OSCR

Facial palsy reveals the sensorimotor contribution to facial-emotion recognition.

Code ↔ Paper

2 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 2 matches
  1. [1] § Results › Severity–Performance Coupling in Congenital Palsy (H2a). ↔ Study 2/Script/Study2_Confirmatory_Analyses_and_Plot.R, lines 789–848 · score 0.54 · Model projections, cross validation, facial motor, correlated, matched, predicted
  2. [2] § Materials and Methods ↔ Study 2/Script/Study2_Confirmatory_Analyses_and_Plot.R, lines 683–746 · score 0.50 · SFGS scores, ADFES accuracy, JeFEE, correlations, CI, bootstrap

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

R · 915 lines · 32 KB · no license · 2 matches

  1. ## HEADER ----
  2. # Experiment: CARIPARO
  3. # Programmer: QUETTIER THOMAS
  4. # Created: 04/06/2023
  5. # Updated: 24/07/2025
  6. # Description:
  7. # This script performs data analysis for the CARIPARO experiment,
  8. # specifically for the Accuracy and Intensity measures of the
  9. # Online student experiment.
  11. ## Clean up ----
  12. rm(list = ls()) # Remove all objects
  14. ## Load libraries ----
  15. library(tidyverse)
  16. library(afex)
  17. library(effectsize)
  18. library(BayesFactor)
  19. library(emmeans)
  20. library(dplyr)
  21. library(performance)
  22. library(brms)
  23. library(bayestestR)
  24. library(tidyverse)
  25. library(cowplot)
  26. library(ggplot2)
  27. library(ggrain)
  28. library(mediation)
  29. library(broom)
  30. library(boot)
  32. ## Define preprocessing function ----
  33. preprocessed_data <- function(file_path, value_col, ...) {
  34. read_csv(file_path) %>%
  35. group_by(...) %>%
  36. summarise(across(all_of(value_col), ~ mean(.x, na.rm = TRUE)), .groups = "drop") %>%
  37. mutate(label = paste0(subject,video_set),
  38. video_set = factor(video_set, levels = c("ADFES", "JeFEE")))
  39. }
  42. figure<-function(data, type){
  44. colors <- c("ADFES" = "#657d9a", "JeFEE" = "#b37269")
  45. colorgroup <- c(Moebius = "#F0E442", Control = "#C07DA5")
  47. # stat summary
  48. if(type == "acc"){
  49. df <- data %>% dplyr::select(-label) %>%
  50. 'colnames<-'(c("subject" , "group" ,"video_set", "mean"))%>%
  51. mutate(mean = mean*100,
  52. match = paste0(parse_number(subject),video_set))
  53. ytitle <- "Mean Accuracy (%)"
  55. } else if(type == "int"){
  56. df <- data %>% dplyr::select(-label) %>%
  57. 'colnames<-'(c("subject", "group" , "video_set", "mean"))%>%
  58. mutate(match = paste0(parse_number(subject),video_set))
  59. ytitle <- "Mean Intensity (px)"
  60. }
  62. descritive <- df %>%
  63. group_by(video_set, group) %>%
  64. summarise(sd = sd(mean),
  65. se = sd(mean)/sqrt(n()),
  66. mean = mean(mean),
  67. .groups = "drop")%>%
  68. mutate(x_pos = case_when(
  69. group == "Moebius" & video_set == "ADFES" ~ 2 - 0.1,
  70. group == "Moebius" & video_set == "JeFEE" ~ 2 - 0.1,
  71. group == "Control" & video_set == "ADFES" ~ 1 + 0.1,
  72. group == "Control" & video_set == "JeFEE" ~ 1 + 0.1
  73. ))
  75. p <- ggplot(df %>% filter(group %in% c("Moebius", "Control")),
  76. aes(x = group, y = mean, fill = video_set)) +
  77. geom_rain(alpha = 0.5, rain.side = 'f2x2', id.long.var = "match", cov = "group") +
  78. geom_point(data = descritive,
  79. aes(x = x_pos, y = mean, fill = video_set),
  80. size = 3, shape = 23, color = "black") +
  81. geom_errorbar(data = descritive,
  82. aes(x = x_pos, ymin = mean - se, ymax = mean + se),
  83. width = 0, color = "black", linewidth = 0.5, inherit.aes = FALSE) +
  84. scale_fill_manual(values = colors) +
  85. scale_color_manual(values = colorgroup) +
  86. labs(x = NULL, y = ytitle, fill = "Expression Type") +
  87. theme_minimal(base_size = 14) +
  88. theme(legend.position = "none") +
  89. theme(text=element_text(size=10, family="Helvetica"),
  90. panel.background = element_blank(),
