Abstract
Complex numerical cognition is a crucial ability in the human brain. Conventional neuroimaging techniques do not differentiate between epiphenomena and neuronal groups critical to numerical cognition. Transcranial magnetic stimulation (TMS) allows defining causal models of the relationships between specific activated or inhibited neural regions and functional changes in cognition. However, there is insufficient knowledge on the differential effects of various TMS protocols and stimulation parameters on numerical cognition. This systematic review aimed to synthesize the evidence that different TMS protocols provide regarding the neural basis of numerical cognition in healthy adults. We included 21 experimental studies in which participants underwent any transcranial magnetic stimulation such as a single pulse TMS, repetitive TMS, and theta-burst stimulation. The primary outcome measures were any change in numerical cognition processes evidenced by numerical or magnitude tasks, measured with any independent variable like reaction times, accuracy, or congruency effects. TMS applied to regions of the parietal cortex and prefrontal cortex has neuromodulatory effects, which translate into measurable behavioral effects affecting cognitive functions related to arithmetic and numerical and magnitude processing. The use of TMS for the study of the neural bases of numerical cognition allows addressing issues such as localization, timing, lateralization and has allowed establishing site-function dissociations and double site-function dissociations. Moreover, this technique is in a moment of expansion due to the growing knowledge of its physiological effects and the enormous potential of combining TMS with other techniques such as electroencephalography, functional magnetic resonance imaging, or near-infrared spectroscopy to reach a more precise brain mapping.