Neuroscience's Habit of Attributing Thought to Individual Neurons

The mereological fallacy — ascribing properties of a whole to one of its parts — is among the most pervasive yet underappreciated errors in contemporary neuroscience. Psychological predicates like intention, recognition, and understanding are properties of whole organisms situated in environments, not of neurons or circuits. Yet the field routinely attributes such capacities to neural elements. The mirror neuron literature is the canonical instance: a neuron that fires during both action execution and action observation is said to "understand" the observed agent's intention. This is a Category error. The neuron is a component; intention, if it is a coherent psychological category at all, belongs to the organism embedded in its social and physical context.

The fallacy is compounded by two reinforcing habits. First, premature psychologizing: assigning rich cognitive roles to neural elements before exhausting simpler mechanistic or statistical explanations of their activity. Second, a systematic equivocation in key theoretical terms — "representation" being the most damaging example. The same word is deployed at both the neural and psychological levels of description, creating the appearance of a bridge between them. In reality, no explanatory work has been done; the psychological meaning has simply been imported to inflate the significance of the neural finding.

This produces what might be called pseudo-explanation: accounts that satisfy intuitive demands for understanding while leaving the actual explanatory gap — how neural processes give rise to psychological phenomena — entirely untouched. The interchangeable use of psychological and neuroscientific vocabulary obscures the fact that the hard problem of connecting levels of description has been sidestepped rather than solved.