  91. axis.line = element_line(colour = "black"),
  92. strip.text.x = element_text(size = 6.6),
  93. strip.text.y = element_text(size = 6.6),
  94. panel.grid.major = element_blank(), # Rimuove le griglie principali
  95. panel.grid.minor = element_blank() )
  97. return(p)
  98. }
  100. # Function to extract results from mediation analysis
  101. extract_med <- function(model, mediator_name, outcome_name) {
  102. summ <- summary(model)
  103. tibble(
  104. mediator = mediator_name,
  105. outcome = outcome_name,
  106. acme = summ$d0,
  107. acme_ci_low = summ$d0.ci[1],
  108. acme_ci_high = summ$d0.ci[2],
  109. acme_p = summ$d0.p,
  110. ade = summ$z0,
  111. ade_ci_low = summ$z0.ci[1],
  112. ade_ci_high = summ$z0.ci[2],
  113. ade_p = summ$z0.p,
  114. total_effect = summ$tau.coef,
  115. total_ci_low = summ$tau.ci[1],
  116. total_ci_high = summ$tau.ci[2],
  117. total_p = summ$tau.p
  118. )
  119. }
  121. # Generic function to bootstrap Pearson r
  122. cor_fun <- function(data, indices, x, y) {
  123. d <- data[indices, ]
  124. return(cor(d[[x]], d[[y]], method = "pearson"))
  125. }
  127. # Helper function to append test results
  128. add_result <- function(var, outcome, test, bf, eff = NA) {
  129. tibble::tibble(
  130. variable = var,
  131. outcome = outcome,
  132. stat_type = test$method,
  133. stat_value = unname(test$estimate),
  134. p_value = test$p.value,
  135. bf01 = bf,
  136. effect_size = eff
  137. )
  138. }
  140. # Pearson r with BCa 95 % CI (10 000 resamples)
  141. bcacor <- function(d, i, var){
  142. d <- d[i, ]
  143. cor(d$Sunnybrook, d[[var]], method = "pearson")
  144. }
  146. get_bca <- function(var){
  147. bt <- boot(dat, bcacor, R = 10000, var = var)
  148. ci <- boot.ci(bt, type = "bca")$bca[4:5]
  149. tibble(
  150. Variable = var,
  151. r = cor(dat$Sunnybrook, dat[[var]]),
  152. CI_low = ci[1],
  153. CI_high = ci[2],
  154. p = cor.test(dat$Sunnybrook, dat[[var]])$p.value
  155. )
  156. }
  158. # Leave-one-out cross-validated R² with percentile CI
  159. calc_loo <- function(y){
  160. n <- nrow(dat)
  161. yh <- numeric(n)
  162. for(i in seq_len(n)){
  163. fit <- lm(reformulate("Sunnybrook", y), data = dat[-i, ])
  164. yh[i] <- predict(fit, newdata = dat[i, ])
  165. }
  166. r2 <- 1 - sum((dat[[y]] - yh)^2) / sum((dat[[y]] - mean(dat[[y]]))^2)
  168. bt <- boot(dat, function(d, ind){
  169. d <- d[ind, ]
  170. n <- nrow(d); yh <- numeric(n)
  171. for(i in seq_len(n)){
  172. fit <- lm(reformulate("Sunnybrook", y), data = d[-i, ])
  173. yh[i] <- predict(fit, newdata = d[i, ])
  174. }
  175. 1 - sum((d[[y]] - yh)^2) / sum((d[[y]] - mean(d[[y]]))^2)
  176. }, R = 10000)
  177. ci <- boot.ci(bt, type = "perc")$percent[4:5]
  179. tibble(Variable = y, R2_CV = r2, CI_low = ci[1], CI_high = ci[2])
  180. }
  182. get_pred <- function(y){
  183. fit <- lm(reformulate("Sunnybrook", y), data = dat)
  184. predict(fit, newdata = grid, interval = "prediction", level = 0.95) %>%
  185. as_tibble() %>%
  186. mutate(Sunnybrook = grid$Sunnybrook, Variable = y)
  187. }
  189. ## Accuracy Analysis ----
  191. ### 0. Data preparation ----
  192. df_acc <- preprocessed_data("data/Study2_Mean_forEmotion_Accuracy.csv", "accuracy", subject, group, video_set, emotion)
  193. df_int <- preprocessed_data("data/Study2_Mean_forEmotion_intensity", "intensity", subject, group, video_set)
  196. df_acc$total <- 8 # Number of trials per observation
  198. ### 1. Descriptive statistics for accuracy ----
  199. df_subject_mean <- df_acc %>%
  200. group_by(subject, group, video_set) %>%
  201. summarise(accuracy = mean(accuracy, na.rm = TRUE), .groups = "drop")
  203. descriptive_accuracy <- df_subject_mean %>%
  204. group_by(video_set, group) %>%
  205. summarise(
  206. mean_acc = mean(accuracy) * 100,
  207. sd_acc = sd(accuracy) * 100,
