Research Proposal

profileBella edwards
Bensons.pdf

1

The Effect of Music Therapy and Progressive Neuromuscular Relaxation on Performance of a Simple

Problem-solving Task

Donald F. Kendrick and Danielle Benson

Middle Tennessee State University

2012

2

The Effect of Music Therapy and Progressive Neuromuscular Relaxation

on the Performance of a Simple Problem-solving Task

Stress and anxiety are two factors that, when outside of our control, become debilitating on our

performance on a variety of tasks. Often the area of the highest interest is the effect of anxiety on

academic test performance. Morris and Eagle (1981) analyzed the relationship between pre- and post

test anxiety as related to college student's coping mechanisms. Questionnaires about emotional state

were given before and after an examination, and a second questionnaire regarding cognitive coping

skills recorded the occurrence of various trains of thought, such as “if I do poorly, it is because the

questions are unfair,” or “I might not be able to finish the test in time.” The authors found that there

were often strong correlations between maladaptive coping, such as denial and preoccupation-with-self,

and high levels of anxiety. Only some of the anxiety/coping pairs led to reliable differences in

performance, however.

Given the relationship of coping, anxiety, and performance, a multitude of studies approach the

problem from a position of intervention, attempting to mitigate the stressful effects of test performance

through anxiety-reducing techniques. One such study examined third-graders in two separate

elementary schools. The children in both schools were given a modified questionnaire to measure each

participant's level of test-anxiety. Over a period of five weeks, children in one school were taught one

of two relaxation techniques: “elevator breathing,” which is a deep-breathing technique with a

visualization aspect especially for children, and “guided relaxation for children,” which is a form of

progressive neuromuscular relaxation, discussed later. In both cases, relaxation music was played while

relaxation scripts were read to the children. After a period of five weeks (training twice a week), a

second test-anxiety measure found that the experimental groups had significantly lower levels of test-

anxiety, however the authors were unable to prove statistically that the experimental groups had lower

general levels of anxiety than the control group after five weeks (Larson, Ramahi, Conn, Estes, and

3

Ghibellini, 2010).

Cohen and Ben-Zur (2008) proposed and found a correlation between test-anxiety, coping

strategies, and a third variable, “sense of coherence.” SOC is the participant's subjective measure of

problem-solving ability, and the authors were able to find strong evidence of a negative correlation of

SOC and test-anxiety, as well as a positive correlation with SOC and self-reported problem-focused

coping strategies. It is unclear if anxiety has a negative effect on SOC, if past success with problem-

focused (or any other) coping has a positive effect on SOC, if SOC affects both other factors, etc.

Regardless, the authors made the important contribution of finding that a feeling of self-efficacy seems

to be present in situations of low-anxiety and strong coping ability.

This idea is also explored by Largo-Wight, Peterson, and Chen (2005). Three separate

questionnaires on stress-related physical symptoms, self-reported problem-solving, and perceived level

of stress revealed again that there was a strong level of correlation between high levels of perceived

problem-solving ability, low levels of anxiety, and, in this case, even physical health.

While it seems apparently logical that anxiety is the overall determinant of physical symptoms

in the Largo-Wight, Peterson, & Chen (2005) study, the above examples also support that effective

coping skills/anxiety reducing techniques can counter the effects of anxiety. Because anxiety has so

many intangible causes, it stands to reason that the best way to combat it as a general ailment is to

practice a proven anxiety-reducing technique.

Music therapy is the practice of listening to certain types of music in an effort to replace anxiety

with relaxation. A session of music therapy is usually conducted in a room with minimal distractions,

and the type of music, both in and out of experimental studies, varies. It has been difficult to determine

if there is a single genre of music more effective than any other, as the personal preferences of

participants seems to have a strong mediating effect (Smith, 2008).

However, in two separate studies published together, Elliot, Polman, and McGregor (2011)

4

made an attempt to isolate factors that contribute to a particular piece's relaxation effects. The aim of

the research was to determine whether there were objectively more relaxing pieces of music, attempt to

rule out the need of the participant liking the music in order for it to be relaxing, and to find patterns of

elements common to music regarded as most relaxing. In the first study, participants were exposed to

thirty pieces of music that fit (sometimes only roughly) previously established standards of “relaxing”

music. During each piece of music, participants rated the songs of various factors related to subjective

experience of relaxation. After having heard every piece of music, participants were then asked to

grade various qualities they felt were the most important factors in determining whether they found a

piece to induce relaxation. It was found that liking or being familiar with the piece had little to no effect

for the participants to consider it relaxing. Other elements, such as tempo and melody, tended to be

rated as having a greater importance.

