Humor comprehension (i.e., getting a joke) and humor appreciation (i.e., enjoying a joke) are distinct, cognitively complex processes. Functional magnetic resonance imaging (fMRI) investigations have identified several key cortical regions but have overlooked subcortical structures that have theoretical importance in humor processing. The dorsal striatum (DS) contributes to working memory, ambiguity processing, and cognitive flexibility, cognitive functions that are required to accurately recognize humorous stimuli. The ventral striatum (VS) is critical in reward processing and enjoyment. We hypothesized that the DS and VS play important roles in humor comprehension and appreciation, respectively. We investigated the engagement of these regions in these distinct processes using fMRI. Twenty-six healthy young male and female human adults completed two humor-elicitation tasks during a 3 tesla fMRI scan consisting of a traditional behavior-based joke task and a naturalistic audiovisual sitcom paradigm (i.e., Seinfeld viewing task). Across both humor-elicitation methods, whole-brain analyses revealed cortical activation in the inferior frontal gyrus, the middle frontal gyrus, and the middle temporal gyrus for humor comprehension, and the temporal cortex for humor appreciation. Additionally, with region of interest analyses, we specifically examined whether DS and VS activation correlated with these processes. Across both tasks, we demonstrated that humor comprehension implicates both the DS and the VS, whereas humor appreciation only engages the VS. These results establish the role of the DS in humor comprehension, which has been previously overlooked, and emphasize the role of the VS in humor processing more generally.

SIGNIFICANCE STATEMENT Humorous stimuli are processed by the brain in at least two distinct stages. First, humor comprehension involves understanding humorous intent through cognitive and problem-solving mechanisms. Second, humor appreciation involves enjoyment, mirth, and laughter in response to a joke. The roles of smaller subcortical brain regions in humor processing, such as the DS and VS, have been overlooked in previous investigations. However, these regions are involved in functions that support humor comprehension (e.g., working memory ambiguity resolution, and cognitive flexibility) and humor appreciation (e.g., reward processing, pleasure, and enjoyment). In this study, we used neuroimaging to demonstrate that the DS and VS play important roles in humor comprehension and appreciation, respectively, across two different humor-elicitation tasks.

Humor is a ubiquitous human experience that serves an adaptive purpose by facilitating social interactions. It is a higher-order ability and requires the integration of multiple cognitive processes. Humor processing can be separated into at least two distinct components—humor comprehension and humor appreciation (Ziv and Labelle, 1984).

Humor comprehension (i.e., getting the joke) is a problem-solving process in which one detects and resolves some incongruity or absurdity to reveal the joke (Suls, 1972). Humor appreciation refers to the subjective amusement or mirth experienced on realizing the joke. Although humor comprehension generally occurs only once, humor appreciation can be experienced repeatedly with further elaboration, explaining why some jokes remain funny even once the punchline is known.

Advances in neuroimaging allow researchers to explore brain regions involved in humor processing. Functional magnetic resonance imaging (fMRI) has revealed many cortical regions as integral to humor processing. Chang et al. (2023) identified blood oxygen level-dependent (BOLD) activation in the inferior frontal gyrus, the medial frontal gyrus, the superior frontal gyrus, the middle temporal gyrus, and the inferior parietal lobule in incongruity detection and resolution (i.e., humor comprehension). In contrast, activation of the ventromedial prefrontal cortex, the amygdala, the anterior insula, the nucleus accumbens, and the midbrain occurred during the elaboration stage (i.e., humor appreciation). Activation of fronto-temporoparietal areas during humor comprehension and of mesocorticolimbic areas during appreciation aligns with the conclusions of two meta-analyses of 20 and 57 fMRI humor processing studies, respectively (Vrticka et al., 2013; Farkas et al., 2021).

The role of the dorsal striatum (DS; i.e., dorsal caudate nucleus and putamen) in humor processing has generally been overlooked. Although activations of the left putamen (Iwase et al., 2002; Filik et al., 2019; Sanz-Arigita et al., 2021), the right putamen (Goldin et al., 2005; Neely et al., 2012; Shibata et al., 2014), the left caudate (Sanz-Arigita et al., 2021), and the right caudate (Goldin et al., 2005; Osaka et al., 2014; Sanz-Arigita et al., 2021) have been identified in studies of humor processing, most authors do not put importance on these findings or discuss their implications. Filik et al. (2019) were the lone authors to discuss the role of the putamen in language processing and how this could contribute to humor comprehension. DS involvement in ambiguity resolution (Crinion et al., 2006; MacDonald and Monchi, 2011; Mestres-Missé et al., 2012), suppression of prepotent responses (Zandbelt and Vink, 2010; MacDonald and Monchi, 2011; Akkermans et al., 2018), working memory (Lewis et al., 2004; MacDonald and Monchi, 2011; Darvas et al., 2014), and set shifting (MacDonald and Monchi, 2011; Darvas et al., 2014), which are essential for humor comprehension, have not been considered in the context of humor processing. The tendency is to ignore these DS activations or explain them in the context of reward processing, although experiencing humorous stimuli as rewarding pertains to humor appreciation, a process that has been shown clearly to implicate the ventral striatum (VS; nucleus accumbens and ventral caudate nucleus and putamen, z ≤ 2 using MRI; Mobbs et al., 2003; Azim et al., 2005; Mobbs et al., 2005; Watson et al., 2007; Neely et al., 2012; Noh et al., 2014; Shibata et al., 2014) and not the DS. Discounting the role if the DS in cognitive functions and misattributing all striatal activations in humor processing to affective/reward functions has caused subregions of the striatum to be excluded in reviews of the literature and theories of humor processing.

