04 September 2019

Rube Goldberg Machines for Motivation

In a recent Edutopia article Anthony Palma said, “[I]n my experience, reform policies [that encourage flexibility about retaking tests and giving multiple chances to complete assignments] benefit a small portion of the student body, whereas traditional policies [that encourage rigidity] better serve a majority of students.” Unfortunately, Mr. Palma misunderstands motivation and is perpetuating the use of Rube Goldberg Motivation Machines. The truth is that the strictness or flexibility of the policies are not going to have direct effects on motivation, they are only going to have indirect effects that are dependent on the psychological conditions in which those policies are used.

His first defense for his point of view was in reference to motivation. I'm a researcher and author in the area of motivation in educational settings and I am familiar with that area of research. Based on what I know, his experience appears to be at odds with research on motivation in school age children. Taking the broadest view, Gallup reported in 2018 that 53% of K-12 students are disengaged; therefore, on the face of it, “traditional policies,” in general, clearly do not serve the majority well. And Gallup's reported rate for disengagement is likely low. (In my book More Joy More Genius I cite other experts who arrived at higher figures and based on published data I estimate 65-75% of the overall student population will ultimately be affected.)

A more specific claim about motivation that is problematic in the article is when Mr. Palma refers to a study of college-age students to justify a claim that deadlines should be strict for K-12 students. In research done within the tradition of Self-Determination Theory (SDT) on K-12 students the opposite was found. Rigid deadlines can be a hindrance not a help. Finding the opposite across these two contexts might seem strange. But it makes sense if you consider the fact that college students have more of a choice about being in their classes than K-12 students and the need for autonomy that can be satisfied through choice making is as psychologically important for motivation as water and food are to physiological health. The published recommendations based on research with K-12 children is for teachers to be flexible, not rigid. But, I will come back to this in a little bit to clarify further because the question that is being answered by relative levels of rigidity to flexibility is more problematic than any given answer.

Let's consider another potentially problematic quote from Mr. Palma, “A key flaw in the reform policies is the assumption about motivation. Many students—not all, but many—require extrinsic motivation. Due dates, one-time assessments, and late penalties provide motivation for the majority of our students.” The problem I have is that Mr. Palma is correct that inaccurate assumptions about motivation are embedded in many policies but he is also making an inaccurate assumption about motivation. This is a bit ironic since he is drawing from motivational research to bolster his misunderstanding. Motivation was the topic of my thesis research and is a central theme in my books. When I look at the school system from a motivational perspective it looks like a Rube Goldberg machine. Just in case you are not familiar with a Rube Goldberg machine it is an extraordinary collection of seemingly random junk cobbled together to accomplish the most mundane of tasks like making breakfast. They are a popular way to convey a cute version of the “mad” scientist image, such as Robin Williams' character in Flubber or the Wallace and Gromit movie Curse of the Were Rabbit. Another popular version is the Mousetrap Game that came out in 1963. The original cover for the game (below) shows a nice diagram of how the trap works in only 16 easy steps: a real world mass-produced Rube Goldberg machine. Each step is itself a simple machine. There is a sense that the inventors of these things operate under the misguided sense that mechanical wizardry should be applied to even the most trivial tasks that can otherwise be completed with only moderate efforts using simple tools.

In order to better reflect an accurate understanding about motivation I would strengthen his first claim about the “key flaw” I quoted above by substituting the phrase “almost all” for the phrase “the reform.” Almost all policies in education are based on assumptions about motivation; and the problem is that those assumptions are almost all wrong. The inaccurate assumptions about motivation embedded in mainstream schools are a constant source of annoyance for those of us who have a deeper understanding of the topic.

Comparing schools to Rube Goldberg machines might not sit well with teachers and their fans. Let's acknowledge the school folks who have invested their lives in creating the best they could come up with given what they had. It would not be surprising for them to bristle at the suggestion that they are off the mark. To be honest, they are not to blame; the fact that their efforts resulted in Rube Goldbergian motivational schemes was just a natural consequence of trying to manage a complex system with intuitions that misguide most school leaders about their influences on the motivation of both children and teachers. Those teachers are also victims of the same misunderstandings and the Rube Goldberg motivational machines that result from them.

