Half-education by design

Excessive cognitive offloading interrupts necessary internalisation, leaving us with superficial schemata—weak mental frameworks that can't adequately support critical thinking or creative problem-solving. Barbara Oakley et al., The Memory Paradox

Reading Barbara Oakley and colleagues' paper entitled The Memory Paradox: Why Our Brains Need Knowledge in the Age of AI this week, combined with proofreading hundreds of reports and seeing yet another potted guide to teaching, has led me to wonder whether we are seeing more than a technological shift in education.

Knowledge has been a key recurring theme in my teaching this week. I marked and added comments to the Year 12 Politics A-level end-of-year assessments and also hosted the final Human Conversations Across Cultures and Times co-curricular session for students in Year 9 and above. On the assessment papers, I made references to case studies involving the UK Supreme Court, the Greensill scandal, and acts of devolution. In the after-school session, we discussed ideas beyond the school curriculum, leading one of the students to ask why they were taught a subject in a particular way because it seemed narrow. It was a good question.

I've used AI tools to support the work in both instances, but to help the internalisation of knowledge so they can call on it in an essay or in a discussion when all they have is their body and their ability to engage with questions from a paper or from other students/myself.

The students could have used the technology to replace their own work, and teachers may use AI tools to write reports or produce lesson plans. There is an obvious seduction in this as there is the problem of having too much to do and not having enough time. Generating lesson plans in seconds, producing detailed reports via clear prompts and rubric-perfect essays lead to impressive outputs, more polished than what a teacher/student may generate at 11pm on a weekday or on a Sunday evening. Like many colleagues, I understand the pull. However, the 'cognitive offloading' involved in using these tools leads to the dulling of individuality and expertise.

In the paper, Oakley et al. suggest that reliance on external tools reduces the capacity to transition from declarative memory to procedural memory; the cognitive wrestling that builds expertise gets bypassed. The lesson is delivered, the report gets produced and the essay is submitted and passes muster successfully, but the intellectual work that is essential for creating expertise is circumvented. More than that, it leads to a flat sameness. Whole books and fields and vast datasets are reduced to outputs devoid of nuance and the outlier expertise, the original turn of phrase that drives innovation and sparks curiosity, are now smoothed out by regression to the mean. The appearance of substantive intellectual engagement masks the omission of engaging in the main intellectual work itself. For the consumers of this tech and productions, they can engage in best practices without understanding why those practices work, when they might fail and how to adapt when they do. This is Adorno's analysis realised: a half-education for teachers replicated and packaged as a half-education for students who replicate that in an assessment system that is flawed.

I'm not suggesting that we reject AI or research summaries. In my view, this is impractical and unwise. What I do think is important is asking a fundamentally different question from the one that typically revolves around time-saving when it comes to essays, teacher development and report-writing. We should ask whether using these summaries and tools preserves student and teacher capacity for independent expertise. The goal here is that the work or the tools serve as amplifiers of cognitive effort rather than replacements for it. In simple terms, it might mean treating AI outputs as a starting point for intellectual engagement rather than an end. Scholarship and professionalism should sit at the heart of what we do as schools; otherwise, education becomes devoid of dignity and expertise, and is cheapened.