Table 5.

Individual studies and review articles investigating the neurobiology of music, pain, anxiety, and depression

Author, year	Description	Findings
Becerra et al., 2001 [79]	fMRI was used to examine changes in neural activation due to painful thermal stimuli (delivered to the skin of the left hand through a Peltier-based thermode at 46 °C) in 8 healthy male subjects, who rated their pain levels using a Visual Analog Scale	Two temporal phases of pain response were observed: early and late. During the early phase, increased signal was seen in reward regions such as the left anterior nucleus acccumbens (NAc), the ventral tegmentum (VT), the periaqueductal gray (PAG), the sublenticular extended amygdala (SLEA) of the basal forebrain, and the orbital gyrus. In the late response, decreased signal was seen in the left posterior NAc, while increased signal was seen in classical pain regions such as the thalamus, S1, insula, and anterior cingulate gyrus
Blood et al., 1999 [110]	PET was used to examine changes in regional cerebral blood flow (rCBF) related to emotional responses to music. Ten musicians were exposed to 6 versions of a novel musical passage varying systematically in degree of dissonance	Subjective pleasantness and unpleasantness ratings were correlated with consonance and dissonance, respectively. Increasing consonance was correlated with increasing activity in the orbitofrontal cortex, frontal polar cortex, and the subcallosal cingulate gyrus. Increasing dissonance was correlated with increasing activity in the precuneus region and the right parahippocampal gyrus
Blood and Zatorre, 2001 [97]	PET was used to examine changes in rCBF related to “intensely pleasant emotional responses to music” manifesting as piloerective chills measured by electrodermal monitoring. Ten musicians were exposed to self-selected music, control music, amplitude-matched noise, and silence. rCBF values were extracted from individual scans and plotted against chill intensity	Increasing chill intensity was correlated with increasing activity in the left ventral striatum (including the NAc) and dorsomedial midbrain (including the PAG), and decreasing activity in the right amygdala, left hippocampus, and ventral medial prefrontal cortex
Engel et al., 2009 [82]	This article reviews findings from multiple neuroimaging studies using fMRI, PET, and radioligand binding experiments to investigate anxiety in healthy subjects as well as patients with panic disorder, generalized anxiety disorder, social anxiety disorder, and specific phobias	Enhancement of activity in the amygdala was a very common finding. Enhanced activity was also commonly found in the prefrontal cortex, insula, and the anterior cingulate cortex
Menon and Levitin, 2005 [98]	High-resolution fMRI was used to examine changes in neural activation due to “passive music listening.” Fourteen non-musicians were exposed to 10 intact samples of music and 10 scrambled samples for control stimuli. Hemodynamic changes were subject to statistical, functional connectivity, and effective connectivity analyses	Passive listening to music resulted in significant activation of multiple specific structures including the NAc, the VTA, and the hypothalamus. Functional connectivity analysis showed that responses in the NAc and the VTA were highly correlated, “suggesting an association between dopamine release and NAc response to pleasant music”
Nestler and Carlezon, 2005 [85]	This article reviews the mesolimbic dopamine reward circuit in depression. Data from animal studies and some human studies are discussed	The authors discuss how abnormalities in the VTA and NAc may be related to depressive symptoms such as anhedonia, reduced motivation, decreased energy level, etc. Roles of specific proteins, such as the transcription factor cAMP response element binding protein (CREB) and the endogenous kappa-opioid receptor agonist Dynorphin are reviewed
Pereira, 2011 [106]	fMRI was used to investigate blood oxygenation level dependence (BOLD) responses to musical stimuli that varied according to participant familiarity and preference. Fourteen non-musicians underwent a listening test which consisted of 15 s excerpts from 110 pop/rock songs which they rated by degree of familiarity and preference. During fMRI, the participants were then exposed to 48 excerpts, with 12 excerpts from each of the following categories based upon the listening test results: familiar liked, familiar unliked, unfamiliar liked, and unfamiliar unliked	Familiarity was found to have a greater impact than preference in triggering BOLD responses in the following emotion-related regions: putamen, amygdala, NAc, anterior cingulate cortex, and thalamus
Salimpoor et al., 2011 [100]	Ligand-based PET was used to estimate dopamine release in the striatum based on the competition between endogenous dopamine and [11C] raclopride for binding to D2 dopamine receptors. Information about the dynamics of dopamine release over time was collected using fMRI. Eight participants were exposed to self-selected pleasurable musical excerpts versus neutral musical excerpts. Electrodermal skin conductance was used to measure piloerective chills while participants provided subjective feedback about degree of listening pleasure	Compared to neutral musical excerpts, pleasurable musical stimuli (concurrently measured by chill intensity and subjective participant ratings) resulted in distinct striatal responses detectable by PET and fMRI. Ligand-based PET data revealed increased endogenous dopamine transmission, as indicated by decreased [11C] raclopride binding potential in the right NAc. fMRI showed increased BOLD response in the right NAc during peak pleasure experience epochs