HUMOR, LAUGHTER, AND RECOVERY FROM STRESSFUL...

HUMOR, LAUGHTER, AND RECOVERY FROM STRESSFUL EXPERIENCES

A Dissertation Submitted to the Graduate Faculty

of theNorth Dakota State University

of Agriculture and Applied Science

By

Amanda Jean Dillard

In Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Major Department: Psychology

August 2006

Fargo, North Dakota

R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.

UMI Number: 3265559

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy

submitted. Broken or indistinct print, colored or poor quality illustrations and

photographs, print bleed-through, substandard margins, and improper

alignment can adversely affect reproduction.

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted. Also, if unauthorized

copyright material had to be removed, a note will indicate the deletion.

®

UMIUMI Microform 3265559

Copyright 2007 by ProQuest Information and Learning Company.

All rights reserved. This microform edition is protected against

unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road

P.O. Box 1346 Ann Arbor, Ml 48106-1346


North Dakota State UniversityGraduate School

Title

Humor, Laughter, and Recovery from Stressful Experiences

B y

Amanda Jean Dillard

The Supervisory Committee certifies that this disquisition complies with North Dakota State University’s regulations and meets the accepted standards for the degree of

DOCTOR OF PHILOSOPHY

SUPERVISORY COMMITTEE:

V C h a ir ~0

f -

" 7"

Approved by Department Chair:

Signature


ABSTRACT

Dillard, Amanda Jean, Ph.D., Department of Psychology, College of Science and Mathematics, North Dakota State University, August 2006. Humor, Laughter, and Recovery from Stressful Experiences. Major Professor: Dr. Kevin D. McCaul.

In three studies, we investigated the effects of humor and laughter on mood and

physiological responses in stressful situations in the laboratory. We tested whether humor

and laughter would be most effective in helping individuals cope before, after, or during a

stressor. Study 1 explored the effects of contagious laughter after participants were

exposed to a laboratory stressor. Study 2 explored the effects of humorous, one-minute

clips before participants were exposed to a stressor, and Study 3 explored the effects of

these clips during a stressor. Results showed that humor and laughter positively influenced

mood in response to a stressor and shortly after the stressor. However, this was only true

when humor occurred before or during a stressor, not after. In addition, all three studies

indicated that laughter was arousing relative to a neutral state. However, in Study 3, humor

had a positive physiological influence. Together, the findings suggest that humor and

laughter before and during a stressor can positively influence mood, and humor during a

stressor can positively influence physiology.


ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Kevin D. McCaul, for all of his support,

encouragement and guidance with this project. Dr. McCaul has been a wonderful mentor,

and I feel lucky to have had him as my advisor. I would also like to thank my committee

members, James Council, Clayton Hilmert, Michael Robinson, and Bryan Christenson, for

their effort and helpful suggestions. In addition, I would like to thank Dr. Robinson for

allowing me to use his laboratory space and equipment, and Dr. Hilmert for encouraging

me to learn about physiological measures. Finally, I would like to express my thanks to

Jeremy Brodersen, who spent many hours creating the mat lab code for this project, Ben

Wilkowski who helped with programming the experiments, Leah Meredith who helped run

participants, create figures, and proof read, and Cali Anicha who helped run participants.


V

TABLE OF CONTENTS

ABSTRACT................................................................................................................................ iii

ACKNOWLEDGMENTS.............................................................................................. iv

LIST OF TABLES...................................................................................................................... ix

LIST OF FIGURES..................................................................................................................... x

INTRODUCTION........................................................................................................................1

Humor.............................................................................................................................. 2

Humor as a Positive Influence....................................................................................... 2

Humor as a Stress-Buffer................................................................................................4

Laughter..................................................... 10

STUDY 1....................................................................................................................... 15

METHOD................................................................................................................................... 15

Participants.................................................................................................................... 15

Design............................................................................................................................. 15

Procedure........................................................................................................................15

Task................................................................................................................................. 16

Questionnaire Measures................................................................................................ 17

Dependent Measures..................................................................................................... 18

Hypotheses..................................................................................................................... 20

RESULTS...................................................................................................................................22

Manipulation Check...................................................................................................... 22

Analyses......................................................................................................................... 22

Moderating Analyses.................................................................................................... 32


DISCUSSION............................................................................................................................ 34

STUDY 2 ....................................................................................................................... 38

METHOD...................................................................................................................................40

Participants....................................................................................................................40

Design................................................................................... 40

Procedure....................................................................................................................... 40

Task................................................................................................................................ 41

Questionnaire Measures................................................................................................42

Dependent Measures.....................................................................................................42

Hypotheses.....................................................................................................................42

RESULTS...................................................................................................................................44

Manipulation Check......................................................................................................44

Analyses......................................................................................................................... 45

Moderating Analyses....................................................................................................52

DISCUSSION............................................................................................................................ 55

STUDY 3....................................................................................................................... 61

METHOD...................................................................................................................................62

Participants.....................................................................................................................62

Design....................................................................................................................... ....62

Procedure....................................................................................................................... 62

Task.................................................................................................................................63

Questionnaire Measures................................................................................................64

Dependent Measures.....................................................................................................64


Hypotheses..................................................................................................................... 64

RESULTS...................................................................................................................................66

Manipulation Check...................................................................................................... 66

Analyses......................................................................................................................... 67

Moderating Analyses.................................................................................................... 76

DISCUSSION............................................................................................................................ 78

GENERAL DISCUSSION....................................................................................................... 82

Limitations..................................................................................................................... 89

Future Directions.......................................................................................................... 91

Conclusion.....................................................................................................................92

REFERENCES.......................................................................................................................... 93

APPENDIX A. EXPERIMENTER SCRIPTS....................................................................... 100

Study 1..........................................................................................................................100

Study 2 ..........................................................................................................................101

Study 3..........................................................................................................................103

APPENDIX B. CONSENT FORM S...............................................................................107

Studies 1 and 2 ............................................................................................................107

Study 3 ..........................................................................................................................108

APPENDIX C. DEBRIEFING FORMS.............................................................................I l l

Studies 1 and 2 ............................................................................................................I l l

Study 3 ..........................................................................................................................112


v ii i

APPENDIX D. HUMOR STYLES QUESTIONNAIRE....................................................113

APPENDIX E. COPING HUMOR SCALE......................................................................... 117

APPENDIX F. ANXIETY..................................................................................................... 118

APPENDIX G. NEUROTICISM..........................................................................................119

APPENDIX H. LIFE ORIENTATIONS TEST-REVISED................................................ 120

APPENDIX I. ADDITIONAL ANALYSES FOR STUDY 1.............................................121

Analyses........................................................................................................................121

APPENDIX J. ADDITONAL ANALYSES FOR STUDY 2..............................................124

Analyses........................................................................................................................124


LIST OF TABLES

Table Page

1. Example of Trial 1 from Study 1....................................................................................... 16

2. Example of Trial 1 from Study 2 .......................................................................................41

3. Timeline of Study 3 ............................................................................................................. 63


X

LIST OF FIGURES

Figure Page

1. Study 1 zygomatic activity by condition collapsed across trials...................................... 24

2. Study 1 corrugator activity by condition collapsed across tria ls..................................... 25

3. Study 1 Noise x Laughter interaction for corrugator activity during Interval 2 ..............26

4. Study 1 SC activity by condition collapsed across trials...................................................27

5. Study 1 Noise x Laughter interaction for SC activity during Interval 2 ...........................28

6. Study 1 SC activity without the confounding variable.....................................................29

7. Study 1 peak mood reports by condition collapsed across trials...................................... 30

8. Study 1 Noise x Laughter interaction for peak mood during Interval 2 ...........................31

9. Study 1 Noise x Laughter interaction for peak mood during Recovery...........................32

10. Study 2 zygomatic activity by condition collapsed across trials....................................46

11. Study 2 corrugator activity by condition collapsed across tria ls...................................47

12. Study 2 SC activity by condition collapsed across trials.................................................48

13. Study 2 peak mood reports by condition collapsed across trials.................................... 50

14. Study 2 mood change by condition collapsed across trials..............................................53

15. Study 3 zygomatic activity for the funny and neutral conditions...................................67

16. Study 3 corrugator activity for the funny and neutral conditions...................................69

17. Study 3 average SC activity for the funny and neutral conditions.................................71

18. Study 3 average mood reports for the funny and neutral conditions...............................73

19. Study 3 HR for the funny and neutral conditions............................................................ 75

20. Study 2 SC change by condition collapsed across trials................................................125


INTRODUCTION

Most research on humor and subjective well-being, although correlational, suggests

that humor is good for one’s health. Sense of humor as a trait, for example, correlates

positively with self-esteem, extraversion, cheerfulness, positive emotions, and intimacy

(Hampes, 1994; Ruch, 1994; Thorson, Powell, Sarmany-Schuller, &Hampes, 1997).

Sense of humor is negatively associated with depression, pessimism, neuroticism, anxiety

about death, and bad moods (Kohler & Ruch, 1996; Ruch, 1994; Thorson & Powell, 1993,

1994; Thorson et al., 1997).

In investigations of how humor influences subjective well-being, researchers have

often focused on humor as a positive influence on responses to stress. Dixon (1980)

believed that humor could protect people from the stress of negative events. Indeed,

researchers have found that individuals who score high on sense of humor trait measures

are less likely to experience negative moods in response to life’s stressors (Labott &

Martin, 1987; Kuiper, Martin, & Olinger, 1993; Martin & Lefcourt, 1983; Newman &

Stone, 1996; Nezu, Nezu, & Blisset, 1988; Perlini, Nenonen, & Lind, 1999).

Unfortunately, this support for a stress-buffering effect of humor has often come in the

form of self-report and correlational studies. Few experiments have tested if humor

actually attenuates negative moods (e.g., depression or anxiety) from stressful events.

The purpose of this proposal was to explore whether humor “protects” individuals

from stressful experiences. In addition to self-report measures, the following studies used

physiological variables to explore the effects of humor and laughter on a stressful

experience. In Study 1, we tested how exposure to contagious laughter after a stressor

influenced psychological and physiological responses. In Study 2, we tested how


experiencing amusement before a stressor influenced these same responses. In Study 3, we

tested how experiencing amusement during a stressor influenced psychological and

physiological responses.

Humor

Humor is one way of achieving amusement and laughter, and often, it involves the

element of surprise. Indeed, humor typically takes place when something unexpected has

occurred—an incongruity between objects or expectations (Ruch, 2001). Thus, humor

produces amusement, and it also may produce smiling, or laughter (Deckers & Ruch,

1992). Martin (2001) suggests that humor can represent a stimulus (e.g., a funny film),

mental process (e.g., reffaming), or response (e.g., mirth), and a “sense of humor” typically

refers to a personality trait or an individual difference variable. Most humor studies have

focused on trait humor, with fewer focused on state humor. The present studies measured

both trait and state humor.

Humor as a Positive Influence

Freud (1928) was one of the first to view humor and laughter as having a positive

influence. He believed humor was a way for individuals to rise above the seriousness of

reality. In his book, “Jokes and their Relation to the Unconscious,” Freud (1960)

conceptualized humor and laughter as one of the most mature defense processes,

suggesting that laughter allowed individuals to release “pent-up” sexual and aggressive

energy. Even today, psychotherapists regard humor as one of the most adaptive defense

mechanisms (Vaillant, 2000). Vaillant, for example, argued that some types of humor

could be adaptive when facing a difficult situation.


Although Freud reflected on humor, it was not until about 25 years ago that others

became interested in the topic. In 1979, Norman Cousins published his book, Anatomy o f

an Illness, detailing his experiences with ankylosing spondylitis, a painful rheumatologic

disease. Cousins wrote that he had improved his condition by treating himself to massive

doses of laughter via Candid Camera episodes and Laurel and Hardy movies. His story

brought humor and its relationship to health to the forefront.

One way humor may positively influence health and well-being is that it elicits

other positive traits and characteristics. For example, studies have found that a good sense

of humor is related to self-reported happiness and expressions of optimism. Lyubomirsky

and Tucker (1998) asked if there were differences in the ways self-rated and peer-rated

“happy” and “unhappy” people thought about and interpreted their past events. In one of

their studies, they discovered that happy individuals were more likely than unhappy

individuals to use humor to cope with negative events in their past. In a different study on

optimism, Carver and others (1993) interviewed women immediately before and after they

had surgery for breast cancer, with follow-ups at three, six, and twelve months later. The

authors found that women who used humor to cope (as measured by the COPE inventory;

Carver, Scheier, & Weintraub, 1989) with their breast cancer were also more likely to

express optimism about their disease at the follow-ups.

Studies have also shown that humor is negatively associated with depression,

pessimism, neuroticism, anxiety about dying, and bad moods (Kohler & Ruch, 1996; Ruch,

1994; Thorson & Powell, 1993, 1994; Thorson et al., 1997). In one interesting study,

Lefcourt and others (1995) asked participants to think about their mortality by completing

“death exercises,” such as writing their own death certificates, composing their eulogies,


and drafting a will. After the task, most participants experienced increases in anger,

tension, and depression. However, participants who scored high on a measure of coping

humor (Martin & Lefcourt, 1983) were less likely to experience increases in these negative

emotions. Keltner and Bonanno (1997) were interested in how humor related to adjustment

after the loss of a spouse. They found that after six months had passed, individuals who

were able to laugh and smile (as measured by orbicularis oculi EMG) while discussing

their former partners reported less anger and more satisfaction with their life than those

who were unable to laugh.

Humor as a Stress-Buffer

Martin and Lefcourt (1983), in three correlational studies, investigated if trait and

state humor interacted with stressful life events to influence mood. In Study 1, participants

completed the Coping Humor Scale (CHS; Martin & Lefcourt, 1983) and the Situational

Humor Response Questionnaire (SHRQ; Martin & Lefcourt, 1984). The CHS assesses the

degree to which individuals use humor to cope, and the SHRQ assesses the degree to which

individuals laugh or smile in different situations. Participants also completed measures of

life events and mood. Results showed that stressful events were associated with less

depression and anxiety for individuals who used humor to cope with their stress, reported

that they valued humor, and laughed and smiled in a variety of situations.

In a second study, Martin and Lefcourt (1983) presented participants with various

objects (e.g., an old tennis shoe, a drinking glass, an aspirin bottle) and asked them to

create a 3-minute comedy routine relating the objects. Each participant’s routine was

scored for overall “wittiness.” Results showed that the correlation between negative life

events and negative moods was stronger for participants who received low wittiness scores.


This finding suggested that the moods of participants who received higher wittiness scores

were less distressed by their stressors.

In their third study, Martin and Lefcourt (1983) instructed participants to create a

humorous narrative while watching the stress-inducing film, Subincision. Similar to Study

2, researchers observed a stronger correlation between negative life events and negative

moods for the participants who performed poorly on the humor task. Interestingly,

participants who were better able to generate humor while watching the film were more

likely to report using humor in actual stressful situations. Although the authors failed to

test a theoretical explanation for their findings, they said that future studies should

investigate the process by which humor has a positive influence.

In a prospective study, Nezu and others (1988) found support for humor as a stress

moderator. Participants in their study completed the CHS, the SHRQ, a life events scale,

and measures of depression and anxiety. Two months later, participants returned to

complete identical measures. Results showed that humor interacted with life stress to

predict depression, but not anxiety. Thus, when they experienced similar stress, those

individuals who scored high on the CHS and the SHRQ reported less depression than those

who scored low on the humor measures. Because they failed to find effects for anxiety, the

authors suggested that humor may be helpful in dealing with past events, but not ongoing

or future events.

In contrast to the study by Nezu et al. (1988), other evidence suggests that humor

influences anxiety along with depression. Abel (2002), in a cross-sectional study,

investigated the coping strategies of high and low humor individuals. In her study, high

and low humor was defined as scoring one standard deviation above or below the mean,


respectively, on the Multidimensional Sense of Humor Scale (MSHS; Thorson & Powell,

1993). The MSHS is a measure of overall sense of humor which assesses use, recognition,

and appreciation of humor. Results revealed that, compared to low humor participants,

high humor participants reported less stress and anxiety during the past two months, despite

experiencing the same number of daily problems. High humor individuals were also more

likely to use positive coping strategies, both emotion-focused (e.g., positive re

interpretation) and problem-focused (e.g., effort to resolve the problem). Based on her

findings, Abel suggested that humor exerts a positive influence on the appraisal of stressful

events by reducing the negative moods in response to these events.

Trait humor does not just have implications for mood; it can also influence illnesses

that have historically been linked to negative emotion. For example, Clark, Seidler, and

Miller (2001) found that the propensity to laugh in various situations (as measured by the

SHRQ) was negatively associated with the incidence of coronary heart disease (CHD).

They had healthy participants along with those who been diagnosed with CHD complete

measures of sense of humor, anger and hostility. The authors found that compared to the

healthy controls, individuals with CHD scored higher on the hostility scale and lower on

the SHRQ. Importantly, this association remained even after controlling for risk factors

such as hypertension and cigarette smoking.

Each of the above correlational studies has suggested that humor may facilitate

coping with stressful situations because it moderates the typical negative emotional

response to such situations. Although the authors of the studies have typically disagreed

about which specific emotions are influenced by humor, all of them have focused on humor

and its relationship with affect. One important issue is whether humor may influence stress


through an avenue other than mood. Next, we discuss research that has shown that humor

can affect the physiology of a stress response.

Trait humor has been associated with positive physiological responses to stressful

tasks. Lefcourt, Davidson, Prkachin, and Mills (1997) were interested in how sense of

humor related to coping with stress in the laboratory. They measured participants’

physiological reactivity to four stressful laboratory tasks. The tasks included the Cold

Pressor task (placing a hand or forearm in cold water), the Favorable Impressions task

(trying to make a good impression on someone of the opposite sex), a Mental Arithmetic

task and the Stroop Color-Word Test. Experimenters used the CHS and the SHRQ to

assess sense of humor, and systolic and diastolic blood pressure to assess cardiovascular

reactivity. The authors expected that participants who scored high on both humor measures

would show less reactivity to the stressful tasks. Findings showed that only one task, the

Cold Pressor, provided evidence for the stress buffering effects of humor. Interestingly,

coping humor interacted with gender to predict blood pressure for all of the tasks; females

who were high in coping humor were less reactive to the tasks than females who were low

in coping humor. An opposite pattern was found for males, suggesting that gender

differences might be important in humor’s role as a stress moderator.

The studies on trait humor and responses to stress, thus far, have shown that people

who score higher on measures of humor may cope more effectively with their stressful

experiences (Martin & Lefcourt, 1983; Newman & Stone, 1996; Nezu et al., 1988).

Researchers have argued that trait humor can inhibit the negative affect that is typically

present in stressful situations. In addition, sense of humor has been associated with

positive physiological responses in stressful situations (Lefcourt et al., 1997). But can


experiencing humor (i.e., state humor) reduce the negative affect and heightened

physiological response that occurs in a stressful situation? Next, we discuss the few

experiments that have tested whether situational humor can be used to moderate the

negative affect and physiological reactivity during a stressful experience.

In one of the few true experiments on humor and responses to stress, Yovetich,

Dale, and Hudak (1990) created stress in the laboratory. After participants (all females)

completed the SHRQ, an experimenter told them they would be receiving an electric shock

in 12 minutes. While they waited, participants listened to a) a humorous tape, b) a non-

humorous tape, or c) no tape. They also rated their anxiety while an experimenter

monitored their heart rate (HR) and zygomatic (i.e., smiling behavior) activity. Results

showed that, as the shock approached, participants who scored low on the SHRQ were

more likely than participants who scored high to feel stressed. Also, participants who

scored high on the SHRQ were more likely to smile (i.e., as evidenced by zygomatic EMG)

while waiting for the shock. Also, the humor tape had a significant effect. Participants

who listened to the humorous tape (compared to those who listened to the non-humorous

tape) reported less anxiety and stress while they were waiting for the shock. Those who

listened to the humorous tape also showed more smiling behavior during the experiment.

However, participants in the humor condition did not show a HR reduction. This study was

one of the first to show that situationally-produced humor may provide relief from a

stressful experience. Although the humorous tape was associated with less self-reported

anxiety and stress, it failed to reduce participants' HR.

Using a similar paradigm to Yovetich et al. (1990), Danzer, Dale, and Klions (1990)

investigated the effect of humor on depression. They showed female students depressive


slides from the Velten mood statements. After the slides, participants listened to a) a

humorous tape, b) a non-humorous tape, or c) no tape. Experimenters recorded HR,

zygomatic (i.e., smiling) and corrugator (i.e., frowning) EMG. Participants who listened to

the humorous tape had the lowest depression scores following the slide presentation. Also

zygomatic activity was more frequent in the humorous condition than the other two

conditions. While viewing the depressive slides, all participants’ HR levels declined, but

afterward, HR increased for participants who listened to the humorous or non-humorous

tapes. Heart rate, however, was unaffected for participants in the no-tape condition. The

findings showed that humor, relative to non-humor, was unsuccessful in influencing HR

change: Listening to anything produced arousal.

The studies above suggest that when people experience amusement in the context of

a stressful situation, humor can attenuate their typical negative mood response. This

finding has been observed even for those who score low on trait humor measures.