  208. se_acc = sd(accuracy) / sqrt(n()) * 100,
  209. min_acc = min(accuracy) * 100,
  210. max_acc = max(accuracy) * 100,
  211. .groups = "drop"
  212. )
  214. print(descriptive_accuracy)
  216. # Compute % difference between ADFES and JeFEE
  217. diff_percent <- descriptive_accuracy %>%
  218. dplyr::select(video_set, group, mean_acc) %>%
  219. pivot_wider(names_from = video_set, values_from = mean_acc) %>%
  220. mutate(percent_diff = (ADFES - JeFEE) / ADFES * 100)
  222. ### 2. Fit GLMM on accuracy (binomial with weights) ----
  223. mod_glmer <- glmer(
  224. accuracy ~ video_set * group + (1 | subject),
  225. family = binomial(),
  226. weights = total,
  227. data = df_acc
  228. )
  230. summary(mod_glmer)
  232. ### 3. Estimated marginal means (response scale) ----
  233. emm_acc <- emmeans(mod_glmer, pairwise ~ video_set, type = "response")
  235. ### 4. Effect size from odds ratio ----
  236. odds_ratio <- summary(emm_acc$contrasts) %>% pull(odds.ratio)
  237. cohens_d <- oddsratio_to_d(odds_ratio)
  238. print(cohens_d)
  240. ### 5. R² from frequentist model ----
  241. r2(mod_glmer)
  243. ### 6. Welch's correction (variance == False -> apply correction) ----
  244. # data preparation
  245. ### 0. Data preparation
  246. df_welch <- preprocessed_data("Study2_Mean_forEmotion_Accuracy.csv", "accuracy", subject, video_set, group)
  248. # Subset data first
  249. adfes_data <- df_welch %>% filter(video_set == "ADFES")
  250. jefee_data <- df_welch %>% filter(video_set == "JeFEE")
  252. # Run Welch t-tests
  253. ADFES_ph <- t.test(accuracy ~ group, data = adfes_data, var.equal = FALSE)
  254. JeFEE_ph <- t.test(accuracy ~ group, data = jefee_data, var.equal = FALSE)
  256. # Compute Cohen's d from raw data
  257. ADFES_d <- cohens_d(accuracy ~ group, data = adfes_data)
  258. JeFEE_d <- cohens_d(accuracy ~ group, data = jefee_data)
  262. ### Bayesian Analysis (Accuracy) ----
  264. #### 1. Add success counts ----
  265. df_acc <- df_acc %>%
  266. mutate(successes = round(accuracy * total))
  268. #### 2. Fit full Bayesian model ----
  269. mod_full <- brm(
  270. successes | trials(total) ~ video_set * group + (1 | subject),
  271. family = binomial(),
  272. data = df_acc,
  273. prior = set_prior("normal(0, 5)", class = "b"),
  274. chains = 4, cores = 4, iter = 4000, seed = 123,
  275. save_pars = save_pars(all = TRUE)
  276. )
  278. #### 3. Fit null Bayesian model ----
  279. mod_null <- brm(
  280. successes | trials(total) ~ 1 + (1 | subject),
  281. family = binomial(),
  282. data = df_acc,
  283. prior = set_prior("normal(0, 5)", class = "Intercept"),
  284. chains = 4, cores = 4, iter = 4000, seed = 123,
  285. save_pars = save_pars(all = TRUE)
  286. )
  288. #### 4. Compute Bayes Factor ----
  289. BF <- bayesfactor_models(mod_full, mod_null)
  290. BF10 <- exp(abs(BF$log_BF[BF$Model == "1 + (1 | subject)"]))
  293. ## Intensity Analysis ----
  295. ### 1. Descriptive statistics for intensity ----
  296. descriptive_intensity <- df_int %>%
  297. group_by(video_set, group) %>%
  298. summarise(
  299. mean_int = mean(intensity, na.rm = TRUE),
  300. sd_int = sd(intensity, na.rm = TRUE),
  301. se_int = sd(intensity, na.rm = TRUE) / sqrt(n()),
  302. .groups = "drop"
  303. )
  305. print(descriptive_intensity)
  307. ### 2. Frequentist ANOVA ----
  308. anova_int <- aov_ez(id = "subject", dv = "intensity", within = "video_set", between = "group",data = df_int)
  309. summary(anova_int)
  311. ### 3. Effect size: η² ----
  312. eta2 <- eta_squared(anova_int, partial = TRUE)
  313. print(eta2)
  315. ### 4. Cohen’s d for repeated measures ----
  316. df_wide <- df_int %>%
  317. select(-label) %>%
  318. pivot_wider(names_from = video_set, values_from = intensity) %>%
  319. drop_na() %>%
  320. column_to_rownames("subject")
