Mapping the differential impact of spontaneous and conversational laughter on brain and mind: A fMRI study in autism

doi:10.21203/rs.3.rs-3575382/v1

Download PDF

Research Article

Mapping the differential impact of spontaneous and conversational laughter on brain and mind: A fMRI study in autism

https://doi.org/10.21203/rs.3.rs-3575382/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 May, 2024

Read the published version in Cerebral Cortex →

Version 1

posted

You are reading this latest preprint version

Background While most research on the non-verbal communication challenges encountered by autistic people centres on visual stimuli, non-verbal vocalizations remains overlooked. Laughter serves as a socio-emotional signal for affiliative bonding in interactions. Autistic people seem to experience and produce laughter differently to non-autistic people and are known to have mentalizing difficulties. Neuroimaging evidence suggests that non-autistic people engage in mentalizing to understand the meaning behind conversational laughter. Here, we propose that autistic people’s difficulties lie not in processing all types of laughter, but only in processing conversational laughter.

Methods We used fMRI to explore the neural mechanisms underlying implicit processing of different types of laughter in autistic and non-autistic adults. 23 autistic and 23 non-autistic adults matched for age, gender, and IQ were scanned whilst passively listening to funny words followed by spontaneous laughter, conversational laughter, or noise-vocoded vocalizations. Post-scan, they listened to the word plus laughter pairs again and rated the funniness of each word.

Results Behaviourally, words plus spontaneous laughter were rated as funnier than words plus conversational laughter, and the groups did not differ. However, neuroimaging results showed increased activation in the medial prefrontal cortex only for non-autistic adults during implicit processing of words plus conversational laughter. Additionally, autistic adults showed greater activation in the supplementary motor area, a part of the sensorimotor network, when listening to words plus either type of laughter.

Limitations Our current design does not disentangle the processing of humour and laughter, with insufficient temporal resolution for their neural distinction, suggesting a need for future research with more temporally-precise techniques.

Conclusions While both non-autistic and autistic adults find laughter makes funny words funnier, there are differences in autism in the neural mechanisms that support mentalizing and contagion during implicit laughter processing. This study reinforces the idea that the essence of laughter serves both as an emotional expression and a sophisticated social signal during communication. It also highlights the challenges that autistic people face in understanding the meaning behind the conversational laughter we consistently encounter in everyday life, which may lead them to social vulnerability. Altogether, we advocate for clearer communication with autistic people.

Autism spectrum disorder

laughter

social communication

fMRI

medial prefrontal cortex

Autistic people often encounter difficulties in non-verbal social communication [1]. While most research in this area centres on visual cues, such as eye gaze, gesture and facial expressions, revealing different patterns between autistic and non-autistic people [2, 3], auditory cues have largely been neglected. It is equally important to understand how autistic people experience non-verbal auditory cues differently. For instance, laughter, as a universal non-verbal vocalization, plays a significant role in social bonding during human interactions. However, our understanding of laughter in autistic people remains limited. We have previously found laughter is implicitly processed by both autistic and non-autistic people: not only by its presence but also by its type, namely spontaneous and conversational laughter. Here, by using fMRI, we take a step further to explore whether there are neural differences during the implicit processing of different types of laughter in autism. In other words, we challenge the assumption that autistic people universally experience difficulties with all forms of non-verbal cues; in our case, all types of laughter. In this novel study, we seek to explore neural similarities and differences during the implicit processing of different types of laughter, specifically spontaneous and conversational, among autistic and non-autistic people.

Laughter is often viewed as a spontaneous and uncontrolled emotional vocalisation triggered by tickling and humour [4, 5]. Intriguingly, however, laughter predominantly occurs in conversation as a voluntary communicative signal [6, 7]: people frequently laugh after verbal utterances to signal affiliation and agreement with others, mediating the meaning of utterances, and regulating the flow of conversation [6]. Spontaneous and conversational laughter are therefore both salient social signals but play very different roles in communicating socio-emotional meaning [8], recruit different production systems [9, 10], and are perceived to differ in authenticity, associated with acoustic differences [11]. Understanding the meaning of laughter is therefore essential to establish and maintain social bonds and relationships [12]. Given laughter’s evolutionary roots, it not only promotes group cohesion but also fosters rapport in human interactions [13, 14].

The perception of laughter engages oro-facial mirror networks, including premotor cortex, the pre-supplementary motor area (SMA) and right inferior frontal gyrus [15, 16], consistent with findings that much of human adult laughter is associated with behavioural contagion [17] – we are primed to laugh when we hear laughter, and we are 30 times more likely to laugh when with others than when alone. The contagious-laughter effect is strongly mediated by social contexts, such as the audience size and the intimacy/familiarity of the relationship [17, 18]. However, autistic people tend to express and experience laughter differently to their peers [19–23]. Laughter may be less socially contagious for autistic people: autistic children are less likely to join in others’ laughter and are more prone to laughing by themselves; they seldom try to elicit laughter from others and are less inclined to laugh at funny faces or socially inappropriate act [22]. They perceive cartoons with a laugh track to be less enjoyable than one without, and laugh less when watching these cartoons than non-autistic children [20].

When considering the processing of different types of laughter, passive listening to spontaneous laughter induces greater activity in auditory cortex (superior temporal gyri, STG) than conversational laughter, likely reflecting the processing of acoustic differences in authenticity between these types of laughter. Interestingly, conversational laughter specifically engages mentalizing networks, showing greater activation in the medial prefrontal cortex (mPFC) and anterior cingulate cortex (ACC) compared to spontaneous laughter [24], and the degree of mPFC activation correlates with the perceived authenticity of laughter [25]. This suggests that laughter processing also involves representing the intentions behind the laughter, especially with socially ambiguous conversational laughter. As autistic people experience difficulties in mentalizing, supported by neuroimaging evidence of atypical activation of mPFC [26–28], they may specifically struggle to comprehend the meaning of conservational laughter, processing it more like spontaneous laughter, which may subsequently impact their use of laughter in social interactions. Indeed, autistic children rarely produce unvoiced laughter during social play, which is more closely associated with spontaneous than conversational laughter [21].