Surprisingly, the studies that have used an experimental paradigm to investigate the stress-

buffering effects of humor are few in number. Because there are so few, the findings of

these preliminary experiments are inconclusive. In the experiment by Yovetich and others

(1990), effects for anxiety and zygomatic activity supported the notion of humor as a stress

buffer before a stressful experience, but the HR findings did not. Heart rate failed to show

a decline that would have been consistent with the reduction in anxiety. Also, Yovetich

and others failed to record actual laughter, a problem that exists with many studies on

humor (Martin, 2001). Failing to record laughter is problematic because it is then unclear

whether any effects are due to a) experiencing amusement (which can occur without

laughter), b) laughter, or c) both. Martin (2001) has argued that laughter may mediate the


10

humor-health relationship. It may not be experiencing humor or amusement that accounts

for the beneficial influence on stress, but laughter and its positive influence on physiology

(Martin, 2001).

Laughter

According to Provine (2000), laughter is an ancient phenomenon that existed long

before speech and humor. Despite its longevity, people know little about laughter. Provine

argued that “laughter is a harlequin that shows two faces—one smiling and friendly, the

other dark and ominous” (p. 2). Early theoretical perspectives lend support to this

quotation. The dark, ominous laughter represents superiority theory (Zillmann, 1983), the

idea that humor was used to humiliate and ridicule those in a lower social position. The

smiling, friendly laughter reflects the idea of humor as a social bonding tool.

Research shows that laughter is most likely to occur in social situations. In one

“real-time” investigation of humor, Mannell and McMahon (1982) found that almost 90%

of college students’ humor experiences took place in social situations. In a more recent

study, Kuiper and Martin (1998) found that, for community participants, laughter was also

most likely to occur in the presence of others.

Pro vine (1992, 2000) argued that laughter was an extremely contagious

phenomenon. In his research, he has cited instances of "laughter outbreaks," in which

groups of people experience bouts of uncontrollable laughter. One such laughter outbreak

occurred among teenagers in Tanzania (east Africa), and it was so severe that it shut down

an entire school district. Provine likened contagious laughter to contagious yawning and

said that the only exception was that laughter was an auditory influence while yawning was

a visual influence. In one study on contagious laugher, he presented his students with 10


trials of 18 seconds of canned laughter (i.e., from a laugh box) followed by 42 seconds of

silence. During the 42 seconds, he asked the students to record whether they laughed or

smiled in response to the laughter sounds. Results indicated that almost one-half of the

participants (47%) laughed on the first trial, but laughter declined across trials. Smiling

was more likely, with 91% of the participants smiling after hearing the laughter sounds.

Based on his study, Provine argued that simply hearing laughter could elicit laughter and

smiles, and suggested that “contagious laughter,” provided by a laugh box or another

person, could elicit laughter in the laboratory. In Study 1, we presented participants with

sounds of “contagious laughter.” We pilot tested approximately 20 clips and then chose the

four that elicited the most laughter.

Researchers have often disagreed about how laughter influences physiology

(Provine, 2000; White & Camarena, 1989). Dixon (1980) showed that the physiological

effects of laughter such as an increase in muscle tension, HR, and respiration followed by a

relaxation period was similar to an attenuated stress response. Like Dixon, Berk (2001) in

his review of the physiological effects of laughter, argued that laughter resembles the

benefits incurred by aerobic exercise, by first producing an increase in HR and BP and then

a relaxation period. Fry (1992) showed that increased respiration was also a benefit of

laughter. He found that mirthful laughter was characterized by more expiration than

inspiration, but the benefits also depended on the characteristics of participants' individual

laughs.

In a more recent investigation on laughter and its physiological effects, Miller and

others (2006) linked laughter to a reduction in the risk of cardiovascular disease.

Participants in their study watched scenes from “Saving Private Ryan," and then 48 hours


later, they watched scenes from “There’s something about Mary.” Participants were

instructed to watch as much of the film as they wanted; they were told to watch it until they

felt they had been affected by it (all watched between 15-30 minutes of each film). Before,

during, and after participants watched the films, brachial artery flow-mediated

vasodilatation (FMD) was assessed. Flow-mediated vasodilatation is associated with the

endothelium, which when impaired, contributes to coronary heart disease. Results

indicated that FMD declined in 14 of the 20 participants after they watched the stressful

film, and increased in 19 of the 20 participants after they watched the funny film. These

effects were somewhat immediate; they showed up one minute after participants finished

watching the films. Although the authors stated that, “laughter was readily elicited when

observing selected scenes,” they failed to report how much laughter occurred. Miller and

others concluded that like mirthful laughter, other positive emotions may have an opposite

effect of negative emotions on the endothelium.

Importantly, not everyone has found that laughter has a positive influence on

physiology. In an older study, White and Camarena (1989) failed to find a physiological

benefit of laughter. The authors randomly assigned individuals to take part in a) a laughter

group, b) a relaxation training group, or c) a control group for a period of 6 weeks. The

groups met once each week for 1.5 hours. Heart rate, BP, and mood were assessed before

and after each session. Although participants in the laughter group reported better moods

following each session, they failed to experience a physiological benefit relative to the

other two groups. It actually was the participants in the relaxation training group who

experienced the most physiological benefit from the sessions; they had the lowest HR and

BP levels following each session.


In a more recent investigation, Elliott (2005) found that watching a humorous film

failed to improve older adults’ cardiovascular health. The author randomly assigned 18

older adults to view either 20 minutes of a comedy film, stressful film, or neutral film.

Elliott measured HR, BP, respiratory rate, and oxygen saturation levels immediately before

and after participants watched the film. Results revealed no differences between groups for

the physiological variables. However, compared to participants who watched the stressful

or neutral films, those who watched the comedy film reported less tension and anxiety after

watching the film. Unfortunately, the author failed to monitor actual laughter that

occurred.

The above studies show that although laughter can sometimes have a positive

influence on physiology, it can also fail to change physiology. Thus, the issue of how

laughter influences physiology remains unresolved. According to Scheff and Bushnell

(1984), the inconsistency could relate to the timing of physiological assessment. For

example, early on, laughter may increase respiration, HR, and BP, but after some time has

passed, it may lead to a state of relaxation in which these measures return to pre-laugh

levels. If an investigator only measures HR and BP immediately before and after the study,

he or she may fail to uncover the positive effects. In each of our studies, we measured

physiological responses at least a few minutes before the study began, during, and at least a

few minutes after the study ended.

In summary, a number of correlational studies have suggested that humor may

moderate the negative moods that arise from stressful experiences, but few experimental

studies support this claim. In addition, investigators have suggested that laughter may

positively influence physiology, but the research has been inconsistent. We conducted


three experimental studies to investigate the effects of humor and laughter in stressful

situations in the laboratory. We were interested in how humor and laughter would affect

mood and physiological response in a stressful context. In addition, we were interested in

whether humor and laughter would be most effective in helping individuals cope before,

after, or during a stressor. Study 1 explored the effects of contagious laughter after

participants were exposed to a laboratory stressor. Study 2 explored the effects of

humorous one-minute clips before participants were exposed to a stressor, and Study 3

explored the effects of these clips during a stressor.


15

STUDY 1

METHOD

Participants

Twenty-three college students participated in this study. The sample included 14

females and 9 males. The mean age of participants was 19.74 (SD =1.14), and 73.9% were

Caucasian, 8.7% were Black/African American, 8.7% were Middle Eastern, and 8.7%

other.

Design

This study used a within subjects 2 (Aversive Noise/Neutral Stimulus) x 2

(Laughter/Neutral Clip) x 4 (Trials) design. Participants were randomly assigned to one of

four possible trial presentation orders.

Procedure

Students came into the laboratory to complete a study on “mood and physiology.”

After they signed a consent form, an experimenter informed participants that the goal of the

study was to understand how positive and negative feelings influence mood and

physiological responses. The experimenter explained the task, telling participants that they

would be continuously monitoring their mood while listening to a series of sounds. The

experimenter also told participants that in addition to the sounds, there would be quiet

periods during the task. Participants were instructed to sit back and relax during these quiet

periods, but to continue to monitor their mood. After explaining the task, the experimenter

attached sensors to the participant to measure zygomatic, corrugator, and skin conductance

activity. The experimenter also showed participants how to use the joystick to


16

continuously record their feelings during the task. Finally, the experimenter completed a

sensor check to confirm that all sensors were working properly. When the task was

finished, all participants completed a series of questionnaires assessing sense of humor,

trait anxiety, trait neuroticism, and trait optimism.

Task

Participants completed sixteen trials, with random exposure to each condition every

four trials. During each trial, participants heard an aversive noise or a neutral sound

clip for 15 seconds (Interval 1) followed by a laughter or neutral sound clip for 15

seconds (Interval 2). After Interval 2, there was an inter-trial resting period of

approximately 30 seconds (i.e., Recovery). During this time, participants continued

to record their mood. Table 1 presents the timeline of each trial.

Table 1. Example of Trial 1 from Study 1_____ _____________________Interval 1

(Aversive Noise or Neutral Clip)

Interval 2 (Laughter or Neutral

Clip)Recovery

15 seconds 15 seconds 30 seconds

Throughout the course of the experiment, participants heard 4 different aversive

noises, 4 different laughter clips, and 8 different neutral clips. The clips were fixed within

each trial, but we varied the order of trial presentation. The total time for the listening task

was approximately 19 minutes (this includes three minutes of baseline).

Aversive noise. Research shows that individuals show increased sympathetic

responding when presented with a strong, auditory stimulus (Ekman, Friesen, & Simons,

1985; Hagemann, Levenson, & Gross, 2006). We downloaded several different noise clips

from “Sounddogs.com”, an online sound effects resource (http://www.sounddogs.com/).


http://www.sounddogs.com/

17

We pilot tested the noises, and chose the four that produced the most discomfort in pilot

testing. The noises were, “white noise,” “car alarm,” “alarm clock,” and “nails on

chalkboard.”

Contagious laughter clips. These clips were also downloaded from

“Sounddogs.com.” We pilot tested several different laughter clips and chose the four that

produced the most overt laughter in pilot testing. Three of the laughter clips included

individuals laughing alone and one included two women laughing.

Neutral clips. The neutral clips were also downloaded from “Sounddogs.com,” and

included “knocking on a door,” “plastic bag in the wind,” “scissors clipping hair,”

“dishwasher humming,” “glasses clinking,” “hallway sound,” “bicycle wheel spinning,”

and “typing on a keyboard.”

Questionnaire Measures

Sense o f humor. Trait sense of humor was assessed by two questionnaires

(Appendices D and E). The Humor Styles Questionnaire (HSQ; Martin, Puhlik-Doris,

Larsen, Gray, & Weir, 2003) is a 32-item questionnaire that measures the different ways

individuals express and experience humor. The questionnaire includes items such as, “If

someone makes a mistake, I will often tease them about it,” and “I often try to make people

like or accept me more by saying something funny about my own weaknesses, blunders, or

faults.” Participants rated their agreement with the statements on a 1 (totally disagree) to 7

(itotally agree) scale. The CHS (Martin & Lefcourt, 1983) is a 7-item questionnaire that

measures the degree to which individuals use humor to cope. It includes items such as, “I

usually look for something comical to say when I am in tense situations,” and “I have often


18

felt that if I am in a situation where I have to either cry or laugh, it’s better to laugh.”

Participants rated the statements on a 1 {strongly disagree) to 4 {strongly agree) scale.

Trait anxiety. To measure how anxious individuals feel in general, participants

completed a 10-item Anxiety scale from the International Personality Item Pool (IPIP;

2001; See Appendix F). This scale consists of statements such as “Fear for the worst,” and

“Am afraid of many things.” Participants rated how descriptive the statements were of

themselves on a 1 {very inaccurate) to 5 {very accurate) scale.

Trait neuroticism. A 10-item Neuroticism scale from the IPIP (2001; See Appendix

G) was used to measure neuroticism. This scale consists of statements such as “Often feel

blue,” and “Have frequent mood swings.” Participants rated how descriptive the

statements were of themselves on a 1 {very inaccurate) to 5 {very accurate) scale.

Trait optimism. The 10-item Revised Life Orientations Test (LOT-R; Scheier,

Carver, & Bridges, 1994) was used to assess individual differences in optimism. Examples

of items on this measure (See Appendix H) are, “In uncertain times, I usually expect the

best,” and “If something can go wrong for me, it will.” Participants rated their agreement

with the statements on a 1 {I agree a lot) to 5 {I disagree a lot) scale.

Dependent Measures

Both mood and physiological responses were monitored during the 3-minute

baseline period and the 16-minute sound presentation. Data were recorded by the BioPac

MP150 system and stored via BioPac’s AcqKnowledge software.

Facial electromyography. Participants’ facial electromyography (EMG) was

recorded, including zygomaticus major, which tends to correspond with positive emotions

and corrugator supercilli, which tends to correspond with negative emotions. We measured


19

the zygomatic muscle (“smile muscles”) and the corrugator muscle (“frown muscles”) by

placing two BioPac EL254S electrodes just below the cheekbone and two just above the

inner eyebrow, respectively. In addition, one BioPac EL254S electrode was placed on the

forehead and served as a reference electrode.

The sampling rate of the zygomaticus major and corrugator EMG was 1000 Hz

(i.e., 1000 samples per second). To obtain a measure of the integrated (i.e., temporal

summation) EMG signal, we followed the procedures outlined by Tassinary and Cacioppo

(2000). The Root Mean Square (RMS) procedure was used to quantify the facial EMG

data. We squared each raw EMG sample (i.e., signal) and then computed the mean of the

squared samples for each 1-second period. Finally, we computed the square root of this

value and summed across the 1-second periods to obtain a measure of total EMG activity

for every 5-second period.

Skin conductance level. Participants’ skin conductance (SC) level was monitored

by placing a BioPac TSD203 electrodermal response transducer on the fingertips of the

first and second fingers on the left hand. The sampling rate for skin conductance (SC) was

1000 Hz. We computed the mean and found the peak values for each 5-second interval.

Because a “new” SC response is often elicited before an individual has had time to recover

from an “old” SC response, we followed procedures according to Ben-Shakhar (1985) for

computing an SC change value. Because this value takes into account participants’ SC

levels immediately preceding stimulus onset, it provides a more accurate picture of the SC

response. We computed the change value by subtracting the average SC value of the last 5

seconds of the previous Recovery Interval from the peak value of the present 5 seconds.

The change value was computed for all 12 Periods: Interval 1 (made up of the first three


20

periods), Interval 2 (made up of the second three periods), and Recovery (made up of the

last six periods) for each Trial.

Self-reported mood. Participants continuously recorded their mood during the task

using a joystick apparatus. The joystick was connected to the BioPac via a variable

assessment transducer (TSD115). The joystick had a slide control with a graduated scale.

Participants could indicate neutral affect by moving the slide control to the center. They

could indicate increasingly positive or negative affect by moving the slide control to the

right or left, respectively. On the joystick scale, the center was clearly marked “Neutral,”

the right was clearly marked “Positive,” and the left was clearly marked “Negative.”

The sampling rate for mood was 1000 Hz. To reduce the data, we computed the

mean value of each 1-second period (i.e., across 1000 samples), and the mean value across

each 5-second period. We also found the peak value of each 1-second period, and across

each 5-second period. Because there were few differences between the results with peak

mood and average mood, only peak mood is reported.1

Hypotheses

First, we expected participants to show more corrugator (e.g., frowning) activity

when they heard the aversive noise compared to the neutral clips. We also expected that

participants would show more zygomatic (e.g., smiling) activity when they heard the

contagious laughter clips compared to the neutral clips. Next, we expected that participants

would show higher SC levels and report more negative moods when they heard the noise

1 There was only one difference between the findings for average and peak mood. The

Noise x Laughter interaction during Recovery was not significant for average mood

ip = .14) and significant for peak mood (p = .05).


compared to the neutral clips. We expected that participants would report more positive

moods when they heard the laughter compared to the neutral clips. Finally, during

Recovery, we expected that participants would show lower SC levels and report less

negative moods when the noise had been followed by clips of contagious laughter instead

of neutral clips.


RESULTS

Manipulation Check

At the end of the study, we asked participants about the funniness of the laughter

clips. Participants responded on a 1 (Not at all funny; they did not make me feel like

laughing) to 5 ( Very funny; they made me feel like laughing) scale, with a midpoint of 3

(Somewhat funny; they kind o f made me feel like laughing). The mean response was a 2.71

(SD =1.21), suggesting that participants thought that the clips were “somewhat funny.”

Almost one-half of the participants (41.2%) rated the clips as less than “somewhat funny”

(3 on the 5-point scale). We also asked participants to report how many times they laughed

during the study. Only 3 participants (17.6%) reported that they did not laugh at all during

the study. The rest of the participants reported that they laughed a little under their breath

(64.7%), out loud once or twice (11.8%), or out loud more than once or twice (5.9%). In

addition to the self-report, an experimenter recorded laughter that occurred during the

study, and found that only 1 participant laughed out loud.

Analyses

To analyze the results, we used repeated measures Analysis of Variance (ANOVA).

For each dependent variable, three separate repeated measures ANOVAS were conducted,

focusing on consecutive 5-second periods: one for Interval 1 (Periods 1-3), one for Interval

2 (Periods 4-6), and one for Recovery (Periods 7-12). For Periods 1-3 and 4-6, the repeated

measures analysis was a 2 (Aversive Noise/Neutral Stimulus) x 2 (Laughter/Neutral Clip) x

4 (Trials) x 3 (Periods within Trials). For Periods 7-12, the repeated measures analysis was

a 2 (Aversive Noise/Neutral Stimulus) x 2 (Laughter/Neutral Clip) x 4 (Trials) x 6 (Periods

within Trials). If, for any one of the dependent variables, the analysis failed to produce a


23

meaningful interaction with either Trials or Periods, we collapsed across the interval for

analysis purposes.

Because the analysis for each interval (i.e., Interval 1, Interval 2, Recovery)

produced a total of 45 possible effects, only the significant interactions that were relevant

to the manipulation are reported. For the interactions that were significant, but unrelated to

the manipulation, we report the findings in Appendix I. For each dependent measure, a

figure is displayed that presents all of the data, collapsed across individuals and trials, for

all 12 Periods.

Zygomatic-Interval 1 (Periods 1-3). Because participants heard only the aversive

noise or neutral clips, there was no reason to expect differences in zygomatic activity

during the first interval. However, as Figure 1 shows, during the first interval, the aversive

noise produced more zygomatic activity than the neutral sounds (Ms = .11 vs. .09), and this

difference was significant, F (1, 22) = 6.09,p = .02. Also, activity for the aversive noise

tended to fade over time; for the neutral clips, activity started to fade, but then increased.

This pattern of results produced a weak Noise x Periods interaction, F (2, 44) = 2.88, p =

.07.

Zygomatic-Interval 2 (Periods 4-6). During this interval, we expected that

participants who heard the laughter would show more zygomatic activity than participants

who heard the neutral clips. However, as illustrated in Figure 1, the laughter clips

produced only slightly more zygomatic activity than the neutral clips (Ms = .10 vs. .09),

and the effect was non-significant, F (1, 22) = 2.22, p = .15. Figure 1 shows that laughter

was producing more activity than neutral in Periods 5 and 6, but not in period 4. However,

the Laughter x Periods interaction was not significant, F (2,44) = 2.73, p = .08.


24

;erval 2 Recovery;erval 1

0.13

0.12

0.09

0.08

0.07

0.06

0.05

0.04

Periods

Figure 1. Study 1 zygomatic activity by condition collapsed across trials.Note. For all figures, NL = aversive noise followed by laughter clips; NO = aversive noise followed by neutral clips; OL = neutral clips followed by laughter clips; and OO = neutral clips followed by neutral clips.

Zygomatic-Recovery (Periods 7-12). Results showed no significant main effects or

interactions during Recovery.

Corrugator-Interval 1 (Periods 1-3). We hypothesized that participants would

show more corrugator activity during the aversive noise compared to the neutral clips. As

can be seen from Figure 2, the noise produced more corrugator activity than the neutral

clips (Ms = .25 vs. .21), but the difference was small, F (1, 22) = 3.91,/? = .06. Also, as

periods continued, activity declined for the noise, but stayed the same or increased for

neutral clips. This pattern of results produced a significant Noise x Periods interaction, F

(2, 44) = 6.59,/? <.01.

Corrugator-Interval 2 (Periods 4-6). We hypothesized that the laughter, compared

to the neutral clips, might produce less corrugator activity in participants, but there were no


25

;erval 1 Interval 2 Recovery

0.35

0.3 ■— NL„ NO— OL- - 00

0.25

0.2

0.15

PeriodsFigure 2. Study 1 corrugator activity by condition collapsed across trials.

differences between the two clips (both Ms = .21). There was, however, a significant

Laughter x Periods interaction, F (1, 22) = 4.86, p = .04 (Figure 3). The interaction showed

that following an aversive noise, laughter produced less corrugator activity than neutral

clips (Ms = .20 vs. .22), but following neutral clips, laughter produced more corrugator

activity than other neutral clips (Ms = .21 vs. .19).

Corrugator-Recovery (Periods 7-12). We did not have expectations about

corrugator activity during this interval because we were unsure if activity would carry over.

As can be seen from Figure 2, participants showed less corrugator activity after listening to

laughter compared to neutral clips (Ms = .20 vs. .21), but the effect was not significant, F

(1,21) = 3.40,/? = .08.

