In our study, however, we did not detect any correlation between motor cortex excitability and cortical thickness on the premotor areas. Therefore, since the correlation between the cortical thickness on the precuneus and the cuneus with the sensorimotor cortex excitability was significant in our study population of AD and MCI subjects, it might, at least in part, reflect actual pathophysiological alterations specifically in these areas. The thicknesses on the cuneus and the precuneus correlated positively with the thickness on the sensorimotor cortex in all groups. If the correlation with the EFMT was determined solely by the amount of the neurons, the correlations between the cortical thickness and the EFMT would be similar in all ROIs as would be the relationship between the groups as well. In our results, however, this was not the case. The correlation between the EFMT and the cortical thickness varied between the groups especially on the sensorimotor cortex thus implying that there is a Rapamycin 53123-88-9 difference in the function of the neurons or neuronal circuits in different ROIs between the groups. Healthy controls displayed no correlation between cortical thickness of the ROIs and EFMT. Thus, it seems that normal cortical excitability is not determined solely by the number of excited neurons but instead the healthy brain may have its own individual threshold, depending on local facilitatory and inhibitory interactions. Previous studies have found rMT to be slightly higher in older healthy subjects than in young subjects. One explanation could be that the individual excitability threshold is determined at a young age. Therefore, as a consequence of natural grey matter degeneration related to aging, elderly people have slightly higher motor thresholds than their younger counterparts to counteract the subtle neuronal loss. However, the atrophy rate is slow in normal aging and, thus, there is no direct correlation between the motor threshold and cortical thickness. In AD subjects the grey matter is clearly atrophied and the severity of the atrophy depends on the progression of the disease. Thus, based on the findings of a negative correlation between EFMT and cortical thickness on the sensorimotor cortex in all of the subjects, one would expect that the EFMT required to elicit a MEP would be highest in the AD group since the cortex is thinnest in AD patients. However, this was not the case in our study, as the mean EFMT of AD patients was lower than that of controls or MCIs although the differences were not statistically significant. Moreover, the correlation between EFMT and cortical thickness in the sensorimotor cortex in the AD group was not statistically significant although it was near the threshold of significance. It has been previously shown that motor cortex excitability is increased in AD patients as compared to controls, i.e. a lower stimulation intensity is required to generate MEPs. This hyperexcitability has been hypothesized to reflect impairment in both cholinergic activity and abnormal N-methyl-D-aspartic acid transmission. Furthermore, this hyperexcitability has been hypothesized to counterbalance the neuronal loss in the sensorimotor cortex occurring in AD.