The second study attempted to address issues related to the lack of musical expertise present in

the sample of study one (i.e., participants often reported not knowing how to rate elements such as key

or articulation). A smaller group of well-trained music students participated in a focus group. Elliot,

Polman, and McGregor (2011) reviewed the reports from the previous study, and were then allowed to

listen to the same pieces rated as most and least relaxing. Afterward, the music experts discussed and

came to a consensus on what made each piece more or less relaxing, especially with regard to particular

musical elements that previous participants were unable to rate. It was found that pieces with tempos

between eighty and one hundred beats per minute, with predictable and easily hummed melodies,

subtle but consistent 4/4 beats, low complexity, and smoothly transitioning notes (most often from low

to high) tended to be pervasive throughout relaxing music, and the opposite of each element tended to

be rated as distracting (i.e., strong beats tended to induce an inclination to dance or move around).

Progressive neuromuscular relaxation is the practice of making focused efforts (often guided by

an instructor) to tense the body's various muscle groups, and subsequently relax each group, releasing

5

tension in the process. The key to guidance or practice in PNR is that many of the muscle groups are

commonly neglected, which does not exclude them from harboring stress-induced tension. It is

common for the release of a previously neglected muscle group to provide deep relaxation where the

participant was not consciously aware of tension (Naquin, Vest, Hurtt, and Cole, 2011).

Perhaps because music therapy and PNR do not interfere with one another, they are often

studied and practiced simultaneously. For example, Smith (2008) conducted a study on telemarketers in

the workplace in which she intended to explore to effectiveness of a single session of music therapy.

All of the participants completed the session together, having selected as a group live improvised

guitar. There was a significant decrease in state-anxiety from the beginning and end of the workday for

participants who received the treatment, suggesting that music therapy was highly effective, however

Smith included guided PNR in the session, making it unclear which technique produced the result.

Another study examines PNR and music therapy explicitly, and considers the combined

technique such that Smith used. In this study, participants were simply given a state-anxiety survey

before and after one of three experimental conditions (Music only, PNR only, Music & PNR). All three

conditions showed significant within-subjects increases from the pretest to the post-test, but there were

no significant differences between each group (Robb, 2000).

These results may serve as a sort of theoretical “control” group, but it does not address the

effect either of these techniques have performance under conditions of anxiety. Scheufele (1999)

addressed this in a study done with an element of stress-manipulation. Groups in each treatment

condition (PNR, music therapy, and control) were surprised with the task of preparing and delivering

an oral report under time constraints, and were given treatment immediately following the

announcement. After the treatment, participants were instead given a simple, repetitive task to

complete, and finally told that there would be no oral report. The experimenters recorded various

measures of attention, relaxation, and even observable stress responses throughout the experiment. It

6

was found that participants in the PNR group reported the greatest reduction in anxiety immediately

following the stress manipulation, as well as the highest level of relaxation at the end of the

experiment. Additionally, participants in the music therapy groups reported/exhibited the highest level

of distraction during the simple task, despite also receiving relaxation benefits from treatment.

Scheufele's (1999) study suggests that PNR could be more effective than music therapy in high-

stress situations, despite being equal with music therapy in Robb's study. However, neither study

addresses whether or not these techniques contribute to greater performance outcomes on difficult

tasks. It is also difficult to assess the effect either technique may have on subjective self-efficacy

measures, such as perceived problem-solving ability or Cohen and Ben-Zur's SOC. For that matter,

none of these studies seem to suggest that any one factor discussed here (anxiety level, SOC, perceived

problem-solving ability, choice of coping/relaxation techniques) has any direct effect on performance

on various tasks.

An extensive, if somewhat outdated, literature review (Zeidner, 1995) concurs that there is little

or no established evidence that any coping intervention has a consistently effect on the outcome of an

academic exam. The exceptions are usually negative, for instance, avoidance as a coping mechanism

has occasionally been shown to correlate with poorer scores on exams than any other mechanism or

control group.

