History Embedded in Physical Spaces - Study 5: Positive v Negative Impressions of History

Link here (private link at the time of writing this)

Hypothesis

In general, this project investigates the effect of physical proximity to places where past institutional wrongdoing occurred on people’s perceptions of temporal distance of the wrongdoing from the present and the continuity of the institution over time. Generally, when engaging with history that is not threatening (e.g., history of institutional achievement), we expect that physical and temporal proximity should be positively correlated, whereby learning about events in an institution’s history that happened in the place where a person is physically present should prompt the historical event to feel closer to the present (versus someone who is further away). However, we expect that this space-time relationship should be attenuated or diminished when the institutional history implicates institutional wrongdoing. In particular, we expect that people will feel threat when learning about past wrongdoing committed by valued (i.e., self-relevant) institutions and are physically at (versus distant from) the location of wrongdoing.

Therefore, we expect that the nature of the history described (wrongdoing/positive) and participants’ physical proximity to the location of the history (proximal vs. distal) will shape participants’ perception of the history and the deployment of defensive reactions (e.g., Peetz et al., 2010). We aim to test identification as an exploratory predictor.

In this study, we measure two key relevant perceptions of history and institutions, both of which we measure at baseline and then again after history exposure. We measure subjective temporal distance of the history and the perceived continuity/discontinuity of the institution across time (i.e., is the present institution perceived as connected to or a distinct entity from previous versions of itself).

When the institutional history that people read about is positive, we expect that participants will not engage in defensive mechanisms and will perceive the institution as more continuous and the history as closer to the present compared to baseline, particularly when learning about the history while physically proximal (vs. distant).

However, we expect that when learning about a history of institutional wrongdoing (i.e., dark/negative history), participants will be more likely to employ defensive mechanisms. That is, we expect that after learning about a history of institutional wrongdoing, people will be more likely to perceive the institution as more discontinuous over time (i.e., as a distinct entity from the historical institution that committed the wrongdoings) and to perceive the history as temporally further away from the present.

Moreover, we expect that perceived geographical distance from the history will not be as strongly related or predictive of temporal distance perceptions in the dark/negative history condition as in the positive history condition.

Given we imagine that it is important for the described institution to be self-relevant, we utilize the history of an educational institution and examine how current students at the institution perceive the history. Formally, we will employ a 2 x 2 x 2 experimental design with 2 between-participant factors (physically close V physically far; positive history V dark/wrongdoing history) and one within-participant factor (timepoint) that will allow us to examine change from baseline.  We predict a significant two-way interaction of experimental condition on change in DV. We hypothesize that participants who read a positive history piece about their institution will report greater institutional continuity and will feel this historical time period as happening closer to the present (compared to baseline). We expect that these effects will be exacerbated when they are close to (vs far from) the physical space where the positive history took place. In contrast, we predict that people who read a dark/negative history piece in the same space it occurred will perceive greater institutional discontinuity and will perceive the historical time period as more distant from the present (compared to baseline). We expect that the effect of physical proximity condition on these perceptions will be mitigated in this dark/negative history condition.

Dependent variable

Temporal distance and institutional continuity (detailed below) will be measured on a department wide “pretest” before students participate in the study, and will be measured again after participants read the manipulation materials. The other two DVs listed below will only be collected after the manipulation.

Subjective geospatial distance: Please think about the physical location of where UVA hospital was first opened. Remember, UVA hospital was where eugenics/medical research was originally practiced on Grounds. Thinking about the physical location of where UVA hospital was first opened, how far away does that feel to you? Use the slider below to indicate how far away from where UVA hospital was first opened feels to you currently. 0 = Feels very close to 100 = Feels very distant

Temporal distance (z-scored):

  1. The past may feel quite close or far away, regardless of the amount of time that has actually passed. Please think about when UVA hospital first opened. Remember, UVA hospital was where eugenics/medical research was originally practiced on Grounds. How far away in time does the opening of UVA hospital feel to you? Use the slider below to indicate how far away in time the opening of UVA hospital feels from the present. 0 = Feels very recent to 100 = Feels very distant
  2. On the below scale, please indicate how far away in time the opening of the first UVA hospital feels to you. 1 = Feels like yesterday to 100 = Feels very far away

Institutional continuity [1 = totally disagree, 7 = totally agree]: Below we list all items and plan to treat the first 9 items as an omnibus measure per previous studies (dropping items that do not hold together, if alpha >= .7), and test the defined subscales separately (if alphas >= .7).