Our aim was to directly investigate the distinct contributions of the DS and VS in humor processing. We predicted that humor comprehension will involve the DS, whereas humor appreciation will engage the VS. We investigated these hypotheses using both a traditional behavior-based humor processing task and a naturalistic sitcom viewing method in fMRI with striatal regions of interest (ROIs).

Using fMRI and two independent measures of humor processing, performed by the same healthy young participants, we investigated BOLD activity associated with humor comprehension and humor appreciation. In whole-brain analyses, for both tasks we found significant activation of the inferior frontal gyrus, the middle frontal gyrus, the supplementary motor area, the middle temporal gyrus, the temporal poles, and the midbrain for humor comprehension. We found common activations in the temporal cortex [i.e., Brodmann area (BA) 37 and BA 38] for humor appreciation in both tasks. In addition to whole-brain analyses, we examined BOLD signal in the DS and VS associated with humor comprehension and appreciation with an ROI approach. In both tasks, we found that humor comprehension seems to implicate both the DS and VS, whereas humor appreciation preferentially engages the VS. These findings align with our expectations that different brain regions underlie humor comprehension and appreciation and that the striatum is involved in humor processing.

Our whole-brain and conjunction analyses corroborated the findings of previous studies regarding cortical regions that are involved in humor comprehension and appreciation. For the humor comprehension contrast in the Joke task, we found significant activation in the left inferior frontal gyrus, the left middle frontal gyrus, the left superior frontal gyrus, the bilateral middle temporal gyrus, the bilateral temporal pole, the left angular gyrus, the left supplementary motor area, the left precentral gyrus, the left putamen, the left midbrain, the left thalamus, and the right amygdala. We corroborated these results with our Seinfeld viewing task, albeit with a slight shift in hemispheric lateralization, finding significant clusters of activation in the left inferior frontal gyrus, the right middle frontal gyrus, the bilateral middle temporal gyrus, the right superior temporal gyrus, the right temporal pole, the right supramarginal gyrus, the left fusiform gyrus, the right supplementary motor area, the bilateral insula, the left hippocampus, the left midbrain, and the right amygdala. Many of these cortical regions (e.g., the inferior frontal gyrus, the middle temporal gyrus) have been identified in previous studies of humor comprehension (Bartolo et al., 2000; Goel and Dolan, 2001; Wild et al., 2006; Samson et al., 2008, 2009; Bekinschtein et al., 2011; Chan et al., 2013; Vrticka et al., 2013; Osaka et al., 2014). The shift in hemispheric lateralization between the Joke task and Seinfeld viewing might be because of the differences in humor modality between these tasks. Verbal humor, which was measured in the Joke task, is associated with greater activation in the left hemisphere, whereas visual/situational humor as assessed in the Seinfeld viewing task, is associated with greater activation in the right hemisphere (Moran et al., 2004; Vrticka et al., 2013). We also found activation in a cluster encompassing the left putamen in this contrast, which is consistent with a previous meta-analysis of 28 studies that identified coactivation of the left anterior putamen and cortical regions such as the inferior frontal gyrus and precentral gyrus during language processing tasks (Viñas-Guasch and Wu, 2017). Finally, we also observed significant activation of the left midbrain for humor comprehension in both tasks, which, coupled with our striatal ROI findings, could indicate that humor comprehension involves dopamine signaling. For humor appreciation, we found activations of BA 38 (temporal pole) in the Joke task and BA 37 (inferior temporal gyrus) in the Seinfeld viewing task. Our conjunction analysis confirmed that temporal regions, among others, were activated by humor appreciation across both tasks. The temporal cortex has been implicated in laughter associated with mirth (Swash, 1972; Satow et al., 2003; Wildgruber et al., 2013; Caruana et al., 2015; Yamao et al., 2015) as opposed to nonmirthful laughter, which implicates the anterior cingulate cortex (Caruana et al., 2015), a region that was not identified in our whole-brain analyses of humor appreciation. Importantly, the temporal cortex has been identified in previous fMRI studies of humor appreciation (Mobbs et al., 2003; Kipman et al., 2012; Amir et al., 2015). Interestingly, our humor appreciation conjunction analysis also revealed activation of medial occipital regions (i.e., the lingual gyrus and the cuneus). These regions have been implicated in nonvisual functions such as language processing (Palejwala et al., 2021).