The purpose of this study will be to uncover whether music therapy and/or PNR will reduce

observable stress-response behaviors and facilitate problem-solving. Using a classic “matchstick

arithmetic” puzzle, we will determine if music therapy, PNR, or a combination of the two results in

facilitation in solving the puzzle as opposed to a control group. This study will also examine observable

stress behaviors, determining whether the treatment groups show fewer signs of frustration with the

puzzle than the control group.

Hypothesis One. Music therapy will decrease the time it takes to solve the matchstick puzzle

7

relative to the control group.

Hypothesis Two. PNR therapy will decrease the time it takes to solve the matchstick puzzle

relative to the control group.

Hypothesis Three. Combined ttherapy will decrease the time it takes to solve the matchstick

puzzle relative to the control group.

Hypothesis Four. Music therapy will decrease stress-related behavior relative to the control

group.

Hypothesis Five. PNR will decrease stress-related behavior relative to the control group.

Hypothesis Six. Combined therapy will decrease stress-related behavior relative to the control

group.

Method

Participants

Eighty subjects will volunteer from the Department of Psychology subject pool of freshmen

enrolled in General Psychology. Students are required to volunteer for projects and receive credit

towards their grade in the class. The participants will be randomly assigned to one of four groups, a

no-treatment control group, a group given the music therapy treatment, a group given PNR, and a

group given both treatments. A counterbalancing random assignment procedure with equating for

gender to ensure there are equal numbers of males and females in each group will be followed. As the

participants arrive they will be assigned to groups as follows: control, music, PNR, both; both, PNR,

music, control; PNR, music; control, music, PNR, both; etc. This will be done for the males and

females separately. For example, the subjects will arrive at approximately the same time and will be

assigned a number based on their arrival time: First male, second male, third male, first female, second

female, third female, etc. The males will be then be assigned to groups and then the females will be

assigned to groups.

8

Materials

All participants must read and sign the University's informed-consent form (see Appendix A). A

simple five-item demographic survey will also be given to participants to collect data on age, gender,

year of education, whether they have ever tried music therapy prior to the study, and whether they have

ever tried PNR prior to the study (see Appendix B).

The experimenter will read instructions to each group (see Appendix C). Daniel Kobialka's

song Going Home Medley, will be played during music therapy and combined therapy (Robb 2000). A

Toshiba Model XC-IV CD player with stereo sound will be used to play the music. A script of guided

exercises used in PNR therapy will also be used to assist students with the therapy (see Appendix D).

The students will also be instructed in the matchstick arithmetic puzzle (see Appendix E). A

box of toothpicks (at least 13 toothpicks) will be used for participants to perform the matchstick

arithmetic puzzle. Also, a video camera (Panasonic digital palmcorder, Model PV-GS59) will be used

to record the behaviors of the participants during the task phase only. The behaviors observed will

include shifting in the seat, playing with hair, and nail-biting. See Appendix F for the behavioral

checklist. Scores on this checklist will be simply the number of exhibited behaviors per minute on

average. Scores will range from 0, indicating no stress-related behavior to some undetermined

maximum number per minute. Higher values indicate more stress. Finally, a stopwatch will be

necessary to record the time it takes participants to complete the matchstick arithmetic puzzle.

Procedures

Students will be assigned a subject number as they enter the meeting room (to be used to assign

them to groups). They will then sign and return the consent form. Those that consent will then be

given the demographic questionnaire to fill out. Once that is completed, they will be assigned to

groups as described above. Four counseling rooms will be used, one for each group. The subjects will

be escorted into the room by the experimenter and the instructions will be read. Students in the control

9

group will be allowed to do whatever they wish, study, read, use their phones, etc. Students n the

music therapy group will sit quietly and listen to the music. Students in the PNR group will engage in

the PNR exercises. Students in the Both group will engage in PNR exercises while listening to the

music. Sessions will last 30 minutes.

The students will then be instructed in the matchstick arithmetic puzzle, the video recorder

turned on, and they will then begin to solve the puzzle. They will be given 10 minutes to solve the

puzzle at which time they will be thanked for their participation, released , and the next four

participants will be brought into one of the four counseling rooms.

After all data is collected, the video recordings will be analyzed by two assistants who will be

blind to the experimental treatments given. A simple count of stress-related behavior will be recorded

by the two assistants and a reliability index calculated.