  322. d_rm <- cohens_d(df_wide$ADFES, df_wide$JeFEE, paired = TRUE, within = TRUE)
  323. print(d_rm)
  327. ### 5. Bayesian ANOVA ----
  328. df_int$subject <- as.factor(df_int$subject)
  329. df_int$video_set <- as.factor(df_int$video_set)
  330. df_int$group <- as.factor(df_int$group)
  332. bf_intensity <- anovaBF(
  333. intensity ~ video_set * group * subject,
  334. data = df_int ,
  335. whichRandom = "subject",
  336. iterations = 10000
  337. )
  339. BF <-extractBF(bf_intensity)
  343. ### 6. Welch's correction (variance == False -> apply correction)----
  344. # Subset data first
  345. adfes_data <- df_int %>% filter(video_set == "ADFES")
  346. jefee_data <- df_int %>% filter(video_set == "JeFEE")
  348. # Run Welch t-tests
  349. ADFES_ph <- t.test(intensity ~ group, data = adfes_data, var.equal = FALSE)
  350. JeFEE_ph <- t.test(intensity ~ group, data = jefee_data, var.equal = FALSE)
  352. # Compute Cohen's d from raw data
  353. ADFES_d <- cohens_d(intensity ~ group, data = adfes_data)
  354. JeFEE_d <- cohens_d(intensity ~ group, data = jefee_data)
  357. ### 7. Plot Figure 2S ----
  358. # Plots fig. 2S supp
  359. df_acc <- preprocessed_data("Study2_Mean_forEmotion_Accuracy.csv", "accuracy", subject, group, video_set)
  360. df_int <- preprocessed_data("Study2_Mean_forEmotion_intensity.csv", "intensity", subject, group, video_set)
  362. p1 <- figure(df_acc,"acc") # plot accuracy
  363. p2 <- figure(df_int,"int") # plot intensity
  364. p3 <- plot_grid(p1, p2, nrow = 1,labels = "AUTO") #combine plots
  366. # Export
  367. ggsave("figures/Fig_2S_finale.png", # export for draft
  368. plot = p3,
  369. device = "png",
  370. dpi = 300,
  371. width = 18,
  372. height = 12,
  373. units = "cm")
  375. ggsave("figures/Fig_2S.tiff", # export for journal
  376. plot = p3,
  377. device = "tiff",
  378. dpi = 1200,
  379. width = 18,
  380. height = 12,
  381. units = "cm")
  383. ## Normative sample comparison ----
  385. ### 1. Normative values ----
  386. df_acc_n <- read_csv("Study2_Mean_forEmotion_Accuracy.csv")%>%
  387. group_by(subject,video_set) %>%
  388. summarise( accuracy = mean(acc, na.rm = TRUE)) %>%
  389. mutate(video_set = as.factor(video_set),
  390. label = paste0(subject,video_set))
  392. df_int_n <- read_csv("Study2_Mean_forEmotion_intensity.csv")%>%
  393. group_by(subject,video_set) %>%
  394. summarise(intensity = mean(intensity, na.rm = TRUE)) %>%
  395. mutate(video_set = as.factor(video_set),
  396. label = paste0(subject,video_set)) %>%
  397. ungroup() %>%
  398. dplyr::select(-c(subject,video_set))
  400. df_norm <- left_join(df_acc_n, df_int_n, by = "label") %>%
  401. dplyr::select(-label) %>%
  402. group_by(video_set) %>%
  403. summarise(norm_acc = mean(accuracy),
  404. norm_int = mean(intensity),
  405. sd_acc = sd(accuracy),
  406. sd_int = sd(intensity))
  407. norm_ADFES <- df_norm$norm_acc[df_norm$video_set == "ADFES"]
  408. norm_JeFEE <- df_norm$norm_acc[df_norm$video_set == "JeFEE"]
  410. norm_ADFES_int <- df_norm$norm_int[df_norm$video_set == "ADFES"]
  411. norm_JeFEE_int <- df_norm$norm_int[df_norm$video_set == "JeFEE"]
  414. ### 2. Subset your data ----
  415. df_acc <- preprocessed_data("Study2_Mean_forEmotion_Accuracy.csv", "accuracy", subject, group, video_set)
  416. df_int <- preprocessed_data("Study2_Mean_forEmotion_intensity.csv", "intensity", subject, group, video_set)%>%
  417. select(intensity, label)
  419. df <- left_join(df_acc, df_int, by = "label") %>%
  420. select(-label)
  422. controls <- df %>% filter(group == "Control")
  423. moebius <- df %>% filter(group == "Moebius")
  425. ### 3. Compute per-subject means for each video_set ----
  426. control_means <- controls %>%
  427. group_by(subject, video_set) %>%
  428. summarise(acc = mean(accuracy),
  429. int = mean(intensity),
  430. .groups = "drop")
  432. moebius_means <- moebius %>%