Mr. Palma is, in a certain sense, correct to suggest that extrinsic motivation is OK. However, he appears to be relying on an outdated and inadequate view of motivation as being made up of merely the component pair of intrinsic and extrinsic aspects. He is correct to be skeptical of assuming that only intrinsic motivation is good and all extrinsic motivation is bad. This oversimplified view of a binary pair was abandoned long ago when SDT psychologists realized that extrinsic motivation has a whole spectrum of manifestations.

The current view is that extrinsic motivation has four components which are each a form of regulation: external, introjected, identified, and integrated. That list takes them in ascending order of effectiveness in terms of the sustainability of the behaviors that follow from each of these forms of extrinsic motivation. External regulations (typically punishments and rewards) are the least sustainable. You may get compliant behavior but the long-term prospects for the behavior continuing are not good. On the other hand, if the motivation is a form of integrated regulation, then that is nearly indistinguishable from intrinsic motivation, so the prospects of sustaining those behaviors are very good. The problem with Mr. Palma's claim is that it is not sufficiently specific to be either true or false. It's just too vague. Students, as with all people, are motivated in a variety of ways. They can have both intrinsic and extrinsic motivations for the same behavior at the same time. While it is sort of true that the practices he listed may “provide” motivation, they are unlikely to be “providing” the kind of motivation that is necessary for the deeper learning that is needed for effectively adapting to our global society today. Mr. Palma's strategies will likely result in short term compliance, but lower the quality of learning in the long term.

I'll have more to say about deeper learning at the end but I'd like to return now to the issue of how helpful deadlines and “reform policies” that Mr. Palma addressed can be. The key motivational variable is not really the flexibility or rigidity of the rules and expectations. The arguments made by Mr. Palma and the author he is criticizing are not relevant unless you are making another assumption about the context in which those policies are playing out. To be generous let's make the best possible assumption: teachers are working in schools in which the primary human needs of both them and their students are supported and those provisions for support actually result in the satisfaction of those needs. If this is true, then what we know from many decades of research into human motivation is that they will all be internalizing the aspects of the situation that are consistent with getting their needs met.

Notice the emphasized word “internalizing.” That is where the rubber meets the road in terms of motivation and engagement. Remember that motivation occurs on a spectrum from external regulation through integrated regulation to intrinsic motivation. Internalizing is the process of moving your motivations from the more external end to the more internal end of the spectrum. Are the people in a given situation internalizing the requirements of that situation? The more they internalize the requirements of the situation, the more they will be motivated to meet those requirements and more likely to engage in sustainable behaviors and learning strategies that help them achieve desirable results. This is what has been revealed by decades of research in SDT. You will not have an effective learning environment unless primary needs are being satisfied; a result of need satisfaction is that the people in that situation are internalizing the requirements of the situation. The fact is that merely getting short-term compliance is likely to produce only shallow learning; full engagement with the available activities is the secret to deeper learning. More internal motivations lead to fuller engagement. The internalizing process occurs in the context of primary need satisfaction. So the challenge is to ensure that the pedagogical choices about the rigidity vs. flexibility of testing procedures and deadlines occur in the context of school and classroom climates that are primarily need satisfying. If the need satisfactions are in place then the students and teachers will internalize those situational requirements.

The reason that the flexibility vs. rigidity arguments are ultimately not the right terms for this debate is that those arguments for and against are only pointing at issues that are tangential to the causal path towards deeper learning. If you want better motivation and engagement then you need to create a situation in which the primary human needs of the people in that situation are externally supported and internally satisfied. Those needs happen to include the psychological needs for relatedness, competence, and autonomy. And the implication of the fact that those are needs on a par with air, water, food, shelter, and sleep means that the challenge is not as simple as getting the right curriculum to be delivered by a skilled instructor. The psychological conditions for the learner, it turns out, are more important for deeper learning than the material and organizational conditions of the classroom and school. Don't get me wrong, there are some material and organizational conditions that can prevent deeper learning, but all material and organizational conditions will be rendered irrelevant if the psychological conditions are such that the causal chain for deeper learning is broken (I'll explain the casual chain in a moment).

The point I'm making here is that the retest policies that sparked this debate are one of those Rube Goldberg mechanisms in school; those policies are not direct leverage points for changing motivation and behavior. Let's take a moment to imagine the difference between a Rube Goldberg machine and using physical leverage directly to accomplish a real world task before we return to using it as a metaphor to understand how schools currently work. Moving forward, this distinction between Rube Goldberg machines and leverage can help us understand how to discern a more useful debate from a less useful debate in education.