Consistent with other studies of humor processing, our whole-brain analyses showed sparse subcortical activation. Although whole-brain analysis is a popular approach for analyzing fMRI data, the height and extent thresholds that are routinely applied to correct for multiple comparisons favor larger cortical regions, making it difficult for activation in small brain regions (e.g., DS and VS) to survive these corrections. Illustrating this, most of our striatal and midbrain clusters barely exceed 10 contiguous voxels, with our largest measuring only 112 voxels in extent. Failing to account for these small-volume regions either through ROI analyses or small-volume correction might have led to omission of the striatum and midbrain in theories of humor processing.

For our striatal ROI analyses, we found significant activation in the DS and VS for humor comprehension in both the Joke and Seinfeld viewing tasks. This supports and extends our initial hypothesis that the DS is involved in humor comprehension. First, the DS is implicated in cognitive functions that underlie humor comprehension, including inhibition of prepotent responses (Zandbelt and Vink, 2010; MacDonald and Monchi, 2011; Akkermans et al., 2018), cognitive flexibility (Crinion et al., 2006; MacDonald and Monchi, 2011; Mestres-Missé et al., 2012), and working memory (Lewis et al., 2004; MacDonald and Monchi, 2011; Darvas et al., 2014). Furthermore, the DS is functionally and structurally connected to frontotemporal cortical regions that have been implicated in humor comprehension and related processes, such as the inferior frontal gyrus (Kireev et al., 2015; Haber, 2016; Graff-Radford et al., 2017). For example, the right putamen demonstrates functional connectivity with the left inferior frontal gyrus, the left superior temporal gyrus, the left precentral gyrus, and the left middle temporal gyrus during language processing (Viñas-Guasch and Wu, 2017), and the left caudate head and the inferior frontal gyrus demonstrate increased functional connectivity during deliberate deception in young, healthy humans (Kireev et al., 2015). Finally, there is evidence that patients with Parkinson's disease, in which the DS is dopamine depleted, experience deficits in humor comprehension but not humor appreciation (M. Prenger, K. Van Hedger, K. Seergobin, AM Owen, PA MacDonald, unpublished observations). Together, this body of literature supports the notion that the DS is intricately involved in social and cognitive functions, such as humor comprehension, via its connections with cortical areas that have a demonstrated role in humor comprehension. Here, we have demonstrated that the DS indeed plays a role in humor comprehension and have replicated this result across two different humor processing elicitation methods.

The involvement of the VS in humor comprehension was somewhat unanticipated. There are a few studies that implicate the VS and the ventral tegmental area (VTA; the region that supplies dopamine to the VS) in humor comprehension (Chan et al., 2012, 2023). It is possible that the VS contributes to humor comprehension by motivating the resolution of incongruities. In their discussion, Chan et al. (2012) suggest that VS activation during humor comprehension might be related to a feeling of relief associated with incongruity resolution that might be separate from the amusement feeling of humor appreciation. The VS is also implicated in reward expectation (Knutson et al., 2001; de la Fuente-Fernández et al., 2002; Filimon et al., 2020; Pool et al., 2022). Given that humor comprehension is an effortful process, activation of the VS in anticipation of a potential humor-appreciation-related reward might help to drive the humor comprehension process forward. This could be related to the role of the VS in humor generation (another effortful process), demonstrated by Amir and Biederman (2016).

Unsurprisingly, we observed significant activation of the VS during humor appreciation in the Joke task and the Seinfeld viewing task. Activation of the VS during humor appreciation has been well established in the previous literature (Mobbs et al., 2003; Azim et al., 2005; Mobbs et al., 2005; Watson et al., 2007; Bekinschtein et al., 2011; Neely et al., 2012; Noh et al., 2014; Shibata et al., 2014; Chang et al., 2023), and aligns with the role of the VS in reward processing and prediction error (Schultz, 2016). Importantly, our Bayesian one-sample t tests supported the null hypothesis that the DS is not activated during humor appreciation. DS activation during humor processing appears not to be linked to the rewarding nature of humor appreciation. Rather, activation of the DS during humor processing seems related to the cognitive processes that support humor comprehension.

Our findings represent an advancement in the field of humor research by establishing roles for both the DS and VS in humor comprehension and for the VS only in humor appreciation. This could suggest that midbrain dopaminergic signaling is an important component of humor processing. So far, only behavioral research has demonstrated humor comprehension deficits in dopamine-related disorders, such as Parkinson's disease (Benke et al., 1998; Thaler et al., 2012; Mensen et al., 2014). Further research using neuroimaging, clinical cohorts, and pharmacological manipulation would provide further support for the hypothesis that dopamine signaling is involved in humor comprehension and appreciation.