Studies focusing on the processing of laughter in autistic adults are rare. We have demonstrated that laughter increases the funniness of ‘dad jokes’, by adding either spontaneous or conversational laughter to spoken jokes [29]. Spontaneous laughter made the jokes even funnier than conversational laughter, likely due to its greater authenticity; intriguingly, we found this effect in both autistic and non-autistic adults. Despite typical behavioural responses to laughter, an fMRI study of laughter in autistic adults found reduced mPFC activation to written jokes followed by laughter; while laughter increased the pleasantness of jokes for all participants, this effect was smaller in autistic adults [30]. However, this study did not report what type of laughter was used. It is therefore unclear whether autistic adults would show reduced mentalizing-related activation during the processing of all types of laughter due to their atypical non-verbal communication, or whether autistic adults might only show atypical neural activity to conversational laughter that contains a degree of social ambiguity, whilst perceiving spontaneous laughter similarly to non-autistic people. To address this gap in the literature, we aim to further explore the neural systems recruited in the implicit processing of different types of laughter. Using fMRI, both autistic and non-autistic adults passively listened to funny words followed by either spontaneous laughter, conversational laughter, or noise-vocoded human vocalization. Post-scan, participants listened to the word plus laughter pairs again and rated the funniness of each word. We hypothesized that there would be differences in the neural correlates of implicit processing of these different types of laughter between autistic and non-autistic adults, specifically in the mPFC when listening to conversational laughter, and in the sensorimotor network to both types of laughter, despite typical behavioural responses to laughter.

Participants

Twenty-five autistic adults and 23 non-autistic adults participated in this study, they were right-handed and had no speech, hearing or neurological difficulties. The groups were comparable for sex (χ²(1) = .060, p = .807), age (t(46) = − .134, p = .894), and verbal (t(46) = .720, p = .475), performance (t(46) = − .875, p =. 386), and full-scale IQ (t(46) = .031, p = .975) as measured by four subtests of the Wechsler Adult Intelligence Scale (WAIS-IV UK; [31]; Matrix Reasoning; Block Design; Similarities; Vocabulary). The groups differed on Autism Spectrum Quotient (AQ; [32], t(46) = 11.879, p < 0.001. Although two autistic males were unable to complete the scan due to discomfort and health concerns, the scan groups remained comparable for sex (χ²(1) = .000, p = 1.000), age (t(44) = − .258, p = .798), and verbal (t(44) = .392, p = .697), performance (t(44) = -1.123, p =. 268), and full-scale IQ (t(44) = − .303, p = .763); and differed on AQ, t(44) = -12.325, p < 0.001. See Table 1.

The autistic participants had received an official diagnosis from a qualified clinician. Due to testing restrictions during the COVID-19 pandemic, we were unable to administer ADOS [33] to confirm their diagnosis. Nonetheless, the AQ was used in the pre-screening assessment, and non-autistic participants with an AQ score below the cut-off of 32 were included in the study. Non-autistic participants were recruited from the Autism@ICN and UCL SONA subject database. Autistic participants were recruited through advertisements placed at the Cambridge University Autism Research Centre (ARC) and various university disability services throughout the United Kingdom, in addition to the aforementioned subject databases. Informed written consent was obtained prior to testing, and the project received approval from the UCL research ethics committee.

Table 1

Participant Demographic Information
	NA	Autism	Autism for scan
N (male: female)	23 (13:10)	25 (15:10)	23 (13:10)
Age (years)	28.348 (9.203)	28.720 (9.969)	29.087 (10.238)
Verbal IQ	121.435 (15.048)	117.800 (19.416)	119.435 (19.315)
Performance IQ	109.478 (15.985)	113.560 (16.289)	114.783 (16.059)
Full Scale IQ	119.000 (15.895)	118.840 (19.429)	120.565 (19.023)
AQ	13.652 (5.928)	35.560 (7.292)	37.348 (7.062)
Note. Values are given as mean (standard deviation). NA = non-autistic; AQ = autism-spectrum quotient.

Experimental Design and Procedure

Word Stimuli. A subset of 300 words was selected from the original pool [34]. The words were carefully chosen to avoid floor and ceiling effects, as per the results of the baseline ratings task (Funniness: Mean = 3.309, SD = .586; Intelligibility: Mean = 92.959%, SD = 10.619%), and recorded by a professional male comedian in a comedy performance (Duration: Mean = .736 seconds, SD = .225 seconds; Root-mean-square: Mean = .031, SD = .000; Pitch: Mean = 189.677 Hz, SD = 82.154 Hz). Full details of word selection are given below.

To avoid the floor effect, we initially selected 719 words with a humour rating higher than 2.8 from a 5-point Likert scale (1 - humourless; 5 - humorous) from original pool [34]. Subsequently, a list of 621 words were audio recorded by a professional male comedian in a comedy performance following annotated by four native English speakers to ensure for appropriateness. The raw audio file was downsampled at a rate of 44100 Hz to mono.wav files with 32-bit resolution, and each word was trimmed and edited into a 1-second sound file (.wav) using version 2.3.3 of Audacity(R) recording and editing software. The files were then normalised for root-mean-square (RMS) amplitude using PRAAT [35]. We further conducted an online task to establish baseline ratings for the funniness of words. In this task, 621 words were assigned to three lists, each containing 207 words. The three lists were matched on the humour ratings from the original pool (List 1: M = 3.19, SD = 1.23; List 2: M = 3.20, SD = 1.23; List 3: M = 3.18, SD = 1.23). Fifty-eight native English speakers were randomly assigned to one of the three lists (list 1 n = 18; list 2 n = 19; list 3 n = 21). Participants were instructed to listen to the recordings of a comedian named Ben performing some funny words and were asked to rate the funniness of each word on a 7-point rating scale (“How funny was the word the way that Ben said it?” 1 - Not funny at all, 4 - Neutral, 7 - Extremely funny). Additionally, participants indicated whether they understand the meaning of each word. A practice trial was given before the actual task. The task was built and presented using Gorilla Experiment Builder [36].

Sound Stimuli. This study used 150 sound stimuli: 50 spontaneous laughter stimuli, 50 conversational laughter stimuli, and 50 noise-vocoded (NV) human vocalisations. The spontaneous and conversational laughter stimuli were recorded using a method previously validated in behavioural and neuroimaging experiments [16, 24] and were selected from a previous study, as detailed in [29]. We created NV stimuli by applying noise-vocoding to various human emotional vocalizations, including expressions of anger, pleasure, disgust, surprise, etc. The resulting NV stimuli lack emotional meaning and are not perceived as contagious by normal-hearing listeners. We normalised all stimuli for root-mean-square (RMS) amplitude using PRAAT [35], and extracted the acoustic parameters. One-way ANOVA indicated that spontaneous laughter, conversational laughter and NV sound stimuli were comparable in duration, root-mean-square (RMS), and intensity (Table 2). For the detailed comparation of acoustic properties between spontaneous and conversational laughter, see Table S1.

