Automatic and intentional number processing both rely on intact right parietal cortex: a combined FMRI and neuronavigated TMS study

Abstract

Practice and training usually lead to performance increase in a given task. In addition, a shift from intentional toward more automatic processing mechanisms is often observed. It is currently debated whether automatic and intentional processing is subserved by the same or by different mechanism(s), and whether the same or different regions in the brain are recruited. Previous correlational evidence provided by behavioral, neuroimaging, modeling, and neuropsychological studies addressing this question yielded conflicting results. Here we used transcranial magnetic stimulation (TMS) to compare the causal influence of disrupting either left or right parietal cortex during automatic and intentional numerical processing, as reflected by the size congruity effect and the numerical distance effect, respectively. We found a functional hemispheric asymmetry within parietal cortex with only the TMS-induced right parietal disruption impairing both automatic and intentional numerical processing. In contrast, disrupting the left parietal lobe with TMS, or applying sham stimulation, did not affect performance during automatic or intentional numerical processing. The current results provide causal evidence for the functional relevance of right, but not left, parietal cortex for intentional, and automatic numerical processing, implying that at least within the parietal cortices, automatic, and intentional numerical processing rely on the same underlying hemispheric lateralization.

Keywords: automaticity; brain stimulation; congruity effect; distance effect; intentional processing; lateralization; numerical cognition; parietal lobe.

Figure 1

Experimental paradigm. Two tasks were presented in separate blocks: a physical size comparison task and a numerical comparison task. The numerical values appeared in these tasks can affect automatic and intentional numerical processing, respectively. Each pair of stimuli was preceded by a fixation point and a blank of 500 ms each, and remained visible for 1 s. After an inter-trial interval of 6 s a new trial began with the presentation of a fixation point. Responses were indicated by a button press on the side corresponding to the larger relevant dimension. Triple-pulse TMS at 60% of the maximum stimulator output was applied at 220, 320, and 420 ms after stimulus onset.

Figure 2

Behavioral effect for automatic and intentional numerical processing [(A,B) respectively]. (A) The size congruity effect for the right IPS, sham, and left IPS stimulation. Only TMS over the right IPS decreased automatic numerical processing as indicated by a significantly reduced size congruity effect. (B) The numerical distance effect for the right IPS, sham, and left IPS stimulation. Only TMS over right IPS decreased intentional numerical processing as indicated by a significantly reduced numerical distance effect. Hence, the same functional parietal asymmetry seems to underlie automatic and intentional numerical processing. Error bar reflects one SE of mean.

Figure 3

Plotting individual activation hot spot for the size congruity effect and the numerical distance effect. Individual activation hot spot for the size congruity effect (sphere) and the numerical distance effect (rectangular) as used for the TMS neuronavigation study plotted on a flat brain using Caret (Van Essen et al., ; Van Essen, 2002). Different colors represent different individuals. As can be seen, the average coordinates for the size congruity effect and the numerical distance effect in each hemisphere show a substantial overlap. This is indicated by the black and white rings in each hemisphere (size congruity and numerical distance, respectively) and by the bottom graphs (each graph depicts the average coordinates from the hemisphere above it).