Data from: Differential involvement of the senses in disgust memories
Data files
Feb 27, 2024 version files 228.16 KB
Abstract
One prediction derived from the disease avoidance account of disgust is that proximal disgust cues (smells, tastes, touches), should elicit this emotion more intensely than distal disgust cues (sights, sounds). If correct, then memories of disgusting experiences should involve smelling, tasting or touching to a greater degree, than seeing or hearing. Two surveys were conducted on university students to test this idea, drawing upon their naturalistic experiences. Survey one (N = 127) asked participants to detail their most memorable disgusting, fear-provoking, morally-repulsive, and yucky/gross experience, with each recollection self-rated for sensory involvement. Survey two (N = 89) employed the same task, but this time participants recollected their most common disgusting, fear-provoking, morally-repulsive, and yucky/gross experience in the preceding week. For core disgusts, the proximal and distal sensory cues contributed equally for most memorable disgust experiences, but proximal exceeded distal for most common. When the same comparison was made for core disgusts, relative to moral disgusts and fear-provoking experiences, the proximal sensory cues contributed more to core disgusts. The implications of these findings for a disease avoidance account of disgust, for multisensory disgust research, and core disgust’s classification as an emotion or a drive, are discussed.
README: Differential involvement of the senses in disgust memories
https://doi.org/10.5061/dryad.70rxwdc4m
This survey examines which senses are most involved in participants' most memorable (across their lives) and most-common (in their past week), disgusting, yucky/gross, fearful and morally repulsive experiences.
We had two main aims:
(1) Core disgust experiences (disgust, and yucky/gross) will have more involvement of proximal senses (smell, taste, passive touch, active touch) compared to the distal senses (sound, sight). To achieve this aim we compare whether any proximal sense was involved more than any distal sense (i.e., maximum distal vs. maximum proximal).
(2) Core disgust experiences (disgust, and yucky/gross) will have more involvement of proximal vs distal sensory involvement, relative to other forms of disgust and emotions (morally repulsive and fear*). To test this we compare the difference in proximal and distal sensory involvement (proximal sensory minus distal sensory), across the different affective states (disgust, yucky/gross, fear, morally repulsive)*
Description of the data and file structure
All datafiles attached are excel documents. They contain 2 sheets, the first provides a variable key, and the second the dataset for analyses. All data are deindentified.
Analyses pertaining to Aim 1 and Aim 2 (outlined above) are detailed below
AIM 1: Are the proximal senses more involved in core disgust experiences, relative to the distal senses?
- Aim 1 Analysis Survey 1 (most memorable) > To assess aim 1 in survey 1, the data set titled "Dryad_Survey_One" needs to be opened > Wilcoxon tests then need to be run (i.e., paired sample t-tests, non parametric), to compare the involvement of the maximum distal sense to the maximum proximal > The variables to be compared are 'DisEvMaxDistal' vs. 'DisEvMaxProximal'
- Aim 1 Analysis Survey 2 (most common experiences) > To assess aim 1 in survey 2, the data set titled "Dryad Survey2_Aim1Aim2" needs to be opened > Wilcoxon tests then need to be run (i.e., paired sample t-tests, non parametric), to compare the involvement of the maximum distal sense to the maximum proximal > The variables to be compared are 'DisgustMaxDistal' vs. 'DisgustMaxProx'
AIM 2: Is the difference between Proximal and Distal sensory involvement, greater for core disgust experiences, than moral disgust and fear experiences?