Cultural subscale 1) The university’s founding values and worldview have endured over time. 2) The University of Virginia will always be characterized by specific traditions and beliefs 3) UVa has changed its traditions and customs throughout history - reversed

Temporal subscale 4) There is a causal link between different events in UVa history 5) Major phases in UVa history are linked to one another 6) There is no connection between past, present, and future events at the university reversed Diversity value subscale 7) The value UVa places on diverse people and ideas has been stable over time 8) The rules about who can enroll as a student are different now than when the university was founded reversed 9) Institutional messaging about who is welcome at UVa has been continuous throughout its history

Institutional continuity (exploratory subscales). We plan to test these 2 subscales separately from the prior omnibus measure of continuity (alpha/pearson’s r >= .7). These measures were only measured at Time 2.

Physical subscale 10) The physical aesthetic of UVA has remained constant over time 11) UVA’s hallmark buildings have endured over time 12) UVA has significantly changed its architecture throughout its history 13) UVA has changed the way it looks throughout history Leadership subscale 14) The people who run UVA today have a different vision for the university than people who have run UVA in the past 15) UVA leadership has largely taken the same approach to running the university throughout the university’s history

Indirect consequences measure (Ripple effect) For how many years did the original eugenics/medical research practices of UVA hospital likely have a direct or indirect impact on [Black/women/people with lower-income] residents living in Charlottesville? Scale is 1901 to 2025

Estimated actual temporal and physical distance: Temporal 1) Please think about the year UVA hospital first opened. Use the slider below to estimate how many years ago UVA hospital first opened. 0 = 0 years ago to 200 = 200 years ago

Physical distance 1) Please think about the physical location of where UVA hospital was first opened. Use the slider below to estimate how many feet away you currently are from where UVA hospital was first opened. 0 = 0 feet away to 10,000 feet away

Conditions

Four experimental conditions in a 2x2. Factor 1: Positive v dark/wrongdoing history Factor 2: Physical proximity (close v far from where described history occurred)

We also measure participants responses to key DVs pre and post the experimental manipulation as a within-participants factor.

Analyses

For the DVs where we collect pretest measures (i.e. institutional continuity and temporal distance), we will employ a pre/post analysis. We will estimate a regression model predicting the DV from two experimental factors - history condition (two levels: positive and dark/wrongdoing) and proximity (two levels: close and far). The model will include participants’ pretest scores as a covariate to adjust for baseline differences.

We will also examine whether these effects of subjective distance hold when covarying estimates of actual geographical and temporal distance (measured at Time 2), and whether the DV is predicted by two experimental predictors in a similar manner.

Outliers and Exclusions

In some cases, participants submit multiple responses to the departmental pretest. In those cases, we prioritize the submissions in the following order: 1) complete submissions, 2) earliest submission (if both/all submissions are complete) If participants complete our study but do not have complete pretest data on our two measures of interest, we will not be able to analyze their data.

Sample Size

We will collect as many participants as we can afford per 1) our psychology department’s participant pool of undergraduate students, and 2) time until the semester ends. The amount credited per researcher varies by semester per size of the pool. In the Fall of 2025, we were allotted credits to collect 400 participants. We will continue collecting participants until we reach our credit limit or until the semester ends.

Other

In exploratory analyses, we will examine other variables that we collected, including: time spent reading the manipulation, geographic distance as a moderator for the other two DVs of interest, ingroup identification as a moderator of experimental condition, physical and leadership continuity as a DV, the perceived ripple effects of the historical events that participants read about, and measures of group loyalty as a DV and moderator.

