Power & Pain
In the current project, we investigate whether people in power seem less vulnerable to experiences of pain, i.e. superhuman. We hypothesized that power manipulations make those in high-power roles seem dehumanized and invulnerable to pain.
In study 1, we test this with an in-lab dyad study. We predicted that participants induced into a low-power perspective will report that their high-power counterparts experience less physical and social pain compared to how their high-power counterparts rate their experiences of physical and social pain.
In study 2, we test this hypothesis by measuring how people expect a political candidate (high-power condition) or a campaign worker (low-power condition) to experience the same medical emergency.
IRB materials:
These studies all operated under IRB protocol #2311 “Social Interaction”
Pre-registration
- it’s complicated
- Hypothesis: Power manipulations make those in high-power roles seem dehumanized and invulnerable to pain. Specifically: after completing a role-playing task, participants induced into a low-power perspective will report that their high-power counterparts experience less physical and social pain compared to how their high-power counterparts rate their experiences of physical and social pain.
- Dependent variable:
Humanization: 10-item scale (0 not at all - 6 extremely) adapted from Weisman et al. 2017. Scale is composed of 3 subscales: bodily experiences (items 1 - 3), heartfelt experiences (items 4 - 7), and mind experiences (items 8 - 10).
- How capable is s/he of experiencing fear?
- How capable is s/he of experiencing joy?
- How capable is s/he of being self-aware?
- How capable is s/he of feeling love?
- How capable is s/he of experiencing guilt?
- How capable is s/he of knowing right from wrong?
- How much embarrassment would s/he feel in an embarrassing situation?
- How capable is s/he of working toward a goal?
- How capable is s/he of remembering?
- How capable is s/he of reasoning about things?
Physical pain sensitivity: 10-item scale (1 not at all painful - 4 extremely painful) Answer the following questions about the [manager/assistant] you just participated with. To what extent would that person find the following events painful?
- This person drops a pot on their toe
- This person slams the car door on their finger
- This person gets an injection in their arm
- This person gets shampoo in their eye
- This person fractures their ring finger
- This person burns their palm on an oven set to 425F
- This person steps on a 3” piece of glass and gets an incision that requires 4 stitches
- This person fractures their shin bone
- This person gets a migraine
- This person gets a poison ivy rash
Physical pain tolerance: 10-item scale
Answer the following questions about the [manager/assistant] you just participated with. How do you think that person should cope with their pain in each of the following events? [painful events from physical pain sensitivity scale repeated below]
1 = No action necessary
2 = Use non-medication coping strategies (e.g. massaging the area, resting, relaxing, or meditating)
3 = Take a small dose of over the counter pain medications (e.g. one Tylenol or Ibuprofen)
4 = Take the max dosage of multiple over the counter pain medications (e.g. alternate Tylenol & Ibuprofen every 2 hours)
5 = Take powerful pain medication (e.g. Oxycodone or Vicodin) prescribed by their primary care doctor
Social pain sensitivity: 10-item scale (1 = not at all painful, 4 = extremely painful)
Answer the following questions about the [manager/assistant] you just participated with. To what extent would that person find the following events painful?
1. This person’s best friend moves across the country
2. This person realizes after walking around all day that a pair of underwear was stuck to the back of their shirt.
3. Strangers laugh at this person’s haircut
4. This person’s best friend gossips about them behind their back
5. A friend makes fun of this person in front of others
6. This person invites friends over to celebrate their birthday and no one comes
7. This person overhears a coworker talking about their incompetence at work
8. This person’s family pet passes away
9. This person trips and falls and people laugh at them
10. This person’s romantic partner asks them for some ‘space’
Social pain tolerance: 10-item scale
Answer the following questions about the [manager/assistant] you just participated with. How much support do you think that person should seek out to cope with each of the following events? [painful events from social pain sensitivity scale repeated below]
1 = no action necessary
2 = Use personal coping strategies (e.g., do a calming activity, take a walk, pray/meditate)
3 = Seek minor informal support from friends and family
4 = Seek maximum informal support from friends and family
5 = Request formal support from a mental health professional (e.g., clinical psychologist, counselor, psychiatrist)
Conditions Two conditions: low-power and high-power. Participants randomly assigned to the low-power condition will be assigned the ‘assistant’ role in the role-playing task. Participants randomly assigned to the high-power condition will be assigned the ‘manager’ role in the role playing task.
Analyses. First we will test if participants’ reports of their partners’ pain tolerance differs significantly based on their partner’s gender. We will fit a multilevel mixed effects model for each of the 5 DVs with a fixed effect for partner’s gender and a random effect for each dyad’s intercept. lmer(DV ~ partnerGender + (1|dyadID)) If any of the models reveal a significant effect of gender, we will include partner gender as a control variable in the models that test our hypotheses, detailed below.
To test our main hypotheses, we will fit a multilevel mixed effects model for each of the 5 DVs with a fixed effect for condition and a random effect for each dyad’s intercept. lmer(DV ~ condition + (1|dyadID))
Outliers and exclusions. We will not analyze the data of any participants who do not grant permission to use their data after they were debriefed and informed of the deception. We will exclude any participants who knew each other (within dyad) as indicated by the research assistants’ or the participants’ notes. We will also review the research assistants’ and the participants’ notes and exclude any participants in cases where the notes suggest that the data may be compromised (e.g. a fire alarm interrupted the study).
Sample size. We will collect a minimum of 150 participants, but we aim to collect as many participants as we are afforded by our university’s student participant pool credit limit. If the credit limit is lifted during the semester that we are collecting, then we will collect as many participants as possible until the last day of the semester.
Anything else? In response to question 1: We have begun collecting data in-lab but have not analyzed or downloaded any data yet.
