How to Teach Judgment in the Age of Instant Answers
Dr. Rachel Besing, Professor of Psychology and Neuroscience, Brescia University
In an era when students can generate a confident answer in seconds, Dr. Rachel Besing teaches a harder skill: how to distinguish a convincing answer from a defensible one.
Using LLMs, students can now produce polished explanations before they have learned how to examine the evidence beneath them. If higher education focuses only on policing the final paper, it risks missing the more important task: teaching students how to move from a question to a conclusion they can actually support.
Rachel teaches that process in the classroom. When a student asks whether something has been studied and she does not know the answer offhand, she does not guess or defer the question until later. She pulls Consensus onto the classroom screen and says: “Well, here, let’s look.”
She searches alongside her students, showing them how to move from a question into the research and reminding them that what they find is a starting point for investigation, not simply an answer to accept.
Her students are watching more than a search. They are watching a scientist encounter uncertainty without disguising it: ask a question, examine the evidence, qualify what can be said, and change course when the research demands it. At a moment when AI can make uncertainty look like certainty, Rachel models how to investigate before deciding what to believe.
Rachel is a professor of psychology and neuroscience at Brescia University, a small liberal arts institution in Owensboro, Kentucky. Trained in behavioral neuroscience, she teaches courses including biological psychology, psychopharmacology, statistics, research methods, clinical neuroscience , sleep and circadian rhythms, and the psychology of fear.
Across all of them, she wants students to do more than find information. She wants them to judge whether it is credible. She first heard about Consensus through the Society for the Teaching of Psychology, where instructors were discussing how to guide students’ use of generative AI. Her frustration was direct: “I got so sick of reading plagiarized papers, and I wanted to give them a tool that was useful.” Instead of spending her energy catching misuse, she wanted to give students a better way to do the work.
Making statistical thinking visible
One of Rachel’s live searches began with a public claim about acetaminophen and autism. The timing was fortuitous: her statistics students were learning about effect sizes. Rachel opened Consensus in class and pulled up the relevant studies. Together, they examined the research behind the claim and discussed a distinction that can disappear in headlines: a result can be statistically significant without being large enough to carry the meaning assigned to it in public debate.
The point was to give students access to the studies, then show them what a trained reader notices: the population, the context, the size of an effect, and the distance between a paper’s finding and the claim someone makes about it. That kind of lesson is difficult to reproduce with a static slide. By searching live, Rachel can model how an expert evaluates evidence, and how they change course when the evidence complicates the original question. Her students see that scientific literacy is more than locating a peer-reviewed paper. It is knowing what the paper can and cannot support.
Rachel sees the same problem far beyond a single news story. Research increasingly reaches people as a headline, a bullet point, or a simplified claim. The temptation is to absorb the conclusion and keep moving. Rachel believes that students are less likely to develop sound research habits if the path from curiosity to credible literature remains opaque. Students need access to credible papers, but they also need to know how to form a question, evaluate a source, and build an argument.
Consensus became one way Rachel could make that process more approachable. In class, Consensus lets her turn spontaneous questions into live literature searches. In research methods, it helps students locate real papers connected to questions of their own. The reading, interpretation, and scholarly decisions remain theirs.
Building a proposal one decision at a time
In Rachel’s 400-level Research Methods course, students produce an original research proposal and present it as if she were reviewing the science for an NIH study section. Rather than face one enormous deadline, they move through the project in stages: analyzing the structure of an empirical article, generating a novel hypothesis, identifying five relevant sources, and writing the proposal section by section. The structure slows the process down enough for Rachel to teach it, making each intellectual decision visible before students move to the next.
They also participate in what Rachel calls “research paper speed dating,” preparing short elevator pitches and exchanging drafts with classmates. The exercise pushes them to explain their reasoning, hear where it is unclear, and refine the connection between evidence and argument. The scaffolded approach gives students a way to build on what they are learning rather than confronting the entire research process at once.
But that structure also reveals a challenge many faculty will recognize: Rachel can see the sources students submit and respond to their drafts, but she cannot easily see the search process that produced them. “I don’t know what search terms they’re using. I don’t know what engines they’re always using. I don’t know what databases they’re on,” she says.
She wants to give feedback not only on the finished paper, but on the decisions that produced it. That is where academic research tools could be especially useful: making more of the research process visible enough to evaluate, discuss, and teach.
Make the lesson hard to forget
Rachel’s teaching can also be playful. In her Psychology of Fear course, the major project does not take place in a lecture hall. She takes students to a purportedly haunted location in Kentucky, where they go on ghost tours and even stay overnight. Then they bring the course with them, using what they have learned about fear and perception to examine the stories and explanations they encounter.
The assignment says something about how Rachel thinks about teaching. If she wants students to practice evaluating a claim, she looks for a setting where that question will feel real, surprising, and worth remembering. Rather than simply explaining critical thinking, she creates situations in which students have to practice it for themselves.
What students carry out of the classroom
Rachel has been thinking about what makes the classroom valuable when information and answers are easier than ever to produce. She talks about the importance of making the “covert curriculum” visible: not just what students learn, but the habits of thinking they develop along the way. For higher education, she says, part of the challenge is helping students understand “the value of why they’re in the seats in the classroom.”
That may be one of the more important questions AI is forcing faculty to confront. If students can get to an answer faster than ever, the value of education cannot rest on the answer alone. It also has to lie in what students learn by getting there: how to follow a question, encounter evidence that changes their mind, make sense of uncertainty, and develop confidence in their own ability to investigate something for themselves.