Pretest

show code
#check eligiblity: load pretest and compile averages for the UVA institutional continuity scale ("UVA History 1 - 9") and the two temporal distance measures that correspond to the opening of UVA's first hospital (temporal distance 3 and 4) so that we can check for students who completed those measures in particular
study5_eligibility = read.csv('Prescreen and Pretest data/F25 Pretest.csv') %>% 
  mutate(institutionalContinuity = rowMeans(select(.,matches("UVaHistory")), na.rm=FALSE),
         temporalDistance_hospital = rowMeans(select(.,c('temporalDistance_3', 'temporalDistance_4')), na.rm=FALSE)) %>% 
  mutate(i5_eligibility = ifelse(is.na(institutionalContinuity), 0, 
                                 ifelse(is.na(temporalDistance_hospital), 0, 1))) %>% 
  select(computID, i5_eligibility)

#load pretest data for analysis
study5_pretest = read.csv('Prescreen and Pretest Data/F25 Pretest.csv') %>% 
  mutate(across(matches(c("UVaHistory_3","UVaHistory_6","UVaHistory_8")), ~ 8 - .), #reverse code items 3, 6, 8 in historical continuity scale 
         across(matches(c("physicalContinuity_3", "physicalContinuity_4", "physicalContinuity_5")), ~ 7-.)) %>%  #reverse code items 2, 3, 5 from the physical continuity scale NOTE the design mistake in Qualtrics that resulted in a missing scale option (somewhat disagree) which reduces the scale to 1-6  
  mutate(pretest_institutionalContinuity = rowMeans(select(.,matches("UVaHistory"), matches("physicalContinuity_5"), matches('physicalContinuity_6')), na.rm=TRUE),
         pretest_institutionalContinuity_cultural = rowMeans(select(.,c('UVaHistory_1', 'UVaHistory_2', 'UVaHistory_3')), na.rm=TRUE),
         pretest_institutionalContinuity_temporal = rowMeans(select(.,c('UVaHistory_4', 'UVaHistory_5', 'UVaHistory_6')), na.rm=TRUE),
         pretest_institutionalContinuity_diversity = rowMeans(select(.,c('UVaHistory_7', 'UVaHistory_8', 'UVaHistory_9')), na.rm=TRUE),
         pretest_institutionalContinuity_leadership = rowMeans(select(.,c('physicalContinuity_5', 'physicalContinuity_6')), na.rm=TRUE),
         pretest_ingroupID = rowMeans(select(.,matches("IngroupID")), na.rm=TRUE),
         pretest_physicalContinuity = rowMeans(select(., c('physicalContinuity_1','physicalContinuity_2','physicalContinuity_3','physicalContinuity_4'))),
         pretest_temporalDistance = (as.vector(scale(temporalDistance_3)) + as.vector(scale(temporalDistance_4))/2)) 

First, potential participants responded to four measures on a departmental pretest to serve as pretest scores. These four measures were captured again when participants were taking part in the study to serve as the post-test scores, which allows us to employ a repeated-measures design for 4 DVs.

Participants were eligible to participate in Study 5 if they responded to all of the items on the following two pretest measures:

pretest_institutionalContinuity [1 = totally disagree, 7 = totally agree]

  1. The university’s founding values and worldview have endured over time.
  2. The University of Virginia will always be characterized by specific traditions and beliefs
  3. UVa has changed its traditions and customs throughout history reversed
  4. There is a causal link between different events in UVa history
  5. Major phases in UVa history are linked to one another
  6. There is no connection between past, present, and future events at the university reversed
  7. The value UVa places on diverse people and ideas has been stable over time
  8. The rules about who can enroll as a student are different now than when the university was founded reversed
  9. Institutional messaging about who is welcome at UVa has been continuous throughout its history

pretest_temporalDistance [z-scored]

  1. Think about when UVA hospital first opened (1901). How far away in time does the opening of UVA hospital feel to you? Use the slider below to indicate how far away the opening of UVa hospital feels from the present [0 - 100 pt slider]
  2. On the below scale, please indicate how far away in time the opening of UVA hospital feels to you. [likert: 1 = feels like yesterday, 7 = feels very far away]

** The following scale was also on the pretest, but was not required for eligibility: **

pretest_physicalContinuity [1 = totally disagree, 6 = totally agree]

  1. The university’s aesthetic has endured over time
  2. The University of Virginia has a hallmark buildings and aesthetic has remained constant over time
  3. UVa has changed its architecture throughout history reversed
  4. UVa has changed the way it looks throughout history reversed
  5. The people who run UVA today have a different vision for the university than people who have run UVA in the past reversed
  6. UVA leadership has largely taken the same approach to running the university throughout the university’s history

Note: items 5 and 6 from physical continuity are considered part of the institutional continuity scale: leadership subscale

pretest_ingroupID [1 = not at all, 7 = very much]

  1. How important is UVa to your own personal identity?
  2. How similar do you feel in attitudes and opinions to other UVa students?
  3. How strongly do you identify as a UVa student?

