Role of the anterior temporal lobes in semantic representations: paradoxical results of a cTBS study

According to the 'Semantic Hub' model, which was developed from data gathered in the moderate to advanced stages of semantic dementia (SD), a unitary amodal mechanism, located in the anterior parts of both temporal lobes (ATLs), should support the interactive activation of semantic representations in all modalities and for all semantic categories. This model has been challenged by clinical findings, which show that in the early stages of SD, when important asymmetries can be observed at the level of the right and left ATLs, the semantic impairment can be modality-specific, mainly affecting lexical-semantic knowledge when the left temporal lobe is more atrophic and pictorial representations when atrophy prevails on the right side. On the other hand, findings of experiments conducted in normal subjects with repetitive transcranial magnetic stimulations (rTMS), support the unitary model. In the most compelling of these studies, rTMS was used to investigate the role of right and left ATLs directly, by comparing semantic processing of the same concepts, presented as written words or pictures. The efficiency of semantic processing for words and pictures was reduced to the same degree by rTMS applied to the left and right ATLs. However, to consider more in depth some methodological inconsistencies of these studies and with the aim of discussing the effects of rTMS on high-level cognitive functions, we decided to repeat that experimental paradigm, using the continuous theta burst stimulation (cTBS) protocol over the right ATL, left ATL and vertex (as control site). A significant interaction was found between side of cTBS application and type of stimulus, but, contrary to our predictions, we observed significantly faster (rather than slower) responses to pictures after application of cTBS to the right ATL and no difference between responses to written words after application of cTBS to the left ATL in comparison with the vertex. These unexpected results are discussed with respect to the nature of the semantic representations supported by the right and left ATLs and to re-appraisal of the 'virtual lesion' account to explain results obtained with rTMS experiments on high-level cognitive functions.