Ask most people how the brain works and you get the same three-step story. You take something in through your senses. You weigh it up and decide. Then you act. Perception, then decision, then action, each step handing off to the next like a relay. It feels obviously true because it matches how choosing feels from the inside.

One neuroscientist thinks that middle step, the deciding, is where the story goes wrong. His argument: your brain has sensory systems and movement systems, but no separate part whose job is to make decisions. What feels like a chooser in your head may be nothing of the kind.

We are not neuroscientists or psychologists, and nothing here is advice about your own mind or behavior. This is us reading one researcher’s argument and thinking it through. The paper it rests on is a single perspective piece, not settled science, and it makes a claim about theory rather than a finding about you.

That tidy three-step picture has a nickname among cognitive scientists: the “sandwich model,” with the decision as the filling between perception and action. The term is usually credited to philosopher Susan Hurley, who used it to describe the standard view in which thinking sits neatly between what comes in and what goes out. It is the mental model behind a lot of research, and behind the everyday sense that somewhere in your head, a chooser is doing the choosing.

The sandwich model has a lot going for it. Sensing maps onto real machinery in the brain. So does movement. There are systems that handle input and systems that produce output, and you can point to where they live. So it seems natural that the middle step, the deciding, must work the same way, with its own dedicated region running the show.

Tom James, a cognitive neuroscientist at Indiana University Bloomington, thinks that is where the story quietly goes wrong. He does not deny how it feels. “Our actions feel like they are caused by decisions based on desires, beliefs, and intentions,” he says, describing the intuition most of us share. His disagreement is not about the feeling. It is about whether the feeling points to a real mechanism.

In a paper called “Sensorimotor Mechanisms of Decisions and Actions,” published in June 2026 in the Journal of Cognitive Neuroscience, James argues that the brain has sensory systems and movement systems, but no matching, dedicated decision system between them.

In his account, behavior comes out of sensory and movement processes running at the same time and looping back on each other, spread across brain, body, and environment. He prefers to call the whole thing “action selection” rather than decision-making.

He is not throwing out the word “decision.” Asked whether decisions exist, James answers, “Of course they do. We use this language all the time and it’s very helpful in terms of describing behavior.” What he objects to comes next. “The leap, I think, is to say that the brain works by having decision-making or control processes,” he continues. Describing behavior as a series of decisions is fine. Assuming the brain contains a part that makes those decisions is the step he wants to question.

The distinction is subtle and worth slowing down on. James’s view is that behavior can look decision-shaped without any component inside the head whose job is to decide. As he puts it, the brain “produces behavior that is well described in that way. But it doesn’t need a process that does that to make it look that way.”

The two pictures predict different things about what you would find inside a brain. The sandwich model is staged: input arrives, a central stage weighs it, output follows, and each stage should match up with something distinct in the brain tissue. Find the deciding region and you have found the decider.

James’s picture is circular and simultaneous. There is no handoff from a sensing stage to a deciding stage to a moving stage, because sensing and moving happen together and feed each other continuously. To make the point concrete, he offers the analogy of a robot: “The robot does not have decisions built into it. It just senses its environment and moves around accordingly.” You can describe the robot’s behavior in the language of choices even though nothing in it is choosing. That shows what is possible, not that human brains work the same way, and James frames it only as the simpler explanation to consider.

There is an older objection underneath all this, and James reaches for it directly. If you explain behavior by putting a controller in the brain that makes the decisions, you have not really explained anything. You have just moved the problem: “Explaining that the brain works by way of a central controller suggests that you haven’t figured out how the brain works, because you’ve just put a person inside your brain.” This is the old worry about a little person watching a screen and pulling levers inside your head, which the philosopher Daniel Dennett named the Cartesian Theatre. It is a conceptual objection with a long history, not a new experiment, and James uses it as a reason to be suspicious of the central-controller idea rather than as a knockout blow.

If the difference were only philosophical, it would matter less. But James argues that the staged view is built into how a lot of decision research actually gets done. When the model assumes a middle stage that piles up evidence until it crosses a threshold and triggers a choice, the experiments get built to isolate exactly that stage: a subject sits still, watches a screen, and presses a button, with sensing and acting deliberately pulled apart so the “decision” can be studied on its own.

The worry is that this setup can bake in its own answer. Strip out the constant loop between perceiving and moving, and you have removed the very thing James thinks does the work, then measured what is left and called it decision-making.

He argues for more lifelike methods: tests that give people real control over their actions and let them learn by actively exploring the world, rather than reacting to isolated cues on a screen. The quarrel is partly about theory and partly about the shape of the experiment, and the two are harder to separate than they look.

It would be easy to hear all this as a claim that choice is an illusion, or that no one is really in charge of anything. That is not what James is saying, and the difference matters. He agrees that decision language is accurate and useful for describing what people do. What he questions is the next leap: that this useful description must match up with a dedicated piece of brain machinery.

None of this is settled. This is one researcher’s argument in one paper, a perspective that challenges the standard model rather than overturning it, and plenty of scientists will read the same evidence and keep the sandwich. But the reframe is an interesting one. “Action selection” describes behavior without hunting for a control room, and if James is right, the missing control room was never the person doing the choosing. It was only ever a picture we drew of ourselves.