Summary: Researchers describe how brain activity shifts when people create humor.
Source: USC
USC study identifies brain regions engaged when telling a funny story
What happens in your brain when you try to be funny? That appears to depend on experience. Scientists in the Image Understanding Laboratory at the University of Southern California examined professional improvisers, amateur comedians and non-comedian controls as they generated captions for wordless cartoons. Their goal was to map the neural changes that occur during real-time humor creation and to see how expertise shapes those processes.
Study design and methods
Participants—ranging from seasoned improvisers (many affiliated with the Los Angeles Groundlings troupe) to hobbyist comics and people with no comedy background—viewed New Yorker-style cartoons without captions. For each cartoon, they were instructed to produce two captions: one intended to be humorous and one deliberately mundane.
While subjects formulated captions, researchers recorded brain activity using functional magnetic resonance imaging (fMRI). Each caption produced during scanning was later rated for funniness by an independent panel, allowing the team to link brain activity during generation to the perceived humor of the final product.
Key findings: temporal regions versus prefrontal control
The study revealed two primary neural systems involved during joke generation: the medial prefrontal cortex (mPFC) and temporal association regions. How much each area contributed depended on the comedian’s level of experience.
Experienced improvisers showed stronger activation in temporal lobe association areas. The temporal lobe integrates sensory input, supports visual cognition and language comprehension, and converges abstract and semantic associations—functions well suited to producing surprising, contextually rich captions.
By contrast, amateur comedians and non-comedians relied more on the medial prefrontal cortex, a region associated with top-down executive functions such as planning, decision-making and deliberate search strategies. In other words, less experienced participants engaged more conscious, controlled processes when trying to be funny.
As lead author Ori Amir and co-author Irving Biederman explain, professionals tend to trust spontaneous free associations that arise from temporal association networks, while novices depend more on mPFC-guided, effortful search. The study also found that higher temporal region activation during caption generation correlated with higher funniness ratings assigned by outside judges.
Creativity and the medial prefrontal cortex
Across multiple studies of creativity, the medial prefrontal cortex consistently appears as a common node related to creative tasks. Amir notes that mPFC may act as a cognitive control director—coordinating and guiding the creative process—rather than being the sole origin of creative insight. The locus of the creative spark itself varies with task and expertise, emerging in different networks depending on the nature of the creative challenge.
Why humor is a useful model for creativity
Humor provides a compact, well-defined form of creative production: each joke has a clear beginning, middle and end and typically unfolds quickly—features that make it especially suitable for neuroimaging research. Unlike more open-ended creative tasks (for example, composing music or writing poetry), humor produces an end product that is easy to evaluate: it either elicits amusement or it does not. This clarity makes humor an effective testbed for studying the neural mechanisms of creativity and for investigating how those mechanisms change with practice.
Relation to visual recognition and novelty
The temporal association regions activated by humor generation overlap with brain areas previously implicated in aesthetic appreciation and high-level visual recognition. Biederman notes that the pleasure associated with these activations diminishes with repetition: the same joke, image or story loses impact on subsequent exposures. That diminishing return helps explain humans’ drive to seek novel, richly interpretable experiences—what Biederman calls our status as “infovores,” naturally inclined to pursue new information and fresh perspectives.
Irving Biederman holds the Harold Dornsife Chair in Neurosciences at USC Dornsife College of Letters, Arts and Sciences. Ori Amir led the study as a USC doctoral student and is now a postdoctoral researcher. Their work, titled “The Neural Correlates of Humor Creativity,” was published in Frontiers in Human Neuroscience.
Abstract (condensed)
The study contrasted professional and amateur improv comedians and control participants as they generated humorous versus mundane captions for cartoons while undergoing fMRI. Greater comedic experience was associated with increased activation in temporal association regions and decreased activation in the striatum and medial prefrontal cortex. Less experienced participants showed greater mPFC activation, reflecting a deliberate top-down search through associative space. Professionals relied more on spontaneous associations emerging from temporal networks and less on guided control.
Reference: “The Neural Correlates of Humor Creativity” by Ori Amir and Irving Biederman, Frontiers in Human Neuroscience. Published online November 25, 2016. DOI: 10.3389/fnhum.2016.00597