  433. group_by(subject, video_set) %>%
  434. summarise(acc = mean(accuracy),
  435. int = mean(intensity),
  436. .groups = "drop")
  438. # --- Split by ADFES / JeFEE ---
  439. control_adfes <- control_means %>% filter(video_set == "ADFES") %>% pull(acc)
  440. control_jefee <- control_means %>% filter(video_set == "JeFEE") %>% pull(acc)
  442. moebius_adfes <- moebius_means %>% filter(video_set == "ADFES") %>% pull(acc)
  443. moebius_jefee <- moebius_means %>% filter(video_set == "JeFEE") %>% pull(acc)
  445. control_adfes_int <- control_means %>% filter(video_set == "ADFES") %>% pull(int)
  446. control_jefee_int <- control_means %>% filter(video_set == "JeFEE") %>% pull(int)
  448. moebius_adfes_int <- moebius_means %>% filter(video_set == "ADFES") %>% pull(int)
  449. moebius_jefee_int <- moebius_means %>% filter(video_set == "JeFEE") %>% pull(int)
  452. ### 4. Welch one-sample t-tests vs normative means X---
  453. # ADFES ACCURACY
  454. t_ctrl_adfes <- t.test(control_adfes, mu = norm_ADFES)
  455. d_ctrl_adfes <- cohens_d(control_adfes, mu = norm_ADFES)
  456. bf_ctrl_adfes <- ttestBF(x = control_adfes, mu = norm_ADFES)
  458. t_moeb_adfes <- t.test(moebius_adfes, mu = norm_ADFES)
  459. d_moeb_adfes <- cohens_d(moebius_adfes, mu = norm_ADFES)
  460. bf_moeb_adfes <- ttestBF(x = moebius_adfes, mu = norm_ADFES)
  462. # JeFEE ACCURACY
  463. t_ctrl_jefee <- t.test(control_jefee, mu = norm_JeFEE)
  464. d_ctrl_jefee <- cohens_d(control_jefee, mu = norm_JeFEE)
  465. bf_ctrl_jefee <- ttestBF(x = control_jefee, mu = norm_JeFEE)
  467. t_moeb_jefee <- t.test(moebius_jefee, mu = norm_JeFEE)
  468. d_moeb_jefee <- cohens_d(moebius_jefee, mu = norm_JeFEE)
  469. bf_moeb_jefee <- ttestBF(x = moebius_jefee, mu = norm_JeFEE)
  471. # ADFES INTENSITY
  472. t_ctrl_adfes_int <- t.test(control_adfes_int, mu = norm_ADFES_int)
  473. d_ctrl_adfes_int <- cohens_d(control_adfes_int, mu = norm_ADFES_int)
  474. bf_ctrl_adfes_int <- ttestBF(x = control_adfes_int, mu = norm_ADFES_int)
  476. t_moeb_adfes_int <- t.test(moebius_adfes_int, mu = norm_ADFES_int)
  477. d_moeb_adfes_int <- cohens_d(moebius_adfes_int, mu = norm_ADFES_int)
  478. bf_moeb_adfes_int <- ttestBF(x = moebius_adfes_int, mu = norm_ADFES_int)
  480. # JeFEE INTENSITY
  481. t_ctrl_jefee_int <- t.test(control_jefee_int, mu = norm_JeFEE_int)
  482. d_ctrl_jefee_int <- cohens_d(control_jefee_int, mu = norm_JeFEE_int)
  483. bf_ctrl_jefee_int <- ttestBF(x = control_jefee_int, mu = norm_JeFEE_int)
  485. t_moeb_jefee_int <- t.test(moebius_jefee_int, mu = norm_JeFEE_int)
  486. d_moeb_jefee_int <- cohens_d(moebius_jefee_int, mu = norm_JeFEE_int)
  487. bf_moeb_jefee_int <- ttestBF(x = moebius_jefee_int, mu = norm_JeFEE_int)
  489. ## Mediation Analysis ----
  490. ### 1. Data preparation----
  491. data <- read_csv("data/Study2_Complete_Dataset.csv") %>%
  492. filter(group == "moebius") %>%
  493. dplyr::select(subject, AQ_score, TAS_score, Sunnybrook, acc_ADFES, acc_JeFEE)
  495. ### 2. Total effect models ----
  496. model_adfes <- lm(acc_ADFES ~ Sunnybrook, data = data)
  497. model_jefee <- lm(acc_JeFEE ~ Sunnybrook, data = data)
  499. ### 3. Mediator Models -----
  500. model_tas <- lm(TAS_score ~ Sunnybrook, data = data)
  501. model_aq <- lm(AQ_score ~ Sunnybrook, data = data)
  504. ### 4. Mediation Models -----
  505. #### AQ mediation for ADFES ----
  506. med_model_aq_adfes <- lm(AQ_score ~ Sunnybrook, data = data)
  507. out_model_aq_adfes <- lm(acc_ADFES ~ Sunnybrook + AQ_score, data = data)
  509. mediation_aq_adfes <- mediate(med_model_aq_adfes, out_model_aq_adfes,
  510. treat = "Sunnybrook", mediator = "AQ_score",
  511. boot = TRUE, sims = 1000)
  513. #### AQ mediation for JeFEE ----
  514. med_model_aq_jefee <- lm(AQ_score ~ Sunnybrook, data = data)
  515. out_model_aq_jefee <- lm(acc_JeFEE ~ Sunnybrook + AQ_score, data = data)
