Prosocial Regulatory Focus

Study 2 Examining Seven Mindsets

Authors

Affiliations

Zihan Hei

Binghamton University

Shane McCarty

Binghamton University

Promote Care & Prevent Harm

Jordan Booker

University of Missouri

Kyle Pacque

Promote Care & Prevent Harm

Heather Orom

University at Buffalo

Abstract

Promotion motivation and prevention motivation are two distinct motivational orientations, rooted in the promotion and prevention motivation systems, that influence whether one focuses on approaching gains and avoiding non-gains as well as approaching non-losses and avoiding losses, respectively. Individuals with a chronic preference for promotion focus are more likely to select promotion goals (that produce gains) and eager strategies; whereas people with a dominant prevention focus are more likely to select prevention goals (that produce non-losses) and vigilant strategies. More than two decades of reseach on regulatory focus theory shows motivational orientation influences health behavior, work behavior, and prosocial behavior (Scholer et al., 2019). However, promotion/prevention motivation has not been integrated with prosocial motivation to explain the goal pursuit strategy for prosocial behavior that benefits the health, safety, and wellbeing of others. In this pilot study, we examine the seven mindset types of prosocial regulatory focus: promote care, prevent harm, fail to care, fail to harm, maintain care, respond to harm, and flipping from harm to care.

Keywords

prosocial regulatory focus, pilot study

1 Descriptives

1.1 Distribution of Promoting Care Mindset

• The distribution of PROMOTECARE scores is left-skewed. Most scores are concentrated toward the higher end of the scale (5-7).This indicates that the PROMOTECARE mindset, which focusing on actively creating positive outcomes and wellbeing for others is strongly present in the majority of respondents. The concentration of scores at the upper end suggests promote care might be a common characteristic among the sample population.

## Distribution of PROMOTECARE
distri_of_procare <- ggplot(df, aes(x=PROMOTECARE)) + 
  geom_histogram(binwidth=0.5, fill="#8297ce", color = "black", alpha=0.7) +
  labs(title="Distribution of PROMOTECARE Scores", x="Score", y="Count")

distri_of_procare

1.2 Distribution of Preventing Harm Mindset

• The PREVENTHARM mindset histogram displays a bimodal distribution with peaks at scores 5 and 7. This pattern suggests two distinct subgroups exist within the sample population regarding their tendency to focus on preventing negative outcomes and protecting others.

## Distribution of PREVENTHARM
distri_of_preharm <- ggplot(df, aes(x=PREVENTHARM)) + 
  geom_histogram(binwidth=0.5, fill="#453a98", color = "black", alpha=0.7) +
  labs(title="Distribution of PREVENTHARM Scores", x="Score", y="Count")

distri_of_preharm

1.3 Distribution of Failing to Care Mindset

• The FAILCARE mindset histogram shows a bimodal distribution with primary peaks at scores 5 and 7. This distribution pattern suggests considerable variability in how participants experience anxiety about inadequately caring for others.

distri_of_failcare <- ggplot(df, aes(x=FAILCARE)) + 
  geom_histogram(binwidth=0.5, fill="#FF7F7F", color = "black", alpha=0.7) +
  labs(title="Distribution of FAILCARE Scores", x="Score", y="Count")

distri_of_failcare

1.4 Distribution of Failing to Prevent Harm Mindset

• The FAILHARM distribution does approximate a normal distribution centered around score 5. While there is some irregularity, the overall pattern shows a central peak with gradually decreasing frequencies toward both ends of the scale. This suggests that anxiety about failing to prevent harm to others follows a natural variation in the population.

distri_of_failharm <- ggplot(df, aes(x=FAILHARM)) + 
  geom_histogram(binwidth=0.5, fill="#FFA500", color = "black", alpha=0.7) +
  labs(title="Distribution of FAILHARM Scores", x="Score", y="Count")

distri_of_failharm

1.5 Distribution of Maintaining Care Mindset

• The MAINTAIN score histogram shows a bimodal distribution, with two distinct peaks, one around a score of 5 and another near 7. The presence of two modes indicates variability in how individuals approach the maintenance of positive impact.