Results

Descriptive Statistics

The mean ages, year of education, and gender breakdown of the participants will be presented

for each group and in total in the text as well as in a table (see Table 1). Also, Table 2 will show the

proportion of participants in each group who solved the puzzle, the mean solution time and standard

deviations for those in each group who solved the puzzle, and the mean number of stress-related

behaviors. Median solution times may be shown also, if there are sufficient numbers of participants

who do not solve the puzzle in one or more of the groups.

Inferential Statistics

A 2x4 factorial analysis of variance will be used to determine whether the genders and groups

(no therapy, PNR, music therapy, or a combination) differed significantly from one another in terms of

time to solve the puzzle. The main effect of gender, main effect of group, and the interaction effect will

be reported in the form, F(X,XX) = X.XX, p<.05 for each effect. Given a significant interaction effect,

10

Table 1

Age and Gender Make-Up of the Groups

N Mean Age Males Females

Control XX XX XX XX

PNR XX XX XX XX

Music XX XX XX XX

PNR+Music XX XX XX XX

Total XX XX XX XX

Note: XX indicates a number will be provided once data is available.

11

Table 2

Number of Participants Who Solved the Puzzle, Time to Solution, and

Number of Stress-Related Behaviors

Number of

Solved

Mean Solution

Time (seconds)

SD

Stress Behaviors

Per Minute

Control XX XX XX XX

PNR XX XX XX XX

Music XX XX XX XX

PNR+Music XX XX XX XX

Note: XX indicates a number will be provided once data is available.

12

simple effects analysis will then be reported. Hypothetical results are shown in Figure 1, which shows

a significant interaction effect in which the males solve the puzzle sooner than the females in the no

therapy control group and the music therapy group, but the females solve the puzzle sooner than the

males in the PNR and combined therapies groups.

A 2x4 factorial analysis of variance will be used to determine whether the gender and groups

differed significantly on the number of stress-related behaviors. Main effects and the interaction effect

will be reported with the F-ratios, degrees of freedom, and the p-values. Simple effects analysis will be

conducted if the interaction effect is found significant, otherwise the main effect of the groups will be

further analyzed using Tukey's post hoc test. Hypothetical results are shown in Figure 2, which is

similar to the hypothetical results presented in Figure 1.

The Pearson product-moment correlation coefficient will be used to determine the correlation of

stress-related behaviors to the number of minutes it took the participants to solve the puzzle. It is

expected that no matter which group they were in, the correlation will be positive, indicating that more

stress-related behavior is associated with longer times to complete the puzzle. This is shown in Figure

3.

These fake findings show that music therapy improved the males' problem-solving speed and

reduced stress-related behaviors during the problem-solving task. However, PNR and combined

therapy did not differ from the no therapy control group. For the females however, music therapy,

PNR, and combined therapy showed marked improvement in problem-solving speed and reduced

number of stress-related behaviors.

Discussion

Basing this discussion on the hypothetical findings, the major result was the sex difference in

the effectiveness of the three therapies. Also interesting was the finding that speed of problem-solving

was negatively related to the number of stress-related behaviors during the problem-solving task.

13

14

15

16

In general the purpose of the study was to determine if a short one-session therapy could improve

cognitive ability. Music therapy is the one that works for males and females alike, while both PNR and

the combination of PNR and music therapy work for females too. The hypotheses were largely

supported with all three therapies reducing problem-solving time and number of stress-related behavior

for the females (only music therapy did so for the males).

One issue raised in the introduction was the relationship between stress and performance. In

this study a correlation was found between the two such that fewer stress-related behaviors during

problem-solving indicated faster times at solving the problem. Admittedly this is only correlational,

but nonetheless the findings show that those who do the relaxation protocols show reduced stress-

related behavior and enhanced problem solving ability.

These therapies are relative easy and inexpensive; they can readily be used in a dorm, an

apartment, or home. Students, and others engaged in problem-solving tasks in which enhance

cognitive ability would be of benefit, are encouraged to pursue these home therapies. Future research

might want to consider whether the results obtained here in the campus laboratory setting apply to the

real-world life of the dorm, the apartment, and the home. Groups of students could be shown how to

self-administer the therapies and then do so for one or more semesters to see if there grades improve. A

pretest-posttest design could be used (AB), or perhaps the ABA, return to baseline design, would be

more likely to establish a cause-effect relationship between the therapies and improved grades.

