The Learning Crisis in Universities: Redesign or Redundancy......by Pushpinder Singh Gill

​The real crisis in higher education is not employability, funding, or technology. It is that universities continue to confuse teaching with learning. As long as this confusion remains unaddressed, no policy reform, digital platform, or curriculum revision will produce meaningful change.

Universities are still organised around a model that assumes learning happens when content is delivered—through lectures, syllabi, and credit hours—and verified through examinations conducted under artificial constraints.

This model endured for centuries not because it reflected how humans learn, but because it was administratively efficient. It allowed institutions to scale, standardise, and certify large populations. What it did not guarantee was understanding.

That distinction now determines relevance.

In most universities, learning remains time-based rather than mastery-based. Students progress because semesters end, not because understanding has been demonstrated. Courses are “completed” when content is covered, not when it is internalised.

Examinations reward recall under pressure rather than reasoning under uncertainty. These practices persist because they are familiar, not because they work. What makes this model increasingly redundant is not technology, but knowledge. We already know how learning actually happens.

Decades of research across cognitive science and skill acquisition converge on a few settled truths: learning requires effortful engagement, timely feedback, iteration, and application across contexts. Understanding develops unevenly. Forgetting is predictable. Mastery cannot be rushed, but it can be measured. None of this aligns with lecture-driven instruction followed by episodic testing.

Yet universities continue to privilege delivery over diagnosis. They are structured to broadcast information, not to detect misunderstanding. Grades tell students where they rank, not what they failed to grasp. By the time gaps surface, they are often too deep to repair within the system. This is not a marginal inefficiency. It is a design failure.

If universities wish to remain learning institutions rather than credentialing agencies, they must abandon several entrenched assumptions.

The first is that lectures are the primary vehicle of learning. Lectures can frame ideas and inspire curiosity, but they are among the weakest tools for building durable understanding. They privilege fluency over comprehension and reward passive familiarity rather than active use.

The second assumption is that uniform pace equals fairness. In practice, uniform pacing ensures that learning works well for only a narrow segment of students. Everyone else is either disengaged or left behind. Fairness in learning does not mean sameness; it means responsiveness.

The third assumption is that assessment can simultaneously measure learning and sort people efficiently. These goals often conflict. Sorting rewards speed and test familiarity. Learning requires time, exploration, and revision. When institutions prioritise sorting, learning becomes secondary.

Once these assumptions are discarded, the redesign task becomes clearer.

A serious learning system must be organised around mastery, not exposure. Subjects must be broken into conceptual units that require demonstration of understanding before progression. Learners must be allowed to move at different speeds without stigma. Confusion must be treated as diagnostic information, not as failure.

Assessment must follow suit. Instead of relying primarily on high-stakes, time-bound examinations, universities must shift toward continuous, performance-based evaluation. Explanation, application, and synthesis must carry more weight than recall.

Portfolios of work, oral defences, live problem-solving, and iterative projects provide far richer evidence of competence than a single exam ever can.

This is not an argument against standards. It is an argument for better ones.

Technology, particularly artificial intelligence, becomes meaningful only within this redesigned architecture. Used properly, intelligent systems can diagnose misconceptions early, generate targeted practice, track learning trajectories, and schedule revision before forgetting sets in. These tools do not replace academic judgment; they support it. They free teachers from repetitive evaluation and allow them to focus on mentoring and intellectual guidance.

Without structural change, however, technology merely accelerates dysfunction. If exams reward memorisation, AI will be used to memorise faster. If grades remain the dominant currency, students will optimise for grades. Tools do not transform systems; incentives do.

Faculty roles must therefore evolve. Universities must invest seriously in retraining teachers as learning designers and mentors, not merely subject experts.

Designing problems, sequencing concepts, and interpreting learning data require different skills from delivering lectures. This is not a demotion of the academic role. It is a professional elevation—but one that demands institutional commitment.

Adult and professional learning further exposes the inadequacy of existing structures. Adults do not need exhaustive syllabi disconnected from immediate purpose. They need targeted learning tied to real problems, applied quickly, and validated through performance. Institutions that insist on long, inflexible programs for short-term learning needs will simply be bypassed.

At a deeper level, universities must confront a truth they have long avoided: the most valuable outcome of education is not subject mastery alone, but the capacity to learn independently. In fast-changing domains, no curriculum remains current for long. What endures is learning velocity—the ability to identify gaps, seek resources, test ideas, fail productively, and adapt. This capacity must be cultivated deliberately.

India’s New Education Policy acknowledges many of these issues in principle. Its emphasis on flexibility, multidisciplinary learning, and critical thinking is directionally sound. But it stops short where redesign becomes disruptive. It does not dismantle the dominance of high-stakes examinations. It remains vague on assessment reform. It underestimates the scale of faculty retraining required. And it assumes compliance without altering incentives.

So what should be done now?

Universities should begin not with sweeping declarations, but with measurable pilots. One multidisciplinary college under the NEP framework could replace end-semester exams in a single program with mastery-based portfolios, where students demonstrate understanding through projects, explanations, and applied work. Faculty time should be redirected from grading scripts to mentoring and feedback. Learning outcomes should be made explicit and auditable.

In parallel, institutions should pilot mastery progression in a small number of foundational courses, allowing students to advance only when concepts are understood, not when the calendar dictates. These pilots would generate evidence—of learning quality, student confidence, and faculty workload—that policy documents alone cannot provide.

These steps are modest, testable, and reversible. But they would signal seriousness.

Universities that continue to optimise for content delivery and credential issuance will survive for a time on reputation and inertia. But they will no longer be central to learning. Those that redesign themselves around how humans actually learn can reclaim their relevance—not as factories of certificates, but as institutions that develop genuine capability.

The choice is no longer abstract. Redesign learning—or drift toward redundancy.

February 3, 2026

​