Skin Conductance-Interval 1 (Periods 1-3). We expected that the aversive noise

would produce more SC activity than the neutral clips. Consistent with expectations,

Figure 4 shows that the noise produced greater SC activity than the neutral clips (Ms = .52

and .12), F (1, 22) = 105.32,/) < .001. Additionally, there was a main effect of Periods, F


(2, 44) = 12.30,/? < .001, showing an increase and then a decline across periods. Figure 4

also shows that the noise produced a spike in SC activity in the beginning, but then activity

began to fade across periods. Neutral clips produced slightly more SC activity in the

beginning, and then stayed the same or declined slightly across periods. This pattern

produced a significant Noise x Periods interaction, F (2, 44) = 16.94, p < .001.

0.23

0.22

0.21

0.2

0.19

0.18

0.17

Laughter Neutral

Clips

NoiseNeutral

Figure 3. Study 1 Noise x Laughter interaction for corrugator activity during Interval 2.

Skin Conductance-Interval 2 (Periods 4-6). Figure 4 indicates that the aversive

noise continued to produce higher SC levels than the neutral clips (Ms = .37 vs. .05), and

this effect was significant, F (1, 22) = 23.51,/? < .001. Although not predicted a priori,

laughter produced more SC activity than neutral clips (Ms =.32 vs. .11), F (1, 22) = 13.59,

p < .01. In addition to the main effects, a significant Noise x Laughter interaction, F (1, 22)

= 4.91,/? = .04, is presented in Figure 5. The interaction showed that the noise followed by

laughter clips produced higher SC than the noise followed by neutral clips (Ms = .54 vs.

.20). A neutral clip followed by laughter also produced higher SC than a neutral clip


followed by other neutral sounds (Ms = .09 vs. .01), but the effect was smaller than the one

observed for the aversive noise. Finally, results indicated that activity declined across

Periods, F (2, 44) = 15.41,/>< .001.

Interval 1 Interval 2 Recovery

0.8

0.6■—NL— NO— OL - - OO

0.4

0.2

- 0.2

-0.4

Periods

Figure 4. Study 1 SC activity by condition collapsed across trials.

Skin Conductance-Recovery (Periods 7-12). We hypothesized that, during

Recovery, individuals who had just heard the laughter would show lower SC levels than

individuals who had just heard the neutral clips. This was because we expected the

laughter to show a buffer effect. However, during this interval, the aversive noise, relative

to the neutral clips, continued to produce greater SC activity (Ms = .23 vs. -.08), F (1, 21) =

24.51,/? < .001 (Figure 4). The effects of laughter, relative to the neutral clips, also carried

over and continued to produce greater SC activity (Ms = .15 vs. -.01), F (1, 21) = 7.08,

p = .02. Results also indicated that SC activity declined across periods, F (5, 105) = 3.16,

p = .01.


28

0.5 Noise

Neutral0.4

0.3

0.2

Laughter Neutral

Clips

Figure 5. Study 1 Noise x Laughter interaction for SC activity duringInterval 2.

During the analysis phase of Study 1, we discovered a potential confound: We had

failed to pair each aversive noise with both laughter and neutral clips. This was a problem

because results showed that one of the noises was particularly aversive compared to the

other three. Figure 4 presents the difference in SC levels produced by the aversive noises.

Because the noises failed to produce equal amounts of SC in Interval 1, we were unclear

about the influence of the neutral clips compared to the laughter clips during Interval 2. To

examine the results without the confounding variable, we re-ran analyses without the

“white noise.” Figure 6 presents the SC change activity across 12 Periods, without the

white noise. It is evident from the figure that hearing clips of contagious laughter did not

reduce SC levels relative to hearing neutral clips. Indeed, laughter was actually arousing

following a neutral clip.

Peak Mood-Interval 1 (Periods 1-3). We expected that participants who heard the

aversive noise would report worse moods than participants who heard the neutral clips.


Figure 7 shows that during Interval 1, the noise produced more negative moods than the

neutral clips (Ms = -2.73 vs. 1.77), F (1, 22) = 103.58,/? < .001. A Noise x Periods

interaction, F (2, 44) = 74.26, p < .001, showed that across periods, the noise produced

increasingly more negative moods, but the neutral clips did not. In addition, a significant

Noise x Trials interaction, F (3, 66) = 13.06,/? < .001, showed that, across trials, the noise

produced increasingly negative moods, but neutral clips produced the same or increasingly

positive moods.

Recovery

Periods

Figure 6. Study 1 SC activity without the confounding variable.

Peak Mood-Interval 2 (Periods 4-6). We expected that the aversive noise might

carry over and continue to produce worse moods than the neutral clips. Results showed

that the noise continued to produce more negative moods than the neutral clips (Ms = .08

vs. 2.33; Figure 7), F (1, 22) = 44.66,/? < .001. During the second interval, we also

expected that the laughter clips would produce more positive moods than the neutral clips.


30

However, there were no significant differences in self-reported mood for the laughter

versus the neutral clips (Ms = 1.36 vs. 1.05). Figure 7 also shows that mood improved

across Periods, another significant main effect, F (2, 44) = 28.20,/? < .001. When

participants heard a noise that was followed by laughter, they reported worse moods than

when they heard a noise that was followed by neutral clips (Ms = -.14 vs. .30). However,

when participants heard a neutral clip that was followed by laughter, they reported better

moods than when they heard a neutral clip that was followed by other neutral clips (Ms =

2.85 vs. 1.81). This pattern of results produced a significant Noise x Laughter interaction,

F ( l , 22) = 12.19,/? < .01 (Figure 8).


OL- - - O O

Periods

Figure 7. Study 1 peak mood reports by condition collapsed across trials.

Peak Mood-Recovery (Periods 7-12). During Recovery, we expected that

participants who had just heard the laughter would report better moods than participants

who had just heard the neutral clips. The results for Recovery showed a similar pattern to

those obtained for Interval 2. There were no differences in self-reported mood for the


31

laughter versus the neutral clips (Ms = 2.03 vs. 1.74). Also, the aversive noise continued to

produce more negative mood than the neutral clips (Ms = 1.62 vs. 2.14; Figure 7), F (1, 21)

= 6.06, p = .02. Whether participants heard a noise followed by laughter clips or a noise

followed by neutral clips failed to differentially influence their moods (Ms = 1.64 vs. 1.61).

However, when participants heard a neutral clip that was followed by laughter, they

reported better moods than when they heard a neutral clip that was followed by other

neutral clips (Ms = 2.43 vs. 1.86). This pattern of results produced a significant Noise x

Laughter interaction, F (1, 21) = 4.15, p = .05 (Figure 9).

2.5Noise

Neutral

NeutralI aughter-0.5

Clips

Figure 8. Study 1 Noise x Laughter interaction for peak mood duringInterval 2.

Peak Mood-Recovery (Periods 7-12). During Recovery, we expected that

participants who had just heard the laughter would report better moods than participants

who had just heard the neutral clips. The results for Recovery showed a similar pattern to

those obtained for Interval 2. There were no differences in self-reported mood for the

laughter versus the neutral clips (Ms = 2.03 vs. 1.74). Also, the aversive noise continued to


produce more negative mood than the neutral clips (Ms = 1.62 vs. 2.14; Figure 7), F (1,21)

= 6.06, p = .02. Whether participants heard a noise followed by laughter clips or a noise

followed by neutral clips failed to differentially influence their moods (Ms = 1.64 vs. 1.61).

However, when participants heard a neutral clip that was followed by laughter, they

reported better moods than when they heard a neutral clip that was followed by other

neutral clips (Ms = 2.43 vs. 1.86). This pattern of results produced a significant Noise x

Laughter interaction, F (1, 21) = 4.15,/) = .05 (Figure 9).

3

2.5

2

1.5

1

0.5

0NeutralLaughter

Noise

Neutral

Clips

Figure 9. Study 1 Noise x Laughter interaction for peak mood during Recovery.

Moderating Analyses

Because research (Lefcourt et al., 1997) has shown that women may be more likely

than men to use humor to cope, we conducted moderator analyses to determine if gender

interacted with the laughter manipulation to influence any of the dependent measures.

Analyses showed that gender interacted with laughter to influence only one dependent

measure: self-reported mood. Gender interacted with laughter to influence mood during


33

Interval 1, F (\, 21) = 5.02,/? = .04, Interval 2, F {\, 21) = 4.57,/? = .04, and Recovery, F

(1, 20) = 7.64,/? = .01. In all cases, the interactions showed that women reported more

positive moods when they heard the laughter compared to the neutral clips. Men, on the

other hand, reported more negative moods when they heard the laughter compared to the

neutral clips.

We also conducted a series of moderator analyses to determine if various trait

measures (e.g., coping and other types of humor, anxiety, neuroticism, and optimism)

interacted with laughter to influence the dependent measures. For these analyses, we

conducted median splits on each of the trait measures. Then we used the median split

variables as between-subject variables in the same repeated measures ANOVAs used

above. The trait measures failed to interact with laughter to influence mood or physiology

(all ps > .06).

In addition to the moderator analyses, we were interested in whether participants’

scores on any of the sense of humor questionnaires would be associated with their self-

reported mood during the laughter manipulation. We found only one significant pattern of

associations: Self-reported mood during the laughter manipulation was negatively

correlated with scores on the self-defeating humor subscale. The pattern (r > -.40) was

consistent for 6 out of the 8 trials in which participants heard clips of contagious laughter.

The pattern suggested that the more likely a participant was to report engaging in self-

defeating humor (i.e., putting himself down for the amusement of others), the less likely he

or she was to report positive moods during the laughter manipulation.


34

DISCUSSION

Although there have been several correlational studies, few experiments have

investigated humor's stress-moderating effects. Study 1 was a careful laboratory

investigation of the influence of humor and laughter in a stressful situation, created by

blasting participants with aversive noise. After participants heard the noise, they heard

either a clip of contagious laughter or a clip of a neutral sound. Results showed that the

aversive noise produced more corrugator activity relative to the neutral clips, although the

finding was weak. Also, the noise produced much higher SC levels and greater negative

mood compared to the neutral clips, an effect that carried over to Interval 2 and Recovery.

The contagious laughter clips that followed produced only slightly more zygomatic activity

(not significant) than the neutral clips. Surprisingly, the laughter produced higher SC

levels than the neutral clips, and this effect carried over to Recovery. Following the noise,

the laughter clips failed to produce more positive moods relative to the neutral clips. The

findings from Study 1 showed that listening to contagious laughter after a stressor failed to

inhibit the negative mood and physiological arousal from the stressor.

Consistent with expectations, the laughter sounds produced more zygomatic activity

than the neutral sounds. Although the effect was weak, it became stronger across periods

as was indicated by a significant Laughter x Periods interaction. The interaction suggested

that it took participants a few seconds to “catch on” to the laughter. This finding was

consistent with the idea that laughter is a contagious phenomenon (Provine, 1992, 2000).

Surprisingly, the results showed that the aversive noise produced more zygomatic activity

than the neutral clips. Although this was an unexpected finding, it suggested that the noise


35

may have startled or surprised participants, a facial expression that affects zygomatic

activity (Hu & Wan, 2003).

Another finding that was consistent with expectations was that the aversive noise

produced more corrugator activity than the neutral clips, although the difference was not

significant (p < .06). Interestingly, during Recovery, participants who had just heard

laughter showed a small effect (Cohen’s d = .44) of less corrugator activity than

participants who had just heard neutral clips. This finding showed, with respect to

corrugator EMG, that laughter produced an effect during Recovery. Because EMG reflects

rapid and low-level changes (Tassinary & Cacioppo, 2000), this finding cannot be

attributed to activity subsiding from Interval 2 (i.e., the laughter clips). Instead, it suggests

that there were genuine changes in corrugator activity during Recovery. However, more

research should be conducted before drawing strong conclusions about this finding.

The aversive noise produced more SC activity than the neutral clips, but this effect

faded across periods, suggesting that participants habituated to the noise. Unexpectedly,

laughter also produced more SC activity than the neutral clips. A significant Noise x

Laughter interaction showed that compared to neutral clips, laughter clips always produced

more SC activity, and this was regardless of whether the laughter followed neutral clips or

noise. This same pattern emerged during the Recovery Interval, but the interaction was not

significant (p = .16). In summary, the results for SC showed that laughter a) increased SC

activity relative to neutral, and b) failed to lower SC activity following an aversive noise.

Together, these findings suggest that simply hearing laughter is arousing in and of itself.

The aversive noise produced more negative moods than the neutral clips, and this

effect became stronger across periods, suggesting that the longer participants had to endure


the noise, the worse moods they reported. A significant interaction during Interval 2

showed that if participants heard the noise followed by laughter, they tended to report

worse moods than if they heard the noise followed by neutral clips. The pattern was

reversed, however, for a neutral clip: If participants heard a neutral clip that was then

followed by laughter, they tended to report better moods than if they heard a neutral clip

that was then followed by other neutral clips. This same pattern (main effects along with

the interaction) was observed during Recovery. These findings for mood suggest that

laughter benefited mood only when it came after a neutral clip compared to after a noise.

Laughter may have been interpreted differently depending on what preceded it; when the

laughter was preceded by a noise, it was perceived as strange, but when the laughter was

preceded by a neutral clip, it was perceived as pleasant and enjoyable. In any case,

laughter clearly failed to attenuate participants’ negative moods stemming from the

stressor.

The findings for Study 1 are somewhat inconsistent with other research showing

that humor helped individuals recover from negative moods. In 1990, Danzer and others

found that listening to tapes of comedians for 12 minutes reversed the effects of a

depression induction. The authors found that compared to those who listened to a taped

introductory seminar on geology or nothing, the depression scores of those who listened to

a humorous tape returned to baseline levels. Danzer et al. also found that participants in

the humorous tape condition had higher heart rates while listening, suggesting that humor

was arousing. The authors concluded that humor may have alleviated the depression by

providing insight or releasing tension, but they did not test these explanations directly.


The most important question stemming from Study 1 is why the laughter clips,

compared to the neutral clips, led to more arousal and worse moods after the noise? One

theory that could be used to explain these findings is Zillmann's (1971) excitation transfer

theory. According to this theory, residual, or "leftover," excitement from one arousing

stimulus can lead to an exaggerated emotional response to a later stimulus. Sympathetic

activity reflected by SC levels is particularly sensitive to excitation transfer because it is a

slow-response system (Dawson, Schell, & Filion, 2000; King, 2003). In our study, it is

possible that participants experienced increased physiological arousal in response to the

laughter clips because the clips came after an aversive noise. Compared to the neutral

clips, the laughter clips elicited a strong emotion that was intensified due to residual arousal

from the noise.

It is possible that we failed to find beneficial effects of laughter in a stressful

situation because of our design (i.e., laughter came after the stressor). Thus, we conducted

a second study in which we presented the humor manipulation first and then followed it

with the stressor manipulation. If excitation theory can explain the results, then in Study 2,

the good excitement experienced during the humor manipulation might make the stressor

seem less negative.


STUDY 2

The findings from Study 1 suggested the need for three important methodological

changes. First, it was unclear whether the laughter clips in Study 1 produced amusement

per se. For example, three participants reported that they did not find the clips funny and

they did not laugh at all. Also, although eleven participants reported at least laughing

under their breath, and three reported laughing out loud once or twice, an experimenter

using a tape recorder in the participant room, detected laughter from only one participant.

This finding was surprising given that during pilot testing, the clips produced laughter.

Because the purpose of this project was to explore the effects of actual laughter, we

changed the manipulation in Study 2 to increase the likelihood of producing overt laughter.

A second change that we made in Study 2 related to the stressor. In Study 1, we

discovered that the "white noise," produced significantly higher SC levels than the other

three noise (stressor) manipulations. Because it was the most stressful out of the four

noises from Study 1, we chose to use the white noise and two variations of it for our

stressor manipulation in Study 2.

The third change in Study 2 related to the timing of the humor manipulation. The

goal of these studies was to test if humor and laughter can improve recovery from stressful

experiences. It is possible that humor and laughter can both help individuals cope with past

stressful events and prepare them for coping with upcoming stressful events. In Study 1,

the laughter was presented after the stressor had occurred. Thus, we tested whether

laughter could help individuals cope with a past event. Although, in this context, humor

and laughter failed to help individuals cope with a stressor, it raised the question of whether

humor and laughter might have a positive influence on coping if it precedes a stressor. The


positive effects of the humor manipulation could carry over to influence the stressor. In

other words, experiencing humor could cause individuals to experience the later stressor

less negatively. The paradigm in Study 2 provided a second and slightly different test of

humor and laughter's influence on the stress response.

Study 2 was based on the idea that humor may be able to help individuals cope with

an anticipated stressor. The paradigm in Study 2 was similar to a study by Yovetich et al.

(1990), in which participants anticipated a shock. During the anticipation period,

participants listened to a humorous tape, neutral tape, or no tape. Results showed that

participants in the humorous tape condition rated themselves as less anxious and stressed as

the shock approached. Study 2 was different from this earlier study because in addition to

exploring the effects of humor on the anticipation of a stressor, it also explored the effects

of humor during an actual stressor and immediately after.


METHOD

Participants

Twenty-seven college students participated in this study. The sample included 13

females and 14 males. The mean age of participants was 20.56 (SD =1.70), and 70.4%

were Caucasian, 18.5% were Black/African American, 11.1% were other ethnicity.

Design

This study used a within subjects 2 (Funny/Neutral Clip) x 3 (Trials) design, with a

randomized presentation order.

Procedure

Students again came into the laboratory to complete a study on “mood and

physiology.” After they signed a consent form, an experimenter informed participants that

the goal of the study was to understand how positive and negative feelings influence mood

and physiological responses. The experimenter explained the task, telling participants that

they would be continuously monitoring their mood while listening to a series of sounds.

We told participants that "we are interested in testing the physiological differences between

sounds that are meaningful, like a story, and sounds that are meaningless, like a buzzing

noise." The experimenter also told participants that in addition to the sounds, there would

be quiet periods during the task. Participants were instructed to sit back and relax during

these quiet periods, but to continue to monitor their mood. After explaining the task, the

experimenter attached sensors to the participant to measure zygomatic, corrugator, and skin

conductance activity. The experimenter also showed participants how to use the joystick to

continuously record their feelings during the task. Finally, the experimenter completed a

sensor check to confirm that all sensors were working properly. When the task was


41

finished, all participants completed a series of questionnaires assessing sense of humor,

trait anxiety, trait neuroticism, and trait optimism.

Task

Participants completed six trials, with random exposure to each condition every two

trials. Approximately 10 seconds before participants went through each trial, they read, on

the computer, “In the next minute, you may or may not hear a burst of noise.” We thought

that this prime might increase the anticipation, and thus stressful experience of the

upcoming aversive noise. During the first interval of each trial, participants heard a funny

or neutral clip for 60 seconds, followed by an aversive noise for 15 seconds (Interval 2).

Following the second interval, there was a Recovery Interval for 60 seconds which served

as an inter-trial resting period. During this time, participants continued to record their

mood. Table 2 presents the timeline of each trial.

Table 2. Example of Trial 1 from Study 2________________________________________

Anticipation primeInterval 1

(Funny or Neutral clip)

Interval 2 (Aversive Noise) Recovery

10 seconds 60 seconds 15 seconds 60 seconds

Throughout the course of the study, participants heard 3 different funny clips, 3

different neutral clips, and 3 different aversive noises. The clips within each trial were

randomized. The total time for the listening task was approximately 18 minutes (this

included three minutes of baseline).

Aversive noise. The noise was created by combining lOOd of white noise with lOOd

of microphone feedback and lOOd of static.


42

Funny clips. The funny clips included excerpts from “Jerry Seinfeld Live on

Broadway.”

Neutral clips. These clips included excerpts from the audio book, “Orchard.”


Similar to Study 1, the HSQ (Martin et al., 2003), and the CHS (Martin & Lefcourt,

1983) were used to asses trait sense of humor. Also, we again used the IPIP (2001) scales

to assess anxiety and neuroticism, and the LOT-R (Scheier et al., 1994) to assess optimism.

Dependent Measures

Mood and physiological responses were monitored during a 3-minute baseline

period and the 15-minute sound presentation. Similar to Study 1, participants in Study 2

continuously recorded their mood, and we monitored facial EMG and SC levels. Data

were recorded by the BioPac MP150 system and stored via BioPac’s AcqKnowledge

software. The measures used the same sampling rates, and the data were reduced using the

same procedures outlined in Study 1.

Hypotheses

First, we expected participants to show more zygomatic activity when they heard

the funny clips compared to the neutral clips. Because in Study 1, we found that

contagious laughter produced higher SC levels than neutral clips, we were uncertain about

what might happen for SC in Study 2. On the one hand, the findings showed that laughter

was arousing, but on the other hand, our manipulation in Study 1 may have been

problematic. After all, an experimenter detected laughter from only one person in Study 1,

and participants did not report significantly more positive moods during the laughter clips

compared to the neutral clips. For these reasons, we did not have clear expectations about


how humor and laughter would influence participants' SC levels in Study 2. However, if

humor and laughter are indeed protective, then, during Interval 2 (i.e., Stressor) and

Recovery, participants should show lower SC levels when they hear the funny clips instead

of the neutral clips in Interval 1. As for mood, we expected that participants would report

more positive moods when they heard the funny compared to the neutral clips. Also, if

humor and laughter are indeed protective, then, during Interval 2 and Recovery,

participants should report less negative affect when they had hear the funny clips compared

to the neutral clips in Interval 1.


