What motivates people to learn? How could the structure of education be changed to promote student motivation? Could these changes be made outside of a traditional classroom or course setting?
Theorists of motivation have put forward two main types: intrinsic and extrinsic motivation. Intrinsic motivation is a desire for a thing for its own sake. Extrinsic motivation is desire for something as a means to another goal, or to fulfill someone else’s expectations. Edward Deci and other pedagogical theorists argue that intrinsic motivation leads to deeper comprehension of a field and overall better learning outcomes, and that students who are motivated only by grades and social expectations are less likely to strive for excellence, retain knowledge, and make future contributions.
While I think it is clear that students who truly desire learning for its own sake will learn more, in practice, personal motivation is rarely clearly separated into these different types. For the great mass of people, even those who will likely go into the professional world, motivation to learn is a mixture of desires: curiosity, enjoyment of the subject, wishing to enter a well-paying or respected field, fulfillment of expectations of parents or teachers, to keep up with one’s peers, to be better than one’s peers, to be a good student as a matter of personal identity, to make good on one’s investment in education, intending to use the skills gained to help others, etc.
Further, there are very few people who have sufficient internal motivation to carry them through an entire course of study, regardless of initial interest. Consider those who enroll in Massive Online Open Courses. Some may wish to gain a skill for practical career growth, but most do start with meaningful intrinsic motivation. Why do so many fail to complete these courses? For most people, either they had a mistaken impression of the field, or their interest didn’t match up to the effort required.
Online learning lacks the typical sources of extrinsic motivation that a traditional classroom setting has. Physical presence engenders types of motivation that cannot be replicated online:
- Mimetic Desire: When students sit in a class and face a lecturer or each other in a studio setting, they have learning as a socially acknowledged common object of desire. Online lectures, even if synchronous, do not create the same psychological response.
- Cohort Maintenance: students wish to keep up with others in the same cohort who are sitting around them. People don’t perceive other participants in a video conference in the same way.
- Friendship: In Aristotle’s framing, they can form friendships both of utility in their learning and peer tutoring, and of the cultivation of virtue as students. Being a good student becomes part of being a good friend.
- Location-based Habit: a classroom is a physical setting for learning, as a space that is marked apart, and presence in it helps to set one’s mind to the task. Presence is an investment in the pursuit of learning that builds to further effort.
- Integrity of Identity: Further, presence commits one to the identity of being a student. Once you say to yourself, “I am a student”, it should naturally lead to thinking “I want to be a good student”, then that learning and studiousness are good, and that it is desirable to learn the subject for its own sake. This is a ladder from extrinsic to intrinsic motivation.
These are some of the reasons, which in my view are under-appreciated, for why in-person lectures help to carry one’s interest through the troughs of motivation. A self-motivated learner can watch a series of video lectures, take notes, do the readings, work through the exercises, seek out help on forums online, and even devise projects to apply the skills. But for most learners who do not have the same level of interior motivation, the sources based in a physical community are invaluable to sustain interest, even though they are mostly extrinsic.
With all of these benefits of physical presence, is there any hope for the efficacious adoption of online courses outside of the highly motivated minority?
Here I will try to sketch an answer that gains the benefits of an in-person community while not requiring the traditional classroom structure. Content distribution, in itself, has effectively zero marginal cost. The best static content: readings, video lectures, problem sets, and project prompts can be distributed for free. This was the promise of MOOCs – but they lacked the motivational structure. The online ‘community’ was not meaningfully better than that of existing forums or Stack Exchange. How can we join this content to a real community?
“Company” Cohort Formation through Pizza Subscription:
Eating together is the root of companionship, and growing together in learning makes a cohort. What if we could form “learning pacts” with “pizza subscriptions”?
In the same way that people find training buddies to go to the gym at regular times, students naturally also form study groups with others of similar skill. Training buddies in a “gym pact” help to create the kinds of motivation formed through community and social expectation discussed above. Going to a gym with the same set of friends builds a habit. The college cafeteria serves a similar purpose: it builds companionship among a group of learners. Enrollment is not a purchase of content: it is buying entrance into a learning pact and community. But we should be able to create such pacts at much lower cost than college today.
