由于功能磁共振成像（functional magnetic resonance imaging，fMRI）软件存在bug ，广泛利用该技术产生的大脑及其活动图像全错了，这意味着过去15年利用该技术的文章都错了。|
Here’s the problem: if the software doesn’t work correctly, the images generated may be incorrect: a part of your brain that appears active may not actually be active. In a study published in the journal PNAS, researchers from Linköping University in Sweden, Anders Eklunda, Thomas E. Nicholsd, and Hans Knutssona, examined fMRI data from 499 healthy patients and found that the software (i.e., SPM, FSL, and AFNI) used to generate the fMRI images often showed parts of the brain lighting up when it shouldn’t have, in some cases up to 70% of the time (i.e., a false positive rate of up to 70%). These software packages had bugs or glitches in them that were leading to faulty images and may have existed for 15 years until they were recently found and corrected. This means that up to around 40,000 fMRI studies published in the scientific literature over this period could have shown incorrect results.
Could Brain Research For The Past 15 Years Be Wrong?
A major discovery (and not the good kind) can mean that up to 40,000 research studies about how the brain functions published over the past 15 years could go out the window. There has been a bug (and not the insect kind) in the software used to create functional magnetic resonance imaging (fMRI) images of the brain. fMRI is an imaging technology widely used to generate pictures of the brain and its activity. If you have ever seen a picture of the brain with different colors in different parts of the brain showing brain activity, that’s fMRI. So, many of the studies that have told us what happens in your brain when you work, play, exercise, have sex, etc. may be wrong. This is (in technical language)