Agency is a Trap

Defining agency as something to possess, display, or improve makes it a status symbol. Predictably, this shifts attention from the task itself to how your actions reflect on you. As a result, performance declines, learning slows, and external forces like incentives, shame, or low status become more influential.

Psychology has had a name for this for decades: choking under pressure. Pressure pushes attention on itself, and skilled action very often falls apart when the mind starts supervising what used to run smoothly. Baumeister's model puts it bluntly: pressure "increases the conscious attention to the performer's own process of performance," and attention "disrupts the automatic or overlearned nature of the execution." [1]

The agency definition that has taken over founder culture and self-improvement circles recreates that pressure on purpose. It tells people to keep two scoreboards running in their head: not only whether the action worked, but whether the action proves something about the actor. This secondary layer is the killer behind what, otherwise, is a good thing.

The true problem is self-monitoring

Having skills, whether in sports, writing, negotiating, coding, or leading a meeting, relies heavily on routines that run without proper narration. Beilock and Carr describe expert performance as "encoded in a procedural form that supports performance without the need for step-by-step attentional control." [2]

When you start checking yourself mid-action, you do two damaging things at once. First, you take away attention from the task. Second, you change a live situation into a test of identity. That is why "high agency" talk so often produces fragile people who are loud about intent and inconsistent in results. They are busy proving they are the kind of person who does things, which is very different from doing the actual thing well.

DeCaro and colleagues summarize the same split: "Explicit monitoring theories claim essentially the opposite: that pressure prompts individuals to attend closely to skill processes in a manner that disrupts execution." The point here is that the failure mode is well-mapped. Attention turns inward, execution falls apart. [3]

Why "trying to have agency" doesn't succeed

Wanting to be effective is fine. The backfire comes from treating effectiveness like a trait you must hold onto in real time.

If the mind is asking, even quietly, "Am I being agentic right now?", it has already introduced a second job. That second job is image management. Image management competes with perception. You start seeing the world through the lens of what holds the story instead of what updates the loop.

Measuring agency often reduces it. Metrics create hierarchy, hierarchy creates self-consciousness, and self-consciousness stalls real progress.

What actually predicts improvement

The clean loop is simple and unglamorous: do your best, notice what happens, learn, adjust. No identity ceremony required.

Research on attention in motor learning makes the same point from another angle. When people focus on the effects of their actions in the world instead of their body mechanics, they often learn faster and perform better. In one paper, the authors put it directly: focusing on distant effects enhances learning by "promoting the utilization of more natural control mechanisms."

A review by Wulf and Prinz reports a consistent pattern across studies: instructions that direct attention to movement effects "are more effective than instructions directing their attention to the movements themselves." [4]

Translate that out of sports science and into normal life: the more attention is spent on the work and its feedback, the less attention is spent on self-narration, and the better the system updates.

A practical way to use agency without turning it into a new identity game

The trick is to aim at conditions, not labels. Agency talk fails because it makes "being a certain kind of person" the goal. A better goal is reducing interference during your task, and afterwards, improving the feedback loop.

Here are changes you can make, which actually work well:

Stop doing this:

• Treating decisiveness as a personality trait instead of a skill.
• Turning every choice into a showcase of your worth.
• Protecting a self-made story when the output of your actions disagree with it.

Do more of this:

• Make smaller bets that actually return feedback, then iterate on it.
• After a task, ask what changed in the world, not what the action implies about you or your image.
• Keep your promises to yourself, and keep them small and frequent, so your feedback stays honest.
• Let the results teach you, even when they're embarrassing.

Notice what is missing: there is no requirement to "feel agentic", identify as agentic, or talk about agency at all.

The uncomfortable part

My view insults a lot of popular motivational culture. It also removes a lot of the excuses. There is no caste system of "agentic" people versus everyone else. There is only a person in a situation, with some mix of clarity, skill, fear, ego, and incentives, producing a result, then self-improving or not.

People who look "high agency" from the outside usually have one boring advantage: they spend less time negotiating with themselves. They do something, they see what happened, and they change course.