maxjmartin.com

Staying motivated

Or, Why your productivity system might be counter-productive

Maybe you use many tools and techniques to overcome akrasia and achieve you goals. If you are like me though, you find yourself at your most productive when you simply really want to do something.

Why do we use productivity systems in the first place

It is hard to make someone want to do something 1.

Making yourself excited about doing something is even more difficult. How might you trick yourself into liking something?

This type of intrinsic motivation is also liable to weaken, or fluctuate.

For a certain type of person, there is a temptation to try and shore up our intrinsic motivation by adding some form of extrinsic motivation to everything we do. Relying only on our changeable moods and desires to get ourselves to do things feels fragile. This was definitely my approach for a long time.

If only I could build the perfect system to push myself to stick to all my goals and habits and hobbies and projects every day, I felt I would be unstoppable. It seemed like a perfect solution.

We turn to the host of available tools and techniques. We set SMART goals, publicly commit to them, create a calendar, create check-lists of daily habits, put ourselves on the line with beeminder, make bets with friends, create point systems for ourselves, etc. ad nauseum.

Of course, these methods can be effective. But I now think we should use them sparingly.

Why extrinsic motivation is sometimes bad

Using rewards or punishments on yourself often produces compliance for only as long as they continue. Once the extrinsic motivation is removed, you tend to lose interest again.

In Influence, Cialdini writes:

Social scientists have determined that we accept inner responsibility for a behavior when we think we have chosen to perform it in the absence of strong outside pressures. A large reward is one such external pressure. It may get us to perform a certain action, but it won’t get us to accept inner responsibility for the act. Consequently, we won’t feel committed to it. The same is true of a strong threat; it may motivate immediate compliance, but it is unlikely to produce long-term commitment. All this has important implications for rearing children. It suggests that we should never heavily bribe or threaten our children to do the things we want them truly to believe in. Such pressures will probably produce temporary compliance with our wishes. However, if we want more than just that, if we want the children to believe in the correctness of what they have done, if we want them to continue to perform the desired behavior when we are not present to apply those outside pressures, then we must somehow arrange for them to accept inner responsibility for the actions we want them to take. […] The important thing is to use a reason that will initially produce the desired behavior and will, at the same time, allow a child to take personal responsibility for that behavior. 2 3

To cause intrinsic motivation in someone you need to make them think they are doing something for their own reasons, not because of any outside influence.

You need to perform inception.

Extrinsic motivation generally does not cause intrinsic motivation, in fact, studies 4 5 have shown that extrinsic motivation can often decrease intrinsic motivation. The results have been replicated many times both within the lab and without:

These results were replicated in a controlled field experiment (Deci, 1971) which took place over a 16-week period in a college newspaper. Subjects were staff members who wrote headlines for the college newspaper and who were unaware that an experiment was being performed. As in the laboratory study, subjects who were paid for their performance ($.50 per headline written) showed a decrease in intrinsic motivation which was evident as much as eight weeks after the payments had stopped. 4

One theory explaining this finding is the overjustification effect (part of self-perception theory), whereby a person will decide how they feel about a task by observing their own behaviour, much like they would evaluate someone else:

When an individual observes another person engaging in some activity, he infers that the other is intrinsically motivated to engage in that activity to the extent that he does not perceive salient, unambiguous, and sufficient extrinsic contigencies to which to attribute the other’s behaviour. Self-perception theory proposes that a person engages in similar processes of inferance about his own behaviour and its meaning. To the extent that the external reinforcment contigencies controlling his behaviour are salient, unambiguous, and sufficient to explain it, the person attributes his behaviour to these controlling circumstances. But if external contigencies are not percieved, or if they are unclear, invisible and psycologically insufficient to account for his actions, the person attributes his behaviour to his own dispositions, interests and desires. 5

The promise of punishment or reward upon completion of a task will change how you view that task. It will come to feel as though you are performing the task because of the stimuli.