  517. mediation_aq_jefee <- mediate(med_model_aq_jefee, out_model_aq_jefee,
  518. treat = "Sunnybrook", mediator = "AQ_score",
  519. boot = TRUE, sims = 1000)
  521. #### TAS mediation for ADFES ----
  522. med_model_tas_adfes <- lm(TAS_score ~ Sunnybrook, data = data)
  523. out_model_tas_adfes <- lm(acc_ADFES ~ Sunnybrook + TAS_score, data = data)
  525. mediation_tas_adfes <- mediate(med_model_tas_adfes, out_model_tas_adfes,
  526. treat = "Sunnybrook", mediator = "TAS_score",
  527. boot = TRUE, sims = 1000)
  529. #### TAS mediation for JeFEE ----
  530. med_model_tas_jefee <- lm(TAS_score ~ Sunnybrook, data = data)
  531. out_model_tas_jefee <- lm(acc_JeFEE ~ Sunnybrook + TAS_score, data = data)
  533. mediation_tas_jefee <- mediate(med_model_tas_jefee, out_model_tas_jefee,
  534. treat = "Sunnybrook", mediator = "TAS_score",
  535. boot = TRUE, sims = 1000)
  537. ### 5. Extract summary stats ----
  538. results <- bind_rows(
  539. extract_med(mediation_tas_adfes, "TAS", "ADFES"),
  540. extract_med(mediation_tas_jefee, "TAS", "JeFEE"),
  541. extract_med(mediation_aq_adfes, "AQ", "ADFES"),
  542. extract_med(mediation_aq_jefee, "AQ", "JeFEE")
  543. )
  545. ### 6. FDR correction
  546. results <- as.data.frame(results) %>%
  547. dplyr::select(
  548. mediator, outcome,
  549. total_effect, total_ci_low, total_ci_high, total_p, total_q,
  550. acme, acme_ci_low, acme_ci_high, acme_p, acme_q
  551. ) %>%
  552. print()
  554. # Extract tidy regression summaries
  555. tas_sum <- tidy(model_tas, conf.int = TRUE)
  556. aq_sum <- tidy(model_aq, conf.int = TRUE)
  558. # Extract only Sunnybrook rows
  559. tas_sunny <- filter(tas_sum, term == "Sunnybrook")
  560. aq_sunny <- filter(aq_sum, term == "Sunnybrook")
  562. # Combine and compute FDR-corrected q-values
  563. mediator_pred <- bind_rows(tas_sunny, aq_sunny) %>%
  564. mutate(mediator = c("TAS", "AQ"),
  565. q = p.adjust(p.value, method = "fdr"))
  567. print(mediator_pred)
  570. # Demography ----
  571. data <- read_csv("data/Study2_Complete_Dataset.csv") %>%
  572. filter(group == "moebius")
  574. ### 1. Inizialyse results matrix ----
  575. results <- tibble::tibble(
  576. variable = character(),
  577. outcome = character(),
  578. stat_type = character(),
  579. stat_value = numeric(),
  580. p_value = numeric(),
  581. bf01 = numeric(),
  582. effect_size = numeric()
  583. )
  585. ### 2. Palsy laterality (monolateral vs bilateral) ----
  586. for (outcome in c("acc_ADFES", "acc_JeFEE")) {
  587. x <- data[[outcome]][data$lateralisation == "monolateral"]
  588. y <- data[[outcome]][data$lateralisation == "bilateral"]
  590. t_res <- t.test(x, y)
  591. d <- cohens_d(x, y)$Cohens_d
  592. bf <- 1 / extractBF(ttestBF(x = x, y = y))$bf
  594. results <- dplyr::bind_rows(results, add_result("palsy_laterality", outcome, t_res, bf, d))
  595. }
  597. ### 3. Age (continuous predictor) ----
  598. for (outcome in c("acc_ADFES", "acc_JeFEE")) {
  599. cor_res <- cor.test(data[[outcome]], data$age, method = "pearson")
  600. bf <- 1 / extractBF(correlationBF(data[[outcome]], data$age))$bf
  602. results <- dplyr::bind_rows(results, add_result("age", outcome, cor_res, bf))
  603. }
  605. ### 4. Education (years) ----
  606. for (outcome in c("acc_ADFES", "acc_JeFEE")) {
  607. cor_res <- cor.test(data[[outcome]], data$y_ed, method = "pearson")
  608. bf <- 1 / extractBF(correlationBF(data[[outcome]], data$y_ed))$bf
  610. results <- dplyr::bind_rows(results, add_result("education", outcome, cor_res, bf))
  611. }
  613. ### 5. Gender (m vs f) ----
  614. for (outcome in c("acc_ADFES", "acc_JeFEE")) {
  615. x <- data[[outcome]][data$gender == "m"]
  616. y <- data[[outcome]][data$gender == "f"]
  618. t_res <- t.test(x, y)
  619. d <- cohens_d(x, y)$Cohens_d
  620. bf <- 1 / extractBF(ttestBF(x = x, y = y))$bf