Table 2

Acoustic properties of sound stimuli set
Acoustic measure	Conditions	Mean	SD	F	p
Total duration (sec)	Spont	2.376	.406	1.297	.276
	Conver	2.269	.361
	NV	2.307	.216
Root-mean-square	Spont	.317	.000	.908	.406
	Conver	.317	.000
	NV	.317	.000
Intensity (dB)	Spont	64.000	.000	.945	.391
	Conver	64.000	.000
	NV	64.000	.000

Note. df = (2, 147). p values are given in two tailed. Spont = Spontaneous laughter. Conver = Conversational laughter. NV = Noised-vocoded human vocalization.

fMRI Experimental Design. An event-related paradigm was utilized, with each trial consisting of a jittered inter-trial interval (ITI) ranging from 2 to 4 seconds. In the sound stimuli conditions, a funny word was presented, followed by a sound stimulus from one of three conditions (Spont Laugh, Conver Laugh, NV) with a fixed duration inter-stimulus interval (ISI) of 0.09 seconds. The rest condition included a 2-second period of silence following the ITI. Vigilance trials involved a 0.5-second beep sound, requiring participants to press a button within 3 seconds. Each functional run, approximately 14 minutes long, comprised 25 trials per condition and five vigilance trials to assess attentiveness. The entire experiment consisted of four functional runs, each comprising 105 trials, during which participants passively listened to 300 words paired with sound stimuli. The sound stimuli were used twice each. Trial conditions were pseudorandomized to prevent more than two consecutive trials of the same condition involving words plus sound stimuli. Furthermore, neither rest nor vigilance conditions were the first trial of each run, nor were they presented consecutively. The pairs of words plus sound stimuli were pseudorandomized and counterbalanced across both runs and participants (see Fig. 1A).

Behavioural Experimental Design. Participants listened to the 200 word plus laughter pairs again and rated the funniness of each word. Due to time constraints imposed by COVID testing restrictions, the 100 word plus NV pairs were excluded. The pairs were the same as in the previous scan conditions, but the order of the pairs was shuffled. Participants rated each word on a 7-point rating scale (‘How funny was the word the way that Ben said it?’, 1 – Not funny at all, 4 - Neutral, 7 - Extremely funny). For each trial, participants had up to 6 seconds to give a rating. There was a short practice session before the real task to let participants become familiar with the structure of the task. The post-scan behavioural task lasted approximately 25 minutes (See Fig. 1B).

Procedure. Participants were informed that the fMRI study was about humour processing. Notably, any mention of laughter was intentionally avoided during the recruitment and testing. Before the scan, participants were instructed to listen to humorous words spoken by a comedian, and people's reactions. They were told to press a button on a button-box whenever they heard a "beep" sound. A practice sequence at the beginning ensured the volume was adequate and that participants could clearly hear the stimuli by recalling the words they heard. The testing lasted approximately 2 hours, split between a one-hour scan at the Birkbeck-UCL Centre for Neuroimaging (BUCNI) and a one-hour behavioural session at the UCL Institute of Cognitive Neuroscience, which encompassed the post-scan behavioural study, WAIS testing, and questionnaires. Both fMRI and behavioural experiments were presented using MATLAB [37] with the psychophysics toolbox [38].

Neuroimaging, Pre-processing and Analysis

Acquisition. We employed continuous event-related fMRI, acquiring blood-oxygen-level-dependent (BOLD) images using a Siemens Avanto 1.5-Tesla MRI scanner with a 32-channel head coil. The study involved four runs of 260 echo-planer whole-brain volumes (TR = 3s; TE = 50ms; TA = 86ms; Slice tilt = 25 ± 5 degrees; flip angle = 90°; 3 mm × 3 mm × 3 mm in-plane resolution). Auditory stimuli were delivered via an MR-compatible insert earphone connected to a Sony STR-DH510 digital AV control centre. After two functional runs, we obtained high-resolution anatomical images using a T1-weighted magnetisation prepared - rapid acquisition gradient echo sequence (MPRAGE; 176 sagittal slices, TR = 2730ms; TE = 3.57ms; flip angle = 7°, acquisition matrix = 224 × 256 × 176, slice thickness = 1 mm, 1 mm x 1 mm x 1mm).

Pre-processing. The first three volumes of each EPI sequence were discarded. The remaining volumes underwent spatial alignment along the AC-PC axis for each participant, followed by slice time correction using the last slice as a reference. The corrected images were then spatially realigned and registered to the? mean. The structural image was co-registered with the mean of the corrected images, aligning structural scans with SPM12 [39] tissue probability maps during segmentation. The forward deformations image from the segmentation was used to normalise the functional images to standard MNI space. Finally, the normalised functional images were resampled into 2 × 2 x 2 mm voxels and spatially smoothed using an isotropic 8 mm full width at half maximum (FWHM) Gaussian kernel.

Analysis. fMRI data were analysed in an event-related manner. Variance in each time series was decomposed in a voxelwise general linear model with the following regressors: onsets and durations of 1) words plus spontaneous laughter (Spont Laugh), 2) words plus conversational laughter (Conver Laugh), 3) words plus NV stimuli (NV), 4) vigilance trials. These regressors, along with six additional regressors representing realignment parameters calculated by SPM12 [39], constituted the full model for each session. The data were high-pass filtered at 128 seconds.

Individual design matrices were contrasted per participant [All Laughs (Spont Laugh & Conver Laugh) > NV, Spont Laugh > NV, Conver Laugh > NV], modelling the three experimental conditions across four runs and including movement parameters as nuisance variables. These contrasts were entered into a second level, two-sample t-test for the group analysis. Whole-brain analysis results were corrected for multiple comparisons using a cluster-extent-based thresholding approach [40]: a voxel-wise threshold of p < 0.001 combined with a cluster extent threshold determined by SPM12 [39] (p < 0.05 family-wise-error (FWE) cluster-corrected threshold). All reported clusters exceeded this cluster-corrected threshold. Reported cluster coordinates corresponded to the Montreal Neurological Institute (MNI) coordinate system and were labelled using the AAL labelling atlas in SPM12 [39].

Software. Pre-processing and statistical analysis were conducted in SPM12 [39], implemented in MATLAB R2018B [37]. The MarsBaR toolbox [41] was used for creating ROIs, building spherical 8 mm radius ROIs around the peak voxels in selected contrasts. Beta values were extracted for analysis.

Behavioural Analysis

The linear mixed model analysis was conducted using R Studio [42] with lme4 package [43] to estimate fixed and random coefficients. Model term selection was guided by the Akaike information criterion (AIC; [44]). The car package [45] was used to obtain t-statistics, with significance of fixed effects determined using Satterthwaite degrees of freedom. The lmerTest package [46] was used to calculate significance, while the emmeans package [47] was used for Tukey’s honestly significant difference (HSD) tests and computed the estimated marginal means.

