TOKYO —

What we experience when making a decision may be very different from what actually happens inside the brain, according to Indiana University Professor Tom James. For decades, both scientific theories and everyday intuition have tended to describe decision-making as a sequential process: perception, thought, choice, and action. This "sandwich model," as Professor James terms it, posits distinct brain functions for each stage, moving from sensory processing to cognition and then to motor activity. Many influential approaches in cognitive neuroscience are built around this linear framework, which also aligns with our subjective feeling of decision-making, where actions appear to be directly caused by desires, beliefs, and intentions. However, Professor James argues that this familiar explanation, while intuitively appealing and useful for describing behavior, does not accurately reflect the intricate neural processes at play.

The Limitations of the "Sandwich Model"

Professor James, a distinguished figure in the Department of Psychological and Brain Sciences at Indiana University, presents a compelling challenge to the conventional understanding of decision-making in his recent article, "Sensorimotor Mechanisms of Decisions and Actions," published in the prestigious Journal of Cognitive Neuroscience. He contends that while sensory and motor mechanisms are well-defined, the supposed distinct cognitive stage responsible for decision-making lacks a corresponding, identifiable neural process that functions as an independent "decision maker."

Instead of a dedicated system that directs behavior from a central command post, Professor James proposes that what we perceive as decision-making is, in reality, a complex interplay of sensory, sensorimotor, and motor processes that collectively give rise to what he terms "action selection." In this revised paradigm, behavior emerges not from a linear sequence of events but from continuous, dynamic interactions among the brain, the body, and the surrounding environment. These processes, he emphasizes, can occur simultaneously and feed back into one another, creating a fluid and integrated system rather than a step-by-step computation.

A Physicalist Framework and the Nature of Decisions

Professor James’s argument is firmly rooted in a "physicalist" framework, a philosophical approach associated with thinkers like Daniel Dennett. This perspective posits that physical phenomena can cause both physical and nonphysical phenomena, but critically, nonphysical phenomena cannot, in themselves, cause physical events. Applying this to decision-making, Professor James highlights a key implication: if decisions are considered nonphysical mental phenomena, they cannot, in a literal sense, directly cause physical actions.

To elucidate this complex idea, Professor James employs a series of insightful analogies that draw upon established philosophical and scientific concepts. These analogies serve to demystify the notion that what we subjectively experience as a "decision" might be more accurately understood as an abstract description of a physical process rather than an active causal agent.

Decisions as Abstract Descriptions: Analogies for Understanding

One of the most potent analogies Professor James draws is from Daniel Dennett’s comparison between the concept of a "self" and a "center of mass" (CoM) or "center of gravity." A center of mass is an indispensable mathematical tool for understanding and predicting the motion of physical objects. However, it is crucial to recognize that a center of mass is not a physical entity that can independently exert force. One cannot move an object’s center of mass without simultaneously moving the object itself.

Similarly, Professor James suggests that a "decision" may function as an abstract description, a conceptual shorthand, rather than a distinct physical entity that initiates action. The term "decision" effectively summarizes a complex set of underlying physical processes that lead to a particular behavior, but it does not possess independent causal power in the physical world.

Another analogy employed by Professor James highlights how useful high-level concepts can become less informative when attempting to understand events at a more granular, mechanistic level. Consider the term "the university." This term is immensely useful in everyday discourse, conveniently encapsulating a vast collection of people, buildings, departments, and complex organizational processes. When we say, "the university took certain actions during a campus protest," we are using a high-level abstraction. However, for a scientist seeking to understand the precise events that unfolded, this statement offers little explanatory power. A detailed analysis would require examining specific interactions: meetings between administrators, phone calls to law enforcement, the actions of individual security personnel, and so forth.