RESULTS

Manipulation Check

At the end of the study, we asked participants about the funniness of the Seinfeld

clips. Participants responded on a 1 (Not at all funny) to 7 (Very funny) scale, and the mean

response was 5.11 (SD = 1.58). Approximately 84% of the participants rated the funny

clips as a “4” or higher. We also asked participants about the funniness of the neutral clips,

and the mean response was 1.85 (SD = 1.49). This difference was significant, t (1, 26) =

8.59, p < .001, and showed that participants found the funny clips more humorous than the

neutral clips. Next, we asked participants to report how distracting it was to listen to each

type of clip. Participants responded on a 1 (Not at all distracting) to 7 (Extremely

distracting) scale. The mean response for the funny clips was a 1.96 (SD = 1.48), and the

mean response for the neutral clips was a 3.22 (SD = 1.55). This difference was

significant, t (1, 26) = 3.53, p < .01, and showed that participants found it more distracting

to listen to the neutral clips.

Finally, we asked participants to report how many times they laughed during the

study. Only one participant reported that he did not laugh at all during the study. The rest

of the participants (n = 26) reported that they laughed at least a little under their breath

(59.3%), laughed out loud once or twice (25.9%), or laughed out loud more than once or

twice (11.0%). In addition to the self-report, an experimenter recorded laughter that

occurred during the study, and found that 19 of the 27 participants laughed out loud at least

once.


45

Analyses

To analyze the results, we used repeated measures ANOVAS. For each dependent

variable, we conducted three separate repeated measures ANOVAS, focusing on

consecutive 5-second periods: one for Interval 1 (Periods 1-12), one for Interval 2 (Periods

13-15), and one for Recovery (Periods 16-27). For Periods 1-12 and 16-27, the repeated

measures analysis was a 2 (Funny/Neutral Clip) x 3 (Trials) x 12 (Periods within Intervals)

design. For Periods 13-15, the repeated measures analysis was a 2 (Funny/Neutral Clip) x

3 (Trials) x 3 (Periods within Intervals) design. If, for any one of the dependent variables,

the analysis failed to produce a meaningful interaction with either Trials or Periods, then

we collapsed across the interval for analysis purposes.

Because the analysis for each interval (i.e., Interval 1, Interval 2, or Recovery)

produced a total of 21 possible effects, we report only the significant interactions that were

relevant to the manipulation. For the interactions that were significant, but unrelated to the

manipulation, we report the findings in Appendix J. For each dependent measure, a figure

is displayed that presents all of the data, collapsed across individuals and trials, for all 27

periods.

Zygomatic-Interval 1 (Periods 1-12). During this interval, we expected that

participants would show more zygomatic activity when they listened to the funny clips

compared to when they listened to the neutral clips. As Figure 10 shows, during the first

interval, the funny clips produced more zygomatic activity than the neutral clips (Ms = .17

vs. .13) and this difference was significant, F (1, 20) = 11.01,/? < .01. A main effect of

Periods showed that, overall, activity tended to stay the same or increase slightly, F (11,

220) = 2.10,/? = .02. A closer look showed that, across periods, zygomatic activity failed


to change for the neutral clips but increased for the funny clips. This pattern of results

produced a significant Clips x Periods interaction, F (11, 220) = 3.31,/? < .001.

interval 1 Interval 2 Recovery

0.15

0.25

0.2-— ON— LN

0.1

0.05

01 3 5 7 9 11 13 15 17 19 21 23 25 27

Periods

Figure 10. Study 2 zygomatic activity by condition collapsed across trials.Note. For all figures, ON = Neutral Clip followed by Aversive Noise; and LN = Funny Clip followed by Aversive Noise.

Zygomatic-Interval 2 (Periods 13-15). Because participants heard only aversive

noise during Interval 2, we did not expect to see differences in zygomatic activity. As

Figure 9 shows, the noises produced equal amounts of zygomatic activity (Ms = .15 and

.16). Results showed that activity decreased across Periods, F (2, 48) = 11.86,/? < .001.

Zygomatic-Recovery (Periods 16-27). During Recovery, there were no differences

in zygomatic activity, (Ms = .13 and .14), F ( l , 24) = 1.41,/? = .25, showing that

differences in activity that occurred during the first interval failed to carry over to the

Recovery Interval.


47

Corrugator-Interval 1 (Periods 1-12). As can be seen from Figure 11, there were

no differences between the funny and the neutral clips for corrugator activity (Ms = .15 and

.16). There were also no significant main effects or interactions during the first interval.

interval 1 Interval 2 Recovery

0.5

0.4ONLN0.3

0.2

Periods

Figure 11. Study 2 corrugator activity by condition collapsed across trials.

Corrugator-Interval 2 (Periods 13-15). Consistent with expectations, the second

interval produced more corrugator activity than the first (Overall Ms = .28 vs. .15). We did

not expect to see differences in corrugator activity during the second interval because

participants only heard the aversive noise during this interval. During this interval, there

was amain effect of Periods, F (2, 48) = 12.62,/? < .001, which showed that corrugator

activity faded as periods continued.

Corrugator-Recovery (Periods 16-27). Results for Recovery revealed only a

significant main effect of Periods, F (11, 264) = 1.80,/? = .05, showing that overall

corrugator activity declined over time.


Peak Skin Conductance2-Interval 1 (Periods 1-12). Based on the findings from

Study 1, we expected that the funny clips might produce more SC activity than the neutral

clips. As can be seen from Figure 12, this expectation was supported. The funny clips

produced higher SC levels than the neutral clips (Ms = 10.19 vs. 9.99), although this

finding was weak, F (1, 24) = 3.32,p = .08. Additionally, there was a main effect of

Periods, F (11, 264) = 10.14,/? < .001, and a significant Clips x Periods interaction, F (11,

264) = 2.88,p = .001. The latter effect indicated that, across periods, SC activity declined

for the neutral clips, but it stayed about the same for the funny clips.


10.610.510.410.310.210.1

ONLN

9.99.89.79.69.5

Periods

Figure 12. Study 2 SC activity by condition collapsed across trials.

2 The only differences between the findings for Peak SC and SC change were that the

weak main effects observed for the funny clips during Intervals 1 and 2 disappear for the

SC change measure. The significant Clips x Periods interactions during Intervals 1 and 2

remain significant in SC change.


Peak Skin Conductance-Interval 2 (Periods 13-15). Consistent with expectations,

the second interval produced higher SC levels than the first (Overall Ms =10.31 vs. 10.09).

Importantly, there was no benefit of listening to the funny clips during Interval 1; aversive

noises that came after the funny clips actually produced higher amounts of SC activity than

aversive noises that came after the neutral clips (Ms = 10.39 vs. 10.23), but this effect was

weak, F (1, 24) = 3.34,/? = .08. Results during this second interval also revealed a main

effect of Periods, F (2, 48) = 11.26 ,p < .001, indicating that activity increased (highest

during period 14) but then declined (Figure 12). In addition, a significant Clips x Periods

interaction, F (2, 48) = 4.16,/? = .02, indicated that across periods, SC activity increased

slightly when participants had just listened to the funny clips, and increased to a greater

extent when participants had just listened to the neutral clips. After this increase, SC levels

began to decline.

Peak Skin Conductance-Recovery (Periods 16-27). During Recovery, we expected

that participants would show lower SC levels when they had heard the funny clips instead

of the neutral clips in Interval 1. However, there were no differences between SC levels

during Recovery (Ms = 9.97 and 10.10), F (1, 24) = 2.04,/? = .17. Thus, listening to the

funny clips in Interval 1 failed to reduce arousal one minute later. Also, SC activity

declined across Periods, F (11, 264) = 11.79,/? < .001.

Peak Mood-Interval 1 (Periods 1-12). Consistent with expectations, during Interval

1, the funny clips produced more positive moods than the neutral clips (Ms = 4.09 vs.

2.41), F (1, 24) = 34.37,/? < .001 (Figure 13). Results also showed a main effect of

Periods, F (11, 264) = 44.24, /? < .001, such that mood improved as time continued. Also,

as periods continued, the moods of participants who listened to the neutral clips increased


50

slightly, suggesting that the clips were interesting, but the moods of participants who

listened to the funny clips increased to a much greater extent. This pattern produced a

significant Clips x Periods interaction, F (11, 264) = 14.43,/? < .001.

Figure 13. Study 2 peak mood reports by condition collapsed across trials.

Peak Mood-Interval 2 (Periods 13-15). Although we expected that the aversive

noise during Interval 2 would produce negative moods, we believed that type of clip heard

in Interval 1 would interact with the noise to influence mood. We expected that

participants would report less negative moods during this stressful interval when they had

heard the funny clips instead of the neutral clips in Interval 1. Results supported this

hypothesis with a significant main effect of Clips: Participants who had just finished

listening to the neutral clips reported worse moods than those who had just finished

listening to the funny clips (-1.59 and -.52), F ( l , 24) = 7.66,/? = .01. A large Periods main

effect, F (2, 48) = 58.66, /? < .001 showed that mood worsened across periods, but this

main effect was qualified by a significant Clips x Periods interaction, F (2, 48) = 6.78,


-4

Periods


p < .01. The interaction indicated that participants' moods became more negative and at a

faster rate when they listened to neutral clips during Interval 1 compared to funny clips

(Figure 13). Follow-up simple main effects analyses showed that participants reported less

negative moods during all three periods of the aversive noise when they had heard funny

rather than neutral clips in Interval 1.

PeakMood-Recovery (Periods 16-27). We hypothesized that mood effects would

carry over to Recovery such that participants would be in more positive moods when they

had listened to the funny clips during Interval 1 instead of the neutral clips. Consistent

with expectations, participants reported less negative moods during Recovery when they

heard a funny clip followed by an aversive noise compared to a neutral clip followed by an

aversive noise (Ms = 1.01 and .42), and this effect was significant, F (1, 24) = 7.33,/? = .01.

Results also revealed a main effect of Periods, F (11, 264) = 21.14,/? < .001, showing that

mood improved across time. This main effect was qualified by a Clips x Periods

interaction, F (11, 264) = 3.32,/? < .001, showing that participants’ moods improved at

faster rates when they had listened to funny clips before the aversive noise compared to

when they had listened to neutral clips before the aversive noise.

Peak Mood change. One could argue, based on Figure 13, that the differences in

mood observed during Interval 2 (i.e., noise) and Recovery were simply a function of the

differences observed in Interval 1. In other words, maybe humor and laughter did not

protect individuals during the Noise and Recovery Intervals, but self-reported mood had

not had the chance to return to “normal.” To test this idea, we computed mood change

scores, in which we a) subtracted the average mood value of the last 5 seconds of Interval 1

from the peak values of each five-second period in Interval 2, and b) subtracted the last 5


seconds of Interval 2 from the peak values of each five-second period in Recovery. Then,

we re-ran the repeated measures ANOVAS. The differences observed during Interval 2

remained significant even after controlling for the last 5 seconds of Interval 1. The analysis

showed that mood changed significantly more when participants had just listened to the

funny compared to the neutral clips, F (1, 24) = 18.57, p < .001 (Figure 14). Although the

change score was significantly greater when participants had just listened to the funny

clips, participants were still reporting significantly less negative moods after they had just

heard the funny clips. We failed to find the same pattern for Recovery: The analysis

showed that the differences observed during Recovery became non-significant after

controlling for the last 5 seconds of Interval 2. Thus, the degree of mood change was

similar when participants listened to both types of clips, F < 1. Although we failed to find

the difference in Recovery, the change analysis for Interval 2 (i.e., the stressor) suggests

that there was a genuine influence of the funny clips that carried through to influence the

stressor.

Moderating Analyses

As for Study 1, we conducted moderator analyses to determine if gender or various

trait measures interacted with the humor manipulation to influence the dependent variables.

Analyses showed that gender failed to interact with the funny clips to influence any of the

dependent variables.

To test for interactions with the trait measures, we conducted median splits on these

measures and then re-ran the ANOVAs. Analyses showed that coping humor interacted

with the funny clips to influence mood during Interval 1, F (1, 23) = 5.90, p = .02, Interval

2, F (1, 23) = 5.1%,p = .03, and Recovery, F (1, 23) = 5.19,p = .03. The interactions


53

Interval 2 Recovery

ON

LN

Periods

Figure 14. Study 2 mood change by condition collapsed across trials.

showed the same pattern during each interval; individuals who scored higher on a measure

of coping humor always reported better moods when they heard the funny compared to the

neutral clips. However, individuals who scored lower on coping humor reported no change

or worse moods when they heard the funny compared to the neutral clips. In addition,

scores on aggressive humor interacted with the funny clips to influence corrugator activity.

The interaction was weak during Interval 2, F (1, 23) = 3.25, p = .09, but significant during

Recovery, F (1, 23) = 4.20, p = .05. Both interactions showed the same pattern:

Participants who scored high on aggressive humor showed more corrugator activity when

they had heard the funny compared to the neutral clips. However, participants who scored

low showed less corrugator activity when they had heard the funny compared to the neutral

clips. Finally, scores on self-enhancing humor interacted with the funny clips to influence

corrugator activity. The interaction was significant during Interval 2, F (1, 23) = 4.97,/? =

.04, and showed that participants who scored high on self-defeating humor showed more


54

corrugator activity after they heard the funny compared to the neutral clips. Those who

scored low showed less corrugator activity after they heard the funny compared to the

neutral clips. The other trait measures failed to interact with clips to influence mood or

physiology (all ps > .10).


DISCUSSION

Study 2 provided a second test of the hypothesis that humor may help individuals

recover from a stressful experience. Based on the findings from Study 1, we modified

Study 2 in three ways. First, we changed the humor manipulation from clips of contagious

laughter in Study 1 to clips of a Jerry Seinfeld stand-up routine. The change was a success;

the results suggested that the funny clips in Study 2 produced more amusement than the

contagious laughter clips in Study 1. For example, the manipulation in Study 2 produced

stronger zygomatic activity than the manipulation in Study 1 (Ms = .17 vs. .10). Also, an

experimenter detected laughter from 19 participants in Study 2. In comparison, only one

person laughed out loud in Study 1. Participants also reported more positive moods while

listening to the funny clips in Study 2 compared to listening to the contagious laughter clips

in Study 1 (Ms = 4.09 vs. 1.36).

A second change made in Study 2 related to the stressor manipulation. Because the

white noise in Study 1 produced the highest SC levels, we used variations of this noise for

Study 2. Results indicated that the noise in both studies produced nearly equal amounts of

corrugator activity (Overall Ms = .25 vs. .28). Unexpectedly, the noise in Study 1 elicited

higher levels of SC activity (Ms = 10.72 vs. 10.31) and poorer moods (Ms = -2.73 vs. -

1.05) than the noise in Study 2. This was a surprising finding given that the manipulation

in Study 2 was the noise that tended to produce the most SC activity in Study 1. It is

important to note that, although the stressor in Study 2 was not as strong as the stressor in

Study 1, it was still successful. For example, corrugator activity was higher in Interval 2

(i.e., noise) than Interval 1 (i.e., funny or neutral clips). Also, the noise produced worse

moods than the funny or neutral clips.


5 6

A third change in Study 2 was that instead of the humor manipulation coming after

the stressor (Study 1), the humor manipulation came before the stressor. It is possible that

the difference in SC levels between Studies 1 and 2 was a result of changing the placement

of the noise. Participants were less physiologically aroused by the noise when it came after

rather than before a funny or neutral clip.

Although the funny clips in Study 2 produced stronger zygomatic activity than the

neutral clips, the effect did not carry over to the other intervals. The funny clips produced

higher SC levels than the neutral clips and they failed to reduce SC activity during either

the stressor (i.e., Interval 2) or the Recovery Interval. Unlike the findings for SC, the

findings for mood indicated that humor and laughter were beneficial. The funny clips

produced more positive moods than the neutral clips, and this strong mood effect carried

over through Interval 2 and Recovery. In other words, after participants heard the funny

clips, they reported a better mood during the stressor and afterward.

Consistent with Study 1 findings, the funny clips in Study 2 produced higher SC

levels than the neutral clips. A Clips x Periods interaction during Interval 2 showed that

there was a jump in activity when participants heard the aversive noise. This jump was

higher when participants had just listened to the neutral clip compared to the funny clip.

However, analyses for SC change showed that the jump was about equal; it only looked

higher for participants when they heard the neutral clips because they started out lower

during Interval 1. Skin conductance activity began to decline after the initial jump, and this

decline occurred regardless of what participants had just listened to in Interval 1. Together

the results for SC change and SC peak showed that experiencing humor and laughter before

a stressor failed to reduce SC levels relative to experiencing a neutral state.


Consistent with expectations, the funny clips produced significantly more positive

moods than the neutral clips. Although the noise in Interval 2 produced negative moods,

the extent of that negativity depended on what participants heard in Interval 1. Participants

reported less negative moods during the noise interval if they had listened to the funny clips

instead of the neutral clips in Interval 1. This finding suggests that laughter served as a

buffer against the noise. Importantly, during the Recovery Interval, participants also

reported more positive moods if they had heard the funny clips instead of the neutral clips

in Interval 1. This finding suggests that the positive effects of laughter were strong; they

carried over and were present even after a stressor had occurred. The mood differences in

Interval 2 were not simply due to the differences in Interval 1; a mood change score

produced the same significant result. This finding for mood during the Noise Interval

suggests that humor and laughter can improve coping with a stressful experience.

One of the biggest differences between the results for Studies 1 and 2 relates to the

mood findings. Based on Study 2 findings, we could conclude that laughter, at least with

respect to mood, protects individuals from a stressor. However, this conclusion would be

inconsistent with the mood findings from Study 1. In Study 1, clips of contagious laughter

only improved mood when they came after a neutral clip, not after the stressor. Two

explanations may account for the differences in mood findings between the two studies.

First, it is possible that the differences relate to the placement of the humor manipulation in

the two studies. In Study 1, the contagious laughter clips came after the stressor, and in

Study 2, funny clips came before the stressor. It may be that humor is more likely to help

individuals cope with upcoming stressors instead of stressors that have already passed. It

could also be the case that in an experimental context, humor will only help with a stressor


58

that has already passed if the humor can be related to that stressor (e.g., Newman & Stone,

1996). In other words, if one can use humor to cognitively reframe a stressor that has

already passed, the humor will be helpful. Otherwise unrelated humor can be beneficial for

an upcoming stressor, as Study 2 has demonstrated.

A second explanation for the different mood findings in Studies 1 and 2 relates to

how we manipulated humor in Study 1. The contagious laughter clips in Study 1 may have

failed to improve mood because they were not all that humorous. After all, the clips only

produced a weak zygomatic finding. The only problem with this explanation is that the

laughter clips did produce more positive moods when they followed a neutral clip, a

finding which suggests that the laughter clips were not totally unsuccessful. Nonetheless, it

would be worthwhile in future research to replicate Study 1 using the funny clips from

Study 2.

Using similar designs to Study 2, others have investigated the effects of humor on

an upcoming stressful experience. In the study by Yovetich et al. (1990), participants who

listened to humorous tapes rated themselves as less anxious as a shock approached. Unlike

Study 2, the aversive event in this other study never occurred; participants never received a

shock. Study 2 tested if humor could help individuals cope during an actual stressful event

and after. We found that listening to funny clips helped individuals cope with respect to

mood, but not SC. Similar to Yovetich and others, we failed to answer the question of why

humor buffers mood; we only showed that it could.

In both studies, we found that laughter was arousing. Compared to the neutral clips,

SC activity was elevated during the contagious laughter clips in Study 1 and the funny clips

in Study 2. According to Provine (2000), researchers have often disagreed about whether


humor and laughter increase or decrease physiological arousal. Our findings are

inconsistent with studies that have found that humor may decrease arousal in a stressful

context. For example, in the study by Newman and Stone (1996), participants who were

asked to create a humorous monologue while watching a stressful film showed lower HR,

lower SC levels, and higher skin temperature than participants asked to create a serious

monologue. However, as the authors pointed out, their study lacked a control group so the

findings could be attributed to a stress-inducing effect of creating a serious monologue.

Our studies are different from Newman and Stone’s because we did not ask participants to

use humor to reffame a stressor; we simply explored the effects of humor and laughter on

an unrelated stressor. Unlike Newman and Stone’s study, our studies clearly suggest that

humor and laughter are arousing.

We explained Study 1 with reference to Zillmann's (1971) excitation transfer

theory. The theory could also account for the findings in Study 2. For example,

participants listened to a funny clip, which produced arousal along with a positive mood.

This positive excitement (increased SC levels and positive mood) from Interval 1 carried

over to Interval 2 (i.e., Stressor) and caused participants to experience the stressor less

negatively. Like the contagious laughter clips in Study 1, the funny clips in Study 2 also

produced arousal. Thus, the laughter clips in Study 1 were not arousing because they were

perceived negatively; they were arousing because humor and laughter are arousing.

Finally, one could argue that humor buffered negative mood from the stressor

because it distracted participants. Distraction, however, probably cannot account for our

findings. In Study 2, we asked participants how distracting it was to listen to the neutral

and funny clips. Results showed that participants rated the neutral clips as more distracting


than the funny clips. Although we are unsure why participants found the neutral clips more

distracting, one possibility is that the neutral clips were more difficult to make sense out of.

Unlike the funny clips, there was no “punch-line” to the neutral clips.