  622. results <- dplyr::bind_rows(results, add_result("gender", outcome, t_res, bf, d))
  623. }
  625. ### 6. FDR correction ----
  626. results <- results %>%
  627. dplyr::mutate(q_value = p.adjust(p_value, method = "fdr"))
  629. # Correlation and Figure 2 ----
  630. ### 1. Data preparation ----
  631. moebius_data <- read_csv("data/Study2_Complete_Dataset.csv")%>%
  632. filter(group == "moebius") %>%
  633. mutate(acc_ADFES = acc_ADFES*100,
  634. acc_JeFEE = acc_JeFEE*100)
  636. ### 2. ADFES correlation ----
  637. cor_test_adfes <- cor.test(moebius_data$Sunnybrook, moebius_data$acc_ADFES, method = "pearson")
  638. print(cor_test_adfes)
  639. bf_test_adfes <- correlationBF(x = moebius_data$Sunnybrook, y = moebius_data$acc_ADFES)
  640. posterior_samples <- posterior(bf_test_adfes, iterations = 10000)
  641. p_r_greater_0 <- mean(posterior_samples[, "rho"] > 0)
  642. sprintf("P(r > 0) = %.3f", p_r_greater_0)
  644. ### 3. JeFEE correlation ----
  645. cor_test_jefee <- cor.test(moebius_data$Sunnybrook, moebius_data$acc_JeFEE, method = "pearson")
  646. print(cor_test_jefee)
  647. bf_test_jefee <- correlationBF(x = moebius_data$Sunnybrook, y = moebius_data$acc_JeFEE)
  648. posterior_samples <- posterior(bf_test_jefee, iterations = 10000)
  649. p_r_greater_0 <- mean(posterior_samples[, "rho"] > 0)
  650. sprintf("P(r > 0) = %.3f", p_r_greater_0)
  652. ### 4. Extract raw p-values ----
  653. p_values <- c(cor_test_adfes$p.value, cor_test_jefee$p.value)
  654. # Apply BH correction to get q-values
  655. q_values <- p.adjust(p_values, method = "BH")
  657. ### 5. Bootstrap ----
  658. set.seed(123) # for reproducibility
  660. # Bootstrap ADFES
  661. boot_adfes <- boot(data = moebius_data, statistic = cor_fun, R = 10000, x = "Sunnybrook", y = "acc_ADFES")
  662. ci_adfes <- boot.ci(boot_adfes, type = "perc")$percent[4:5]
  664. # Bootstrap JeFEE
  665. boot_jefee <- boot(data = moebius_data, statistic = cor_fun, R = 10000, x = "Sunnybrook", y = "acc_JeFEE")
  666. ci_jefee <- boot.ci(boot_jefee, type = "perc")$percent[4:5]
  668. # Prepare data frame
  669. df_boot <- data.frame(
  670. r = c(boot_adfes$t, boot_jefee$t),
  671. test = rep(c("ADFES", "JeFEE"), each = length(boot_adfes$t))
  672. )
  674. # Define colors
  675. color_map <- c("ADFES" = "#3271AD", "JeFEE" = "#C56637") # blue and violet
  677. # update data from R script
  678. r_JeFEE <- round(cor_test_jefee$estimate,2)
  679. r_ADFES <- round(cor_test_adfes$estimate,2)
  680. CI_JeFEE <- c(0.28, 0.81)
  681. CI_ADFES <- c(0.39, 0.8)
  683. ### 6. Plots Figure 2 ----
  684. p1 <- ggplot(moebius_data, aes(x = Sunnybrook, y = acc_ADFES)) +
  685. geom_point(color = "#3271AD") +
  686. geom_smooth(method = "lm", color = "#3271AD", fill = "#3271AD", alpha = 0.3) +
  687. labs(title = " ", x = "SFGS Score", y = "ADFES Accuracy (x)") +
  688. theme_minimal() +
  689. coord_cartesian(ylim = c(0, 100))+
  690. theme_minimal(base_size = 14) +
  691. theme(legend.position = "none",
  692. text = element_text(size = 10,
  693. family = "Helvetica"), # use Helvetica, size 16 for all text
  694. panel.background = element_blank(), # clean white background
  695. axis.line = element_line(colour = "black"), # black axis lines
  696. strip.text.x = element_text(size = 6.6), # facet labels (x): size 20
  697. strip.text.y = element_text(size = 6.6),
  698. panel.grid.major = element_blank(), # Rimuove le griglie principali
  699. panel.grid.minor = element_blank())
  701. p2 <- ggplot(moebius_data, aes(x = Sunnybrook, y = acc_JeFEE)) +
  702. geom_point(color = "#C56637") +
  703. geom_smooth(method = "lm", color = "#C56637", fill = "#C56637", alpha = 0.3) +
  704. labs(title = "", x = "SFGS Score", y = "JeFEE Accuracy (%)") +
  705. theme_minimal() +
  706. coord_cartesian(ylim = c(0, 100))+
  707. theme_minimal(base_size = 14) +