Professor James argues that "decisions" present a comparable challenge for cognitive neuroscience. While they provide a valuable and intuitive way to describe and categorize behavior from an external perspective, they may obscure the actual physical mechanisms responsible for generating that behavior. As he articulates, "As mental phenomena, they are defined on too abstract a level for the goals of cognitive neuroscience." In essence, simply stating that someone "made a decision" does not explain the underlying neural computations and physical interactions that culminated in the observable action.

The Robot Analogy: Simulating Intent Without a Decision-Maker

Pushing his argument further, Professor James introduces a third illustrative example: a relatively simple robot designed with a limited number of sensory, motor, and sensorimotor modules. This machine can exhibit sophisticated behaviors, such as "wall-following," which, from an external vantage point, can appear remarkably purposeful. The robot’s actions might seem to indicate goals, strategies, and even something akin to intentions.

However, Professor James points out a critical detail: this robot contains no dedicated system designed for decision-making. "The robot does not have decisions built into it," he 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 observation leads to a provocative and far-reaching question: If a comparatively simple machine can generate behavior that convincingly mimics intentionality and strategic choice without an internal decision-making system, could human behavior also appear to result from centralized decisions, even if no such distinct central process exists in the brain? Professor James suggests that this explanation is not only plausible but also more parsimonious than positing the existence of a hypothetical "higher-level, central controller that monitors and regulates sensory and motor processes."

The Philosophical Quagmire of the "Central Controller"

The concept of a central controller, an entity within the brain that orchestrates decisions and actions, faces significant philosophical hurdles, ones that have been debated since the era of René Descartes. If such a higher-level entity is responsible for processing information and making choices, then the question inevitably arises: how does this controller itself operate?

Professor James echoes the concerns raised by philosophers like Dennett, who have termed this the "Cartesian Theater." The problem, he explains, is that positing a central controller merely shifts the explanatory burden. If this controller makes decisions, then it, too, must have a mechanism for making decisions. This leads to an infinite regress: the controller would require another controller, which would require another, and so on, ad infinitum. This "problem of the homunculus," as it is sometimes called, fails to provide a genuine explanation for how decision-making functions. It is akin to explaining how the brain works by proposing that there is a "little person" inside the brain making the decisions, who in turn would need another little person inside their head, and so forth. The problem is never solved; it is simply deferred endlessly.

Instead of invoking such an internal, often metaphorical, decision-maker, Professor James advocates for a more direct scientific approach: focusing on the observable, interacting sensory and motor systems that demonstrably generate behavior. This perspective shifts the focus from abstract mental states to the concrete neural and bodily mechanisms that underpin our actions.

An Experimental Path Forward: Embracing Complexity

If decision-making indeed emerges from the continuous, dynamic interactions among the brain, body, and environment, then the scientific investigation of this phenomenon must adapt to capture this inherent complexity. Professor James acknowledges that this presents both exciting new avenues for research and significant methodological challenges. Researchers will need to move beyond strictly linear models of cognition and explore processes that occur simultaneously, exert mutual influence, and evolve dynamically as an individual interacts with their surroundings.

Professor James’s own laboratory is actively pursuing this research agenda, drawing inspiration from fields such as embodied cognition and ecological psychology. These approaches emphasize the inseparable link between an organism and its environment and the role of bodily interactions in shaping cognitive processes.

This shift in perspective holds the potential to unlock a deeper understanding of the mechanisms that give rise to what we commonly describe as decision-making. Furthermore, it may offer novel frameworks for investigating a wide array of other cognitive and mental phenomena that have traditionally been compartmentalized as distinct, internally processed functions within the brain. By embracing a more dynamic and integrated view, cognitive neuroscience can move closer to unraveling the intricate symphony of biological processes that guide our actions and shape our experience of the world. The implications of this research extend beyond understanding human behavior, potentially informing the design of more sophisticated artificial intelligence and robotics that can interact with their environments in more nuanced and adaptive ways. The journey from subjective experience to objective neural mechanisms is complex, but Professor James’s work offers a promising new map for navigating this challenging but vital scientific frontier.