Behavioural ratings of words plus laughter

Linear mixed model analysis was conducted to investigate how different types of laughter modulated the perceived funniness of words. Laughter Type (spontaneous vs conversational) and Group (Autism vs NA) were included as fixed effects. We also included participants and word items as two crossed random effects. The models were fitted by restricted maximum likelihood (REML), statistical significance was established via Satterthwaite's method. There were a significant main effect of Laughter Type (\(\beta\) = 0.08, t(9496) = 2.94, p = .003), but not of Group (\(\beta\) = 0.22, t(9496) = 0.68, p = .498). This suggests that participants rated words plus spontaneous laughter (M = 3.34, SEM = .164, 95% CI [3.02, 3.67]) as significantly funnier than the words plus conversational laughter (M = 3.27, SEM = .164, 95% CI [2.95, 3.59]). We further included the interaction between Laughter type and Group as a fixed effect to explore whether the laughter modulation effect changes depending on groups. Neither significant interaction effect (\(\beta\) = 0.05, t(9495) = 0.89, p = .546) nor significant main effects (Laugh Type: \(\beta\) = 0.05, t(9495) = 1.48, p = .138; Group:\(\beta\) = 0.20, t(9495) = 0.60, p = .546) were detected from this model. This suggests that autistic and NA participants experienced the laughter modulation effect on the perceived funniness of words similarly (Table 3). Our findings replicate our previous study of the perceived funniness of jokes [29].

Table 3

Fixed and random effects estimated with linear mixed model
		M1			M2
Predictors	Estimates	CI	p	Estimates	CI	p
(Intercept)	3.16	2.71–3.60	< .001	3.17	2.73–3.61	< .001
LaughterType	0.08	0.03–0.13	0.003	0.05	-0.02 – .013	0.138
Group	0.22	-0.42–0.85	0.498	0.20	-0.44–0.83	0.546
LaughterType x Group				0.005	-0.06–0.15	0.372
Random Effects
σ²	1.64			1.64
τ₀₀	0.15 _Words			0.15 _Words
	1.25 _Participants			1.25 _Participants
ICC	0.46			0.46
N	48 _Participants			48 _Participants
	300 _Words			300 _Words
Observations	9502			9502
Marginal R² / Conditional R²	0.004 / 0.464			0.004 / 0.464

Note. p values are given in two tailed.

Neural responses to implicit processing of words plus laughter versus noise-vocoded human vocalizations

We first investigated the neural responses associated with the implicit processing of laughter. This was accomplished by contrasting the activations elicited during the laughter conditions (spontaneous and conversational laughter) with those evoked during the noise-vocoded human vocalizations (NV) condition. Across all participants, whole-brain analyses indicated that hearing words plus both types of laughter versus noise-vocoded human vocalizations was associated with activation in bilateral Heschl’s gyrus (HG), bilateral superior temporal gyrus (STG), bilateral supplementary motor area (SMA), bilateral posterior (PCC), mid (MCC) and anterior cingulate cortex (ACC), bilateral precuneus, left middle temporal gyrus (MTG), and left superior medial frontal gyrus (mPFC). Additionally, widespread responses were detected in bilateral calcarine, cerebellum, lingual, thalamus, insula, vermis, and inferior frontal gyrus, though the peak coordinates were not located in these regions (See Table 4; Fig. 2A).

Table 4

Brain regions showing significant activity in response to implicit processing of laughter versus noised-vocoded human vocalization
Contrast	No. of voxels	Region	Coordinates			T	Z
			x	y	z
All Laugh > NV	30686	Left Heschl's gyrus	-36	-28	14	20.80	> 8
		/Left superior temporal gyrus	-44	-24	8	19.69	> 8
		/Left middle temporal gyrus	-66	-26	4	15.63	> 8
		/Right Heschl's gyrus	48	-14	4	18.29	> 8
		/Right superior temporal gyrus	60	-12	4	18.86	> 8
	2642	Right supplementary motor area	2	8	64	7.07	5.75
		/Right middle cingulate cortex	6	16	38	6.37	5.33
		/Left supplementary motor area	-4	14	54	6.33	5.31
		/Left middle cingulate cortex	-6	14	40	6.01	5.11
		/Left superior medial gyrus	-8	18	40	5.87	5.02
		/Left anterior cingulate cortex	-6	26	22	5.64	4.87
			0	24	30	5.14	4.52
	833	Left precuneus	-10	-48	42	5.96	5.08
		/Left posterior cingulate cortex	-4	-38	30	4.16	3.80
		/Right precuneus	2	-48	38	4.97	4.33
		/Right posterior cingulate cortex	4	-38	28	3.92	3.61
	236	Left middle cingulate cortex	-8	-12	40	4.80	4.28
		/Right middle cingulate cortex	4	-8	36	3.95	3.63
Note. All Laugh = spontaneous and conversational laughter. NV = noise-vocoded human vocalization.

Neural responses to implicit processing of words plus spontaneous and conversational laughter

Next, we investigated the neural responses associated with implicit processing of different types of laughter. This was accomplished by contrasting the activations elicited during each type of ‘word plus laughter’ condition with those during the noise-vocoded human vocalization condition (Spont Laugh > NV; Conver Laugh > NV) across all participants. Whole-brain analyses revealed that the spontaneous laughter versus NV condition showed activation in bilateral superior temporal gyrus (STG) and bilateral Heschl’s gyrus (HG) (See Table 5; Fig. 2B), while the conversational laughter versus NV condition was associated with widespread responses in bilateral superior medial frontal gyri (mPFC), left anterior cingulate cortex (ACC), left medial orbitofrontal cortex (mOFC), right superior frontal gyrus (SFG). (See Table 5; Fig. 2C). No significant group differences (NA > Autism; Autism > NA) were found within these clusters for any of the contrasts in the whole-brain-corrected analysis.

Table 5

Brain regions showing significant activity in response to implicit processing of spontaneous laughter and conversational laughter
Contrast	No. of voxels	Region	Coordinates			T	Z
			x	y	z
Spont Laugh > NV	1030	Left superior temporal gyrus	-58	-14	2	7.50	5.99
			-58	-22	4	7.15	5.79
		/Left Heschl's gyrus	-42	-18	4	5.33	4.66
	1397	Right superior temporal gyrus	64	-10	4	8.15	6.33
			60	-10	2	8.12	6.32
		/Right Heschl's gyrus	34	-26	16	5.21	4.57
Conver Laugh > NV	1812	Left superior temporal gyrus	-56	-12	4	11.25	7.68
		/Left Heschl's gyrus	-34	-26	16	8.84	6.67
	165	Left anterior cingulate cortex	-12	42	4	4.01	3.68
		/Left superior medial gyrus	-4	52	6	3.77	3.49
		/Left medial orbitofrontal cortex	-4	46	-14	3.65	3.39
			-8	46	-10	3.49	3.26
	2064	Right superior temporal gyrus	64	-10	4	10.97	7.57
		/Right Heschl's gyrus	36	-26	16	7.47	5.97
		/Right superior temporal pole	56	10	-14	4.36	3.95
		/Right middle temporal pole	54	10	-24	3.92	3.61
	234	Right superior frontal gyrus	16	26	60	5.06	4.47
			16	12	50	4.72	4.42
	290	Right superior frontal gyrus	18	54	38	4.57	4.11
			16	48	46	4.17	3.81
		/Right superior medial gyrus	6	54	34	3.43	3.21

Note. Spont Laugh = spontaneous laughter. Conver Laugh = conversational laughter. NV = noise-vocoded human vocalization.

