Laughter in everyday life: an event-based experience sampling method study using wrist-worn wearables

Laughter, a universal nonverbal expression, is significant in human communication and social interaction. However, research on laughter frequency and its relationship to personality and well-being is limited. Previous studies, primarily using paper-and-pencil diaries or daily laughter records (DLR), have methodological limitations such as small sample sizes, short data collection periods, and reliance on retrospective self-reporting. These limitations may lead to inaccurate and incomplete data on laughter frequency, particularly given the fleeting nature of laughter and challenges of recalling instances accurately. The current study uses event-contingent experience sampling methodology (ESM) with a novel approach using wrist-worn wearables and a physical analogue scale (PAS) to overcome these methodological issues. The wearables allow for real-time recording of laughter events, improving accuracy and reducing recall bias. The PAS, a sensor-based scale, measures happiness levels associated with laughter events. This study aimed to address the limitations of prior research by employing this improved methodology, and therefore to answer two research questions. First, is happiness higher when laughing compared to when not laughing? Second, are personality traits (Big Five), well-being (happiness, life satisfaction), cheerfulness, and gelotophobia associated with laughter frequency?

Existing research on laughter in everyday life has primarily relied on paper-and-pencil diaries or daily laughter records (DLRs). Early studies, such as those by Mannell and McMahon (1982) and Kuiper and Martin (1998), used ESM designs but did not specifically focus on laughter incidences. Martin and Kuiper (1999) measured laughter frequency using paper-and-pencil diaries, reporting an average of 18 laughs per day. Vettin and Todt (2004) used tape recordings to analyze laughter in natural conversations, finding a higher frequency than reported in self-report studies, potentially due to participants not noticing every instance of laughter or difficulties in recording events promptly with paper diaries. Valeri (2006) also found similar patterns of laughter frequency. However, all these studies used relatively small timespans and paper diaries which limit their generalizability due to the potential of recall bias, motivational factors (participants might not always record each event), and procedural issues (forgetting to record the event). This study overcomes these shortcomings by employing wearables and PAS, leading to a more accurate and complete measurement of laughter and its associations with various psychological constructs.

This study used an event-contingent ESM design with 52 participants who wore wrist-worn wearables for four weeks. The wearables were programmed to record laughter events (belly laughs—single button press; fits of laughter—double button press) and happiness ratings (using a PAS based on forearm angle) concurrently. Participants were also prompted at three random times daily to provide baseline happiness ratings. After the four-week period, participants completed an online questionnaire assessing demographics, Big Five personality traits (BFI), life satisfaction (SWLS), gelotophobia (Gelotophobia Scale), and cheerfulness (STCI). The PAS, calibrated from 0° (not happy) to 90° (very happy), provided a continuous measure of happiness. Multi-level modeling and recurrent event regression (Cox-type proportional rate model) were used to analyze the data. A total of 9,261 assessments were collected, with 779 cases (8.4%) having angle measurements below 0° due to participants' imperfect leveling of their forearm; these values were set to 0 for a winsorized measure. The study was conducted in accordance with the Declaration of Helsinki and relevant ethical guidelines, obtaining informed consent from all participants.

Participants reported an average of 155 belly laughs and 16 fits of laughter over the four weeks, representing approximately 2.5 belly laughs per day and one fit of laughter every four days. Happiness ratings were significantly higher during laughter events compared to baseline measurements (belly laugh: 8.6 points higher; fit of laughter: 12.9 points higher). Recurrent event regression revealed significant positive associations between laughter frequency and happiness (p < 0.001), openness (p < 0.001), and conscientiousness (p = 0.009). Seriousness was negatively associated with laughter frequency (p = 0.008), while bad mood (p = 0.017) and life satisfaction (p = 0.014) showed positive associations. Women laughed more frequently than men (p < 0.001), with a significant gender-by-age interaction showing women's laughter frequency increasing with age, while men's decreased. No significant association was found between laughter frequency and gelotophobia. The results replicated previous findings on the diurnal pattern of laughter (more laughter in the evening) and day-of-week effect (more laughter on weekends).

This study's findings support the validity of using wrist-worn wearables and a PAS to assess laughter in everyday life. The results replicate previous findings on the relationship between laughter and happiness, while also extending existing research by investigating the association of laughter frequency with various personality traits and aspects of well-being. The positive association between laughter and openness and conscientiousness suggests that individuals who are more open to experience and responsible are more likely to experience frequent laughter in their daily lives. The unexpected positive association between bad mood and laughter frequency requires further investigation, but may suggest the use of laughter as a coping mechanism or social interaction tool even when feeling down. The lack of significant association with gelotophobia and cheerfulness, although descriptively negative and positive respectively, warrants additional studies with larger samples and exploration of contextual factors influencing laughter behaviour. The significant gender differences, particularly the interaction between gender and age, highlight the need to consider gender in future studies of laughter.

This study successfully employed a novel measurement approach using wrist-worn wearables and a PAS to investigate laughter frequency and its correlates. The results replicate previous findings while offering new insights into the relationship between laughter and personality, well-being, and demographic factors. Future research should focus on exploring the role of contextual factors, using more sophisticated wearables capable of recording richer data (e.g., social context of laughter) and on addressing the limitations of this study such as the sample size and potential influence of the COVID-19 pandemic.

Several limitations should be acknowledged. The use of one-button wearables limited the amount of contextual information collected. The study focused on only two types of laughter, potentially missing subtler forms. Although the sample was community-based, there was an overrepresentation of students. The data collection period overlapped with the COVID-19 pandemic, which might have affected participants' behaviors. Finally, some analyses, particularly those involving interactions, might have been underpowered due to sample size.