Connecting Critical Reading and AI Literacy

Shevaun Watson and Cassandra Phillips opened April’s Active Teaching Lab with two provocative arguments: reading is more difficult today than it was in the past, and the tools promising to make it easier are part of the problem. In response, Watson and Phillips reorganized their English and Humanities courses to focus on reading comprehension, help students identify why reading is difficult, and provide them the tools to find non-AI solutions specific to their reading needs. 

The Problem with Digital Reading

Watson and Phillips identify two primary culprits. The first is digital reading itself. When we read digitally, we’re encouraged to skim, constantly seek more interesting content, and stay within the platform we’re using. That’s the playbook of social media, and it carries over as we read other digital material. AI reading tools worsen the issue. Social media influencers and technology companies market AI as the smart path to academic success. They promise speed, volume, and frictionless learning, but extract a heavy cost. 

The second culprit follows from the first: students lack sufficient training to critically navigate different types of writing. In decades past, we adapted our reading to match the content we read: we scanned a dictionary, skimmed an article, and slowly soaked in our favorite novel. We understood (perhaps implicitly) that differing genres had different purposes and required different strategies. Today, because of the digital context in which students read, most enter college not knowing how to engage with long-form text, or even how to identify a text’s argument and purpose. Students weren’t given an opportunity to develop a reading toolkit. They therefore tend to read uniformly, line by line, the way they were taught to read a high school textbook, and expect each section’s main point to be clearly identified in a text-pullout. Yet, in college, the K-12 guardrails disappear. Students are thrown into a sea of genres without a map, and assigned digitized content almost entirely stripped of context. We should not be surprised when they flounder.

Improving Reading Strategies and Student Metacognition

Watson and Phillips argue that our response to students’ reading struggles isn’t to tell them to read more carefully. Students already know they’re perceived as bad readers – they’ve been told that for years. Instead, as detailed in the session recording, Watson and Phillips’ worked to develop reading skills and metacognition. Throughout the semester, the instructors explain to students the purpose and genre of each assigned text, they model multiple ways to engage with it (for content, for rhetorical purpose, for citations), and have students reflect in daily reading notebooks to build their metacognitive awareness.

The results from Watson and Phillips’ fall pilot (33 sections across English 101, as well as first-year bridge courses) are instructive for all instructors who find it difficult to get students to read. By semester’s end, students reported better understanding of how to read in a college context, a better ability to identify AI’s capabilities and limits, and subsequently relied on AI tools less.

Next Steps

Watch the recording for a fuller overview of Watson and Phillips’ teaching strategies, their research context, and a detailed walkthrough of the curriculum.

Review the session slide deck and digital resources below, all shared by Watson and Phillips.