We hypothesized there would be a difference in the neural correlates of the implicit processing of different types of laughter within mPFC and sensorimotor network between groups, as suggested by previous findings. To examine this, we employed a region of interest (ROI) approach. The MNI peak coordinates from a prior fMRI study on the perception of conversational laughter vs spontaneous laughter in non-autistic adults [24] were used to define the ROIs: left superior medial frontal gyrus (mPFC; x = -3, y = 54, z = 9); left temporal thalamus (x = -3, y = -6, z = 9); and right anterior cingulate cortex (x = 0, y = 30, z = 30). For each participant, the mean signal across all voxels was then extracted for the spontaneous laughter versus NV contrast and the conversational laughter versus NV contrast from each of the three ROIs. While significant brain activations in the mPFC were identified in the whole-brain results of implicit processing of conversational laughter across all participants, we adopted an independent approach for hypothesis testing. This ensured that the data used to define the ROIs were not related to the hypothesis testing data, in line with the principles of null hypothesis testing [48]. This unbiased approach also maximized our statistical power to detect group differences within our prior hypotheses.

The 3 (ROIs) x 2 (Contrasts) x 2 (Groups) ANOVA showed a significant three-way interaction (ROIs x Contrasts x Groups), F[1.473,64.797] = 5.898, p = .009, \({\eta }_{p}^{2}\) = .118, indicating that the interaction among contrasts and ROIs was different between autistic and non-autistic adults. There were also a significant main effect of ROI, F[2,88] = 5.499, p = .006 < .01, \({\eta }_{p}^{2}\) = .111, indicating that the brain activation in these three ROIs was different. Follow-up tests were conducted on each ROI: No significant main effects or interactions were detected in the analysis of the right anterior cingulate cortex; a significant main effect of Group was observed in the left temporal thalamus, F[1, 44] = 4.681, p = .036, \({\eta }_{p}^{2}\) = .096, indicating greater activation on both perceptual contrasts in non-autistic (M = .173, SEM = .143) than autistic adults (M = − .266, SEM = .143); and a significant two-way interaction (Contrasts x Groups) was detected in the left superior medial gyrus, F[1, 44] = 7.111, p = .011, \({\eta }_{p}^{2}\) = .139. Post hoc analyses revealed significantly greater mentalizing-related BOLD signal change during the implicit processing of conversational laughter vs NV compared to spontaneous laughter vs NV in the non-autistic group, F[1, 22] = 9.720, p = .005, \({\eta }_{p}^{2}\) = .306 (Spontaneous laughter vs NV: M = .162, SD = 1.103; Conversational laughter vs NV: M = 1.048, SD = 1.120). However, there was no significant difference between the two perceptual contrasts in the autistic group, F[1, 22] = .661, p = .425, \({\eta }_{p}^{2}\) = .029 (Spontaneous laughter vs NV: M = .637, SD = 1.709; Conversational laughter vs NV: M = .374, SD = 1.903) (Fig. 3).

To rule out the potential confounding effect of the NV condition on the observed mentalizing-related BOLD signal change in mPFC (x = -3, y = 54, z = 9), we further compared the direct contrasts of implicit processing of conversational laughter versus spontaneous laughter between the two groups. ROI analyses using small-volume family-wise error correction (SVC FWE) revealed that non-autistic adults exhibited greater BOLD signal change in the mPFC (MNI coordinates for peak voxel: x = 2, y = 50, z = 10; t(1, 44) = 3.42, z = 3.20, p = .030, SVC FWE; cluster size = 226 voxels). Together, these findings suggest that we replicated previous results on the engagement of the mPFC during the perception of conversational laughter in non-autistic adults [24]. By using an implicit measure and comparing neural differences between autistic and non-autistic groups, we provided further evidence for the significant role of the mPFC in autistic differences in interpreting different types of laughter but not all types of laughter.

Exploratory Time-course Analysis

We additionally conducted an exploratory analysis to investigate group differences in the hemodynamic response of brain areas during implicit processing of the words plus laughter conditions and the NV condition, to unpack humour and laughter; we did not have strong predictions about neural differences in the autistic group. Four regions were selected based on a prior interest from the whole-brain results: these correspond to regions associated with hearing sounds and processing speech and emotional vocalisations (HG and the STG respectively; [24, 49]); responses to heard laughter (SMA; [16, 24]) and responses to humour (precuneus; [50–52]). The finite impulse response (FIR) event-related time courses were extracted, with an FIR length of 30 sec and time bins of 3 seconds, across all trials of interest. The graphs in Fig. 4 illustrate the greater sensitivity in BOLD signal changes within the SMA and bilateral precuneus areas (Fig. 4), but not in HG and STG (Figure S1), in the non-autistic group during the laughter compared to the NV condition. This contrasted with the autistic group, who demonstrated similar BOLD signal changes in response to both conditions.

In comparing the groups within these regions, we adopted a method similar to that used by White et al. [28]. A significant difference was observed only in the SMA area (MNI peak coordinate: x = 2, y = 8, z = 64; 8mm sphere) between non-autistic (M = .982, SD = 1.175) and autistic (M = 2.287, SD = 2.099) adults during implicit processing of words plus laughter versus words plus the noise-vocoded human vocalization, t(34.566) = -2.601, p = .014, d = − .767.

This is the first study to use fMRI to investigate the neural mechanisms behind the implicit processing of different types of laughter in autistic and non-autistic adults. Despite no behavioural differences between the groups, we found that non-autistic adults, but not autistic adults, exhibited increased activation in the mPFC when listening to words plus conversational than spontaneous laughter. Conversely, autistic adults showed greater activation in SMA than non-autistic adults when listening to words paired with either type of laughter. These findings suggest a critical role for the mPFC and sensorimotor network in the implicit processing of laughter, respectively engaging mentalizing to understand socially ambiguous laughter and eliciting laughter contagion. Our findings suggest autistic people exhibit specific similarities and differences compared to non-autistic people in the implicitly processing laughter. In particular, autistic people experience difficulties in understanding conversational laughter with sophisticated meaning and social ambiguity. The challenge autistic people encounter in non-verbal communication therefore seems to not be in processing all social signals, but rather in processing specific types of social signals during communication.

