The use of AI tools is becoming increasingly prevalent in educational settings. Many teachers have found that working with AI tools can make their teaching more efficient and a study by the EEF has found using AI tools can cut time spent on lesson planning by 30%.
There is also no doubt that students regularly use AI to support their academic work. A recent survey of students found that almost all of them (92%) are now using AI in some form with 88% having used generative AI tools for assessments.
But is there potentially a deeper and darker cost to this prevalence of AI use when it comes to student learning?
How AI impacts student memory and creativity
AI can impact the memory and creativity of students in a few ways, as shown below:
1. AI and memory
A recent study from MIT looked at essay writing and the participants were split into several groups. In one of these groups, the participants weren’t allowed access to any external sources of information meaning they had to write it totally themselves. In another group the participants were allowed to use Chat GPT to assist in their writing process.
After completing the essay, participants were interviewed about their experiences and the essay content. The most interesting finding was participants’ responses to the question “Can you quote any sentence from your essay without looking at it”. Whilst 11% of the participants who wrote it all themselves failed to answer this question an astonishing 83% of the ChatGPT group couldn’t remember a single quote from the essay they had just written.
To quote from one of the participants in the ChatGPT group:
“I tried quoting correctly but the lack of time made it hard to really fully get into what ChatGPT generated”.
The researchers concluded that “while AI can be used for support during a task, there may be a trade-off between immediate convenience and long-term skill development”.
2. AI and Variation
Another interesting finding from the study discussed earlier was that participants using ChatGPT “produced statistically homogeneous essays… showing significantly less deviation compared to the other groups”. This suggests that when groups of students use AI tools in their learning, it can decrease the overall creativity of the group as the outputs all begin to look very similar.
Truly creative ideas are often the outliers and come from people who don’t follow the tried and tested path. This is echoed in a study looking at creativity in a completely different field – making paper aeroplanes. The researchers compared paper aeroplanes made by participants who were allowed to use the internet to learn how make the planes and those who had no access to the internet.
They found that while the planes made by the internet-supported group flew further on average, the planes that flew the furthest came from the group who were not allowed to use the internet. Those participants who followed the tried and tested pathways produced very similar aeroplanes whereas those who used their intuition, instinct and pure imagination broke conventions around what a paper aeroplane should look like.
Taken together, these two studies suggest that whilst using technology might raise the floor and lead to fewer low-quality outputs, this could come at a cost to innovation. As a society we need outliers and innovators – without them, would we have ever seen the genius of Freddie Mercury producing Bohemian Rhapsody?
Implications of AI in the classroom
Although teachers may find that using AI tools can increase their productivity and improve the quality of their work, this does not necessarily mean we should endorse the widespread use of AI tools by students. Students are by definition novices and previous research has shown that novices and experts engage with AI tools in very different ways. Whilst experts might be able to work with AI as a co-intelligence, using the AI as an assistant to help refine ideas, novices are more likely just to ask an AI to quickly provide answers.
With any task designed for learning, it is the process that is important, not the output. We are more interested in the quality of the thinking that went in to generating the essay rather than the actual essay itself. Students who produce high-quality essays using AI, but who subsequently can’t remember anything about the topic, are bypassing this important learning process.
It therefore seems pertinent to discourage the use of AI tools until students have gained sufficient expertise to work with the AI and effectively critique its output. When novices rely on AI to produce work, they run the risk of offloading the important cognitive processes necessary for effectively encoding information into their long-term memory.
Final Thoughts
AI can be a useful tool in the hands of experts, but in the hands of novices there is a risk that it can result in cognitive offloading, decreased learning and lower creativity. As teachers, it is important that we discuss both the opportunities AI can bring but also the threats it poses to learning. By doing so, we can help students to make the most appropriate choices as to when they should be using AI and when they are better off using their brain only.
To find out more about how AI in education, have a look at our Guide to Using AI in Education or enquire about our workshop on Using AI in Education.