STUDY 3

In Study 1, we found that contagious laughter following a stressor failed to improve

coping with the stressor. In Study 2, we found that funny clips before a stressor, at least

with respect to mood, improved coping with the stressor. The purpose of Study 3, then,

was to test if funny clips during a stressor could improve coping. For Study 3, we used a

between-subjects design for two reasons. First, using a between-subjects design might

reduce experimenter demand. Although no one guessed the hypothesis in Studies 1 and 2,

participants guessed that we were studying positive and negative mood in response to

pleasant and unpleasant sounds. A few participants said that they knew when they were

about to hear the different sounds (i.e., positive versus negative). By using a between-

subjects design, we could reduce experimenter demand, and conduct a stronger test of our

hypothesis. A second reason for using a between-subjects design was related to our

stressor manipulation. We changed the stressor in Study 3: Instead of the aversive noise,

we decided to use stressful films that have been shown to produce strong negative emotion

(Gross & Levenson, 1995). The new manipulation was not appropriate for a within-

subjects design.


METHOD

Participants

Forty-nine individuals from the university community participated in this study.

The sample included 28 females and 21 males. The mean age of participants was 30.22

(SD =11.94), and 71.4% were Caucasian, 10.2% were Black/African American, and 18.4%

listed “other” ethnicity.

Design

This study used a between subjects (Funny/Neutral Clip) design.

Procedure

Participants came into the laboratory to complete a study on “mood and

physiology.” After they signed a consent form, an experimenter explained the task, and

told participants that the goal of the study was to examine the physiology behind mixed

feelings or emotions. Participants were told that they would continuously monitor their

mood while listening to audio clips and watching films. We told participants that the audio

clips were not connected to the films and that they were separate sources of feelings. We

asked them to try their best to pay attention to both. After explaining the task, the

experimenter attached sensors to the participant to measure zygomatic, corrugator, SC

activity, and heart rate (HR). The experimenter also showed participants how to use the

joystick to continuously record their feelings during the task. Finally, the experimenter

completed a sensor check to confirm that all sensors were working properly. When the

task was finished, all participants completed a series of questionnaires assessing sense of

humor, trait anxiety, trait neuroticism, and trait optimism.


6 3

Task

Participants were randomly assigned to listen to funny or neutral audio clips while

watching stressful films. The task consisted of a four-minute baseline period, followed by

a three-minute presentation. During baseline, participants were instructed to sit back, relax,

and monitor their mood. During the presentation, participants heard either three funny

clips or three neutral clips, one right after the other. In the first minute of the task,

participants heard the funny or neutral clip, and nothing was displayed on screen. During

the second minute, participants heard another funny or neutral clip, while the first film was

displayed on screen. During the third minute, participants heard a third funny or neutral

clip, while the second film was displayed on screen. After the third minute, there was a

Recovery Interval for 2 minutes, in which we asked participants to sit back, relax, and

continue to monitor their mood. During this Recovery Interval, participants did not hear or

watch anything. Table 3 presents the timeline for the presentation.

Table 3. Timeline of Study 3___________________________________________________

Audio clip 1 (Funny or Neutral)

Audio clip 2 (Funny or Neutral) +

Stressful film 1

Audio clip 3 (Funny or Neutral) +

Stressful film 2Recovery

60 seconds 60 seconds 60 seconds 120 seconds

Stressful films. We used the films, "Bum," and "Amputation," from Gross and

Levenson (1995) to elicit negative emotions. These are silent films (1.03 minutes each)

that depict medical procedures, and they have been shown to produce disgust in other

studies (e.g., Rottenberg, Ray, & Gross, in press).

Funny and neutral clips. We used the same funny clips and neutral clips from

Study 2.


6 4


Similar to Studies 1 and 2, the HSQ (Martin et al., 2003), and the CHS (Martin &

Lefcourt, 1983) were used to asses trait sense of humor. Also, we again used the IP IP

(2001) scales to assess anxiety and neuroticism, and the LOT-R (Scheier et al., 1994) to

assess optimism.

Dependent Measures

Mood and physiological responses were monitored during a 4-minute baseline

period and the 5-minute presentation. Participants in Study 3 continuously recorded their

mood, and we monitored facial EMG and SC levels. In addition, we recorded participants'

heart rate (HR) in Study 3. Data were recorded by the BioPac MP150 system and stored

via BioPac’s AcqKnowledge software.

Because we added a channel for HR, we changed the sampling rates of some of the

other measures. We continued to sample EMG and mood at 1000 hz, but we reduced the

sampling rate of SC to 125 hz, which is still an appropriate sampling rate for this measure

(Dawson et al., 2000). Heart rate was sampled at 250 hz. The data for EMG, SC, and

continuous mood were reduced using the same procedures outlined in Studies 1 and 2.

However, instead of 5-second periods, we averaged across and found the peaks for

10-second periods. For HR, we used a beats per minute calculation, also separated into

10-second periods.

Hypotheses

In the first minute of the task, we expected participants who heard the funny clips to

show more zygomatic activity than participants who heard the neutral clips. Also, based on

the findings from Studies 1 and 2, we expected participants who heard the funny clips to


show higher SC levels than participants who heard the neural clips. Finally, we expected

that, in that first minute, participants who listened to the funny clips would report more

positive affect than participants who listened to the neutral clips.

During the stressful films, we hypothesized that all participants would show

corrugator activity. We expected that participants who listened to the funny clips during

the stressful films might show less corrugator activity than those who listened to the neutral

clips while watching. Also, we expected that participants who listened to the funny clips

during the films would show higher SC levels and report less negative moods than

participants who listened to the neutral clips during the films. This pattern of findings

would fit with Studies 1 and 2, showing that although laughter is physiologically arousing,

it is psychologically calming.


RESULTS

Manipulation Check

At the end of the study, we asked participants about the funniness of the clips they

listened to during the presentation. Participants responded on a 1 (Not at all funny) to 7

(Very funny) scale. Compared to participants who listened to the neutral clips, those who

listened to the funny clips were more likely to report that the clips were funny (Ms = 2.27

vs. 4.56), and this difference was significant, t (43) = 4.81 ,P < .001. We also asked

participants to report how distracting it was to listen to each type of clip. Participants

responded on a 1 (Not at all distracting) to 7 (Extremely distracting) scale. There was no

difference on this measure; participants in the funny condition were just as likely as those

in the neutral condition to rate the clips as distracting (Ms = 3.13 and 2.86). However,

participants in the funny condition were more likely to rate the clips as interesting (Ms =

4.83 and 3.50), a difference which was significant, t (43) = 2.55,/? = .01.

Finally, we asked participants to report how many times they laughed during the

study. They responded by checking one of four possible responses: “I didn’t laugh at all,”

“I laughed a little under my breath,” “I laughed out loud once or twice,” or “I laughed out

loud more than once or twice.” Results showed that participants in the funny condition

reported laughing more than those in the neutral condition (Ms = 1.76 vs. 1.38), and this

difference was significant, t (47) = 2.45,/? = .02. In addition to the self-report, an

experimenter recorded laughter that occurred during the study, and found that 10 of the 25

participants in the funny condition laughed out loud at least once. This result compares to

0 of the 24 participants in the neutral condition.


6 7

Analyses

To analyze the results, we used repeated measures Analysis of Covariance

(ANCOVA), controlling for baseline. For each dependent variable, we conducted five

separate ANCOVAS, focusing on consecutive 10-second periods: one for Interval 1

(Periods 1-6; funny or neutral clip only), one for Interval 2 (Periods 7-12; first stressful

film along with second funny or neutral clip ), one for Interval 3 (Periods 13-18; second

stressful film along with third funny or neutral clip), one for Recovery Interval 1 (Periods

19-24), and one for Recovery Interval 2 (Periods 25-30). For each dependent measure, a

figure is displayed that presents all of the data.

Zygomatic-Interval 1 (Periods 1-6). Compared to those in the neutral condition, we

expected participants in the funny condition to show more zygomatic activity during the

first minute of the task. However, there were no significant differences between conditions

for zygomatic activity during this interval, F < 1 (Figure 15). Additionally, there was no

main effect of Periods and the interaction was not significant, Fs < 1.

Interval 1 Interval 2 Interval 3 Recovery 1 Recovery 20.25

FunnyNeutral0.15 -

0.05 -

Periods

Figure 15. Study 3 zygomatic activity for the funny and neutral conditions.Note. All Study 3 figures represent means that have been adjusted for baseline values.


68

Zygomatic-Interval 2 (Periods 7-12). Results indicated that there were no

significant differences between conditions during Interval 2, F (1, 46) = 2.41,/? = .13.

Also, there was no main effect of Periods, F< 1. However, a significant Condition x

Periods interaction, F (5, 46) = 2.85,p = .02, suggested that zygomatic activity failed to

change for participants who listened to the funny clips, and it increased slightly for

participants who listened to the neutral clips. Follow-up simple main effects analyses

showed that participants in the neutral condition showed significantly more zygomatic

activity than those in the funny condition during all 6 periods, all ps < .01.

Zygomatic-Interval 3 (Periods 13-18). As can be seen from Figure 15, there were

no significant differences between conditions during Interval 3, F < 1. Also, the main

effect of Periods and the Condition x Periods interaction were not significant, Fs < 1.

Zygomatic-Recovery 1 (Periods 19-24). Results revealed no differences between

conditions for zygomatic activity during the first Recovery Interval, F (1, 46) = 1.59, p =

.21. Also, there was no main effect of Periods and the interaction was not significant, ps >

.29.

Zygomatic-Recovery 2 (Periods 25-30). There were no significant differences

between conditions during the second Recovery Interval F (1, 46) = 2.2%, p = .14. In

addition, there was no main effect of Periods, F< 1. However, a significant Condition x

Periods interaction, F (5, 46) = 2.30, p < .05, showed that zygomatic activity declined

slightly for participants who had listened to the funny clips, but failed to change for those

who listened to the neutral clips (Figure 15). Follow-up simple effects analyses indicated

that participants in the neutral condition showed significantly more zygomatic activity than

participants in the funny condition during each of the 6 periods, all ps < .01.


6 9

Corrugator-Interval 1 (Periods 1-6). As can be seen in Figure 16, results indicated

that there were no differences between conditions for corrugator activity during Interval 1,

F < 1. However, a main effect of Periods, F (5, 46) = 2.36, p = .04, showed that activity

increased in the beginning and then declined. The interaction was not significant, F < 1.

Interval 1 Interval 2 Interval 3 Recovery 1 Recovery 2

0.12

FunnyNeutral

0.08

0.06 -

0.04

0.02

Periods

Figure 16. Study 3 corrugator activity for the funny and neutral conditions.

Corrugator-Interval 2 (Periods 7-12). We hypothesized that participants in the

funny condition might show less corrugator activity than participants in the neutral

condition during the stressful films. Results revealed no differences between conditions

during Interval 2, F < 1. There was a main effect of Periods, F (5, 46) = 2.39, p = .04,

which showed that activity started out high and then declined. The interaction was not

significant, F < 1.

Corrugator-Interval 3 (Periods 13-18). Results revealed no significant differences

between conditions for corrugator activity during this interval, F < 1. Also, there was no

main effect of Periods and the interaction was not significant, ps > .35.


70

Corrugator-Recovery 1 (Periods 19-24). There were no differences between

conditions during the first Recovery Interval, F < 1. Also, there was no main effect of

Periods, F < 1. However, there was a weak but non-significant Condition x Periods

interaction, F (5, 46) = 2.11, p = .15, which showed that corrugator activity declined for

participants who listened to the funny clips, but stayed the same or increased for

participants who listened to the neutral clips (Figure 16).

Corrugator-Recovery 2 (Periods 25-30). Results indicated that participants in the

funny condition showed less corrugator activity than participants in the neutral condition

(Ms = .04 vs. .06). However, this small difference was not significant, F (1, 46) = 2.58, p =

.12. There was a Periods main effect, F (5, 46) = 32.17,/? < .001, which showed that

activity increased in the beginning, and then declined. However, the interaction was not

significant, F < 1.

Average Skin Conductance -Interval 1 (Periods 1-6). Based on Studies 1 and 2, we

hypothesized that participants would show higher SC levels while listening to the funny

clips compared to the neutral clips. However, as is illustrated in Figure 17, there were no

differences in SC levels between the two conditions, F< 1. Also, the Periods main effect

was not significant, F (5, 46) = 1.51,/? = .19. A significant Condition x Periods interaction,

F (5, 46) = 2.51,/? = .03, that was somewhat consistent with our hypothesis, showed that

SC levels stayed about the same for participants who listened to the funny clips, but began

to decline for participants who listened to the neutral clips. Simple main effects analyses

3 There were two differences between the findings for Average SC and Peak SC. The

significant Condition x Periods interactions observed for Average SC during Intervals 1

and 2 were marginal in Peak SC findings.


showed that SC levels were not significantly different between conditions for any of the 6

periods, all/?s > .14.

Interval 1 Interval 2 Interval 3 Recovery 1 Recovery 2

10 . 5 - FunnyNeutral10 -

9.5 -

8.5

Periods

Figure 17. Study 3 average SC activity for the funny and neutral conditions.

Average Skin Conductance-Interval 2 (Periods 7-12). Based on the findings from

Studies 1 and 2, we expected that participants in the funny condition, compared to those in

the neutral condition, might show higher SC levels while watching the first film. However,

results revealed that there were no differences between conditions for SC levels during

Interval 2, F (1, 46) = 1.47,/? = .23. There was a main effect of Periods, F (5, 46) = 3.61,/?

< .01, which showed that activity increased in the beginning, but then declined (Figure 17).

Inconsistent with findings from Studies 1 and 2, a significant Condition x Periods

interaction, F (5, 46) = 2.33, /? = .04, showed that SC levels increased to a greater extent for

participants who listened to the neutral clips compared to participants who listened to the

funny clips. After the increase, SC levels declined for participants in both conditions.

Simple main effects analyses showed that during the first period, SC levels were not

significantly different between the two conditions. However, for the remaining five


72

periods, SC levels were significantly higher for participants in the neutral condition

compared to participants in the funny condition, all ps < .04.

Average Skin Conductance-Interval 3 (Periods 13-18). There were no differences

between conditions for SC levels during Interval 3, F (1, 46) = 1.61, p = .21. In addition,

there was no main effect of Periods and the interaction was not significant, ps > .13.

Average Skin Conductance-Recovery 1 (Periods 19-24). There were no differences

between conditions for SC levels during Recovery 1, F (1, 46) = 1.25,p = .27. However,

there was a main effect of Periods, F (5, 46) = 2.83, p = .02, which showed that SC levels

declined across periods. The interaction was not significant, F < 1.

Average Skin Conductance-Recovery 2 (Periods 25-30). Results showed no

significant differences between conditions for SC levels during Recovery 2, F (1, 46) =

1.40,/? = .24. In addition, there was no main effect of Periods and the interaction was not

significant, Fs < 1.

Average Mood4-Interval 1 (Periods 1-6). For this first minute, we expected

participants who listened to the funny clips to report more positive moods than participants

who listened to the neutral clips. Consistent with expectations, results showed that

participants in the funny condition reported more positive moods than those in the neutral

condition (Ms = 2.45 vs. 1.72), and the difference was significant, F (1, 46) = 4.05,/? = .05

(Figure 18). Also, there was a Periods main effect, F (5, 46) = 4.45,/? < .01, which showed

that moods became better over time. A significant Condition x Periods interaction, F (5,

4 There was only one difference between Average mood and Peak mood. The significant

Condition x Periods interaction observed for Average mood during Interval 3 was marginal

in Peak mood findings.


7 3

46) = 2.43, p = .04, showed that moods improved for participants who listened to the funny

clips, but slightly increased and then declined for participants who listened to the neutral

clips. Follow-up simple effects analyses revealed that, during the first four periods, mood

reports did not differ between the two conditions. However, for the remaining two periods,

mood reports were significantly more positive for participants in the funny condition

compared to participants in the neutral condition, all ps < .03.

Interval 1 Interval 3Interval 2 Recovery 1 Recovery 2

FunnyNeutral

25 27

Periods

Figure 18. Study 3 average mood reports for the funny and neutral conditions.

Average Mood-Interval 2 (Periods 7-12). We expected participants in the funny

condition to report less negative moods than participants in the neutral condition during the

first stressful film. However, there were no significant differences between conditions for

average mood during this interval, F < 1. There was a large Periods main effect, F (5, 46)

= 20.23, p < .001, whereby moods became worse over time. The interaction was not

significant, F < 1.

Average Mood-Interval 3 (Periods 13-18). Although participants in the funny

condition reported less negative mood than participants in the neutral condition during this


7 4

interval, (Ms = -.77 vs. -2.03), the main effect was not significant, F (1, 46) = 2.35,p = .13.

Similar to Interval 2, there was a large Periods main effect, F (5, 46) = 6.09, p < .001,

which showed that moods became worse over time. Consistent with expectations, there

was a significant Condition x Periods interaction, F (5, 46) = 2.98, p = .01, which showed

that the moods of those in the neutral condition were worse and became so at a faster rate

than the moods of those in the funny condition (Figure 18). Follow-up simple effects

analyses indicated that, during the first three periods, mood reports did not differ between

the two conditions. However, for the remaining three periods, participants in the funny

condition reported less negative moods than those in the neutral condition, all ps < .01.

Average Mood-Recovery 1 (Periods 19-24). Results showed no significant

differences between conditions for average mood during this interval, F < 1. There was a

Periods main effect, F (5, 46) = 10.58,p < .001, which showed that moods improved over

time. The interaction was not significant,/) = .27.

Average Mood-Recovery 2 (Periods 25-30). Results showed no significant

differences between conditions for average mood during Interval 2, F < 1. There was a

Periods main effect, F (5, 46) = 6.28,/) < .001, which showed that moods improved over

time. The interaction was not significant, F < 1.

Heart Rate-Interval 1 (Periods 1-6). Because we found that humor in Studies 1 and

2 was arousing (with respect to skin conductance), we expected that listening to the funny

clips in Study 3 would increase heart rate relative to listening to the neutral clips. Although

the means for HR during Interval 1 suggested that this was the case (Ms = 85.40 vs. 80.49),

the condition effect was not significant, F ( l , 46) = 1.98,/ = .17 (Figure 19). Also, there

was no main effect of Periods and the interaction was not significant, Fs < 1.


75

Interval 3 Recovery 1 Recovery 2Interval 2nterval 190

85 -

FunnyNeutral

80 -

75 -

70 -

6523 25 27 29

Periods

Figure 19. Study 3 HR for the funny and neutral conditions.

Heart Rate-Interval 2 (Periods 7-12). Results showed no significant differences

between conditions for Interval 2, F < 1. Also, there was no main effect of Periods, p >

.22, and the interaction was not significant, F < 1.

Heart Rate-Interval 3 (Periods 13-18). There were no significant differences

between conditions for HR in Interval 3, F < 1. Also, there was no main effect of Periods,

p > .31, and the interaction was not significant, F < 1.

Heart Rate-Recovery 1 (Periods 19-24). There were no significant differences

between conditions for this interval and no main effect of Periods, Fs < 1. However, there

was a weak Condition x Periods interaction, F (5, 46) = 1.99,p = .08, which showed that

HR was higher and more variable (e.g., it increased, then declined, and then increased

again) for participants who had listened to the neutral clips compared to those who had

listened to the funny clips. The HR of participants who had listened to the funny clips

declined during this interval. Follow-up simple main effects analyses showed that HR


7 6

levels were not significantly different between conditions for any of the 6 periods, all ps >

.68.

Heart Rate-Recovery 2 (Periods 25-30). Results showed no significant differences

between conditions, F < 1. There was a small but non-significant effect of Periods, F (5,

46) = 1.83, jc? = . 11, whereby HR increased slightly, but then stayed about the same. The

interaction was not significant, F < 1.

Moderating Analyses

As for Studies 1 and 2, we conducted moderator analyses to determine if gender or

various trait measures interacted with condition to influence the dependent measures.

Analyses showed that gender interacted with condition to influence heart rate during

Recovery Interval 1, F (1, 44) = 3.92,p — .05. The interaction showed that the women and

men in the neutral condition had similar HR scores during Recovery (Ms -11.53 and

78.89). However, women in the funny condition had higher HR scores than men in the

funny condition (Ms = 82.59 and 72.76).

To test for interactions with the trait measures, we conducted median splits on these

measures and then re-ran the ANCOVAs. First, analyses showed that scores on affiliative

humor interacted with condition to influence HR during Recovery Interval 1, F (1, 44) =

3.22,p = .08, and Recovery Interval 2, F (1, 44) = 3.97,p = .05. Although the first

interaction was weak, they both showed the same pattern. Heart rates of those who scored

high and low on affiliative humor did not differ in the neutral condition, but they differed in

the funny condition. Compared to participants who scored low on affiliative humor, those

who scored high had lower HR readings during Recovery. Analyses also showed that

scores on trait anxiety interacted with condition to influence mood during Recovery


7 7

Interval 1, F ( l , 44) = 8.12,/) = .01, and Recovery Interval 2, F ( l , 44) = 7.40,p = .01.

Both interactions showed the same pattern: Compared to those who were low, those who

were high in anxiety reported better moods when they were in the funny condition. The

pattern was flipped in the neutral condition: Compared to those who were low, those who

were high in anxiety reported worse moods when they were in the neutral condition. The

other trait measures failed to interact with condition to influence mood or physiology (all

ps > .10).


DISCUSSION

In Study 3, we tested whether humor before and during a stressor could improve

coping with the stressor. For the humor manipulation, we used the same funny and neutral

clips from Study 2. We changed the stressor manipulation in Study 3 from clips of

aversive noise to aversive films that have been shown in other studies to produce strong

negative emotion (Rottenberg et al., in press). Results showed that the films were

successful in producing physiological arousal. The films produced greater SC activity than

the noise in Study 2, but a little less activity than the noise in Study 1 (Overall Ms for SC

levels = 10.65, 10.31, and 10.72, respectively).

