Different neural representations for detection of symmetry in dot-patterns and in faces: A state-dependent TMS study

The chronometry of symmetry detection in the lateral occipital (LO) cortex

The lateral occipital cortex (LO) has been shown to code the presence of both vertical and horizontal visual symmetry in dot patterns. However, the specific time window at which LO is causally involved in symmetry encoding has not been investigated. This was assessed using a chronometric transcranial magnetic stimulation (TMS) approach. Participants were presented with a series of dot configurations and instructed to judge whether they were symmetric along the vertical axis or not while receiving a double pulse of TMS over either the right LO (rLO) or the vertex (baseline) at different time windows (ranging from 50 ms to 290 ms from stimulus onset). We found that TMS delivered over the rLO significantly decreased participants' accuracy in discriminating symmetric from non-symmetric patterns when TMS was applied between 130 ms and 250 ms from stimulus onset, suggesting that LO is causally involved in symmetry perception within this time window. These findings confirm and extend prior neuroimaging and ERP evidence by demonstrating not only that LO is causally involved in symmetry encoding but also that its contribution occurs in a relatively large temporal window, at least in tasks requiring fast discrimination of mirror symmetry in briefly (75 ms) presented patterns as in our study.

Nonlinear interaction between stimulation intensity and initial brain state: Evidence for the facilitatory/suppressive range model of online TMS effects

The effects of online Transcranial Magnetic Stimulation (TMS) can qualitatively vary as a function of brain state. For example, TMS intensities which normally impair performance can have a facilitatory effect if the targeted neuronal representations are in a suppressed state. These phenomena have been explained in terms of the existence of distinct facilitatory and suppressive ranges as a function of TMS intensity which are shifted by changes in neural excitability. We tested this model by applying TMS at a low (60 % of phosphene threshold) or high (120 % of phosphene threshold) intensity during a priming paradigm. Our results show that state-dependent TMS effects vary qualitatively as a function of TMS intensity. Whereas the application of TMS at 120 % of participants' phosphene threshold impaired performance on fully congruent trials (in effect, reducing the benefit of priming), TMS applied at a lower intensity (60 % of phosphene threshold), facilitated performance on congruent trials. These results demonstrate that behavioral effects of TMS reflect a nonlinear interaction between initial activation state and TMS intensity. They also provide support for the existence of facilitatory/suppressive ranges of TMS effects which shift when neural excitability changes.