Advantages and Disadvantages of fMRI

Like any technique, fMRI has advantages and disadvantages, and in order to be useful, the experiments that employ it must be carefully designed and conducted to maximize its strengths and minimize its weaknesses.

Advantages of fMRI

• It can noninvasively record brain signals without risks of radiation inherent in other scanning methods, such as CT or PET scans.
• It has high spatial resolution. 2–3 mm is typical but resolution can be as good as 1mm.
• It can record signal from all regions of the brain, unlike EEG/MEG which are biased towards the cortical surface.
• fMRI is widely used and standard data-analysis approaches have been developed which allow researchers to compare results across labs.
• fMRI produces compelling images of brain "activation".

Disadvantages of fMRI

• The images produced must be interpreted carefully, since correlation does not imply causality, and brain processes are complex and often non-localized.
• Statistical methods must be used carefully because they can produce false positives. One team of researchers studying reactions to pictures of human emotional expressions reported a few activated voxels in the brain of a dead salmon when no correction for multiple comparisons was applied, illustrating the need for rigorous statistical analyses[28].
• The BOLD signal is only an indirect measure of neural activity, and is therefore susceptible to influence by non-neural changes in the body. This also means that it is difficult to interpret positive and negative BOLD responses[29]
• BOLD signals are most strongly associated with the input to a given area rather than with the output. It is therefore possible (although unlikely) that a BOLD signal could be present in a given area even if there is no single unit activity.[30]
• fMRI has poor temporal resolution. The BOLD response peaks approximately 5 seconds after neuronal firing begins in an area. This means that it is hard to distinguish BOLD responses to different events which occur within a short time window. Careful experimental design can reduce this problem. Also, some research groups are attempting to combine fMRI signals that have relatively high spatial resolution with signals recorded with other techniques, electroencephalography (EEG) or magnetoencephalography (MEG), which have higher temporal resolution but worse spatial resolution.
• fMRI has often been used to show activation localized to specific regions, thus minimizing the distributed nature of processing in neural networks. Several recent multivariate statistical techniques work around this issue by characterizing interactions between "active" regions found via traditional univariate techniques.
• The BOLD response can be affected by a variety of factors, including: drugs/substances[31]; age, brain pathology[32]; local differences in neurovascular coupling[33]; attention[34]; amount of carbon dioxide in the blood[35]; etc.
• For these reasons, Functional imaging provides insights into neural processing that are complementary to insights of other studies in neurophysiology.