Skip to content
Open access

TMS Reveals Hybrid Neural Functions Underlying the Sentence Superiority Effect

Aug 2026 · Neurobiology of Language · 0 citations

Abstract

Humans remember sentences far better than random word lists—a phenomenon known as the sentence-superiority effect. Traditional views attribute this effect to content-general mechanisms, in which syntactic–semantic processing facilitates memory across all types of semantic content. However, emerging evidence suggests that specific regions of the lateral temporal lobe selectively support working memory for sentences with social semantics. Here, we employed transcranial magnetic stimulation (TMS) to investigate whether the lateral temporal lobe contributes to the sentence-superiority effect via content-general or social-specific functions. In Experiment 1, offline TMS was delivered to the left ventral temporoparietal junction (vTPJ), lateral anterior temporal lobe (lATL), and posterior superior temporal sulcus (pSTS), with vertex as a control, prior to a recall task. Stimulation led to a decline in the sentence recall advantage over word lists. This decline was more pronounced for high-socialness sentences compared to low-socialness ones, although both conditions were significantly affected. The most substantial sentence recall impairments were observed following vTPJ stimulation. In Experiment 2, online TMS delivered to the vTPJ during sentence maintenance to examine its role in sentence working memory. Disruption again resulted in greater reduction in sentence-superiority effect for high-socialness compared to low-socialness stimuli, with both conditions showing significant declines. Together, our findings provide causal evidence that the left lateral temporal cortex, especially the vTPJ, plays critical roles in sentence working memory. The results tend to favor, though are not exclusive to, a hybrid account in which the lateral temporal cortex contributes to sentence representation through both content-general and social-specific functions.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.