During the last decades, confusions were made in the domain of cognitive neuroscience, between the concepts, the measurements (and their numerical transformations) and the physiological processes that generate the measured phenomenon. The limitations of measurements and their associated theoretical notions led some authors to propose nativist and evolutionist options, such as a coding of space, time and number in the brain. For example, in the neurosciences of movement, the activity of neurons would encode kinematic parameters: when they emit action potentials, neurons would "speak" a language imbued with notions of classical mechanics. Yet, the movement of a body segment is the mere ultimate measured product, the outcome of multiple processes taking place in parallel in the brain and converging on the groups of neurons responsible for muscle contractions. In my presentation, I will present a different viewpoint and expose my criticisms about the so-called "neuronal code for number".