However the actual results turn out, they will confirm the hypothesis if either PNR or Music

therapy (or both) increases the proportion of participants that are able to solve the matchstick arithmetic

puzzle. Furthermore, the mechanics behind the facilitation of problem solving ability (i.e., the reduction

of interference from anxiety) can be supported with the result that observable stress behaviors are

reduced for the groups that received a relaxation treatment and also performed more quickly and more

successfully as a group on the puzzle. A result that would counter the hypothesis is one any in which

17

the treatment groups do not perform better (i.e., higher proportion of solved puzzles and/or faster solve

times) than the control group. It is also vital to the hypothesis that higher performance be accompanied

with fewer displays of anxiety, as this would undermine our proposed mechanism of facilitation of

problem solving ability.

Whatever the results, this study stands to contribute knowledge to the scientific community

about the effects of relaxation approaches on critical thinking and problem solving skills. Should the

results confirm the hypothesis that either PNR or Music therapy can facilitate problem solving ability

by reducing anxiety, many functions of society could benefit, such as executives and legislators making

clearer decisions under pressure, and students could use the therapies to achieve higher grades in

school. Confirmation of the hypothesis will also help to direct future research into exploring the links

between problem solving ability and anxiety, potentially spurring a search for the highest return of

problem solving facilitation for the effort and time of relaxation techniques. It may even become an

objective in the scientific community to find an easily streamlined relaxation therapy for use in various

fields where complex problems arise unexpectedly. If the results indicate that neither PNR nor Music

therapy contribute to problem solving ability, this study can serve to narrow the potential scope of

anxiety-reducing therapies, and in this way prevent further dead-end research.

18

References

Cohen, M., Ben-Zur, H., & Rosenfeld, M. J. (2008). Sense of coherence, coping strategies, and test

anxiety as predictors of test performance among college students. International Journal Of

Stress Management, 15(3), 289-303. doi:10.1037/1072-5245.15.3.289

Elliott, D., Polman, R., & McGregor, R. (2011). Relaxing Music for Anxiety Control. Journal Of Music

Therapy, 48(3), 264-288.

Knoblich, G., Ohlsson, S., Haider, H., & Rhenius, D. (1999). Constraint relaxation and chunk

decomposition in insight problem solving. Journal Of Experimental Psychology: Learning,

Memory, And Cognition, 25(6), 1534-1555. doi:10.1037/0278-7393.25.6.1534

Largo-Wight, E., Peterson, P., & Chen, W. (2005). Perceived problem solving, stress, and health among

college students. American Journal Of Health Behavior, 29(4), 360-370.

Larson, H. A., El Ramahi, M. K., Conn, S. R., Estes, L. A., & Ghibellini, A. B. (2010). Reducing Test

Anxiety among Third Grade Students through the Implementation of Relaxation Techniques.

Journal Of School Counseling, 8(19),

Morris, L. W., & Engle, W. (1981). Assessing various coping strategies and their effects on test

performance and anxiety. Journal Of Clinical Psychology, 37165-171.

Ricks, J., Naquin, M., Vest, A., Hurtt, D., & Cole, D. (2011). Examining the Results of Podcast

Relaxation Techniques in Higher Education. ICHPER-SD Journal Of Research, 6(1), 68-72.

Robb, S. (2000). Music Assisted Progressive Muscle Relaxation, Progressive Muscle Relaxation,

Music Listening, and Silence: A Comparison of Relaxation Techniques. Journal Of Music

Therapy, 37(1), 2-21.

Scheufele, P. M. (2000). Effects of Progressive Relaxation and Classical Music on Measurements of

Attention, Relaxation, and Stress Responses. Journal Of Behavioral Medicine, 23(2), 207.

Smith, M. (2008). The effects of a single music relaxation session on state anxiety levels of adults in a

19

workplace environment. Australian Journal Of Music Therapy, 1945-66.

Zeidner, M. (1995). Adaptive coping with test situations: A review of the literature. Educational

Psychologist, 30(3), 123.

