Here, we assessed the physiological plausibility of these different models by testing their respective predictions regarding event-related BOLD modulations (forward inference using fMRI).
Population-level inferences concerning BOLD signal changes between conditions of interest were based on a random-effects model.
Population-level inferences about BOLD signal changes between conditions of interest were based on a random effects model that estimated the second level t statistic at each voxel.
The use of temporal variability in BOLD signal to make inferences about regional connectivity could be considered hazardous if it were known that interregional differences in hemodynamic delays varied systematically between regions.
However, making causal inferences that spontaneous BOLD activity in the amygdala directly or indirectly produced synchronous activity at targets (or vice versa) would be misleading and the directionality of the effects cannot be inferred from the correlational techniques we applied.
Importantly, we have also previously reported that peripheral inflammation induced with this model in this subject population does not cause a general change in the coupling between neural activity and blood-oxygen-level-dependent (BOLD) signal, the basis of inferences made in fMRI studies (26).
Note that we marked the AUC values in bold to highlight the opposite trend in inference accuracy of the ratio and correlation procedures.
Without such an understanding, inference of spiking activity from the BOLD signal may be seriously misguided.
One can draw inferences either in a "cautious" or "bold" fashion (also known as "skeptical" or, respectively, "credulous").
Inference of neuronal activity from the fMRI BOLD signal thus requires detailed a priori knowledge of how different stimuli or tasks activate the cortical network.