In the whole-brain activation comparing words plus laughter to NV across both autistic and non-autistic groups, we observed neural responses in several regions of the auditory cortex, including bilateral HG and bilateral STG. Additionally, responses were noted in the bilateral PCC, primary visual cortex, precuneus, SMA and mPFC. The recruitment of the dorsolateral temporal lobe fields is consistent with a greater loading on the processing of vocalisations in the words plus laughter conditions [53, 54]. Our design cannot expand upon this, but it is also possible that there is greater processing in auditory areas than would be seen to spoken words or laughter alone: the activation in this contrast is extensive in primary and secondary auditory cortex, despite the high-level baseline containing spoken words and noise-vocoded sounds. It is important to note that the design of our current study differs from previous laughter research [16, 24]: instead of having participants passively listen to laughter stimuli, we used an implicit measure of laughter processing and presented a cover story about people’s responses to a comedian saying funny words, which may have enhanced speech-related processing in auditory fields.

The sensorimotor network, which includes the SMA and ACC as detected in the whole-brain results, has been consistently identified during laughter perception [15, 16, 24]. This can be attributed to its role as part of a simulation mechanism in facilitating social understanding [24], contributing to emotional contagion [16], and its involvement in auditory processing and auditory imagery [55]. Laughter carries greater socio-emotional weight and is more contagious; therefore increased activity in the sensorimotor network was observed when participants heard words plus laughter compared to NV. Through ROI analyses, autistic adults displayed greater activation in the SMA compared to non-autistic adults, indicating that laughter might be differently contagious for autistic people. This observation seems to align with existing findings highlighting differences in the contagiousness of laughter among autistic people: autistic children are less likely to join in others’ laughter and perceive cartoons with a laughter track as less enjoyable than their peers [19, 22].

Similar to the involvement of sensorimotor network, the mPFC was consistently observed in the whole-brain results of words plus laughter versus NV, as well as specifically for conversational but not spontaneous laughter. This finding aligns with previous evidence showing greater mPFC activation while listening to laughter with complex social meaning [30, 56, 57]. In the subsequent ROI results, the non-autistic group showed significantly greater activation in the mPFC during implicit processing of conversational laughter compared to spontaneous laughter. In contrast, no such differential activation was observed in autistic adults. Our finding replicates and extends previous studies on the engagement of the mPFC in processing conversational laughter by including an autistic group [24, 25]. By using an implicit measure, our study allows us to go a step further: it is possible that the observed mPFC activation reflects the socio-emotional ambiguity of hearing conversational laughter together with funny words, suggesting that only non-autistic adults attempt to resolve the reason for the laughter, specifically, by interpreting the meaning and intention behind it.

Further, the lack of mPFC activation in autistic adults during implicit processing of conversational laughter is consistent with Sumiya et al. [30]’s study, which showed lower mPFC activity in autistic adults compared to non-autistic adults to jokes followed by laughter. Although they did not define their laughter as being spontaneous or conversational, it seems likely given our findings that they used the latter. Our results are also consistent with previous findings of atypical function within this mentalizing-related region in autism [26–28]. This group difference in autistic adults might reflect their ‘capacity limits in mentalizing’ as proposed by [28], given that we used an implicit measure to investigate laughter processing, preventing them from using compensatory strategies. Though behaviourally, autistic adults were still able to implicitly differentiate between spontaneous and conversational laughter, this might be due to the inherent acoustic differences between these two types of laughter [11, 12, 24]. However, it seems likely they were processing laughter without fully understanding of the significance of these differences, leading to a subsequent reduction in mentalizing to comprehend the meaning. Hence, autistic people may struggle to identify the meaning and intentions behind laughter in everyday contexts, which could lead to misunderstandings in social situations, leaving them socially vulnerable [58].

In addition to laughter perception-related areas such as the mPFC and sensorimotor network, we also report significant results associated with the processing of words plus laughter in auditory fields associated with the perception of non-verbal emotional vocalisations and speech, and more notably in the precuneus, posterior bilateral cingulate cortex and primary visual fields. The precuneus has been associated with studies both of mentalizing [59, 60] and of humour [50–52], and its involvement in laughter modulating how funny words seem suggests that it may be an important component in the processes underlying this modulation. The laughter may provoke a further rumination on the meaning of the word, its possible humorous associations, and perhaps the intentions of the person laughing. These reflective processes may relate to the role of the precuneus in metacognitive processes [61]. The involvement of primary visual cortex in the absence of visual stimulation implies the recruitment of visual imagery to support these processes. Posterior bilateral cingulate cortex has been implicated in studies of autobiographical memory retrieval and emotional salience and is a key node in the default mode network [62]; its involvement here may indicate both emotional and personal kinds of information being recruited when processing the potentially humorous implications of a word.

Limitations

The present study has two main limitations. Behaviourally, due to the testing restrictions imposed by COVID-19, we were unable to collect baseline funniness ratings of words from both autistic and non-autistic adults. Though we identified a significant main effect for the Type of Words, it is important to note that this effect is predominantly driven by ratings from non-autistic adults. By establishing a matched baseline funniness rating for both groups, we could draw a more robust conclusion about the similarity of laughter modulation effects between autistic and non-autistic adults. Neurally, though we used laughter to modulate the perceived funniness of funny words in the current study, our design does not distinguish between the processing of laughter and humour. Hence, we further explored group differences by performing a time course analysis to determine if any time-related effects existed, considering the temporal dependency in our stimuli (laughter followed the words). The graphs indicate that, where there are differences between the words plus laughter and NV conditions over time, these occur later in the processing of the stimuli in SMA and precuneus, especially for non-autistic participants. This is consistent with the later integration of the laughter with the verbal material, however, future studies with finer temporal resolution, such as through Magnetoencephalography (MEG), will better unpack this integration and its overall effects on perceived funniness.

In sum, autistic people face challenges in processing specific types of social cue during communication, but not in processing all social cues. Our study is the first to use fMRI to probe the neural mechanisms underlying different types of laughter during implicit processing (i.e., how it makes funny words funnier), and further, to compare this in autistic and non-autistic adults. Despite no behavioural differences between the groups, we found that non-autistic adults, unlike their autistic peers, exhibited increased mPFC activation while listening to words plus conversational laughter, whereas autistic adults showed enhanced SMA activation compared to non-autistic adults when listening to words plus any type of laughter. Our results highlight the mPFC and sensorimotor network's crucial roles in understanding socio-emotionally ambiguous laughter via mentalizing, and in evoking laughter contagion respectively. Overall, our findings suggest that autistic people experience difficulties in processing conversational laughter with sophisticated meaning and social ambiguity, which may lead to social vulnerability. These results highlight the challenges that autistic people face in understanding the essence of the laughter we perpetually encounter in everyday life. Therefore, we advocate for clearer communication with autistic people.

Acknowledgements

We thank all participants who took part in this study and the people who assisted with recruitment during COVID-19. We also thank the BUCNI team for their assistance with the fMRI scans. We would like to special thank Manying Chiu and Hannah Partington for their assistance with editing word stimuli, as well as Alexis MacIntyre and Paul MacIntyre for their help with word annotation.

