Summary: Researchers found that both vocalized and imagined jazz improvisation produce similar brain states marked by reduced connectivity to the executive control network and an increased sense of flow, enabling freer musical creativity.
Source: Georgia State University
Researchers at Georgia State University have mapped how the brain shifts during moments of musical “flow,” showing that simply imagining improvised performance can evoke the same flow-like brain states as actual singing.
In a study published in Scientific Reports, the team recruited 21 advanced jazz musicians and asked them to either vocalize or mentally imagine improvising one of four Bebop-era scores built on a standard 12-bar blues progression while undergoing functional magnetic resonance imaging (fMRI). The experimental design compared real-time improvisation with the performance of pre-learned, memorized material to isolate neural patterns linked specifically to creative spontaneity.
The multidisciplinary team — including experts in mathematics, physics, music, neuroscience and computer science — analyzed whole-brain fMRI signals to determine how connectivity between major brain networks reorganizes depending on the level of creativity required. Their primary focus was on two key systems: the default mode network (DMN), often associated with internal thought and resting-state activity, and the executive control network (ECN), commonly engaged during goal-directed tasks and problem solving.
“We measured static functional network connectivity while musicians vocalized, imagined, improvised, or performed memorized scores,” said principal investigator Victor M. Vergara. The results showed a consistent pattern: improvisation corresponded with weaker connectivity to the executive control network, a state the researchers linked to the subjective experience of flow, when creativity proceeds unhindered by conscious control.
Martin Norgaard, associate professor in the School of Music and a co-author, emphasized how striking it was that imagined and vocalized improvisation produced remarkably similar neural profiles. “Whether the musicians were scat singing or simply imagining the improvised performance, we observed comparable brain activity patterns,” Norgaard said, underscoring the neural equivalence between internal musical simulation and externalized performance.
This study extends prior work showing reduced functional connectivity during musical improvisation. The current findings suggest that lowered ECN engagement supports the flow state, allowing musicians to create spontaneously without the interference of heavy executive monitoring. At the same time, activity in sensorimotor and other networks aligned with performance demands showed similar patterns whether music was imagined or vocalized externally.
The research was conducted in collaboration with the Center for Translational Research in Neuroimaging and Data Science (TReNDS), supported by Georgia State University, the Georgia Institute of Technology and Emory University. Vince Calhoun, founding director of TReNDS, noted that this work represents the first whole-brain connectivity analysis during both vocalized and imagined real-time creative production. By treating brain function as dynamic over time, the team captured richer connectivity patterns linked to improvisation and rest.
To minimize interference with the creative process, researchers used non-invasive signal extraction methods and developed a custom algorithm to identify resting-state networks and relevant connectivity patterns. “Brain imaging yields massive, time-varying datasets that require pattern-recognition methods to reveal the areas most involved in creativity,” Vergara explained. The team then compared connectivity patterns across improvisation and pre-rehearsed performance to pinpoint the neural signatures of spontaneous creation.
Because the paradigm accommodates expert jazz performers with diverse instrumental backgrounds, the findings may generalize across improvising musicians rather than being limited to a single instrument or voice. The authors propose that the same experimental approach could be adapted to study other real-time creative activities — such as freestyle rap, spoken-word poetry, or even fast-paced athletic performance — to uncover common neural mechanisms of creative flow.
“The next steps will probe dynamic connectivity more closely to determine which networks switch on and off during improvisation and how those transitions relate to the subjective experience of flow,” said Norgaard. That line of inquiry aims to identify the specific temporal changes in brain networks that support moments of uninhibited creativity.
About this music and neuroscience research news
Author: Noelle Reetz
Source: Georgia State University
Contact: Noelle Reetz – Georgia State University
Image: The image is in the public domain
Original Research: Open access. “Functional network connectivity during Jazz improvisation” by Victor M. Vergara, Martin Norgaard, Robyn Miller, Roger E. Beaty, Kiran Dhakal, Mukesh Dhamala, & Vince D. Calhoun. Scientific Reports.
Abstract
Functional network connectivity during Jazz improvisation
Musical improvisation is one of the most complex forms of creativity, requiring the generation of novel material in real time. Neural responses during music production vary with performance conditions, and creative expression may differ when ideas are externalized through an instrument or voice versus when they remain internalized as imagined music.
This study examines whole-brain functional network connectivity from fMRI data recorded during jazz improvisation and compares it against a baseline of performing prelearned scores. The design also contrasts externally vocalized performance with imagined performance to test whether execution versus mental simulation affects connectivity patterns.
Results indicate that improvisation is associated with a state of weakened connectivity that corresponds to reduced recruitment of the executive control network and aligns with the subjective feeling of flow, facilitating uninhibited musical creation. Additionally, patterns of connectivity in sensorimotor and executive networks were similar whether musicians imagined the performance or vocalized it, suggesting a shared neural substrate for internal and external creative production.