The findings for zygomatic activity were inconsistent with expectations. First, the

funny clips failed to produce more zygomatic activity than the neutral in the first minute of

the task (before the stressful films started). This finding was inconsistent with the results

from Study 2, which showed that the funny clips produced more zygomatic activity than

the neutral clips. However, it is important to note that the design in Study 2 was a within-

subjects design, and zygomatic activity was collapsed across all three funny clips. In Study

3, participants only heard one funny clip during the first minute. The other 2 funny clips

were paired with the stressful films. It is possible that more than one clip was needed to

elicit zygomatic activity. Indeed, compared to Study 2, in which an experimenter detected

overt laughter from 70% of the participants when they listened to the funny clips, in Study

3, an experimenter detected overt laughter from only 40% of participants in the funny

condition.

A second surprising finding for zygomatic activity in Study 3 was that the neutral

clips produced significantly more zygomatic activity than the funny clips during one of the


7 9

films and one Recovery Interval. Simple main effects analyses showed that zygomatic

activity for the neutral condition was significantly higher for every period during the two

intervals. We are uncertain as to why zygomatic activity was higher during these intervals

for participants in the neutral condition. However, it is the case that zygomatic activity has

been somewhat unpredictable in our studies (e.g., main effect for aversive noise in Study

1).

Consistent with expectations, corrugator activity increased when participants began

to watch the stressful films. However, main effects for Periods showed that participants

adapted to the films (i.e., corrugator activity declined over time). In addition, there was

some evidence that the humor was effective in reducing corrugator activity. For example,

during both Recovery Intervals, there were small effects of less corrugator activity if

participants had been in the funny condition instead of the neutral condition.

The findings for SC during Study 3 were somewhat consistent with the findings

from Studies 1 and 2. For example, SC levels during the first interval (i.e., the first minute

of the task) failed to change for participants in the funny condition, but declined for

participants in the neutral condition. This finding was consistent with the other two studies

because it suggested that the funny clips were physiologically arousing relative to the

neutral clips. However, when the stressful films began, a different picture emerged.

Humor appeared to have a protective effect. For example, during the first stressful film

(i.e., Interval 2), an interaction showed that SC levels increased for participants in the

neutral condition, but did not change for participants in the funny condition. Although the

means were always lower for participants in the funny condition compared to those in the

neutral condition, the interactions were not significant for any of the other later intervals.


8 0

Consistent with expectations, the findings for mood showed that during the first

minute of the task, participants in the funny condition reported more positive moods than

participants in the neutral condition. The interaction and simple main effects analyses

showed that it took participants a few moments to "catch on" to the funny clips. These

findings make sense given that it would take individuals a little time to comprehend what

they were hearing. Although mood for the funny clips did not differ from mood for the

neutral clips during the first stressful film, an interaction during the second stressful film

showed that participants in the funny condition reported less negative moods than those in

the neutral condition. Simple main effects analyses showed that compared to the moods of

participants in the neutral condition, the moods of those in the funny condition became less

negative toward the end of the second film. However, these mood effects did not carry

over to either Recovery Interval.

The findings for SC and mood in Study 3 suggest that humor during a stressful

experience may be somewhat beneficial. Study 3 was both consistent and inconsistent with

Studies 1 and 2. For example, like the findings for Studies 1 and 2, humor when presented

alone was arousing relative to a neutral condition. Also, like Study 2, humor in Study 3

appeared to be helpful with respect to mood. Unlike the findings for the first two studies,

participants in Study 3 experienced lower SC levels when they listened to funny rather than

neutral clips during the stressor. Although these findings for mood and SC were not

consistent across all intervals in Study 3, they were present in 2 of the intervals, one of

which included a stressful film interval.

One of the results for HR suggested a protective effect of humor. An interaction

during Recovery Interval 1 revealed that the HR of participants in the funny condition


81

declined faster than the HR of those in the neutral condition. However, the interaction was

weak {p = .08), and it was not present in any of the other intervals. Because the finding

was not robust, and this study was the only one of our studies to include HR, more research

needs to be conducted before drawing conclusions.

Interestingly, the moderating analyses showed that participants who scored high on

using humor to bond with others (i.e., affiliative humor) were most likely to benefit, with

respect to HR, from being in the funny condition. This finding provided partial support for

Martin's (2001) suggestion, that different types of humor could influence responses to

stress differently. More research should distinguish between types of humor and their

differential influence on responses to stress. Another moderating analysis showed that

humor was most likely to benefit mood for those who were high, rather than low in anxiety.

This was an interesting finding because it suggested that humor may be particularly helpful

for people who are high in anxiety. It is possible that anxious people respond more

positively to humor because they tend to use it more to cope with their anxiety.

Similar to Study 2, distraction cannot account for Study 3 results. We asked

participants how distracting it was to listen to the neutral and funny clips, and results

showed that participants in the neutral condition were just as likely as participants in the

funny condition to rate the clips as distracting. However, compared to participants in the

neutral condition, participants in the funny condition were more likely to rate the clips as

interesting.


GENERAL DISCUSSION

Our studies on humor and laughter are unique from others in several ways. First,

unlike the many correlational studies in which participants recall their past stressful

experiences, we conducted three studies in which participants were exposed to short-term

stressors in the laboratory. Similar to the correlational studies, our investigations showed

that humor and laughter can positively influence mood in response to a stressor and shortly

after the stressor. However, this was the case only if humor came before or during a

stressor, not after. In addition, all three studies indicated that laughter was arousing relative

to a neutral state. However, in Study 3, when we presented humorous clips to participants

while they were watching stressful films, participants experienced lower SC levels than

those who listened to neutral clips. Together, the findings suggest that humor and laughter

before and during a stressor can positively influence mood and that humor during a stressor

can have a positive influence on physiology.

Although participants who listened to the funny clips during the stressor in Study 3

experienced lower SC levels, humor was arousing in the first minute (i.e., Interval 1) of the

task before the stressor was introduced. Thus, one conclusion we can draw from the three

studies was that listening to laughter/funny clips increased SC levels relative to listening to

neutral clips. Skin conductance levels are produced by electrodermal activity (EDA) and

are indicative of sympathetic activation (Dawson et al., 2000). Skin conductance levels can

represent emotion, arousal, or attention, but often, it is difficult to identify the precise

underlying psychological process responsible for EDA changes. According to Dawson and

others, researchers can increase their likelihood of accurate conclusions by conducting

highly controlled laboratory studies in which they have only one aspect of a stimulus


83

change over time. Studies 1 and 2 were experiments in which we varied only the valence

of the stimulus over time. Therefore, we believe that during the aversive noise in Studies 1

and 2, the increased SC levels most likely reflected feelings of anxiety.

One question that stems from the findings of the three studies is why humor by

itself was arousing. At first, we believed that the arousal in our studies was simply the

result of the physical characteristics associated with overt laughter, but humor without

overt laughter was also arousing. For example, in Study 1, only one person laughed out

loud, but results showed that SC levels were higher when participants listened to the

laughter as opposed to the neutral clips. These findings suggested that simply hearing

laughter produced arousal. Perhaps this was the case because unlike the neutral clips, the

laughter/funny clips were more likely to elicit strong feelings, whether that was negative

feelings (Study 1) or positive feelings (Studies 2 and 3). The experience of strong feelings

compared to a neutral state led to increased arousal.

Our findings are both consistent and inconsistent with other studies. We found that,

for the most part humor and laughter were arousing. This finding fits with research in

which being amused has failed to have a positive influence. Yovetich et al. (1990) told

participants that they were about to receive a shock, and then had them listen to humorous

tapes or do nothing. Similarly, Danzer et al. (1990) showed participants slides to induce

depressive moods, and then afterward, had them listen to humorous or neutral tapes.

Results for both of these studies indicated that individuals who listened to humorous tapes

had increased HR levels compared to individuals who listened to neutral tapes or nothing.

Thus, these finding suggested that, at least with respect to HR, humor was arousing. The


84

findings also fit with Elliot (2005), who had participants watch a funny or stressful movie

for 20 minutes and failed to find a physiological benefit of laughter.

The finding in Study 3 that humor can have a positive physiological influence

during a stressor, fits somewhat with the study by Newman and Stone (1996). As

discussed earlier, these authors found that participants who created humorous monologues

while watching a stressful film had lower HR, SC levels and higher body temperature than

participants who created a serious monologue while watching. Similarly, in Study 3, we

found that simply listening to humorous clips while watching a stressful film lowered SC

levels relative to listening to neutral clips. Although our findings in Study 3 were

consistent with Newman and Stone's findings, there is some difficulty in comparing the two

studies. For example, the humor in their study was used to reinterpret the stressful situation

whereas the humor in Study 3 was presented as separate and unrelated to the stressors.

It is possible that we failed to uncover robust physiological benefits of humor and

laughter in all three studies primarily because the humor manipulation was unrelated to the

stressor. For example, in the study by Newman and Stone (1996), the authors found that

humor created by participants to re-interpret a stressful situation benefited their physiology.

In contrast, in the study by Yovetich and others (1990), the authors found that listening to

humorous tapes that were unrelated to a stressor failed to benefit physiology while

participants were anticipating a stressor. The issue of whether humor and laughter need to

be relevant to a stressor to benefit physiology is important. Take for example, people who

regularly participate in "Laughter therapy," a therapy which is based on the idea that

laughter, without experiencing humor or relating the laughter to anything, can benefit

physiology and reduce feeling stressed (Provine, 2000). As far as we know, no studies


85

have investigated the physiological effects of laughter therapy. The latter might be an

interesting field study that could provide insight into the question of whether laughter needs

to be relevant to stressors to have a positive physiological influence.

It is also important to recognize that we only found a positive physiological

influence of humor in one of our studies. Why was humor presented during a stressor more

effective in reducing SC levels than humor presented before or after a stressor? Several

alternative explanations could account for these findings. First, it could be that by

presenting the clips at the same time as the stressor, participants were allowed to focus less

on the stressful film and more on the funny clips. This idea is consistent with Newman and

Stone (1996) who concluded that humor allows people to distance themselves from a

stressful situation. It is possible that this distancing property of humor will only benefit

physiology when it occurs concurrently with a stressor. The idea of distancing is different

from the idea of distraction. Participants in the funny condition did not rate the clips as

more distracting than participants in the neutral condition. Participants in the funny

condition, however, rated the clips as more interesting than participants in the neutral

condition. Perhaps the funny clips interested participants to the point that they chose to pay

attention to the clips instead of the stressful films. Based on all three studies, we can

conclude that humor is arousing relative to a neutral state, but when one is exposed to

humor during a stressor, that humor can overpower a stressful experience.

The failure to uncover physiological benefits in Studies 1 and 2 cannot be explained

with reference to the short term nature of the studies. For example, Scheff and Bushnell

(1984) argued that the timing of physiological assessment is an important factor in finding

physiological benefits of laughter. They suggested that laughter produces arousal in the


8 6

beginning, but then later produces a relaxation effect. However, in Study 3, we found that

participants who listened to funny clips during a stressor experienced physiological benefit

immediately (i.e., lower SC levels during the stressor). The immediate benefit is consistent

with a recent study by Miller and others (2006), who found positive physiological changes

one minute after participants watched scenes from a funny movie.

In contrast to the physiological findings, the results for mood in Studies 2 and 3

revealed strong, consistent evidence of a buffering effect of humor. In Study 2, listening to

funny clips before a stressor produced less negative moods during the stressor and after the

stressor had passed. In Study 3, participants who listened to funny clips during a stressor

reported less negative moods than participants who listened to neutral clips during a

stressor. It is important to note that we obtained these positive effects for a study with a

within-subjects and a study with a between-subjects design, strengthening confidence in the

overall effect. The findings for mood are also consistent with a number of correlational

studies (Abel, 2002; Martin & Lefcourt, 1983; Nezu et al., 1988) that have shown that

sense of humor may buffer negative mood from stressful life events. Future studies should

address the question of why situational humor has a positive influence on mood in a

stressful situation.

One reason humor may buffer mood in stressful situations is that it provides

individuals with a "release." Freud (1928, 1960) believed that humor provided relief from

negative emotions, and he suggested that laughter allowed individuals to release “pent-up”

tension. Indeed, evidence of such "comic relief' is ubiquitous in the entertainment

industry. According to King (2003), writers typically use comedy to relieve tension they

have built up in their stories, television shows, and movies. She argues that a small amount


87

of humor is often included in dramatic shows because the audience needs relief from the

tense and serious events that are occurring. Although relief theory could be used to explain

the findings in Studies 2 and 3, it cannot explain the lack of mood effects in Study 1. Study

1 was an ideal context for relief theory to receive support; participants were presented with

an aversive noise which was followed by clips of contagious laughter. The laughter clips

should have allowed participants to "release the tension," but instead they seemed to make

it worse. Participants reported worse moods when they heard the laughter clips after the

noise compared to when they heard the neutral clips after the noise. Because relief theory

was not supported by Study 1 findings, we find it difficult to use the theory to explain the

findings from Studies 2 and 3.

It is a possibility, however, that relief theory failed to receive support in Study 1

because the contagious laughter clips were unsuccessful in producing humor and laughter.

A logical next step with this group of studies would be to repeat Study 1 using the one-

minute humorous clips from Studies 2 and 3. The clips from Studies 2 and 3 produced

more amusement and laughter than the clips from Study 1. If we would repeat Study 1

using these clips and still not find a mood buffer effect, then relief theory could be ruled

out.

Aside from relief theory, another explanation for the mood buffering effect of

humor relates to research on positive emotions. Recent research suggests that humor may

elicit other positive emotions, which then can have the effect of reducing the amount of

negative affect in a stressful situation. Vilaythong, Amau, Rosen, and Mascaro (2003)

found, for example, that humor increases hopefulness. Compared to a control group,

participants who watched 15 minutes of a funny video increased in their scores on a


8 8

hopefulness measure. The authors argued that humor may inhibit negative thoughts and

encourage positive ones. This is an interesting idea that could easily be tested by having

individuals complete a thought listing task following a humor manipulation.

It is unclear whether humor generates other positive emotions or whether humor

itself is a positive emotion. If humor can be defined as a positive emotion, then our studies

fit with other research on positive emotions. This research has shown that positive

emotions have the power to "undo" the effects of negative emotions. Fredrickson and

Levenson (1998), for example, showed all of their participants a fear-eliciting film, and

then followed it with a positive, sad, or neutral film. Results showed that participants who

viewed the positive films instead of the neutral or sad films were quicker to return to

baseline cardiovascular levels after the fearful film. In a second study, the authors

videotaped participants while they watched a sad film, and found that participants who

smiled at least once were faster to return to baseline cardiovascular levels after the film

ended.

More evidence that positive emotions can undo the effects of negative emotions

comes from Fredrickson, Mancuso, Branigan, & Tugade (2000). The authors made

participants feel anxious by telling them they were going to give a speech in a few minutes.

Although no one gave a speech, participants reported negative affect and exhibited

cardiovascular reactivity. Afterward, individuals watched a film clip that elicited sadness,

contentment, amusement, or neutral feelings. Findings indicated that those who viewed the

positive films (contentment and amusement) showed less cardiovascular reactivity after the

film (e.g., less HR acceleration, less of an increase in SBP and DBP, less peripheral

vasoconstriction).


Finally, one other possible reason that humor buffered mood in Studies 2 and 3 may

be that individuals purposefully used the humor to manage their moods. Meadowcroft and

Zillmann (1987) provided evidence that individuals use humor to regulate their moods.

They asked women what type of shows they would prefer to watch if they had 3 hours free.

They found that premenstrual and menstrual women were more likely to choose comedies

than women who were midway through their cycle. The authors argued that the

premenstrual and menstrual women were more likely to choose comedies because they

wanted to alleviate the negative moods they were experiencing due to the hormonal

changes.

Limitations

One limitation of these studies was that we were unable to determine which specific

emotion was occurring when SC levels were elevated. However, this is a problem for most

researchers who investigate sympathetic activation relying on EDA alone. According to

Dawson and others (2000), SC responses to stimuli are typically considered to be part of

the orienting response, unless they are responses to aversive stimuli. Because our studies

involved the presentation of only one type of stimuli at a time, we believe that EDA during

the aversive stimuli most likely represented anxiety, whereas EDA during the funny and

neutral stimuli most likely represented attention and/or orienting. In Study 3, however, we

presented stimuli concurrently, and thus it is more difficult to say what psychological

process was occurring.

We have two suggestions for improving knowledge about processes underlying

EDA and sympathetic activation. First, brain imaging techniques could provide better

insight into the process. For example, when EDA is linked to amygdala activation, it


9 0

represents affective processes, but when EDA is linked to activity in the prefrontal cortex,

it represents orienting and attention (Dawson et al., 2000). An alternative and less

expensive technique might be to use to utilize capnography. Capnography is the recording

and analysis of the amount of carbon dioxide in expired respiration (Campbell, Branson,

Burke, Covington, & Graybeal, 1995). Deficiencies in CO2 represent “over-breathing,”

and are indicative of sympathetic activation due to the stress response. By studying

capnography and EDA together, investigators may better identify the specific emotions

involved (e.g., fear vs. attention).

One other limitation relates to our use of zygomatic EMG to measure laughter. In

Study 1, results indicated a main effect of zygomatic activity, such that participants

exhibited more zygomatic activity when they heard the aversive noise compared to the

neutral clip. We suggested that this main effect may have occurred because participants

were surprised when they heard the aversive noise. Because zygomatic activity can

represent a range of positive emotions (including surprise), we cannot be sure that the

zygomatic activity we observed was indicative of smiling behavior. In retrospect, it would

have been ideal to measure obicularis oculi EMG (a muscle that pulls the skin from the

cheeks and forehead toward the eye; Keltner & Bonanno, 1997) in addition to zygomatic

activity. Together, these measures represent what is known as the Duchenne or genuine

smile. By assessing the Duchenne smile in addition to recording laughter, we could be

certain that genuine smiling or laughter occurred, or was absent, during the observed

zygomatic effects.

Despite the limitations, these three studies have several strengths. First, unlike the

many correlational studies, our three studies were carefully controlled laboratory


investigations in which we tested the stress-buffering hypothesis of humor and laughter.

Second, of the few experiments on humor and stressors, many investigators have asked

participants to report their mood before and after the stressor and humor manipulation. Our

studies were unique because we measured mood continuously. Participants reported how

they were feeling each moment throughout the study. In this way, we were able to

investigate the effects of humor before, during, and after each stressful event. Third, Study

3 was distinctive because of its paradigm. According to Schimmack and Colcombe (1999),

few studies have been successful in eliciting mixed feelings from participants. The results

of Study 3 provide some insight into the processes that occur when people experience

mixed feelings. By presenting a humor stimulus concurrently with a stressful (disgust-

eliciting) stimulus, the stressful stimulus became less intense.

Future Directions

One important problem with studying laughter is that it is difficult to produce in the

laboratory. In Study 1, only one person laughed out loud during the task. This is not

uncommon; according to Provine and Fischer (1989), laughter is a social phenomenon that

almost always disappears when individuals are alone. Although we were able to produce

laughter in Studies 2 and 3 (both solitary situations), other investigators might consider

conducting humor studies in a group setting where laughter is surely more likely to take

place. Of course, laughter in a group setting may influence an individual differently than

laughter experienced alone. Also, different motivations may underlie laughter in a social

situation such as using laughter to bond with others. Also, laughter observed in a social

setting may be less "genuine" than laughter in a solitary situation.


Because we only found a beneficial effect of laughter in one of our three studies, we

cannot draw strong conclusions about laughter's overall influence on physiology. As

indicated earlier, studies on the physiological effects of laughter have been conflicting.

The fact that we attempted to study laughter in a stressful context complicates an already

complex finding. Perhaps if laughter was taken out of this context, and just studied by

itself, its physiological properties could be better determined. Indeed, Provine (2000) has

suggested that laughter be contrasted with other vocalizations such as shouting or cheering.

By studying these other positive emotions, which are physically similar to laughter,

researchers could begin to understand how laughter is unique.

Conclusion

The effects of humor and laughter in stressful situations were investigated in three

laboratory studies. In each of the studies, contagious laughter or funny clips were found to

be physiologically arousing compared to neutral clips. In Studies 2 and 3, humor provided

a buffer against negative moods stemming from the stressful experiences. Thus, humor

was effective, with respect to mood, when it was presented before or during a stressor, but

not after. In Study 3, humor positively influenced physiology, such that individuals who

listened to funny clips during a stressful film experienced lower SC levels relative to

individuals who listened to neutral clips during the film. Thus, humor that occurred during

a stressor protected individuals from negative physiological arousal. Together, the results

from the three studies suggest that allowing oneself to be amused before and throughout a

stressful experience may impede negative emotions and, in some cases, the physiological

arousal associated with such a stressful experience.


REFERENCES

Abel, M. H. (2002). Humor, stress, and coping strategies. Humor, 15, 365-381.

Ben-Shakhar, G. (1985). Standardization within individuals: A simple method to

neutralize individual differences in skin conductance. Psychophysiology, 22, 292-

299.

Berk, R. A. (2001). The active ingredients in humor: Psychophysiological benefits and

risks for older adults. Educational Gerontology, 27, 323-339.

Callner, M. (Director). (1997). Jerry Seinfeld Live on Broadway: I'm Telling You for the

Last Time. United States: HBO Home Video.

Campbell, R. S., Branson, R. D., Burke, W., Covington, J., & Graybeal, J. (1995).

