The human experience of making a choice—weighing options, deliberating, and finally executing an action—feels deeply linear, sequential, and deliberate. For generations, both everyday intuition and mainstream cognitive science have treated decision-making as a cognitive sandwich: sensory perception forms the bottom slice, abstract cognition or decision-making sits in the middle, and motor activity forms the top slice. However, a sweeping new theoretical framework published in the Journal of Cognitive Neuroscience posits that this traditional model is fundamentally flawed.
Dr. Tom James, a professor in the Department of Psychological and Brain Sciences within the College of Arts and Sciences at Indiana University, argues that the brain does not house a centralized decision-making system. Instead, what humans perceive as a deliberate, calculated choice is actually an emergent property resulting from simultaneous, highly complex interactions among the brain, the body, and the external environment. Published under the title "Sensorimotor Mechanisms of Decisions and Actions," James’s paper challenges core assumptions that have guided model-based cognitive neuroscience for decades, opening a fresh debate over how neuroscientists conceptualize human agency, consciousness, and volition.
Main Facts and the Theoretical Shift
At the heart of James’s proposal is a rejection of the "central controller" hypothesis—the long-standing assumption that a distinct neural subsystem evaluates sensory inputs, formulates intentions, and dispatches commands to the motor system. While sensory systems possess clear, identifiable neural mechanisms, and motor systems have corresponding execution pathways, James points out that the purported intermediate cognitive stage lacks a discrete neural counterpart functioning solely as a decision-maker.
Instead of a hierarchical command chain, James introduces the concept of "action selection." In this view, behavior is generated through ongoing, reciprocal loops where sensory, sensorimotor, and motor processes occur concurrently. These systems feed back into one another in real time, making behavior look structured and purposeful from the outside without requiring a dedicated executive control module sitting inside the brain’s neural architecture.
James clarifies that this perspective does not render decisions "unreal" or eliminate the usefulness of decision-related vocabulary. Human beings rely constantly on mentalistic language—desires, beliefs, and intentions—to navigate social and personal landscapes. These terms remain immensely helpful for describing behavior at a macro level. The error, according to James, lies in leaping from the usefulness of this descriptive language to the assumption that physical brain mechanisms must literally mirror it.
Background Context and Philosophical Framework
The historical roots of the central controller model trace back centuries, most famously articulated in the dualistic philosophy of René Descartes, who grappled with how a nonphysical mind could interact with a physical body. Over time, modern neuroscience adopted physicalist frameworks to bypass Cartesian dualism, yet many scientific models covertly retained a homunculus—a little person inside the head—monitoring information and pulling the levers of behavior.
In his recent publication, James utilizes a rigorous physicalist framework heavily informed by the philosophical work of Daniel Dennett. Under this physicalist paradigm, physical phenomena can cause both physical and nonphysical outcomes, but nonphysical phenomena—such as abstract mental decisions—cannot independently exert causal physical force on matter. Because sensory and motor operations are strictly physical, an abstract decision cannot literally cause a physical muscle contraction on its own.
To illustrate this counterintuitive concept, James draws on several powerful analogies. The first is Dennett’s famous comparison between the self and a physical center of mass (CoM), or center of gravity. A center of mass is a remarkably useful mathematical concept for physics and engineering, but it cannot independently exert a physical force; one cannot move an object’s center of mass without physically moving the object itself. Similarly, James suggests that a "decision" is an abstract, high-level description of an event rather than an independent physical entity that triggers the action.
A second analogy examines institutional behavior. When observers state that "the university took certain actions during a campus protest," the phrase provides a convenient, high-level summary representing thousands of individuals, buildings, departments, and bureaucratic procedures. Yet, that broad institutional statement reveals virtually nothing about the concrete physical mechanisms involved—such as specific administrative meetings, electronic communications, or security deployments. A rigorous, granular explanation requires looking past the abstract label ("the university") to examine the constituent physical actions. James argues that neuroscience faces the exact same trap: treating "decisions" as fundamental neurological entities rather than high-level summaries of complex, distributed neural dynamics.
Chronology and Development of the Research
The publication of "Sensorimotor Mechanisms of Decisions and Actions" represents the culmination of years of theoretical refinement and empirical observation within cognitive neuroscience regarding how motor and sensory systems overlap.
For much of the late 20th century, cognitive neuroscience developed in tandem with computational models of the mind, which treated the brain as an organic computer executing serial processing tasks: input, compute, output. This serial framework dominated textbooks, experimental design, and artificial intelligence research.
