Functional magnetic resonance imaging (fMRI) has become a widely used tool to study brain activity by measuring changes in blood oxygen levels, which are assumed to correspond with neuronal activation. Since its invention in 1991, fMRI has provided valuable insights, particularly in clinical applications such as assessing stroke impact and guiding neurosurgical procedures. However, experts caution against overinterpreting fMRI results when it comes to understanding complex human emotions and behaviors.
A notable example is a 2010 study by biological anthropologists at the Kinsey Institute, which observed overlapping brain regions activated during experiences of romantic love and drug addiction, specifically within the caudate, putamen, and ventral tegmentum. This led some to claim that “romantic love is an addiction,” a conclusion that has attracted both interest and skepticism.
Dr. Albert Moukheiber, a cognitive neuroscientist and clinical psychologist, warns that fMRI data must be interpreted with caution. While standard MRI scans visualize brain structures, fMRI detects blood flow changes that indirectly suggest neuronal activity. However, the relationship between blood oxygen levels and actual neural firing is complex and temporally lagged. Each fMRI voxel encompasses millions of neurons, limiting spatial precision and making it impossible to identify activity at the level of individual neurons or determine the exact sequence of neuronal firing.
Further complicating matters, the brain is never truly "inactive"; rather, different areas exhibit varying levels of activation depending on the task. Some researchers, including psychiatrist Sally Satel and neuroscientist Russell Poldrack, argue that certain studies appear to use fMRI findings to support preconceived theories rather than objectively representing brain function.
One illustrative case involves research on fear processing and the amygdalae, almond-shaped brain structures often identified as “fear centers.” A patient known as SM, with Urbach-Wiethe disease causing damage to both amygdalae, demonstrated an absence of fear in some situations but experienced intense fear in others. This suggests that fear arises from distributed brain networks, challenging the notion of localized functions derived solely from fMRI images.
Additionally, experiments have highlighted the risk of false positives in fMRI research. In a notable 2009 study, a dead salmon placed in an fMRI scanner showed apparent brain activity when exposed to social images, underscoring the method’s limitations and the need for rigorous statistical controls.
Critics emphasize that human emotions and behaviors cannot be fully reduced to isolated brain regions lighting up in scans. Environmental, social, and psychological factors profoundly influence these experiences. Moreover, the majority of fMRI studies occur in controlled laboratory settings that may not accurately reflect brain activity in real-life situations. For instance, feelings evoked by viewing a photograph inside a scanner might differ significantly from those experienced during everyday interactions.
Ultimately, while fMRI is a powerful tool for exploring brain function, numerous experts urge restraint in how its findings are interpreted and communicated. The technology offers probabilistic, rather than definitive, mappings of brain activity, and simplifying complex emotions like love into mechanistic terms such as addiction may risk misrepresenting the human experience.