Author contributions

C.Q.C., N.L, S.J.W., and S.K.S designed research; C.Q.C., C.Z.X.W., and O-C.O. performed research; S.H.Y.C created the laughter stimuli; C.Q.C., N.L, and S.J.G analysed data; C.Q.C., S.J.W., and S.K.S wrote the paper; C.Q.C., N.L, S.J.W., S.J.G and S.K.S edited the paper.

Funding

S.J.W. was supported by The Royal Society DH150167; The Royal Society RF\ERE\231054; The Royal Society RF\ERE\210122; The Academy of Medical Sciences: SBF003\1169.

Availability of data and materials

The raw neuroimaging data are not publicly available due to not all participants gave permission. Anonymized behavioural data and beta values extracted from the neuroimaging data is available at: https://osf.io/su5kw/?view_only=dad72c09078d4d2cbb3eb5d245e491be

Ethics approval and consent to participate

This study was approved by the UCL ethic committee.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests

Mundy P, Sigman M, Ungerer J, Sherman T. DEFINING THE SOCIAL DEFICITS OF AUTISM: THE CONTRIBUTION OF NON-VERBAL COMMUNICATION MEASURES. J Child Psychol Psychiatry. 1986;27:657–69.
Senju A, Southgate V, White S, Frith U. Mindblind Eyes: An Absence of Spontaneous Theory of Mind in Asperger Syndrome. Science. 2009;325:883–5.
Trevisan DA, Hoskyn M, Birmingham E. Facial Expression Production in Autism: A Meta-Analysis. Autism Res. 2018;11:1586–601.
Provine RR. Laughing, Tickling, and the Evolution of Speech and Self. Curr Dir Psychol Sci. 2004;13:215–8.
Gervais M, Wilson DS. THE EVOLUTION AND FUNCTIONS OF LAUGHTER AND HUMOR: A SYNTHETIC APPROACH. Q Rev Biology. 2005;80:395–430.
Vettin J, Todt D. Laughter in Conversation: Features of Occurrence and Acoustic Structure. J Nonverbal Behav. 2004;28:93–115.
Provine RR. Laughter Punctuates Speech: Linguistic, Social and Gender Contexts of Laughter. Ethology. 1993;95:291–8.
Neves L, Cordeiro C, Scott SK, Castro SL, Lima CF. High emotional contagion and empathy are associated with enhanced detection of emotional authenticity in laughter. Q J Exp Psychol. 2017;71:2355–63.
Wild B, Rodden FA, Grodd W, Ruch W. Neural correlates of laughter and humour. Brain. 2003;126:2121–38.
Gerbella M, Pinardi C, Cesare GD, Rizzolatti G, Caruana F. Two Neural Networks for Laughter: A Tractography Study. Cereb Cortex. 2020;31:899–916.
Lavan N, Scott SK, McGettigan C. Laugh Like You Mean It: Authenticity Modulates Acoustic, Physiological and Perceptual Properties of Laughter. J Nonverbal Behav. 2016;40:133–49.
Scott SK, Lavan N, Chen S, McGettigan C. The social life of laughter. Trends Cogn Sci. 2014;18:618–20.
Dunbar R, Baron R, Frangou A, Pearce E, Leeuwen EJC van, Stow J, et al. Social laughter is correlated with an elevated pain threshold. Proc Royal Soc B Biological Sci. 2012;279:1161–7.
Manninen S, Tuominen L, Dunbar RI, Karjalainen T, Hirvonen J, Arponen E, et al. Social Laughter Triggers Endogenous Opioid Release in Humans. J Neurosci. 2017;37:6125–31.
Warren JE, Sauter DA, Eisner F, Wiland J, Dresner MA, Wise RJS, et al. Positive Emotions Preferentially Engage an Auditory-Motor “Mirror” System. J Neurosci. 2006;26:13067–75.
O’Nions E, Lima CF, Scott SK, Roberts R, McCrory EJ, Viding E. Reduced Laughter Contagion in Boys at Risk for Psychopathy. Curr Biol. 2017;27:3049–3055.e4.
Provine RR. Contagious laughter: Laughter is a sufficient stimulus for laughs and smiles. B Psychonomic Soc. 1992;30:1–4.
Scott SK, Cai CQ, Billing A. Robert Provine: the critical human importance of laughter, connections and contagion. Philosophical Transactions Royal Soc B. 2022;377:20210178.
Helt MS, Fein DA, Vargas JE. Emotional contagion in children with autism spectrum disorder varies with stimulus familiarity and task instructions. Dev Psychopathol. 2020;32:383–93.
Helt MS, Fein DA. Facial Feedback and Social Input: Effects on Laughter and Enjoyment in Children with Autism Spectrum Disorders. J Autism Dev Disord. 2016;46:83–94.
Hudenko WJ, Stone W, Bachorowski J-A. Laughter Differs in Children with Autism: An Acoustic Analysis of Laughs Produced by Children With and Without the Disorder. J Autism Dev Disord. 2009;39:1392–400.
Reddy V, Williams E, Vaughan A. Sharing humour and laughter in autism and Down’s syndrome. Brit J Psychol. 2002;93:219–42.
Wu C-L, An C-P, Tseng L-P, Chen H-C, Chan Y-C, Cho S-L, et al. Fear of being laughed at with relation to parent attachment in individuals with autism. Res Autism Spect Dis. 2015;10:116–23.
McGettigan C, Walsh E, Jessop R, Agnew ZK, Sauter DA, Warren JE, et al. Individual Differences in Laughter Perception Reveal Roles for Mentalizing and Sensorimotor Systems in the Evaluation of Emotional Authenticity. Cereb Cortex New York Ny. 2015;25:246–57.
Lavan N, Rankin G, Lorking N, Scott S, McGettigan C. Neural correlates of the affective properties of spontaneous and volitional laughter types. Neuropsychologia. 2017;95:30–9.
Frith CD, Frith U. The Neural Basis of Mentalizing. Neuron. 2006;50:531–4.
Gilbert SJ, Meuwese JDI, Towgood KJ, Frith CD, Burgess PW. Abnormal functional specialization within medial prefrontal cortex in high-functioning autism: a multi-voxel similarity analysis. Brain. 2009;132:869–78.
White SJ, Frith U, Rellecke J, Al-Noor Z, Gilbert SJ. Autistic adolescents show atypical activation of the brain′s mentalizing system even without a prior history of mentalizing problems. Neuropsychologia. 2014;56:17–25.
Cai Q, Chen S, White SJ, Scott SK. Modulation of humor ratings of bad jokes by other people’s laughter. Curr Biol. 2019;29:R677–8.
Sumiya M, Okamoto Y, Koike T, Tanigawa T, Okazawa H, Kosaka H, et al. Attenuated activation of the anterior rostral medial prefrontal cortex on self-relevant social reward processing in individuals with autism spectrum disorder. Neuroimage Clin. 2020;26:102249.
Wechsler D. Wechsler Adult Intelligence Scale–Fourth Edition (WAIS-IV). 2008.
Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E. The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Malesand Females, Scientists and Mathematicians. J Autism Dev Disord. 2001;31:5–17.
Hus V, Lord C. The Autism Diagnostic Observation Schedule, Module 4: Revised Algorithm and Standardized Severity Scores. J Autism Dev Disord. 2014;44:1996–2012.
Engelthaler T, Hills TT. Humor norms for 4,997 English words. Behav Res Methods. 2018;50:1116–24.
Boersma P, Weenink D. Praat: doing phonetics by computer [Internet]. 2014. Available from: http://www.praat.org/
Anwyl-Irvine AL, Massonnié J, Flitton A, Kirkham N, Evershed JK. Gorilla in our midst: An online behavioral experiment builder. Behav Res Methods. 2020;52:388–407.
Inc. TM. MATLAB [Internet]. Natick, Massachusetts, United States: The MathWorks Inc.; 2018. Available from: https://www.mathworks.com
Brainard DH. The Psychophysics Toolbox. Spatial Vision. 1997;10:433–6.
Penny WD, Friston KJ, Ashburner JT, Kiebel SJ, Nichols TE. Statistical parametric mapping: the analysis of functional brain images. Elsevier; 2011.
Poline J-B, Worsley KJ, Evans AC, Friston KJ. Combining Spatial Extent and Peak Intensity to Test for Activations in Functional Imaging. NeuroImage. 1997;5:83–96.
Brett M, Anton J-L, Valabregue R, Poline J-B. Region of interest analysis using an SPM toolbox. Sendai, Japan: Presented at the 8th International Conference on Functional Mapping of the Human Brain; 2002.
RStudio. RStudio: Integrated Development for R. [Internet]. RStudio, PBC, Boston, MA; 2020. Available from: http://www.rstudio.com/
Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed effects models using lme4. 2014.
Sakamoto, Ishiguro, Kitagawa G. Akaike information criterion statistics. The Netherlands: In D. Reidel (Ed.); 1986.
Fox, Weisberg, Adler, Bates, Baud-Bovy G, Ellison, et al. Package ‘car’. R Foundation for Statistical Computing.; 2012.
Kuznetsova A, Brockhoff PB, Christensen RHB. lmerTest Package: Tests in Linear Mixed Effects Models. Journal of Statistical Software [Internet]. 2017;82:1–26. Available from: https://www.jstatsoft.org/index.php/jss/article/view/v082i13
Lenth R, Signmann H, Love J, Buerkner P, Herve M. Emmeans: Estimated marginal means, aka least-squares means. 2018.
Kriegeskorte N, Simmons WK, Bellgowan PSF, Baker CI. Circular analysis in systems neuroscience: the dangers of double dipping. Nat Neurosci. 2009;12:535–40.
Mummery CJ, Patterson K, Wise RJ, Vandenberghe R, Vandenbergh R, Price CJ, et al. Disrupted temporal lobe connections in semantic dementia. Brain. 1999;122:61–73.
Li B, Li X, Pan Y, Qiu J, Zhang D. The relationship between self-enhancing humor and precuneus volume in young healthy individuals with high and low cognitive empathy. Sci Rep. 2018;8:3467.
Chan Y, Zeitlen DC, Beaty RE. Amygdala-frontoparietal effective connectivity in creativity and humor processing. Hum Brain Mapp. 2023;44:2585–606.
Brawer J, Amir O. Mapping the “Funny Bone”: Neuroanatomical Correlates of Humor Creativity in Professional Comedians. Soc Cogn Affect Neurosci. 2021;16:nsab049-.
Fecteau S, Belin P, Joanette Y, Armony JL. Amygdala responses to nonlinguistic emotional vocalizations. Neuroimage. 2007;36:480–7.
Sander D, Grandjean D, Pourtois G, Schwartz S, Seghier ML, Scherer KR, et al. Emotion and attention interactions in social cognition: Brain regions involved in processing anger prosody. Neuroimage. 2005;28:848–58.
Lima CF, Krishnan S, Scott SK. Roles of Supplementary Motor Areas in Auditory Processing and Auditory Imagery. Trends Neurosci. 2016;39:527–42.
Szameitat DP, Kreifelts B, Alter K, Szameitat AJ, Sterr A, Grodd W, et al. It is not always tickling: Distinct cerebral responses during perception of different laughter types. Neuroimage. 2010;53:1264–71.
Sumiya M, Koike T, Okazaki S, Kitada R, Sadato N. Brain networks of social action-outcome contingency: The role of the ventral striatum in integrating signals from the sensory cortex and medial prefrontal cortex. Neurosci Res. 2017;123:43–54.
Trundle G, Jones KA, Ropar D, Egan V. Prevalence of Victimisation in Autistic Individuals: A Systematic Review and Meta-Analysis. Trauma, Violence, Abus. 2023;24:2282–96.
Arioli M, Cattaneo Z, Ricciardi E, Canessa N. Overlapping and specific neural correlates for empathizing, affective mentalizing, and cognitive mentalizing: A coordinate-based meta‐analytic study. Hum Brain Mapp. 2021;42:4777–804.
Takahashi HK, Kitada R, Sasaki AT, Kawamichi H, Okazaki S, Kochiyama T, et al. Brain networks of affective mentalizing revealed by the tear effect: The integrative role of the medial prefrontal cortex and precuneus. Neurosci Res. 2015;101:32–43.
Ye Q, Zou F, Lau H, Hu Y, Kwok SC. Causal Evidence for Mnemonic Metacognition in Human Precuneus. J Neurosci. 2018;38:6379–87.
Leech R, Braga R, Sharp DJ. Echoes of the Brain within the Posterior Cingulate Cortex. J Neurosci. 2012;32:215–22.

No competing interests reported.

SILaughfMRI.docx

Download PDF

Journal Publication

published 01 May, 2024

Read the published version in Cerebral Cortex →

Version 1

posted

You are reading this latest preprint version

Mapping the differential impact of spontaneous and conversational laughter on brain and mind: A fMRI study in autism

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Methods

Participants

Experimental Design and Procedure

Neuroimaging, Pre-processing and Analysis

Behavioural Analysis

Results

Behavioural ratings of words plus laughter

Neural responses to implicit processing of words plus laughter versus noise-vocoded human vocalizations

Neural responses to implicit processing of words plus spontaneous and conversational laughter

Exploratory Time-course Analysis

Discussion

Limitations

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1