- Aim 2 Analysis Survey 1 (most memorable) > To assess aim 2 in survey 1 data, the data set titled "Dryad_Survey_One" needs to be opened > A Friedman test (non parametric one way repeated measures ANOVA ) needs to be run on the proximal minus distal sensory scores. The variables to be compared in a Friedman's test are 'DisEvMaxDiffs' ; 'ScaEvMaxDiffs'; 'YukEvMaxDiffs'; 'MorEvMaxDiffs' > Wilcoxon tests then need to be run (i.e., paired sample t-tests, non parametric), to compare the difference scores between each emotion. The variables to be compared in a Wilcoxon's test are the same as above ('DisEvMaxDiffs' ; 'ScaEvMaxDiffs'; 'YukEvMaxDiffs'; 'MorEvMaxDiffs')
- Aim 2 Analysis Survey 2 (most common experiences)
- > To assess aim 2 in survey 2 data, the data set titled "Dryad Survey2_Aim1Aim2" needs to be opened > A Friedman test (non parametric one way repeated measures ANOVA ) needs to be run on the proximal minus distal sensory scores. The variables to be compared in a Friedman's test are 'DisgustProxMinusDistal' ; 'YuckyProximalMinusDistal' ; 'ScaryProximalMinusDistal' ; 'MRproximalMinusDistal' > Wilcoxon tests then need to be run (i.e., paired sample t-tests, non parametric), to compare the difference scores between each emotion. The variables to be compared in a Wilcoxon's test are the same as above ('DisgustProxMinusDistal' ; 'YuckyProximalMinusDistal' ; 'ScaryProximalMinusDistal' ; 'MRproximalMinusDistal')
Extra Analyses
General Descriptives
- Additional variables ( Age, Age during experiences (survey 1), intensity of experiences and frequency (survey 2) are provided in the attached datasetS (Dryad_Survey_One, Dryad Survey2_Aim1Aim2) , so the reader can get descriptive means and SDs.
Human coder data Descriptives
- Survey 1 Reliability data (agreement) on the format of the most memorable experiences can be found in the "FormatExperienceAgreementSurvey1R" and "CoreMoralDisgust_HumanCoderReliabilityS1" - descriptives (frequency) data was generated from these, with variable keys provided in the files.
- Survey 2 Reliability data (agreement) on the format of the most memorable experiences can be found in the "FormatExperienceAgreementSurvey2R" and "CoreMoralDisgust_HumanCoderReliabilityS2" - descriptives (frequency) data was generated from these, with variable keys provided in the files.
Sample Size Justification
In order to calculate sample size needed to get an 80% chance of seeing an effect, the formula (N = [2.8/cohens d]2) was used (Howell, 2007). Based off effect sizes seen in past disgust-survey studies (with cohens d’s ranging between .27 to .40; see Rozin, 2008; Boggs et al., 2020) a sample of around 107 students was required. As such, to cater for participant drop out, we recruit 127 participants in survey 1, with 90 of these participants proceeding to survey 2.
References
Howell, D. C. (2007). Statistical methods for psychology (6th ed.). Belmont, California: Duxbury Press.
Rozin, P. (2008). Hedonic “adaptation”: Specific habituation to disgust/death elicitors as a result of dissecting a cadaver. Judgment and Decision Making, 3(2), 191-194. doi:10.1017/S1930297500001534
Boggs, S. T., Ruisch, B. C., & Fazio, R. H. (2022). Concern about salient pathogen threats increases sensitivity to disgust. Personality and Individual Differences, 186, 111348. https://doi.org/10.1016/j.paid.2021.111348
Sharing/Access information
Data was derived from a qualtric's survey, that recieved ethical vetting. Only deidentifiable data are included.
Code/Software
Graphs
The dataset used to make figure 1 is titled "EmotionDifferenceProximalDataRGraph" and the R code used to generate it , is provided below:
read.table("EmotionDifferenceProximalDataRGraph.csv", sep = ",", header = TRUE, stringsAsFactors = FALSE)
data4 <- read.csv("EmotionDifferenceProximalDataRGraph.csv")
y3 <- ggbarplot(data4, x = "Emotion", y = "Mean", fill = "Survey", position = position_dodge(0.8)) +
labs(fill = "Disgust Experience:", x="Emotion", y = "Proximal minus Distal Involvement (-100 to 100)") +
theme(axis.title.x = element_text(size = 18, vjust = -2, face="bold"),
axis.text = element_text(size = 18, face="bold"),
axis.title.y = element_text(size = 19, vjust = 4, face="bold"),
legend.title = element_text(size = 18, face="bold"), plot.margin = margin(.7, .7, .7, .7, "cm"),
legend.text=element_text(size=18, face="bold"))+
ylim(-70,50)
z3 <- y3 + geom_errorbar(aes(group = Survey, ymax = Mean-SE, ymin = Mean+SE),
position = position_dodge(width = .8), width = 0.25)
z3
z3 <- z3 + scale_fill_manual(values=c("#999999","#FFFFFF"), labels = c("Most Memorable", "Most Common"))
z3
Methods
Data was collected on Qualtrics, and analysed via SPSS and R (for graphical purposes only). Only deidentifiable data is uploaded.