(Conversely, subtly nudging oneself to perform a task frequently without any clear external stimuli might result in you believing that you are performing the task for its own sake — although this is clearly a difficult trick to pull on yourself consciously.)

it seems clear that the effects of intrinsic motivation and extrinsic motivation are not additive. While extrinsic rewards such as money can certainly motivate behavior, they appear to be doing so at the expense of intrinsic motivation 4

Again, this is not to say that rewards are not effective, they definitely can be. But they risk eroding any wish you had to do the task for its own sake. In other words, if we simply build an external motivation system and apply it everywhere, we cannot assume that it will increase our motivation across the board, it might even do the opposite.

Why does this matter? Surely n units of extrinsic motivation equal n units of intrinsic motivation?

If you are working on a truly important project, and your only wish is to complete it as quickly as possible, then yes. I would take 2 extra units of motivation from external sources at the cost of 1 unit of actually-liking-the-task.

Otherwise, I always prefer the actually-liking-the-task units, simply because it is much more enjoyable to work on something you enjoy, and I value that increase in enjoyment more than the potential increase in productivity.

When to avoid the carrot and the stick

In my own experience, using reward (habitica, custom point systems) and punishment (beeminder, self-enforced social pressure) to encourage me to stick to my goals have always increased productivity strongly in the short term. This might well have been due to novelty more than anything. In this long term however they have caused my enjoyment of the task to slowly drop off. Often ending with me giving up entirely.

I can’t know for sure that it was the extrinsic motivation that caused my intrinsic motivation to drop. It might also have dropped had I done nothing at all. Also, had I not put in place any of the above systems, I may have been less productive over all: after all, even if I did fail to continue at a task long-term, I was at least achieving something up until I failed.

I found that the few activities that I did not bring inside the productivity system did not seem to suffer in the same way. Often I would avoid the things I was meant to be doing (things I used to want to do) and procrastinate by working on something else. A guilty pleasure created simply by applying pressure to one task and not to another.

My experience seemed to match the predictions of self-perception theory. At the beginning I was excited about the goal, and looking to find more ways to free up time to work on it. As time went on however, and I was forced to spend time on the goal even if it was 1am and I wanted nothing else but to sleep, I started to feel as though I was being forced to work on the task. This resulted in my resenting the task itself, not just the commitment system.

It is generally a safe bet to be wary of psychology studies 6, and it seems to be the case that people react differently to different types of reinforcement, but I have (anecdotally) been having much more success since reducing the amount of extrinsic motivation I use.

I now split my productivity system between hard and soft motivation.

Hard motivation is any system of reward or punishment that commits me to completing certain tasks by certain times. I use both a point system and beeminder for this purpose.

Soft motivation is a more subtle steering of myself towards doing things that I am already intrinsicly motivated to do, without directly resorting to hard motivation. The idea is to avoid at all costs doing anything which might reduce my productivity long-term by eroding my enjoyment of the task for the reasons discussed above.

When trying to encourage someone else to become interested in a task, it is worth noting that reward systems have been shown to work in many cases: 7 Especially if there is no intrinsic motivation to begin with, and the task is one where the more you perform it, the more rewarding it becomes (learning to ride a bike for example), then using rewards will likely be effective in building some initial interest.

I therefore use hard motivation on tasks when:

  • I don’t enjoy it anyway (anki reps)
  • It is very short and easy to complete (flossing, taking supplements)
  • It is a one-off or short term task, and absolutely must get done (a project with a short deadline, errands, etc.)
  • I want to reduce the amount I do (drinking, snoozing my alarm, etc.)

Whereas I choose soft motivation for:

  • Hobbies that I enjoy (reading, running)
  • Activities I would like to enjoy, since they are good for me (meditation, mobility work)

Soft motivation tactics I use

Removing bad alternatives

For example, one goal that I have been working on recently is to read more books. I was wary of creating a beeminder for pages read or time spent reading, since I could see it turning reading into a chore, which I want to avoid at all costs.