distri_of_maintain <- ggplot(df, aes(x=MAINTAIN)) + 
  geom_histogram(binwidth=0.5, fill="#2AFEB7", color = "black", alpha=0.7) +
  labs(title="Distribution of MAINTAIN Scores", x="Score", y="Count")

distri_of_maintain

1.6 Distribution of Flipping from Harm to Care Mindset

• The FLIP score histogram shows a multimodal distribution, with noticeable peaks around scores of 5, 6, and 8. This pattern suggests that within the group, there are several subgroups with varying strengths in their ability to redirect focus from harm to care.

distri_of_flip <- ggplot(df, aes(x=FLIP)) + 
  geom_histogram(binwidth=0.5, fill="#90EE90", color = "black", alpha=0.7) +
  labs(title="Distribution of FLIP Scores", x="Score", y="Count")

distri_of_flip

1.7 Distribution of Responding to Harm

• The RESPOND score histogram shows a relatively uniform across the range of scores from about 6 to 8, with each score interval containing a similar number of responses. While there are a few lower and higher outliers, the central portion of the distribution appears fairly flat, indicating that participants are spread evenly in their tendency to take action and support recovery after harm.

distri_of_respond <- ggplot(df, aes(x=RESPOND)) + 
  geom_histogram(binwidth=0.5, fill="#FF77FF", color = "black", alpha=0.7) +
  labs(title="Distribution of RESPOND Scores", x="Score", y="Count")

distri_of_respond

1.8 Distribution of Seven Mindsets

• This violin plot displays the distribution of seven healthcare related mindsets (FAILCARE, FAILHARM, FLIP, MAINTAIN, PREVENTHARM, PROMOTECARE, and RESPOND) against their corresponding Z-Scores.

• The violin plots with overlaid boxplots show both the distribution shape and central tendency measures for each mindset. Most mindsets have Z-Scores centered around zero, indicating average performance, with PREVENTHARM showing a slightly lower median.

• The distributions appear relatively symmetrical, though PROMOTECARE shows some left skew with potential outliers at the lower end.

• The violin width suggests varying degrees of response consistency across mindsets, with FLIP and RESPOND showing somewhat wider distributions.

## Distribution of the 7 mindsets

df_long$Mindset <- factor(df_long$Mindset, levels = desired_order)

mindset_colors <- c(
  PROMOTECARE = "#8297ce",
  PREVENTHARM = "#453a98",
  MAINTAIN = "#2AFEB7",
  RESPOND = "#FF77FF",
  FLIP = "#90EE90",
  FAILCARE = "#FF7F7F",
  FAILHARM = "#FFA500"
)

distr_allmindset <- ggplot(df_long, aes(x = Mindset, y = Z_Score, fill = Mindset)) +
  geom_violin(trim = FALSE, alpha = 0.4) +
  geom_boxplot(width = 0.1, outlier.shape = NA, alpha = 0.6) +
  scale_fill_manual(values = mindset_colors) +
  labs(title = "Distribution of the 7 Mindsets",
       y = "Z_Score", x = "Mindset") +
  theme(axis.text.x = element_text(angle = 20))


distr_allmindset

2 Mindset Associations

The correlation heatmap shows the correlations among the seven mindsets with green-yellow indicating the strongest, positive correlations.

• FAILHARM & FAILCARE: Since both mindsets represent anxiety about not meeting one’s caregiving goals, it makes sense that they are positively correlated (r = 0.66).

• PROMOTECARE & MAINTAIN: A strong correlation (r = 0.67) suggests that individuals who emphasize promote care also tend to maintain it.This makes sense because those who focus on creating positive outcomes (PROMOTECARE) are also likely to be committed to sustaining those outcomes (MAINTAIN).