Capnography/capnometry during mechanical ventilation. Respiratory Care, 40,

1321-1324. Retrieved July 16, 2006, from

http://www.rcjoumal.com/online_resources/cpgs/co2mvcpg.html

Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., & Robinson, D. S., et

al. (1993). How coping mediates the effect of optimism on distress: A study of

women with early stage breast cancer. Journal o f Personality and Social

Psychology, 65, 375-390.

Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A

theoretically based approach. Journal o f Personality and Social Psychology, 56,

267-283.

Clark, A., Seidler, A., & Miller, M. (2001). Inverse association between sense of humor

and coronary heart disease. International Journal o f Cardiology, 80, 87-88.

Cousins, N. (1979). Anatomy o f an illness. New York: Bantam Books.


http://www.rcjoumal.com/online_resources/cpgs/co2mvcpg.html

9 4

Danzer, A., Dale, J. A., & Klions, H. L. (1990). Effect of exposure to humorous stimuli

on induced depression. Psychological Reports, 66, 1027-1036.

Dawson, M., Schell, A., & Filion, D. (2000). The electrodermal system. In J. T. Cacioppo,

L. G. Tassinary, & G. G. Bemston (Eds.), Handbook o f psychophysiology (2nd ed.,

pp. 200-223). New York: Cambridge University Press.

Deckers, L., & Ruch, W. (1992). The situational humour response questionnaire (SHRQ)

as a test of “sense of humour”: A validity study in the field of humour

appreciation. Personality and Individual Differences, 13, 1149-1152.

Dixon, N. F. (1980). Humor: A cognitive alternative to stress? In I. Sarason & C. D.

Spielberger (Eds.), Stress and anxiety (Vol. 7, pp. 281-289). New York:

Hemisphere.

Ekman, P., Friesen, W. V., & Simons, R. C. (1985). Is the startle reaction an emotion?

Journal o f Personality and Social Psychology, 49, 1416-1426.

Elliot, H. (2005). The worth of mirth: A pilot study on the physiological and

psychological effect of humour on older adults. Perspectives, 29, 6-9.

Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speed recovery from

the cardiovascular sequelae of negative emotions. Cognition & Emotion, 12,

191-220.

Fredrickson, B. L., Mancuso, R. A., Branigan, C., & Tugade, M. M. (2000). The undoing

effect of positive emotions. Motivation & Emotion, 24, 237-258.

Freud, S, (1928). Humour. International Journal o f Psycho-Analysis, 9, 1-6.

Freud, S. (1960). Jokes and their relation to the unconscious. New York: Norton.


9 5

Fry, W. F. (1992). The physiologic effects of humor, mirth, and laughter. Journal o f

American Medical Association, 267, 1857-1858.

Gross, J. J., & Levenson, R. W. (1995). Emotion elicitation using films. Cognition &

Emotion, 9, 87-108.

Hagemann, T., Levenson, R. W., & Gross, J. J. (2006). Expressive suppression during an

acoustic startle. Psychophysiology, 43, 104-112.

Hampes, W. P. (1994). Relation between intimacy and the Multidimensional Sense of

Humor Scale. Psychological Reports, 74, 1360-1362.

Hu, S., & Wan, H. (2003). Imagined events with specific emotional valence produce

specific patterns of facial EMG activity. Perceptual and Motor Skills, 3, 1091-

1099.

International Personality Item Pool. (2001). A scientific collaboratory fo r the

development o f advanced measures ofpersonality traits and other individual

differences. Retrieved November 10, 2005, from http://ipip.ori.org/

Keltner, D., & Bonanno, G. A. (1997). A study of laughter and dissociation: Distinct

correlates of laughter and smiling during bereavement. Journal o f Personality

and Social Psychology, 73, 687-702.

King, C. M. (2003). Humor and mirth. In J. Bryant, D. Roskos-Ewoldsen, & J. Cantor

(Eds.), Communication and Emotion: Essays in Honor ofDolfZillmann. Mahwah,

NJ: Lawrence Erlbaum Associates, Inc., 2003. Pp. 349-378.

Kohler, G., & Ruch, W. (1996). Sources of variance in current sense of humor

inventories: How much substance, how much method variance? Humor:

International Journal o f Humor Research, 9, 363-397.


http://ipip.ori.org/

Kuiper, N. A., & Martin, R. A. (1998). Laughter and stress in daily life: Relation to

positive and negative affect. Motivation & Emotion, 22, 133-153.

Kuiper, N. A., Martin, R. A., & Olinger, L. J. (1993). Coping humour, stress, and

cognitive appraisals. Canadian Journal o f Behavioural Science, 25, 81-96.

Labott, S. M., & Martin, R. B. (1987). The stress-moderating effects of weeping and

humor. Journal o f Human Stress, 13, 159-164.

Lefcourt, H. M., Davidson, K., Prkachin, K. M., & Mills, D. E. (1997). Humor as a stress

moderator in the prediction of blood pressure obtained during five stressful tasks.

Journal o f Research in Personality, 31, 523-542.

Lefcourt, H. M., Davidson, K., Shepherd, R., Phillips, M., Prkachin, K., & Mills, D. E.

(1995). Perspective-taking humor: Accounting for stress moderation. Journal o f

Social & Clinical Psychology, 14, 373-391.

Lyubomirsky, S., & Tucker, K. L. (1998). Implications of individual differences in

subjective happiness for perceiving, interpreting, and thinking about life events.

Motivation & Emotion. Special Issue: Positive affect and self-regulation, 22,

55-186.

Mannell, R. C., & McMahon, L. (1982). Humor as play: Its relationship to psychological

well-being during the course of a day. Leisure Sciences, 5, 143-155.

Martin, R. A. (2001). Humor, laughter, and physical health: Methodological issues and

research findings. Psychological Bulletin, 127, 504-519.

Martin, R. A., & Lefcourt, H. M. (1983). Sense of humor as a moderator of the relation

between stressors and moods. Journal o f Personality and Social Psychology, 45,

1313-1324.


9 7

Martin, R. A., & Lefcourt, H. M. (1984). Situational humor response questionnaire:

Quantitative measure of sense of humor. Journal o f Personality and Social

Psychology, 47, 145-155.

Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J., & Weir, K. (2003). Individual

differences in uses of humor and their relation to psychological well-being:

Development of the Humor Styles Questionnaire. Journal o f Research in

Personality, 37, 48-75.

Meadowcroft, J. M., & Zillmann, D. (1987). Women's comedy preferences during the

menstrual cycle. Communication Research, 14, 204-218.

Miller, M., Mangano, C., Park, Y., Goel, R., Plotnick, G. D., & Vogel, R. A. (2006).

Impact of cinematic viewing on endothelial function. Heart, 92, 261-262.

Newman, M. G., & Stone, A. A. (1996). Does humor moderate the effects of

experimentally-induced stress? Annals o f Behavioral Medicine, 18, 101-109.

Nezu, A. M., Nezu, C. M., & Blisset, S. E. (1988). Sense of humor as a moderator of the

relation between stressful events and psychological distress: A prospective

analysis. Journal o f Personality and Social Psychology, 54, 520-525.

Perlini, A. H., Nenonen, R. G., & Lind, D. L. (1999). Effects of humor on test anxiety

and performance. Psychological Reports, 84, 1203-1213.

Provine, R. R. (1992). Contagious laughter: Laughter is a sufficient stimulus for laughs

and smiles. Bulletin o f the Psychonomic Society, 30, 1-4.

Provine, R. R. (2000). Laughter: A scientific investigation. New York: Penguin Group.

Provine, R. R., & Fischer, K. R. (1989). Laughing, smiling, and talking: Relation to

sleeping and social contexts in humans. Ethology, 83, 295-305.


Rottenberg, J., Ray, R.R., & Gross, J J . (in press). Emotion elicitation using films. In J.A.

Coan & J J.B Allen (Eds.), The handbook o f emotion elicitation and assessment. New

York: Oxford University Press.

Ruch, W. (1994). Temperament. Eysenck’s PEN system, and humor-related traits.

Humor: International Journal o f Humor Research, 7, 209-244.

Ruch, W. (2001). The Perception of Humor. In A. Kaszniak (Ed.), Emotion, Qualia, and

Consciousness. Tokyo, Japan: Word Scientific Publisher, 2001. Pp. 410-425.

Scheff, T. J., & Bushnell, D. D. (1984). A theory of catharsis. Journal o f Research in

Personality, 18, 238-264.

Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from

neuroticism (and trait anxiety, self-mastery, and self-esteem): A re-evaluation of

the Life Orientation Test. Journal o f Personality and Social Psychology, 67,

1063-1078.

Schimmack, U., & Colcombe, S. (1999). Mixed feelings: Toward a theory ofpleasure and

displeasure. Unpublished manuscript, University of Illinois at Urbana.

Tassinary, L. G., & Cacioppo, J. T. (2000). The skeletomotor system: Surface

electromyography. In J. T. Cacioppo, L. G. Tassinary, & G. G. Bemston (Eds.),

Handbook o f psychophysiology (2nd ed., pp. 163-199). New York: Cambridge

University Press.

Thorson, J. A., & Powell, F. C. (1993). Relationships of death anxiety and sense of

humor. Psychological Reports, 72, 1364-1366.

Thorson, J. A., & Powell, F. C. (1994). Depression and sense of humor. Psychological

Reports, 75, 1473-1474.


Thorson, J. A., Powell, F. C., Sarmany-Schuller, I., & Hampes, W. P. (1997).

Psychological health and sense of humor. Journal o f Clinical Psychology, 53,

605-619.

Vaillant, G. E. (2000). Adaptive mental mechanisms: Their role in a positive psychology.

American Psychologist, 55, 89-98.

Vilaythong, A. P., Amau, R. C., Rosen, D. H., & Mascaro, N. (2003). Humor and hope:

Can humor increase hope? Humor, 16, 79-89.

Watson, L. (2003). Orchard: A novel (Cassette Recording). New York: Random House.

White, S., & Camarena, P. (1989). Laughter as a stress reducer in small groups. Humor, 2,

73-79.

Yovetich, N. A., Dale, J. A., & Hudak, M. A. (1990). Benefits of humor in reduction of

threat-induced anxiety. Psychological Reports, 66, 51-58.

Zillmann, D. (1971). Excitation transfer in communication-mediated aggressive behavior.

Journal o f Experimental Social Psychology, 7, 419-434.

Zillmann, D. (1983). Disparagement humor. In P. E. McGhee & J. H. Goldstein (Eds.),

Handbook o f humor research: Basic issues. (Vol. 1, pp. 86-107). New York:

Springer-Verlag.


1 0 0

APPENDIX A

EXPERIMENTER SCRIPTS

Study 1

Thanks for coming in today. My name is and I’m a student in Psychology. Before

we get started, I need to have you look over this consent form—it tells you a little bit about

the study. So read it over and let me know if you have questions.

{have participant sign consent form)

Our lab is interested in how your feelings (both positive and negative) influence your mood

and physiology. Most people respond in a variety of ways when they experience different

feelings. Some of these responses may be voluntary (like when you report mood) and

some of them may be involuntary (like when your skin temperature increases slightly).

In this study, we’re interested in learning about your feelings and physiology in response to

different types of sounds. So today you’re going to hear a series of sounds and your job is

to continuously monitor how you’re feeling. You will report your feelings using this

joystick {point to apparatus).

Also, because we are interested in physiology, we will be attaching various sensors to

measure more involuntary types of activity. Five sensors will be attached to your face

(here, here, here, here, and here—point to where the face sensors will be) and two will be

attached to your fingertips (here and here—point). Although you may notice the sensors at

first, when we first attach them, you won’t feel anything; they are just a way for us to

measure activity. Any questions? {pause!)

First, I’m going to attach the sensors, and then I’ll show you how the joystick works. (As

you attach each sensor, explain what it measures: muscle “tenseness,” sweating on the

fingers).

After sensors are on: DO SENSOR CHECK!

After sensor check, return to participant and explain joystick!

You will report how you’re feeling by continually moving this joystick to register your

emotions. You will gradually move the joystick to correspond to gradual mood changes

you have. For example, if you feel slightly to extremely positive, you will move the

joystick over here {show participant by moving the joystick from +1 to +10). If you feel


101

slightly to extremely negative, you will move the joystick over here (show participant by

moving joystick from -1 to -10). Negative 10 would represent the worst possible mood but

positive 10 would represent the best possible mood (show participant). If you feel neutral

(neither good or bad) you would move the joystick to the center (show participant by

moving to 0).

We would like you to, as closely as possible, match your moment-to-moment feeling to the

scale on the joystick. If at any time you feel emotion at the extreme ends of the scale, it

may take awhile before your emotion level will return to a neutral state. As best as you

can, move the joystick slowly in the direction you feel your emotions are moving each

moment.

In a minute, I’m going to start the sound presentation. The series of sounds you will hear

during this presentation are not connected in any way; in fact they will be very different

from each other. The sounds will be presented for about 10-15 seconds, but there will also

be some quiet periods during the presentation when you hardly hear anything at all. This is

normal and it may seem longer than it is. It’s important that you continue to monitor your

feelings throughout the sound presentation, even during the quiet times. Remember, you

are NOT rating the sounds; you are simply rating your own feelings throughout the

presentation. Please make your feeling ratings as continuous as possible; this means that

you need to constantly moving the joystick to represent your moment-to-moment feelings.

Now you try it. (Have subject put hand on joystick and move all the way to left and all the

way to right). Start monitoring as soon as the task begins, and make sure to keep your hand

on the joystick the whole time because you will be monitoring your mood each moment.

Also, after you read the directions for the task, there will be a period of approximately 3

minutes when all we want you to do is just sit back and relax. After this period, the sound

presentation will begin.

Study 2

Thanks for coming in today. My name is and I’m a student in Psychology. Before

we get started, I need to have you look over this consent form—it tells you a little bit about

the study. So read it over and let me know if you have questions.

(have participant sign consent form)


1 0 2

Our lab is interested in how your feelings (both positive and negative) influence your mood

and physiology. Most people respond in a variety of ways when they experience different

feelings. Some of these responses may be voluntary (like when you report mood) and


In this study, we’re interested in learning about your feelings and physiology in response to

different types of sounds. Little research has been conducted on how people respond

physiologically to different types of sounds. We are interested in the physiological

differences between sounds that that are meaningful (like a story) and sounds that are

meaningless (like a buzzing sound). Today you’re going to hear a series of sounds and

your job is to continuously monitor how you’re feeling. You will report your feelings

using this joystick {point to apparatus).

Also, because we are interested in physiology, we will be attaching various sensors to

measure more involuntary types of activity. Five sensors will be attached to your face

(here, here, here, here, and here—point to where the face sensors will be) and two will be

attached to your fingertips (here and here—point). Although you may notice the sensors at

first, when we first attach them, you won’t feel anything; they are just a way for us to

measure activity. Any questions? (pause!)



fingers). If Ps have gum, make sure they spit it out. After sensors are on: DO SENSOR

CHECK!



emotions. You will gradually move the joystick to correspond to gradual mood changes


joystick over here (show participant by moving the joystick from +1 to +10). If you feel


moving joystick from -1 to -10). Negative 10 would represent the worst possible mood but


(neither good or bad) you would move the joystick to the center (showparticipant by

moving to 0).


103

We would like you to, as closely as possible, match your moment-to-moment feeling to the

scale on the joystick. If at any time you feel emotion at the extreme ends of the scale, it



moment.

In a minute, I’m going to start the sound presentation. The series of sounds you will hear

during this presentation are going to be very different from each other. They are not

connected in any way. In fact, some of the sounds are going to be meaningful to you, and

others will seem meaningless. The sounds will be presented anywhere from 15 to 60

seconds, but there will also be some quiet periods during the presentation when you hardly

hear anything at all. This is normal and it may seem longer than it is. It’s important that

you continue to monitor your feelings throughout the presentation, even during the quiet

times. Remember, you’re NOT rating the sounds; you’re simply rating your own feelings

throughout the presentation. Also, remember to make your feeling ratings {point to

joystick) as continuous as possible; this means that you should be continuously thinking

about how you’re feeling and moving the joystick to reflect those feelings.

Now you try it. {Have subject put hand on joystick and move all the way to left and all the

way to right). You should start moving the joystick as soon as the task begins, and just

keep your hand on it because you will be monitoring your mood the whole time.

When I start the presentation, the first thing that you’ll see is directions for the task. After

that, there will be a period of about a minute or so when all we want you to do is just sit

back and relax. After this period, the sounds will start. You will want to watch the screen

during the task because it will alert you to when sounds are beginning.

Study 3

Thanks for coming in today. My name i s and I’m a graduate student in Psychology.

Before we get started, I need to have you look over this consent form—it tells you a little

bit about the study. So read it over and let me know if you have questions.

{Have participant sign consent form)


1 0 4

Our lab is interested in how your positive and negative feelings influence your mood and

physiology. Most people respond in a variety of ways when they experience different

feelings. Some of these responses may be voluntary (like when you report your mood) and


In this study, we’re interested in learning about how you respond physiologically to

different feeling states. Most research in this area has focused on the differences between

pleasant and unpleasant feeling states. Thus, researchers have typically studied them

separately. Our study is unique because we’re examining the physiology behind mixed

feelings or emotions. We’re asking the question, what happens to your physiology when

you experience different feelings at the same time. Today you will listen to audio clips

while watching short films. Keep in mind that what you hear, on the audio clips, will be

very different from what you see, in the films. And there isn’t any connection between the

two—they are separate possible sources of feelings.

It’s important that you understand that it will be difficult for you to pay attention to both

the audio clips and the videos, particularly because they are unrelated. However, we would

like you to try your best. If you feel it is too difficult, you can focus on one rather than the

other. During this study, we would like you to continuously monitor how you’re feeling.

You'll report your feelings using this joystick (point to apparatus).

Because we're interested in physiology, we'll be attaching sensors to measure the

involuntary types of activity. Five sensors will be attached to your face (here, here, here,

here, and here—point to where the face sensors will be) and three will be attached to your

fingertips (here, here, and here—point). Although you may notice the sensors at first, like

when I first attach them, you're not going to feel anything; they are just a way for us to

measure activity.

Any questions? {pause!)




105

fingers). If Ps have gum, make sure they spit it out. After sensors are on: DO SENSOR

CHECK!



emotions. You should gradually move the joystick to correspond to gradual mood changes


joystick over here (show participant by moving the joystick from +1 to +10). If you feel


moving joystick from -1 to -10). Negative 10 would represent the worst possible mood and


(neither good or bad) you would move the joystick to the center (show participant by

moving to 0).

We would like you to, as closely as possible, match your moment-to-moment feelings to

the scale on the joystick. If at any time you feel emotion at the extreme ends of the scale, it



moment.

In a minute, I’ll start the presentation. Keep in mind that the audio clips you will hear are

going to be very different from the films you will see. The materials are not connected or

related in any way; remember they may elicit different types of feelings. The audio clips

will be presented for about three minutes, and the films for about 2 minutes. The audio will

begin about a minute before the films. You need to monitor your feelings throughout the

presentation (point to joystick). It's important that you understand that you’re NOT rating

the audio clips or the films; you’re simply rating your own feelings throughout the study.

Also, remember to make your ratings as continuous as possible; this means that you should

be continuously thinking about how you’re feeling and moving the joystick to reflect those

feelings.


1 0 6

Now you try it. (Have subject put hand on joystick and move all the way to left and all the

way to right). Start moving the joystick as soon as I leave the room, and just keep your

hand on it because you will be monitoring your mood the whole time.

When I start the presentation, there will be a period of about 5 minutes or so when all we

want you to do is just sit back and relax. After this period, the audio will start. You need

to watch the screen because soon after, the films will start.


1 0 7

APPENDIX B

CONSENT FORMS

Studies 1 and 2

CONSENT TO PARTICIPATE IN RESEARCH

“Mood and physiology”

Research Study: You are invited to participate in a research study about mood and physiology being conducted by Kevin McCaul, Professor of Psychology. You must be 18 years of age or older to participate in this study.

Purpose of Study: The purpose of the present study is to investigate how different feeling states influence mood and physiological responses.

Explanation of Procedures: If you agree to participate, you will complete questionnaires assessing mood and personality dimensions. You will also listen to various sounds that may cause some arousal. During the listening task, electromyography (facial muscle activity) and skin conductance levels will be measured by attaching sensors to your face and fingertips.

Potential Risks and Discomforts: The sounds may cause some discomfort. If you feel that listening to sounds will make you overly anxious or uncomfortable, please let the experimenter know. You should not participate in this study if you are strongly concerned about being anxious.

Potential Benefits: Potential benefits may include increased knowledge about how feeling states influence mood and physiological responses.

Alternatives to Participation: Your psychology instructor provides alternative ways to earn extra credit.

Compensation for Participation: You will be given 1 extra credit point for every 15 minutes that you are engaged in this study. If you decide to participate, we estimate that you will receive approximately 4 points worth of extra credit (for approximately 50-60 minutes total time) for your psychology course.

Assurance of Confidentiality: Your name will not be associated with your results. We use numbers to identify individuals - not names. This consent form will be kept in a separate file from other questionnaires that you complete, and we will not be able to match names to data. Data and records created by this project are the property of the university and the investigator. In any reports or publications, no participants will be identified or identifiable and only aggregate data will be presented.


1 0 8

Statement of Injury or Special Costs: None.Voluntary Participation & Withdrawal from the Study: Your participation is voluntary. Your decision whether or not to participate will not affect your grade or present or future relationship with the university. If you decide to participate, you are free to withdraw your consent and to discontinue participation at any time.