My solution was to look for ways to indirectly support my pre-existing intrinsic motivation. To do this I used beeminder to reduce:

  • use of reddit anywhere
  • news reading on my phone

By getting rid of these two time sinks, the time they previously occupied was freed-up, and my reading time naturally expanded to fill the gap. I am now reading at least double the amount of books I was previously.

As a very similar example, I have created a limit on the amount of “junk food” I allow myself to eat each week. This was to meant to encourage me to cook my own lunches more frequently, and avoid buying tasteless and unhealthy supermarket “lunch deals” multiple times a week.

As with the book reading goal, reducing the amount of junk going in results in something more nutritious filling the gap.

Giving yourself a choice

My other concern was how to encourage myself to run, meditate, stretch and write. These are all things that I enjoy, but I often fail to do. My intrinsic motivation needs a little extra support to get me started with these tasks. I therefore use hard motivation to ensure that I spend an hour every afternoon on whichever of these tasks I feel like.

By not requiring that I work on any particular task, instead giving myself a choice of what to do, I hope to nudge myself to work on those goals while feeling like I am doing so with no external cause.

The idea of giving yourself choice, thereby increasing your feeling of control, is shown to increase intrinsic motivation. This is widely known in management, where giving an employee more autonomy in their work is thought to be motivating 8. It also appears in this advice from Alan Kazdin:

If a child feels that he has a choice about whether to do what’s asked of him, he’s more likely to do it. The choice may be empty or trivial to you, but not to him. “Put on a jacket, we’re going outside” might be sufficient to get the behaviour you want, but “Please put on your jacket or a sweater — it’s you choice — and we’ll go outside” is much more likely to work. […] You want to look for opportunities to give real choices between alternatives that are both acceptable to you […] As a general rule, lowering the amount of coercion in your lives (The amount of “Because I said so”) will increase overall compliance 9

(There are lot of excellent insights to be found in parenting books. I find it useful to see the part of myself that wants to procrastinate as a child.)

Reminding yourself of your goals

Finally, I have a morning notification to myself on my phone that simply reminds myself of my current goals and priorities.

The smallest nudges can be enough if you enjoy the task already.

  • Leave your running shoes somewhere you can see them
  • Leave your paintbrushes always ready-to-go on a table somewhere
  • Keep a kindle in your bag at all times.

Sometimes this is all you need.

In short

Push yourself to do the things you dislike. Nudge yourself to do the things you enjoy.

  1. Feel free to ask your parents if you want a citation

  2. Influence (Robert Cialdini)

  3. This passage also refers to this interesting study: Compliance Without Pressure: The foot-in-the-door technique (Freedman, J. L & Fraser, S. C.) the effect of which could also be explained by self-perception theory.

  4. The Effects of Contingent and Non-contingent Rewards and Controls on Intrinsic Motivation (Edward L. Deci) 2 3

  5. Undermining Children’s Intrinsic Interest with Extrinsic Reward: A test of the “Overjustification” Hypothesis (Mark R. Lepper and David Greene) 2

  6. Estimating the reproducibility of psychological science (Open Science Collaboration), http://andrewgelman.com/ and many others

  7. For example: Token reinforcement programs in the classroom: A review. (O’Leary, K. Daniel and Drabman, Ronald)

  8. For example: Perceived Control by Employees: A Meta-Analysis of Studies Concerning Autonomy and Participation at Work (Paul E. Spector)

    For all studies combined, it was found that high levels of perceived control was associated with high levels of job satisfaction (overall and individual facets), commitment, involvement, performance and motivation, and low levels of physical symptoms, emotional distress, role stress, absenteeism, intent to turnover, and turnover.

  9. The Everyday Parenting Toolkit: The Kazdin Method for Easy, Step-by-Step, Lasting Change for You and Your Child (Alan E. Kazdin)