Abstract
Transcranial magnetic stimulation (TMS) is a popular tool to study cognition. However, the manner in which TMS actually affects neuronal populations remains unclear. One theory postulates that TMS works similarly to an input-gain model, while opposing views posit that it acts as a noise inducer under the principle of stochastic resonance. Extending the work of Perini et al. (2012), we sought to investigate whether graded levels of visual contrast are differentially affected not merely by TMS, but rather by varying levels of TMS intensity. Methods. Single pulse TMS was delivered to left V1 while participants performed a 2-AFC orientation discrimination (OD) task of one of two Gabor patches presented on either side of fixation at 5 contrast levels and 4 TMS intensities. We analyzed behavioral performance using a 3-way repeated-measures ANOVA and fitted Weibull parameters to the psychometric functions in order to assess whether TMS affected discrimination thresholds. Results. Participants' performance of OD improved with increasing stimulus contrast, irrespective of TMS intensity, and both visual fields were affected by TMS, with increasing TMS intensity yielding decreasing performance at different contrast levels. Additionally, we found an interaction between the effects of TMS intensity and stimulus contrast, wherein increasing TMS intensity decreased performance predominantly at the middle contrast levels. Parameter fitting demonstrated that TMS intensity at 100% yielded decreased contrast sensitivity thresholds in both visual fields compared to the no TMS condition. Summary. Building on the findings of Perini et al. (2012), our results suggest incremental levels of TMS pulse strength yield a gradient-like decrease in performance. These findings contradict previous work which led to the belief that TMS intensity behaved in a relatively linear manner. Overall, our findings cannot corroborate a dichotomous account of TMS in which high and low intensity pulses are confined to their own distinct categories of effects.
Meeting abstract presented at VSS 2017