In lab study

We tested our hypothesis using a 2x2 experimental design. We presented UVA students with an article that depicted either a positive or negative history of UVA’s first hospital, while they were either physically close to or distant from UVA’s first hospital - a building that still stands on campus.

Positive History Article (Medicine) Negative History Article (Race)

After viewing the article, participants responded to the DV and exploratory measures below.

Measures

spatialDistance [0 - 100]

  1. Please think about the physical location of where UVA hospital was first opened. Remember, UVA hospital was where ${e://Field/cond.words} was originally practiced on Grounds. Thinking about the physical location of where UVA hospital was first opened, how far away does that feel to you? Use the slider below to indicate how far away from where UVA hospital was first opened feels to you currently.

physicalContinuity [1 = totally disagree, 7 = totally agree]

  1. The physical aesthetic of UVA has remained constant over time
  2. UVA’s hallmark buildings have endured over time
  3. UVa has significantly changed its architecture throughout its history reversed
  4. UVa has changed the way it looks throughout history reversed

institutionalContinuity [1 = totally disagree, 7 = totally agree]

  1. The university’s founding values and worldview have endured over time.
  2. The University of Virginia will always be characterized by specific traditions and beliefs
  3. UVa has changed its traditions and customs throughout history reversed
  4. There is a causal link between different events in UVa history
  5. Major phases in UVa history are linked to one another
  6. There is no connection between past, present, and future events at the university reversed
  7. The value UVa places on diverse people and ideas has been stable over time
  8. The rules about who can enroll as a student are different now than when the university was founded reversed
  9. Institutional messaging about who is welcome at UVa has been continuous throughout its history
  10. The people who run UVA today have a different vision for the university than people who have run UVA in the past reversed
  11. UVA leadership has largely taken the same approach to running the university throughout the university’s history

temporalDistance [average of scaled scores]

  1. The past may feel quite close or far away, regardless of the amount of time that has actually passed. Please think about when UVA hospital first opened. Remember, UVA hospital was where ${e://Field/cond.words} was originally practiced on Grounds. How far away in time does the opening of UVA hospital feel to you? Use the slider below to indicate how far away in time the opening of UVA hospital feels from the present. [0-100]
  2. On the below scale, please indicate how far away in time the opening of UVA hospital feels to you. [1 - 7]

rippleConsequences [1901 - 2025] For how many years did the original ${e://Field/cond.words} practices of UVA hospital likely have a direct or indirect impact on the lives of [_______] residents living in Charlottesville?

  1. lower-income
  2. women
  3. black

ingroupID [1 = not at all, 7 = very much so]

  1. How important is UVa to your own personal identity?
  2. How similar do you feel in attitudes and opinions to other UVa students?
  3. How strongly do you identify as a UVa student?

ingroupUVAregard [1 = strongly disagree, 6 = strongly agree]

  1. I am proud to be a UVA student
  2. I am happy to be a UVA student
  3. I feel good about UVA
  4. I am loyal to my university
  5. I admire people who show a lot of UVA spirit
  6. It bothers me when someone criticizes our university

heardBefore [1 = yes, 2 = no, 3 = I’m not really sure]

Before reading this article, had you heard of this history before?

familiarity [1 = not at all familiar, 6 = very familiar]

Before reading this article, how familiar were you with this history?

actualDistance_temporal [0 years ago - 200 years ago]

Please think about the year UVA hospital first opened. Use the slider below to estimate how many years ago UVA hospital first opened.

actualDistance_physical [0 feet away - 10,000 feet away (2 miles/3 km)]

Please think about the physical location of where UVA hospital was first opened. Use the slider below to estimate how many feet away you currently are from where the UVA hospital first opened.