(expand.grid(): creates all combinations of the column names (X) and row names (Y) of the matrix.)

mindsets_relation <- plot_ly(
  type = "heatmap",
  x = heatmap_data$X, 
  y = heatmap_data$Y, 
  z = heatmap_data$Z, 
  text = round(heatmap_data$Z, 2),
  texttemplate = "%{text}",
  hovertemplate = "<b>Correlation Details</b><br><b>X:</b> %{x}<br><b>Y:</b> %{y}<br><b>r:</b> %{z}<extra></extra>",
  colorbar = list(title = "Correlation Level (r): <br>")
  ) %>% 
  plotly::layout(
    title = "Mindset Correlation Heatmap",
    xaxis = list(title = ""),
    yaxis = list(title = "")
  )

mindsets_relation

2.1 Promoting Care & Preventing Harm Mindsets

• The scatter plot demonstrates a moderate positive relationship between the PROMOTECARE and PREVENTHARM mindsets, indicating that individuals who tend to focus on actively creating positive outcomes for others are also more likely to prioritize preventing harm.

• Although the relationship was not completely linear, suggesting that there are differences in how these mindsets are expressed between individuals, the overall trend suggests that there may be a complementary link: individuals who focus on promoting well-being are also mindful of minimizing risk.

## Standardize variables to z-scores
df$PROMOTECARE_zscore <- as.numeric(scale(df$PROMOTECARE))
df$PREVENTHARM_zscore <- as.numeric(scale(df$PREVENTHARM))

## compare promotecare & preventharm by scatter plot
ggplot(df, aes(x = PROMOTECARE_zscore, y = PREVENTHARM_zscore)) +
  geom_point(alpha = 0.5) +                       
  geom_smooth(method = "lm", color = "blue") +   
  labs(title = "Comparison of PROMOTECARE vs PREVENTHARM (Z-Scores)",
       x = "PROMOTECARE Z-Score", y = "PREVENTHARM Z-Score")

3 Differences by Sociodemographics

3.1 Promoting Care Mindset by Gender Identity

• According to the data, individuals who identify as cismen tend to score higher on the PROMOTECARE measure compared to other gender identities in the sample.

## PROMOTECARE score distributions across gender identity

gender_procare <- ggplot(df, aes(x = demo1, y = PROMOTECARE, fill = demo1)) + 
  geom_boxplot() +
  labs(title ="PROMOTECARE Scores by Gender Identity", x ="Gender Identity", y ="PROMOTECARE Score")+
  guides(fill = guide_legend(title ="Gender Identity:"))

gender_procare

3.2 Promoting Care Mindset by Age

• Age group does not appear to differ by PROMOTECARE scores, but individuals in the 55-64 age group tend to have higher PROMOTECARE scores, which means they may have a stronger focus on creating positive outcomes for others.

## PROMOTECARE score distributions across age group

age_procare <- ggplot(df, aes(x = demo6, y = PROMOTECARE, fill = demo6)) + 
  geom_boxplot() +
  labs(title ="PROMOTECARE Scores by Age Group", x ="Age Group", y ="PROMOTECARE Score")+
  guides(fill = guide_legend(title ="Age Group:"))+
  theme(axis.text.x = element_text(angle = 20))  


age_procare

3.3 Failing to Care Mindset by Racialized Identity

• FAILCARE scores did differ slight across racial identities.
• Although people who identify as Asian tend to have higher FAILCARE scores, which means they have a greater worries about failing to create positive outcomes for others.

## FAILCARE score distributions across racial group

racial_failcare <- ggplot(df, aes(x = demo2, y = FAILCARE, fill = demo2)) + 
  geom_boxplot() +
  labs(title ="FAILCARE Scores by Racial Group", x ="Racial Group", y ="FAILCARE Score")+
  guides(fill = guide_legend(title="Racialized Group:"))+
  theme(axis.text.x = element_text(angle = 20))  

racial_failcare

3.4 Failing to Prevent Harm Mindset by Educational Level

• As education level increases, individuals tend to endorse lower scores on failing to prevent harm mindset. Simply, this means people with more formal education have a less anxiety about failing to prevent negative outcomes.