There are three parties to this pact: the students, the content creators, and the “cafeterias”. Imagine a “pizza subscription” that is tied to the purchase of an online course: “Buy this online course content, and every Tuesday night for 15 weeks, get 2 pies at a 10% discount if you pay upfront”. Students who want the benefits of such a pact can find others with whom they can jointly purchase this subscription. The subscription is a costly signal of commitment to the group. Once they buy the subscription, eating there is cheaper than any other option, which motivates them to go, in addition to seeing their friends. Students who have purchased such subscriptions are connected together on an online forum to discuss course content. Participants watch lectures and do readings outside of the meeting times, and get together to work on problem sets and projects. Any venue with under-utilized function space and WiFi that serves food can serve as a meeting spot. Restaurants want regular business and to get paid before the goods are delivered. They can also load-balance their business by offering promotions on specific days. Content creators want to sell courses with zero marginal cost. They can take a cut of the subscription. To tie all of the motivation together, when purchasing the subscription, the group also sends a calendar invitation to their friends and members of other such learning pacts for their final project presentation which happens at the end of the course. This functions in the same way as signing up for a future race at the beginning of a training program. The project is an opportunity to practice teamwork and then becomes part of each student’s portfolio. To gain high-level direction and guidance on theoretical depth that is difficult to obtain from forums, hire a graduate student for 5-10 hours of 1:1 or group tutoring through the course. I believe that what I have described here is a Pareto of the college educational experience for a tiny fraction of the cost.
Cost: Suppose 3 friends each pay $200 dollars for a 15 week food subscription, plus access to course content, and the cut of the course is $100 of the sum. The access to the course isn’t a necessary component of this because there are plenty of free courses online, but including it completes the picture. There are plenty of inexpensive restaurants that should be able to profitably serve $11/person/meal especially for regular business that is paid in advance. Posed differently, it is two pizzas for $33 for repeat business on a slow night. Restaurant chains could partner with this “subscription originator” to economize the organizational overhead of coordinating the promotions. The marginal cost to the student is about $8/meal over making one’s own food, or $120 in total. A graduate student tutor would cost a few hundred dollars more.
Team Formation: Forming such groups with friends is ideal, but not everyone has friends looking to study the same things at the same time. What would matchmaking between students require? Indication of interest, availability, and possibly some test of skill. Depending on the subject, there may already be forms of online assessment to automate this testing. Tools like ChatGPT can complicate such assessment, but as a tool that will become part of one’s toolbox, they could also be allowed. Teams should be in groups of 2-4. Even if mismatches of skill occur, they can become opportunities for practicing mentoring for those that advance faster than the others.
Limitations and Challenges: The obvious missing component is credentialing. Having an expert review one’s work costs a meaningful amount of money. The best kind of assessment is an oral exam by an expert. Oral exams don’t have efficiencies of scale. However, we can treat that as a separable problem. Some skills are directly demonstrable without a credential, and some people truly want to learn even without acquiring one.
The other major challenge is that without some kind of binding institutional affiliation and vetting, forming long-term groups with relative strangers is a tough sell. This social pattern could work well for people who are already friends, but it is likely difficult to make work among strangers. It might be suitable among friends of friends, and could benefit from integration with an existing social network. This is hard to estimate without detailed market research.
Could a matchmaking system for these study groups be a viable business? Customer acquisition and retention through repeat subscriptions would be the biggest cost. Convincing people that this is a reasonable social pattern for pursuing learning does not create demand for a specific “subscription originator” which means that the first-mover advantage is small. For these reasons, such a business would not have a moat and is unlikely to be viable by itself.
Instead, online course systems seem like the natural hosts of such a matchmaking system. These platforms already have databases of people interested in such courses. It would be easy to connect them based on location with such promotions. Alternatively, “pizza subscription” independent of matchmaking could be a valuable feature for online order processing systems but would not be a separate business. Such subscriptions need not be driven by study groups. Some people might be willing to buy into regularity for a discount, which is just a different kind of rewards system. But because course platforms have the list of likely customers, it seems like they would be more suited to taking the initiative.
Mike's Engineering Blog 