Demographics

  1. year at UVA
  2. familiarity with West complex
  3. gender
  4. ethnicity
show code
study5_raw = read.csv('study 5/study5.csv') %>% #notes: west complex = 11, gilmer = 99
  mutate(condition_location = ifelse(location ==99, "GILMER", ifelse(location ==11, "WEST", NA)),
         articleTime = coalesce(articleTimingQ19_Page.Submit, Q40_Page.Submit),
         across(c("physicalContinuity_3","physicalContinuity_4", #reverse score relevant items
                  "institutionContinuit_3", "institutionContinuit_6", "institutionContinuit_8", "institutionContinuit_10"), ~ 8 - .)) %>%  
  mutate(
         physicalContinuity = rowMeans(select(.,matches("physicalContinuity")), na.rm=TRUE),
         institutionalContinuity = rowMeans(select(.,matches("institutionContinuit")), na.rm=TRUE),
         institutionalContinuity_cultural = rowMeans(select(.,c('institutionContinuit_1', 'institutionContinuit_2', 'institutionContinuit_3')), na.rm=TRUE),
         institutionalContinuity_temporal = rowMeans(select(.,c('institutionContinuit_4', 'institutionContinuit_5', 'institutionContinuit_6')), na.rm=TRUE),
         institutionalContinuity_diversity = rowMeans(select(.,c('institutionContinuit_7', 'institutionContinuit_8', 'institutionContinuit_9')), na.rm=TRUE),
         institutionalContinuity_leadership = rowMeans(select(.,c('institutionContinuit_10', 'institutionContinuit_11')), na.rm=TRUE),
         temporalDistance = (as.vector(scale(temporalDistance_1)) + as.vector(scale(temporalDistance_2))/2),
         rippleConsequences = rowMeans(select(.,matches('rippleConsequences')), na.rm=TRUE),
         ingroupIdentification = rowMeans(select(.,matches("ingroupUVA")), na.rm=TRUE))

study5 = merge(study5_raw %>% select(c('computID','condition_location','condition_article','articleTime','spatialDistance','rippleConsequences_lowIncome':'rippleConsequences_Black','heardBefore','familiarWest','physicalContinuity':ncol(study5_raw))),
                study5_pretest %>% select(c('computID','pretest_institutionalContinuity':ncol(study5_pretest))), by='computID')

#subset(study5_pretest, computID %in% c("azf8ev", "cza5rb", "ehz9zt", "hgw3bv", "hnj6wj", "jrp5wf", "kbv8hg", "kvu4hz", "kvu4hz", "paz2ru", "syx6mz", "syx6mz", "ttp4kk", "vma3ux", "xdv6gg", "xun3pu", "xzk5eu", "yry3ty", "yus9ba", "zqu7hk"))

We collected a total of 267 complete responses NOTE: There are duplicate computing IDs on the pretest that are resulting in duplicate rows in the data. Will address later

show code
nice_table(study5 %>% group_by(condition_article, condition_location)  %>%  count(), title = 'N per condition')
nice_table(study5 %>% group_by(familiarWest)  %>%  count(), title = "Participants' familiarity with the West complex")
nice_table(study5 %>% group_by(heardBefore)  %>%  count(), title = "Participants' familiarity with the history of the West Complex")

N per condition

condition_article

condition_location

n

MEDICINE

GILMER

52

MEDICINE

WEST

79

RACE

GILMER

64

RACE

WEST

72

Participants' familiarity with the West complex

familiarWest

n

1

203

2

42

3

13

4

7

2

Participants' familiarity with the history of the West Complex

heardBefore

n

1

80

2

163

3

22

2

Preliminary analysis

show code
ggcorrplot(round(cor(study5 %>% select(-condition_article, -condition_location, -familiarWest, -heardBefore,-computID, -articleTime), use='complete.obs'), 3),
           hc.order = TRUE, type = "upper", outline.color = "white", ggtheme = ggplot2::theme_gray, colors = c("#E46726", "white","#6D9EC1" ), lab=FALSE, digits=2)