Over the past decade, however, growing empirical data from neuroimaging, electrophysiology, and embodied cognition began to expose cracks in the serial processing paradigm. Researchers frequently observed motor preparation occurring simultaneously with sensory categorization, suggesting that perception and action are deeply intertwined rather than strictly sequential. James’s work formalizes these observations into a coherent theoretical critique, moving away from purely computational metaphors toward a dynamic, biologically grounded framework rooted in embodied cognition and ecological psychology.
The Robot Experiment: Challenging Intuition
To demonstrate how easily observers can misattribute centralized decision-making to decentralized systems, James employs a thought experiment involving a simple, autonomous robot.
Consider a robot engineered with only a minimal collection of sensory, motor, and sensorimotor modules, programmed to execute a simple "wall-following" behavior. As the machine navigates a room, its trajectory adapts seamlessly to obstacles, corners, and open spaces. To an outside human observer, the robot’s movement appears strikingly purposeful, strategic, and intentional; it looks precisely as though the machine is constantly making deliberate decisions about where to go and how to avoid crashes.
Yet, as James points out, the robot contains no central decision-making algorithm or executive control unit.
"The robot does not have decisions built into it," James explains. "It just senses its environment and moves around accordingly. And based on the environment, wall-following turns out to be a good thing. It looks intentional. It looks strategic. It looks like the robot is making decisions. And yet, it is not. The reason we know it is not is that there are no systems built into it to do that."
This demonstration raises a profound and provocative question for human behavioral science: If a rudimentary machine can generate behavior that appears fully intentional and goal-directed without possessing any internal decision-making apparatus, could human behavior similarly emerge from distributed sensorimotor loops without requiring a central controller? James contends that parsimony favors the simpler explanation, sparing neuroscience from invoking an unnecessary higher-level regulator.
The Infinite Regress Problem of the Cartesian Theater
Retaining a central controller in neural theory also resurrects a notorious philosophical trap that philosophers and scientists have sought to avoid since the Enlightenment.
If neuroscientists argue that a specialized region or entity inside the brain is responsible for reviewing incoming sensory data and making executive choices, they immediately face a foundational explanatory deficit: how does that controller itself operate?
"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," James notes, echoing Daniel Dennett’s critique of the "Cartesian Theater." If a homunculus inside the brain makes decisions, logic dictates that this internal agent would require its own microscopic controller inside its own neural machinery, leading directly to an infinite regress of observers watching observers. By refusing to solve the original problem, the central controller hypothesis merely shifts the burden of explanation deeper into the psyche.
By cutting out the middleman of the central controller, James’s framework anchors explanation directly in the observable, continuous physical interactions occurring between the organism and its environment.
Broader Impact and Methodological Implications
The publication of James’s paper arrives at a time of significant methodological re-evaluation within psychology and neuroscience. As laboratories increasingly recognize the limitations of static, highly controlled laboratory tasks that isolate perception from movement, the demand for dynamic frameworks has intensified.
Adopting a sensorimotor, action-selection model of decision-making carries deep implications for how future experiments must be designed. Traditional experimental paradigms often isolate perception, delay the response, and measure reaction time as a proxy for cognitive processing speed. Under James’s framework, such tasks artificially chop up continuous, flowing loops of interaction into artificial segments.
To validate and explore these ideas, researchers will need to embrace more complex, naturalistic experimental designs capable of capturing real-time, simultaneous neural and behavioral dynamics. James’s laboratory at Indiana University is already pioneering pathways in this direction, integrating principles from embodied cognition—the idea that mental processes are deeply shaped by the body’s physical interactions with the world—and ecological psychology.
Beyond altering how scientists study choice, this shift in perspective could ripple across related disciplines. Psychiatry, neurology, and artificial intelligence stand to benefit from abandoning rigid computational silos. For AI developers, the insight that complex, adaptive behavior can arise without central deliberative architectures offers clues for building more efficient, adaptive robotic systems. For clinical researchers, understanding psychiatric conditions characterized by impaired decision-making or loss of agency through the lens of broken sensorimotor feedback loops—rather than isolated cognitive deficits—could eventually yield novel therapeutic interventions.
As cognitive neuroscience absorbs the implications of "Sensorimotor Mechanisms of Decisions and Actions," the field faces a compelling challenge: to look past the comforting illusion of the mental sandwich and confront the rich, chaotic, and beautifully integrated reality of the living brain in motion.