Title: Power & Pain: In-lab Dyad Study
Type of study: experiment
Data source: university lab
Materials
- run in-lab from 11/27/23 until 02/19/24, across 2 semesters
- Qualtrics title: The Invulnerable Powerful
- Drive folder of materials & participant log
- PIDs that didn’t sign/complete/consent debrief release: 22 (dyad 11), 47 (dyad 24), 109 (dyad 54), 116 (dyad 58),
- PIDs that knew each other: 66 & 67 (dyad 33), 166 & 167 (dyad 83), 182 & 183 (dyad 91)
Results
We collected a total of N = 192 responses. After exclusions for not signing the debrief release form or for two participants knowing each other within dyad, we have a final total N = 176 responses, with 88 in each condition.
Note that I reverse scored the 4 variables for tolerance and sensitivity.
Gender effects
Per the pre-registration, we first tested each of the 5 DVs for gender effects using a nested model: lmer(DV ~ partnerGender + (1|dyadID)). Partner’s gender was a significant predictor for the “social pain tolerance” DV. That is, participants rated males and females were differently on how they should tolerate social pain: people were more likely to suggest more extensive social coping strategies for females than males in the aftermath of a socially painful event. (REMINDER: tolerance items are reverse coded). So we will include partner’s gender as a control in the model that tests our main hypothesis for that particular DV.
Main hypothesis testing
None of the models yielded a significant condition effect. However, the DVs were all trending in the predicted way. Assistants tended to think their managers experienced less pain, needed less coping strategies for painful events, and were less humanized.
Plots
Exploratory Analysis
Break humanization scale into subscales. In particular because we expect that the ‘mind experiences’ subscale performs very differently than the other two.
Qualtrics title: Invulnerable Powerful 2
We collected 300 Prolific workers with a target to achieve 80% power to detect an effect size of d = 0.30.
We launched the study on Prolific at about 4:00PM Eastern time on 09/12/2024.
Procedure
Participants were randomly assigned to evaluate a person in a position of power (power condition) or evaluate someone in a subordinate position (control condition). Participants read that they would be playing the role of a junior emergency room doctor and they will be recommending treatment for an ER patient. In the power condition, the patient was a candidate running for a member of parliament in Canada. In the control condition, the patient was a campaign worker for the MP candidate’s campaign. Participants read about the patient’s medical episode that brought him to the ER and were tasked to choose between two treatment options. One treatment option (Treatment A) was relatively painless but took 4 weeks to administer, and the other treatment option (Treatment B) was a painful surgery - due to the patient’s chronic medical syndrome that makes anesthesia less effective - that was immediately effective and has higher chance of success. After indicating their preference for treatment A or B, participants indicated how many weeks the patient should take off from work to recover for each treatment plan. Next, participants responded to a 10-item humanization scale, a 10-item scale of physical pain sensitivity, a 10-item scale of physical pain tolerance, a 10-item scale of social pain sensitivity, and a 10-item scale of social pain tolerance.
Participants were told that the man pictured above is their patient, and that he is either the candidate for MP (Don Stewart) or a campaign worker for Don Stewart’s campaign.
The main DVs of interest were 1) the participant’s preference for the relatively painless treatment plan vs the painful treatment, 2) the amount of weeks that the participant recommended the patient take off from work following each treatment option, the participant’s ratings of 3) how humanized the patient seems, 4) how much physical pain and 5) social pain the patient can feel, 6) how much support the patient needs to deal with physical pain and 7) social pain.
Results
We collected a total of N = 300 responses from Prolific workers. We excluded the med student, nurse practitioner, 11 nurses, and 3 physicians from analysis on the basis that their experience could change their interpretation and understanding of the medical case that they evaluated.
career | n |
|---|---|
not in med field | 258 |
another role in med field | 21 |
nurse | 11 |
med admin | 4 |
physician | 3 |
1 | |
med student | 1 |
nurse practitioner | 1 |
The final sample consisted of N = 284
condition | n |
|---|---|
candidate | 144 |
worker | 140 |
Race | n |
|---|---|
White | 230 |
Asian | 27 |
Black | 14 |
Not listed | 8 |
Arab | 2 |
1 | |
Amer. Indian/Alaskan,Asian,Polynesian/Pacific Islander,White | 1 |
Black,White | 1 |
gender | n |
|---|---|
Woman | 164 |
Man | 111 |
Non-binary | 3 |
Transgender | 2 |
Woman,Man | 2 |
Not listed | 1 |
Rather not say | 1 |
career | n |
|---|---|
not in med field | 258 |
another role in med field | 21 |
med admin | 4 |
1 |
Plots
Treatment choice and recommended weeks off
Humanization, pain sensitivity and tolerance
variable | statistic | df | p | effsize |
|---|---|---|---|---|
humanization | -1.20 | 282 | .231 | -0.14 |
humanization_body | -1.35 | 282 | .179 | -0.16 |
humanization_heart | -0.72 | 282 | .472 | -0.09 |
humanization_mind | -1.39 | 282 | .164 | -0.17 |
physicalPain_Sensitivity | -0.58 | 282 | .562 | -0.07 |
physicalPain_Tolerance | 0.08 | 282 | .939 | 0.01 |
socialPain_Sensitivity | -0.73 | 282 | .464 | -0.09 |
socialPain_Tolerance | 1.82 | 282 | .070 | 0.22 |
Notes
- people in positions of power seem less humanized and more tolerant of social pain, not superhuman but something else, another form of dehumanization
Study 2.2 was a replication of Study 2, run on the psychology department participant pool.
Qualtrics title: Invulnerable Powerful 2.2
Results
We collected a total of N = 290 complete responses from psychology undergraduate students.
condition | n |
|---|---|
control | 145 |
power | 145 |
Plots
Treatment choice and recommended weeks off
Humanization, pain sensitivity and tolerance
TBD