First, the literal idea of leverage is that when you want to lift a heavy object, say a car or truck, the best way to do that is to use leverage so that you don't hurt yourself when you do the lifting. The more general term for this is mechanical advantage. Levers, pulleys, wheels, and other simple machines are said to provide mechanical advantage when they enable someone to accomplish a task more easily and effectively than without them. The key elements of leverage are 1) the heavy object to be lifted, 2) a lever that can be put under the heavy object, and 3) a fulcrum that is under the lever placed at an optimal distance away from the heavy object. The most efficient way to use this set-up is to apply your effort directly to the end of the lever opposite from the car while the fulcrum is placed much closer to the car end than to your end. Here are two videos that illustrate the principle. One uses the example of lifting the Empire State Building with a notebook. The other is just a high physics class using 2x4's and deck screws to build a lever.

Compare that example of mechanical advantage with the Mousetrap Game. The fact is that the game can become a real mousetrap with a little augmentation: here's video proof (5 min). Shawn Woods, the guy who made the video, points out at the end that it is totally impractical as a real mousetrap. My point here is that you CAN accomplish real world outcomes using Rube Goldberg machines, but it is NOT a desirable method, except for people with time and resources to burn. Using the Mousetrap Game as the basis for a real mousetrap is entertaining, but it would be ultimately wasteful to mass produce them as a solution to an infestation of real-life vermin.

How can the idea of mechanical advantage be applied to the situation of school policies? First, how will we know which policies are Rube Goldberg contraptions? Do we really know effective ways to manipulate the motivation of students, or teachers, or any other humans?

The causal chain for deeper learning that is supported by the SDT research tradition is as follows: a need supportive environment leads to needs being satisfied which leads to more internal motivations which leads to being more fully engaged which leads to deeper learning which finally leads to better outcomes. You will notice that rigidity and flexibility are not involved. They can have an impact on this process when the student's primary psychological needs are affected by the policy and how it is implemented in each classroom. Some teachers may be able to build strong relationships and provide enough supports in other ways to make whichever policy they encounter work for their students. Other teachers may not have that level of skill and may require other forms of support themselves before they will get good results.

The leverage for human motivation is providing primary need support. If you have a motivation problem, primary need support is the most direct causal pathway to affecting change. The reason that flexible retest policies can sometimes be effective is because they can accidentally result in teachers providing support for their students' primary psychological needs for autonomy, competence, and/or relatedness which enables those students to internalize the requirements of the situation. The reason that some strict policies (the opposite) can also sometimes be effective is because they too can be internalized by teachers and students who would have already internalized other means to getting their needs met in their classroom or school.

The problem arises when we assume that some policies could have causal effects on the motivation of students and teachers without having any idea what the psychological conditions in the schools are. We do not currently have any gauges of those conditions in regular use in most schools. The closest we have are measures of school or classroom climate, but most of those do not include measures of primary psychological need supports, patterns of motivation, nor depth of engagement. (The Hope Survey is one climate measure that does include items about need supports and engagement.)

In conclusion, Mr. Palma was right that misunderstandings about motivation are causing problems in schools. Unfortunately, he misunderstands motivation, too, and is perpetuating the use of Rube Goldberg Motivation Machines. The truth is that the strictness or flexibility of the policies are not going to have direct effects on motivation, they are only going to have indirect effects that are dependent on the psychological conditions in which those policies are used. In order to remedy the situation efficiently and effectively schools need to assess their climate using measures that indicate how well primary psychological needs are being supported, what the patterns of motivation are, and the depth of engagement. If these factors of the psychological climate are lacking then they need to take action to better support the psychological needs of the students and teachers before making adjustments to their retest policies.

Relevant Links:
Mr. Palma's Article:
The Article Mr. Palma was responding to:
Lifting the Empire State Building with a Notebook:
https://www.youtube.com/watch?v=IE6hUjjQVSc (2 min)
high physics class using 2x4's and deck screws to build a lever:
https://www.youtube.com/watch?v=fX3doG_z9l4 (5 min)
Mousetrap Game Trap: https://www.youtube.com/watch?v=_ogOkocvldw (5 min)

No comments: