TL;DR
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
The 74 published an opinion article outlining five approaches to help students learn: use active recall, practice through difficulty, change ineffective study methods, revisit older material and use tutors that prompt rather than supply answers. Its author draws on personal experience and cites a 2025 study of about 1,000 high school students, but the article does not provide the study’s full citation or enough detail to independently assess its findings.
The 74 has published an opinion article arguing that students can learn more when they practice recalling and applying material, revisit what they have forgotten, and receive guidance that makes them think rather than giving them answers. The author, a learning-software developer who says he has spent two decades building learning systems, presents five recommendations for parents, educators and developers of educational tools.
The author describes effective learning as a “Learning Goldilocks Zone”: work should be difficult enough to require effort but not so difficult that a student gives up or guesses. The article connects this idea to researchers’ term “zone of proximal development” and argues that productive practice involves retrieving information, using it in different ways and correcting mistakes. It contrasts that process with rereading and highlighting, which can make material feel familiar without showing whether a student can explain or apply it.
The five recommendations are to use active study methods, treat difficulty and mistakes as part of practice, change methods after a poor result rather than simply studying longer, return to older material to address gaps, and choose tutors or software that ask questions and offer hints instead of supplying answers. The author illustrates these points with accounts from his work, including students who had memorized flashcards but struggled to apply the same information in differently worded test questions.
The article also cites a randomized 2025 study involving about 1,000 high school students. According to the author’s description, unrestricted access to GPT-4 improved performance during practice but reduced exam performance, while an AI tutor that gave hints and feedback instead of answers produced better results. The article does not name the study or provide its full methods, so those findings are reported here as the author presents them.
How Study Habits Shape Retention
The article’s central point is that effort alone may not produce durable learning if students use methods that mainly build familiarity. Its advice shifts attention from how long a child studies to whether the child can retrieve information, explain it and use it after time has passed. That distinction matters to families trying to respond to poor grades and to schools choosing tutoring and learning software.
Its discussion of AI adds a practical question for parents and educators: does a tool help a student do the thinking, or does it complete the task for them? If the author’s summary of the 2025 study is accurate, performance while practicing may not predict performance on an exam. The article therefore treats independent recall and application, rather than immediate task completion, as the more meaningful test of learning.
Popular right now
De'Longhi 15-Bar Pump Espresso Machine with Milk Frother & Temperature Control
Dyson Car+Boat Cordless Handheld Vacuum
Shark Rocket Ultra-Light Corded Stick Vacuum, HV301, Gray/Orange
As an affiliate, we earn on qualifying purchases.
From Flashcards to AI Tutors
The author traces his work in learning technology to a study program distributed on a CD-ROM with a McGraw-Hill textbook about 25 years ago. He says the product later became an app and then an AI-based tool. His account is informed by experience developing software, teaching students and studying learning science, but the piece is explicitly an opinion article, not a reported evaluation of all five recommendations.
Several recommendations rest on familiar distinctions in learning: recognizing information is not the same as recalling it, and a grade may not show what a student remembers months later. The author recounts working with high school students he says had substantial gaps in math and English, and says they reached grade level after targeted review over eight months. This is a personal account in the article, not a separately documented study with a comparison group.
“If the work is too easy, the student learns nothing. If it is too hard, the student gives up or guesses.”
— The article’s author
Evidence Behind the Recommendations
The source does not give a publication date or identify the 2025 randomized study by title, authors, journal or institution. It also does not report the study’s exact exam or practice results, how students were assigned to groups, or how long the effects lasted. Those details are needed to judge how broadly its findings apply.
The article’s personal examples, including the reported progress of students who revisited missed material, are not presented with supporting data or a control group. It is also unclear what specific software, prompts or tutoring approaches were tested in the cited AI research. The article offers recommendations and interpretation, not a complete evidence review establishing that every strategy works equally well for every student.
Questions for Families and Schools
The article does not announce a program, policy change or future research milestone. Its immediate recommendation is for adults to ask students what they got wrong, what they understand now and whether their study method is helping them retrieve information later. It also urges families to respond to a disappointing grade by considering a different method rather than defaulting to more hours of the same approach.
For schools and families weighing AI tutors, the next step is to examine the tool’s behavior: whether it gives direct answers or prompts students to try, offers hints and responds to errors. Readers seeking to verify the article’s research claim would need the full citation and study details, which are not included in the supplied source material.
Key Questions
What are the five steps in the opinion article?
The author recommends active recall, learning through effort and corrected mistakes, changing ineffective study methods, revisiting older material, and choosing tutors that guide students instead of giving them answers.
What does the article mean by the “Learning Goldilocks Zone”?
It is the author’s name for work that is challenging but manageable: easy enough for a student to keep trying and difficult enough to require learning. The article relates the idea to the zone of proximal development.
What does the article say about AI and exam performance?
As described by the author, a randomized 2025 study of about 1,000 high school students found that unrestricted GPT-4 access improved practice performance but reduced exam performance, while an AI tutor offering hints and feedback produced better results. The article does not provide the study’s full citation or methods.
Does the article establish that these methods work for every student?
No. It presents the author’s recommendations, experience and summary of one study. The source does not provide enough detail to establish that the approaches work equally well for all students or settings.
Source: the74million.org