20

Appendix A: Consent Form

Principal Investigator: __________________ Study Title: ____________________________________________________________ Institution: Middle Tennessee State University Name of participant: _________________________________________________________ Age: ___________ The following information is provided to inform you about the research project and your participation in it. Please read this form carefully and feel free to ask any questions you may have about this study and the information given below. You will be given an opportunity to ask questions, and your questions will be answered. Also, you will be given a copy of this consent form. Your participation in this research study is voluntary. You are also free to withdraw from this study at any time. In the event new information becomes available that may affect the risks or benefits associated with this research study or your willingness to participate in it, you will be notified so that you can make an informed decision whether or not to continue your participation in this study.

For additional information about giving consent or your rights as a participant in this study, please feel free to contact the MTSU Office of Compliance at (615) 494-8918.

1. Purpose of the study: Your participation in this study will help student researchers practice the elements of study design and execution, as well as data collection and analysis. The results of this study and your participation will never be published or used for any other reason.

2. Description of procedures to be followed and approximate duration of the study:

Participants will be asked to complete a short demographic questionnaire related to the specifics of the study. No personal or discomforting questions will be asked, and any question may be left unanswered.

3. Expected costs:

There are no costs to the participants other than their time.

4. Description of the discomforts, inconveniences, and/or risks that can be reasonably expected as a result of participation in this study: There are no expected discomforts, inconveniences, or risks expected as a result of participation in the study. Participants may feel some frustration with a puzzle task during the experiment, but the participants are not required to solve the puzzle to end the experiment.

5. Compensation in case of study-related injury: There is no compensation. No injury is expected as a result of any phase of this experiment.

6. Anticipated benefits from this study: The relaxation treatments in the study have been proven effective in other research, so you may feel less tense for a period as a result of participation. Participation should also result in good feelings, since you will be helping a fellow student with an assignment.

7. What happens if you choose to withdraw from study participation: Nothing. You may withdraw at any time for any reason with no recrimination.

8. Contact Information. If you should have any questions about this research study or possibly injury,

please feel free to contact Donald F. Kendrick, Professor of Psychology (615 898 2706).

21

9. Confidentiality. All efforts, within reason, will be made to keep the personal information in your research record private but total privacy cannot be promised. Your information may be shared with MTSU or the government, such as the Middle Tennessee State University Institutional Review Board, Federal Government Office for Human Research Protections, but only if you or someone else is in danger or if we are required to do so by law.

10. STATEMENT BY PERSON AGREEING TO PARTICIPATE IN THIS STUDY

I have read this informed consent document and the material contained in it has been explained

to me verbally. I understand each part of the document, all my questions have been answered, and I freely and voluntarily choose to participate in this study.

Date Signature of patient/volunteer

Consent obtained by: Date Signature Printed Name and Title

22

Appendix B: Demographic Survey

Name of participant: _________________________________________________________

1) What is your approximate age? Circle one:

a. younger than 18

b. 18 – 24

c. 25 – 30

d. older than 30

2) What is your gender? Circle one:

Male Female

3) What year of post-secondary education are you currently in? Circle one:

a. 1 st year

b. 2 nd

year

c. 3 rd

year

d. 4 th year

e. 5 th year or later

4) Have you had any previous experience with Music Therapy? If so, please explain:

No

Yes – ___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

5) Have you had any previous experience with Progressive Neuromuscular Relaxation? If so, please

explain:

No

Yes – ___________________________________________________________________

___________________________________________________________________

___________________________________________________________________

23

Appendix C: Instructions

There are no instructions available at this time. But if there were and there should be, then they would

go here. I've deleted them to save pages. You should have the idea by now. For all student projects

complete copies of all materials must be attached - this sample guide violates that rule (do as I say, not

as I do?).

24

Appendix D: The PNR Guide

I've deleted the Guide to save pages. You should have the idea by now. For all student projects

complete copies of all materials must be attached - this sample guide violates that rule (do as I say, not

as I do?).

25

Appendix E: The Matchstick Arithmetic Puzzle

Problem: VI – XII = IV

Solution:

VII = XI – IV Notes Each line in the equations above correspond to one toothpick

(i.e., V consists of two toothpicks). There are 13 total toothpicks in the puzzle. X's should be taken to represent the Roman numeral, not the multiplication sign.

26

Appendix F: Behavioral Checklist

 Nail-biting

 Sighing/huffing

 Shifting in place

 Tapping

 Vocalizing frustration or difficulty

 Grunting/growling

 Putting face/forehead in hands

 Running hands through hair/hair twisting

 Disengaging from the problem (i.e., looking away for 5 or more seconds)