky’s theory of the Zone of Proximal Development (ZPD) using an analogy.
Imagine you’re learning to ride a bicycle. At first, you might need training wheels or someone to hold the bike steady while you get the hang of pedalling and steering. This is something you can’t do on your own yet, but you can do it with help. This is what Vygotsky would call the Zone of Proximal Development.
The ZPD is the difference between what a learner can do without help and what they can do with guidance from someone more knowledgeable or skilled. In our bike-riding example, the ZPD includes all the skills you’re using to ride the bike with help, like balancing, pedalling, and steering.
As you practice and get more comfortable, you’ll start to be able to do these things on your own. The person helping you might let go of the bike, but stay close by in case you start to fall. Eventually, you’ll be able to ride the bike all by yourself. The skills that were once in your ZPD have now become things you can do independently.
This is the essence of Vygotsky’s theory. Learning is a social process that happens when learners are in their ZPD, with the guidance and support of others. As learners master the skills or knowledge in their ZPD, they become more capable and their ZPD expands to include more complex skills or knowledge.
Remember, this is a simplified explanation and Vygotsky’s theory is much more complex and nuanced. For a more detailed understanding, I would recommend reading Vygotsky’s original works or academic texts that delve into his theories. Text created by generative AI: RMIT (2024). Val [Large language model]. https://val.rmit.edu.au The text copied to the clipboard is current as of 07 February 2024.
Prompt:
You are a university lecturer in artificial intelligence. Explain how large language models work, using an analogy.
Output:
Sure, I’d be happy to explain large language models using an analogy.
Imagine you’re trying to learn a new language, let’s say French. You start by learning the basics, like common words and simple sentences. As you progress, you start to understand more complex grammar rules and idioms. You read French books, watch French movies, and even try to have conversations in French. Over time, you become more proficient and can understand and generate French sentences more accurately.
This is similar to how large language models work. These models, like GPT-3, are trained on vast amounts of text data. They start by learning the basics, like common words and their usual contexts. Then they learn more complex structures, like sentence construction and grammar rules. They also learn to predict what word or phrase is likely to come next based on the context.
The more data they are trained on, the better they become at understanding and generating text. They can answer questions, write essays, and even generate creative stories. However, just like a person learning a new language, these models are not perfect. They can make mistakes, misunderstand context, or generate inappropriate responses.
It’s also important to note that these models don’t truly understand the text in the way humans do. Going back to our analogy, it’s like someone who can perfectly mimic French sentences without understanding what they mean.
Remember, this is a simplified explanation. Large language models are complex and involve many more aspects, like neural networks and machine learning algorithms. For more detailed information, I recommend referring to academic resources or reaching out to experts in the field. Text created by generative AI: RMIT (2024). Val [Large language model]. https://val.rmit.edu.au The text copied to the clipboard is current as of 06 February 2024.
Return to Learning through analogies section.