## FAILHARM score distributions across education level

education_preharm <- ggplot(df, aes(x = demo3, y = FAILHARM, fill = demo3)) + 
  geom_boxplot() +
  labs(title = "FAILHARM Scores by Education Level", x ="Education Level", y ="FAILHARM Score")+
  guides(fill = guide_legend(title ="Education Level:"))+
  theme(axis.text.x = element_text(angle = 20))

education_preharm

3.5 Paired Comparisons of PROMOTECARE vs. PREVENTHARM

• The violin plot provides a visual comparison of the distribution of z-scores for the PROMOTECARE and PREVENTHARM mindsets. While both distributions show some overlap, PROMOTECARE generally exhibits higher z-scores, with a larger concentration of participants scoring above the mean compared to PREVENTHARM.

• The individual lines connecting the same participant across both mindsets show varying patterns, but many participants score higher in PROMOTECARE than in PREVENTHARM. This suggests that, on average, participants may be more oriented toward actively promoting positive outcomes than solely focusing on preventing harm.

• The wider spread in PREVENTHARM scores also indicates greater variability in how strongly participants endorse this mindset.

## Compare PROMOTECARE vs. PREVENTHARM using violin plots

procare <- ggplot(df_zscore, aes(x = variable, y = zscore)) +
  geom_violin(aes(fill = variable), alpha = 0.6) +
  geom_point(aes(group = participantID), size = 1.5, alpha = 0.4)+
  geom_line(aes(group = participantID), color = "black", alpha = 0.3) +
  scale_fill_manual(values = c("PROMOTECARE" = "#8297ce", "PREVENTHARM" = "#453a98")) +
  labs(title="Comparison of PROMOTECARE vs. PREVENTHARM", x="Mindset", y="Z-score")

procare  

4 Within-Person Analysis of Mindsets

4.1 Mindsets of Five Randomly Selected Participants

desired_order <- c("PROMOTECARE", "PREVENTHARM", "MAINTAIN", "RESPOND", "FLIP", "FAILCARE", "FAILHARM")

plot_mindsets <- function(person_id) {
  df_person <- df %>% 
    filter(participantID == person_id) %>% 
    select(PROMOTECARE, PREVENTHARM, FAILCARE, FAILHARM, MAINTAIN, FLIP, RESPOND) %>% 
    pivot_longer(cols = everything(), names_to = "variable", values_to = "value") %>%
    mutate(variable = factor(variable, levels = desired_order)) %>%
    mutate(zscore = scale(value))

  ggplot(df_person, aes(x = variable, y = zscore, group = 1)) +
    geom_point(size = 4, color = "#453a98") +
    geom_line(color = "#8297ce") +
    labs(
      title = paste("Comparison of All Mindset Scores for Participant", person_id),
      x = "Mindset", y = "Z-score"
    ) +
    theme_minimal() +
    theme(axis.text.x = element_text(angle = 20, hjust = 1))

}

plot_mindsets(random_ids5[1])
plot_mindsets(random_ids5[2])
plot_mindsets(random_ids5[3])
plot_mindsets(random_ids5[4])
plot_mindsets(random_ids5[5])

4.2 Mindsets of Ten Randomly Selected Participants

ggplot(df_plot, aes(x = variable, y = zscore, color = participantID, group = participantID)) +
  geom_point(size = 3) +
  geom_line(size = 1) +
  labs(
    title = "Mindset Score Comparison (Z-Scores) for 10 Random Participants",
    x = "Mindset Type",
    y = "Z-score",
    color = "Participant ID"
  ) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 20, hjust = 1))

4.3 Combine Multiple Plots with patchwork

plots <- lapply(random_ids5, function(id) {
  plot_mindsets(id) + 
  theme(plot.title = element_blank()) # Remove individual plot titles
})

# Combine them with patchwork
combined_plot <- plots[[1]] + plots[[2]] + plots[[3]] + plots[[4]] + plots[[5]] +
  plot_layout(ncol = 2, nrow = 3) + 
  plot_annotation(title = "Mindset Score Comparison for 5 Random Participants")

combined_plot