Offer to Answer Questions: You should feel free to ask any questions now or during the experiment. If you have any questions about this experiment, or your participation in it, you may contact Dr. Kevin D. McCaul in the Psychology Department in 115 Minard (phone 231-7072). If you have any questions regarding the rights of research participants, or if you experience any problems as a result of this research, you should contact the NDSU Institutional Review Board at 231-8908.

Consent Statement: You are voluntarily making a decision whether or not to participate. Your signature indicates that you have decided to participate, having read the information provided above. You will be given a copy of this consent form to keep.

Please print full name of participant Signature of participant Date

Please print full name of investigator Signature of investigator Date

I wish to earn extra credit for_______________________/___________________Course number Instructor

NAID#

Study 3

CONSENT TO PARTICIPATE IN RESEARCH

“Mood and physiology”

Research Study: You are invited to participate in a research study about mood and physiology being conducted by Kevin McCaul, Professor of Psychology. You must be 18 years of age or older to participate in this study.

Purpose of Study: The purpose of the present study is to investigate how different feeling states influence mood and physiological responses.


1 0 9

Explanation of Procedures: If you agree to participate, you will complete questionnaires assessing mood and personality dimensions. You will also listen to various sounds and watch a brief video that may cause some arousal. During this time, electromyography (facial muscle activity) and skin conductance levels will be measured by attaching sensors to your face and fingertips.

Potential Risks and Discomforts: The video may cause some discomfort. If you feel that watching a video of a medical procedure will make you overly anxious or uncomfortable, please let the experimenter know. You should not participate in this study if you are strongly concerned about being anxious.

Potential Benefits: Potential benefits may include increased knowledge about how feeling states influence mood and physiological responses.

Alternatives to Participation: Your psychology instructor provides alternative ways to earn extra credit.

Compensation for Participation: You will receive either $10.00 for your participation or extra credit for your psychology course. If you choose extra credit, you will be given 1 extra credit point for every 15 minutes that you are engaged in this study. If you decide to participate and choose extra credit, we estimate that you will receive approximately 3 points worth of extra credit (for approximately 40 minutes total time) for your psychology course.

Assurance of Confidentiality: Your name will not be associated with your results. We use numbers to identify individuals - not names. This consent form will be kept in a separate file from other questionnaires that you complete, and we will not be able to match names to data. Data and records created by this project are the property of the university and the investigator. In any reports or publications, no participants will be identified or identifiable and only aggregate data will be presented.

Statement of Injury or Special Costs: None.

Voluntary Participation & Withdrawal from the Study: Your participation is voluntary. Your decision whether or not to participate will not affect your grade or present or future relationship with the university. If you decide to participate, you are free to withdraw your consent and to discontinue participation at any time.

Offer to Answer Questions: You should feel free to ask any questions now or during the experiment. If you have any questions about this experiment, or your participation in it, you may contact Dr. Kevin D. McCaul in the Psychology Department in 115 Minard (phone 231-7072). If you have any questions regarding the rights of research participants, or if you experience any problems as a result of this research, you should contact the NDSU Institutional Review Board at 231-8908.


1 1 0

Consent Statement: You are voluntarily making a decision whether or not to participate. Your signature indicates that you have decided to participate, having read the information provided above. You will be given a copy of this consent form to keep.

Please print full name of participant Signature of participant Date

Please print full name of investigator Signature of investigator Date

I wish to earn extra credit for______________________ /___________________Course number Instructor

NAID #


APPENDIX C

DEBRIEFING FORMS

Studies 1 and 2

Now that the study is over, I would like to tell you a little bit about what we were

hoping to find. First, if you had to guess, what do you think we were studying in this

experiment? {If guess comes close, record subject # and what they said).

Okay, well our goal in this study was to explore the psychological and

physiological effects of humor and laughter. We were specifically interested in whether

experiencing humor and laughter could undo the effects of negative emotions. Some of the

sound clips you listened to were probably sort of annoying and led you to experience some

discomfort. Others were probably pleasant; you heard clips of people laughing [Jerry

Seinfeld for debriefing for Study 2]. We wanted to test if the laughter [funny for debriefing

for Study 2] clips could undo the discomfort you may have experienced during the more

negative sounds.

Before I let you go, I want to ask an important favor of you. Please don’t share

the specifics of this study with your classmates. We want everyone who participates to go

through it the same way you did; not knowing about what sounds to expect and what the

purpose of the study is. So feel free to tell anyone that you listened to different sounds and

that sensors were attached to you, and that it is sort of an interesting study, but please do

not tell others any specifics about the sounds or the purpose of the study. OK?

Do you have any questions?

IF THERE IS TIME, SHOW THEM A PIECE OF THEIR DATA. SAY, this is what some of

your physiological data looks like.


1 1 2

THANK YOU for your participation!

Study 3

Now that the study is over, I would like to tell you a little bit about what we were hoping to

find. First, if you had to guess, what do you think we were studying in this experiment? {If

guess comes close, record subject # and what they said).

Okay, well our goal in this study was to explore the effects of humor and laughter

on a stressful experience. We were specifically interested in whether humor and laughter

could provide a buffer against the negative emotions you may have experienced during the

films. These films have been used in other studies to elicit negative emotions. Most

people feel disgusted when they watch them, and we were looking to see if hearing

something funny (compared to hearing something not funny) could alleviate some of those

negative feelings. Which did you listen to, the funny, or the neutral??

Before I let you go, I want to ask an important favor of you. Please don’t share

the specifics of this study with anyone. We want everyone who participates to go through

it the same way you did; not knowing about what to expect and what the purpose of the

study is. So feel free to tell anyone that sensors were attached to you, but please do not tell

others any specifics about the sounds, video or purpose of the study. OK?

Do you have any questions for me? THANK YOU for your participation!

IF THEY WOULD ASK ABOUT THE VIDEOS, YOU CAN TELL THEM WHAT THEY

ARE, BUT FRAME IT IN A POSITIVE WAY (MAYBE AMPUTATION WAS NEEDED TO

PREVENT INFECTION?? BURN VICTIMS WERE BEING HELPED!!)


1 1 3

APPENDIX D

HUMOR STYLES QUESTIONNAIRE

People experience and express humor in many different ways. Below is a list of statements describing different ways in which humor might be experienced. Please read each statement carefully, and indicate the degree to which you agree or disagree with it. Please respond as honestly and objectively as you can. Use the following scale:

1 2 3 4 5 6 7Totally Moderately Slightly Neither Slightly Moderately Totally

Disagree Disagree Disagree Agree nor Disagree

Agree Agree Agree

1. I usually don’t laugh or joke around much with other people.

1 2 3 4 5 6 7

2. If I am feeling depressed, I can usually cheer myself up with humor.

1 2 3 4 5 6 7

3. If someone makes a mistake, I will often tease them about it.

1 2 3 4 5 6 7

4. I let people laugh at me or make fun at my expense more than I should.

1 2 3 4 5 6 7

5. I don't have to work very hard at making other people laugh — I seem to be a naturally humorous person.

1 2 3 4 5 6 7

6. Even when I’m by myself, I’m often amused by the absurdities of life.

1 2 3 4 5 6 7

7. People are never offended or hurt by my sense of humor.

1 2 3 4 5 6 7


1 1 4

8. I will often get carried away in putting myself down if it makes my family or friends laugh.

1 2 3 4 5 6 7

9. I rarely make other people laugh by telling funny stories about myself.

1 2 3 4 5 6 7

10. If I am feeling upset or unhappy I usually try to think of something funny about the situation to make myself feel better.

1 2 3 4 5 6 7

11. When telling jokes or saying funny things, I am usually not very concerned about how other people are taking it.

1 2 3 4 5 6 7

12. I often try to make people like or accept me more by saying something funny about my own weaknesses, blunders, or faults.

1 2 3 4 5 6 7

13. I laugh and joke a lot with my friends.

1 2 3 4 5 6 7

14. My humorous outlook on life keeps me from getting overly upset or depressed about things.

1 2 3 4 5 6 7

15. I do not like it when people use humor as a way of criticizing or putting someone down.

1 2 3 4 5 6 7

16. I don’t often say funny things to put myself down.

1 2 3 4 5 6 7

17. I usually don’t like to tell jokes or amuse people.

1 2 3 4 5 6 7


1 1 5

18. If I’m by myself and I’m feeling unhappy, I make an effort to think of something funny to cheer myself up.

1 2 3 4 5 6 7

19. Sometimes I think of something that is so funny that I can’t stop myself from saying it, even if it is not appropriate for the situation.

1 2 3 4 5 6 7

20. I often go overboard in putting myself down when I am making jokes or trying to be funny.

1 2 3 4 5 6 7

21. I enj oy making people laugh.

1 2 3 4 5 6 7

22. If I am feeling sad or upset, I usually lose my sense of humor.

1 2 3 4 5 6 7

23. I never participate in laughing at others even if all my friends are doing it.

1 2 3 4 5 6 7

24. When I am with friends or family, I often seem to be the one that other people make fun of or joke about.

1 2 3 4 5 6 7

25. I don’t often joke around with my friends.

1 2 3 4 5 6 7

26. It is my experience that thinking about some amusing aspect of a situation is often a very effective way of coping with problems.

1 2 3 4 5 6 7

27. If I don't like someone, I often use humor or teasing to put them down.

1 2 3 4 5 6 7


28. If I am having problems or feeling unhappy, I often cover it up by joking around, so that even my closest friends don’t know how I really feel.

1 2 3 4 5 6 7

29. I usually can’t think of witty things to say when I’m with other people.

1 2 3 4 5 6 7

30. I don’t need to be with other people to feel amused — I can usually find things to laugh about even when I’m by myself.

1 2 3 4 5 6 7

31. Even if something is really funny to me, I will not laugh or joke about it if someone will be offended.

1 2 3 4 5 6 7

32. Letting others laugh at me is my way of keeping my friends and family in good spirits.

1 2 3 4 5 6 7


1 1 7

APPENDIX E

COPING HUMOR SCALE

Using the following scale, please indicate the extent to which you agree or disagree with the following statements.

1 = Strongly Disagree 2 = Mildly Disagree 3 - Mildly Agree 4 = Strongly Agree

1. I often lose my sense of humor when I’m having problems.

2. I have often found that my problems have been greatly reduced when I tried

to find something funny in them.

3. I usually look for something comical to say when I am in tense situations.

4. I must admit my life would probably be easier if I had more of a sense of

humor.

5. I have often felt that if I am in a situation where I have to either cry or laugh,

it’s better to laugh.

6. I can usually find something to laugh or joke about even in trying situations.

7. It has been my experience that humor is often a very effective way of coping

with problems.


1 1 8

APPENDIX F

ANXIETY

On the following pages, there are phrases describing people's behaviors. Please use the rating scale below to describe how accurately each statement describes you. Describe yourself as you generally are now, not as you wish to be in the future. Describe yourself as you honestly see yourself, in relation to other people you know of the same sex as you are, and roughly your same age. So that you can describe yourself in an honest manner, your responses will be kept in absolutely confidence. Please read each statement carefully, and write in the number that corresponds to your response.

1 = Very 2 = Moderately 3 = Neither 4 = Moderately 5 = VeryInaccurate Inaccurate Inaccurate or Accurate Accurate

Accurate

1. _______ I worry about things.

2. _______ I am relaxed most of the time.

3. _______ I adapt easily to new situations.

4. _______ I get caught up in my problems.

5. I am afraid of many things.

6. I don’t worry about things that have already happened.

7. _______ I am not easily bothered by things.

8. I fear for the worst.

9. _______ I get stressed out easily.

10 . _______ I am not easily disturbed by events.


1 1 9

APPENDIX G

NEUROTICISM

On the following pages, there are phrases describing people's behaviors. Please use the rating scale below to describe how accurately each statement describes you. Describe yourself as you generally are now, not as you wish to be in the future. Describe yourself as you honestly see yourself, in relation to other people you know of the same sex as you are, and roughly your same age. So that you can describe yourself in an honest manner, your responses will be kept in absolutely confidence. Please read each statement carefully, and write in the number that corresponds to your response.

1 = Very 2 = Moderately 3 = Neither 4 = Moderately 5 = VeryInaccurate Inaccurate Inaccurate or Accurate Accurate

Accurate

1. I rarely get irritated.

2. _______ I often feel blue.

3. _______ I feel comfortable with myself.

4. _______ I am not easily bothered by things.

5. _______ I dislike myself.

6. _______ I am often down in the dumps.

7. _______ I seldom feel blue.

8. _______ I have frequent mood swings.

9. _______ I am very pleased with myself.

10 . I panic easily.


1 2 0

APPENDIX H

LIFE ORIENTATIONS TEST-REVISED

Please be as honest and accurate as you can throughout. Try not to let your response to one statement influence your responses to other statements. There are no “correct” or “incorrect” answers. Answer according to your own feelings, rather than how you think “most people” would answer. Place a letter from the following scale by each statement.

A = I agree a B = I agree a C = I neither D = I disagree a E = I disagree alot little agree nor little lot

disagree

1. In uncertain times, I usually expect the best

2. It is easy for me to relax.

3. If something can go wrong for me, it will.

4. I am always optimistic about my future.

5. I enjoy my friends a lot.

6. It is important for me to keep busy.

7. I hardly ever expect things to go my way.

8. I do not get upset too easily.

9. I rarely count on good things happening to me.

10. Overall, I expect more good things to happen to me than bad.


APPENDIX I

ADDITIONAL ANALYSES FOR STUDY 1

Analyses

Zygomatic-Interval 1 (Periods 1-3). During Interval 1, results produced a Noise x

Laughter interaction, F (1,22) = 6.48, p = .02. The interaction showed that an aversive

noise followed by laughter produced slightly more zygomatic activity than a noise followed

by neutral clips (Ms = .11 and .10). In contrast, a neutral clip followed by laughter

produced slightly less activity than a neutral clip followed by other neutral clips (Ms = .087

and .093). Because laughter had not yet occurred during the time of Interval 1 (laughter

occurred in Interval 2), the above interaction most likely represents a carryover effect.

Additionally, there was a main effect of Trials, F (3, 66) = 3.87,/? = .01, showing less

activity across trials.

Zygomatic-Interval 2 (Periods 4-6). Results showed a main effect of Trials, F (3,

66) = 3.17,/? = .03, with activity declining across trials.

Zygomatic-Recovery (Periods 7-12). Results showed only a small effect of Trials,

F (3, 63) = 2.19,/? = .10, such that activity subsided across trials.

Corrugator-Interval 1 (Periods 1-3). During Interval 1, results produced a weak

Laughter x Periods interaction, F (2, 44) = 2.16, p = .07. Similar to the interaction for

zygomatic activity, the interaction occurred during Interval 1, but the laughter had not

occurred yet. Again, this finding might be evidence of a carryover effect. Also during

Interval 1, results showed a significant Laughter x Trials interaction, but again the laughter

had not occurred yet. During Recovery, a significant Laughter x Trials interaction, F (1,


1 2 2

63) = 3.21 ,/? = .03, showed that, across trials, activity increased for both laughter and

neutral clips, except in Trial 4, activity declined for the neutral clips.

Skin conductance-Interval 1 (Periods 1-3). During Interval 1, the laughter clips

produced greater SC activity than neutral clips (Ms = .37 vs. .27), and this effect was

significant, F (1, 22) = 8.58, p < .01. Just like the interactions for the other physiological

measures, this finding possibly reflects a carryover influence. Results also produced a

significant Noise x Laughter interaction, F (1, 22) = 10.50,/* < .01 during Interval 1. The

interaction showed that an aversive noise followed by laughter produced more SC activity

than a noise followed by neutral clips (Ms = .65 and .39). In contrast, a neutral clip

followed by laughter produced less SC activity than a neutral clip followed by other neutral

clips (Ms = .09 and .15). This finding is another example of when carryover effects may

have been present. There was a main effect of Trials, F (3, 66) = 12.68, p < .001, showing

less activity across trials. There was also a significant Noise x Trials interaction, F (3, 66)

= 5.58,/? > .01, showing that SC activity for both the noise and the neutral clips faded

across Trials, but there was a jump on Trial 4 for neutral clips. Finally, there was a

significant Trials x Periods interaction, F (6, 132) = 7.52,/? < .001, showing that SC

activity increased from Period 1 to Period 2, but then declined, and this effect of Periods

became weaker across Trials.

Skin conductance-Interval 2 (Periods 4-6). During Interval 2, results showed a

significant Noise x Periods interaction, F (2, 44) = 4.26,/? = .02, showing that, across

periods, SC activity faded for both the noise and neutral clips, but it faded to a greater

extent for the noise. A significant Laughter x Periods interaction, F (2, 44) = 4.01,/? = .03,

showed that, across periods, activity faded for both the laughter and neutral clips, but it


123

faded faster for neutral clips. During this interval, results also indicated that activity

declined across Trials, F (3, 66) = 10.96, p < .001.

Skin conductance-Recovery (Periods 7-12). Results indicated that SC activity

declined across Trials during Recovery, F (3, 63) = 3.76, p = .02. Also during Recovery, a

significant Noise x Trials, F (3, 63) = 2.11, p = .05, showed that, across trials, SC activity

faded for both the noise and neutral clips, but there was a jump in activity during Trial 3 for

both of these sounds. Finally, results showed a significant Trials x Periods interaction, F

(15, 315) = 4.32,/? < .001.

Peak mood-Interval 1 (Periods 1-3). During Interval 1, there was a main effect of

Trials, F (3, 66) = 18.79, p < .001, showing that mood became more negative across trials.

There was also a main effect for Periods, F (2, 44) = 52.97, p < .001, showing the same

pattern. During Interval 1, there was also a significant Trials x Periods, F (6, 132) = 13.32,

p < .001, showing that mood declined across periods, and this effect became stronger

across trials.

Peak mood-Interval 2 (Periods 4-6). During Interval 2, results produced a main

effect for Trials, F (3, 66) = 8.37,/? < .001, showing that mood tended to decline across

trials. A significant Noise x Periods interaction, F (2, 44) = 17.68,/? < .001, showed that,

across periods, negative moods produced from the noise faded, and neutral clips produced

increasingly more positive moods.

Peak mood-Recovery (Periods 7-12). During Recovery, there was a significant

Noise x Periods interaction, F (5, 105) = 4.88, p < .001, showing that, across periods, the

aversive noise tended to produce less negative moods (i.e., mood improved), and neutral

clips produced less positive moods (i.e., mood declined).


APPENDIX J

ADDITIONAL ANALYSES FOR STUDY 2

Analyses

Zygomatic-Interval 2 (Periods 13-15). Results showed that activity decreased

across Trials, F (2, 48) = 5.94,/? = .01. A significant Clips x Trials interaction, F (2, 48) =

3.14,/? = .05, showed that zygomatic activity was slightly higher for individuals who had

just listened to the funny clips compared to those who had just listened to the neutral clips.

Peak Skin Conductance-Interval 1 (Periods 1-12). There was a main effect of

Trials, F (2, 48) = 5.15,/? < .01, showing less activity over time. Results also showed that

across trials, SC activity faded for both neutral and funny clips, but it faded faster when

participants listened to the funny clips. This pattern produced a significant Clips x Trials

interaction, F (2, 48) = 5.44,/? < .01.

Peak Skin Conductance-Interval 2 (Periods 13-15). Results revealed a main effect

of Trials, F (2, 48) = 6.10,/? < .01, showing that SC activity faded over time.

Peak Skin Conductance-Recovery (Periods 16-27). During Recovery, SC activity

declined across Trials, F (2, 48) = 5.17,/? < .01

Skin conductance change-Interval 1 (Periods 1-12). As can be seen from Figure

12, the funny clips tended to produced more SC activity than the neutral clips (Ms = .12

and .001), but the difference was not significant, F (1, 24) = 1.68,/? = .21. There was a

Periods main effect, F (11, 264) = 10.14,/? < .001, showing that SC faded over time.

Finally, there was a Clips x Periods interaction, F (11, 264) = 2.88,/? = .001, that showed

that SC faded during the neutral clips, but stayed about the same or increased during the

funny clips.


125

interval 1 Interval 2 Recovery0.5

0.4

0.3ONLN

0.2

25 27- 0.1

- 0.2

Periods

Figure 20. Study 2 SC change by condition collapsed across trials.

Skin conductance change-Interval 2 (Periods 13-15), During Interval 2, SC activity

for the funny clips was slightly higher than activity for the neutral clips (Ms = .32 and .24),

although this difference was not significant, F < 1. There was a Periods main effect, F (2,

48) = 11.26, p < .001, showing a jump in SC activity (highest value during period 14), and

then a decline (Figure 20). The main effect was qualified by a Clips x Periods interaction,

F (2,48) = 4.16, p = .02, showing that both the funny and neutral clips produce a jump in

SC activity, and then a decline. Although SC activity was slightly higher in Period 13 for

the funny clips, by period 15, levels of SC were equal for neutral and funny clips. Finally,

there was a Trials main effect, F (2, 48) = 2.89, p = .05, suggesting that SC faded over

time.

Skin conductance change-Recovery (Periods 16-27). During Recovery, there was a

significant main effect of Periods, F (11, 264) = 11.79,/? < .001, which showed that SC

faded over time.


HUMOR, LAUGHTER, AND RECOVERY FROM STRESSFUL...

Documents

Transcript of HUMOR, LAUGHTER, AND RECOVERY FROM STRESSFUL...