Cronbach’s alpha of the institutional continuity measure: 0.7232929

Main analyses

Institutional continuity

show code
#institutional continuity pre v post 

study5 %>% select(condition_article, condition_location, matches('institutionalContinuity')) %>% select(1,2, 3,8) %>% rename(pre = 'pretest_institutionalContinuity', post = 'institutionalContinuity') %>% mutate(PID = 1:nrow(.)) %>% 
pivot_longer(., 3:4, names_to = "time", values_to="institutionalContinuity") %>% mutate(time = factor(time, levels=c('pre', 'post'))) %>% 
  ggerrorplot(x = 'time', y = 'institutionalContinuity', color='condition_article',  alpha=0.25, position = position_dodge(0)) +
  geom_line(alpha=0.15, aes(group = PID, color=condition_article)) +
  facet_wrap(~condition_location)
nice_table(tidy(lm(data=study5, institutionalContinuity~condition_article*condition_location+pretest_institutionalContinuity)), title='institutionalContinuity ~ location * article + pretest')

institutionalContinuity ~ location * article + pretest

Term

estimate

std.error

statistic

p

(Intercept)

2.28

0.24

9.54

< .001***

condition_articleRACE

-0.23

0.10

-2.30

.022*

condition_locationWEST

0.12

0.09

1.28

.201

pretest_institutionalContinuity

0.40

0.06

6.66

< .001***

condition_articleRACE × condition_locationWEST

-0.35

0.13

-2.61

.010**

Main effects: - Reading a negative history article reduces perceived historical continuity

Interaction: - Reading a negative history article while close in proximity to the site of the history reduces perceived historical continuity much more so than reading a positive history article while in the same location.

Interpretation: People are more uncomfortable when they are close in proximity to the site of their ingroup’s dark history, which motivates perceived distance between the self and the history by attenuating perceived institutional continuity.

Temporal distance

show code
#temporal distance pre v post 

study5 %>% select(condition_article, condition_location, matches('temporalDistance'))  %>% rename(pre = 'pretest_temporalDistance', post = 'temporalDistance') %>% mutate(PID = 1:nrow(.)) %>% 
pivot_longer(., 3:4, names_to = "time", values_to="temoralDistance") %>% mutate(time = factor(time, levels=c('pre', 'post'))) %>% 
  ggerrorplot(x = 'time', y = 'temoralDistance', color='condition_article',  alpha=0.25, position = position_dodge(0)) +
  geom_line(alpha=0.15, aes(group = PID, color=condition_article)) +
  facet_wrap(~condition_location)
nice_table(tidy(lm(data=study5, temporalDistance~condition_article*condition_location+pretest_temporalDistance)), title='temporalDistance ~ location * article + pretest')

temporalDistance ~ location * article + pretest

Term

estimate

std.error

statistic

p

(Intercept)

-0.30

0.19

-1.60

.111

condition_articleRACE

0.58

0.26

2.26

.025*

condition_locationWEST

0.27

0.25

1.07

.284

pretest_temporalDistance

0.36

0.06

5.83

< .001***

condition_articleRACE × condition_locationWEST

-0.51

0.35

-1.46

.146

Physical continuity

show code
#institutional continuity pre v post 

study5 %>% select(condition_article, condition_location, matches('physicalContinuity')) %>% rename(pre = 'pretest_physicalContinuity', post = 'physicalContinuity') %>% mutate(PID = 1:nrow(.)) %>% 
pivot_longer(., 3:4, names_to = "time", values_to="physicalContinuity") %>% mutate(time = factor(time, levels=c('pre', 'post'))) %>% 
  ggerrorplot(x = 'time', y = 'physicalContinuity', color='condition_article',  alpha=0.25, position = position_dodge(0)) +
  geom_line(alpha=0.15, aes(group = PID, color=condition_article)) +
  facet_wrap(~condition_location)
nice_table(tidy(lm(data=study5, physicalContinuity~condition_article*condition_location+pretest_physicalContinuity)), title='physicalContinuity ~ location * article + pretest')

physicalContinuity ~ location * article + pretest

Term

estimate

std.error

statistic

p

(Intercept)

3.70

0.29

12.64

< .001***

condition_articleRACE

0.18

0.14

1.30

.196

condition_locationWEST

0.10

0.13

0.78

.436

pretest_physicalContinuity

0.34

0.06

5.33

< .001***

condition_articleRACE × condition_locationWEST

-0.35

0.18

-1.92

.056

Exploratory Measures

Temporal Consequences

show code
ggerrorplot(data = study5, x = 'condition_location', y = 'rippleConsequences', color='condition_article')