  708. theme(legend.position = "none",
  709. text = element_text(size = 10,
  710. family = "Helvetica"), # use Helvetica, size 16 for all text
  711. panel.background = element_blank(), # clean white background
  712. axis.line = element_line(colour = "black"), # black axis lines
  713. strip.text.x = element_text(size = 6.6), # facet labels (x): size 20
  714. strip.text.y = element_text(size = 6.6),
  715. panel.grid.major = element_blank(), # Rimuove le griglie principali
  716. panel.grid.minor = element_blank())
  718. p3 <- ggplot(df_boot, aes(x = r, fill = test)) +
  719. geom_density(alpha = 0.4) +
  720. geom_vline(xintercept = r_ADFES, color = "#3271AD", linetype = "solid", linewidth = 1) +
  721. geom_vline(xintercept = CI_ADFES, color = "#3271AD", linetype = "dashed", linewidth = 0.8) +
  722. geom_vline(xintercept = r_JeFEE, color = "#C56637", linetype = "solid", linewidth = 1) +
  723. geom_vline(xintercept = CI_JeFEE, color = "#C56637", linetype = "dashed", linewidth = 0.8) +
  724. scale_fill_manual(values = color_map) +
  725. labs(
  726. title = "",
  727. x = "Bootstrapped Correlation Coefficient (r)",
  728. y = "Density",
  729. fill = "Test"
  730. ) +
  731. theme_minimal() +
  732. theme(plot.title = element_text(hjust = 0.5))+
  733. theme_minimal(base_size = 14) +
  734. theme(legend.position = "none",
  735. text = element_text(size = 10,
  736. family = "Helvetica"), # use Helvetica, size 16 for all text
  737. panel.background = element_blank(), # clean white background
  738. axis.line = element_line(colour = "black"), # black axis lines
  739. strip.text.x = element_text(size = 6.6), # facet labels (x): size 20
  740. strip.text.y = element_text(size = 6.6),
  741. panel.grid.major = element_blank(), # Rimuove le griglie principali
  742. panel.grid.minor = element_blank()) +
  743. annotate("text", x = 0, y = 3,
  744. label = paste0("ADFES statistics:\nMean r = ",r_ADFES,"\n 95% CI: [",CI_ADFES[1],", 0.80]\nP(r > 0) = 0.987"), color = "#3271AD", size = 3) +
  745. annotate("text", x = 0, y = 1.2,
  746. label = paste0("JeFEE statistics:\nMean r = ",r_JeFEE,"\n 95% CI: [",CI_JeFEE[1],", 0.81]\nP(r > 0) = 0.972"), color = "#C56637", size = 3)
  748. # Plots
  749. row1 <- cowplot::plot_grid(p1, p2, labels = c("A", "B"))
  750. # Then: combine row1 with p3
  751. p4 <- cowplot::plot_grid(row1, p3, nrow = 2, labels = c("", "C"))
  753. # Export
  754. ggsave("figures/Fig_2_finale.png", # export for draft
  755. plot = p4,
  756. device = "png",
  757. dpi = 300,
  758. width = 18,
  759. height = 14,
  760. units = "cm")
  762. ggsave("figures/Fig_2.tiff", # export for journal
  763. plot = p4,
  764. device = "tiff",
  765. dpi = 1200,
  766. width = 18,
  767. height = 14,
  768. units = "cm")
  770. ### 7. R2_cv LOO ----
  771. # 1 Load data and keep Moebius participants
  772. dat <- moebius_data %>%
  773. dplyr::select(Sunnybrook,
  774. acc_ADFES,
  775. acc_JeFEE) %>%
  776. drop_na()
  778. set.seed(1)
  780. # 2 Pearson r with BCa 95 % CI (10 000 resamples)
  782. corr_tbl <- bind_rows(
  783. get_bca("acc_JeFEE"),
  784. get_bca("acc_ADFES")
  785. ) %>% mutate(q = p.adjust(p, method = "BH"))
  787. print(corr_tbl)
  789. # 3 Leave-one-out cross-validated R² with percentile CI
  790. loo_tbl <- bind_rows(
  791. calc_loo("acc_JeFEE"),
  792. calc_loo("acc_ADFES")
  793. )
  795. print(loo_tbl)
  797. # 4 Model projections
  798. grid <- tibble(Sunnybrook = c(20, 80))
  800. pred_tbl <- bind_rows(
  801. get_pred("acc_JeFEE"),
  802. get_pred("acc_ADFES")
  803. )
  805. print(pred_tbl)
  807. # OFMT correlation ----
  808. ### 1. Data preparation ----
  810. df_tot <- read_csv("data/Study2_Complete_Dataset.csv") %>%
  811. dplyr::select(subject, group, Sunnybrook,ofmt ) %>%
  812. mutate(match = parse_number(subject))
  815. df <- df_tot %>%
  816. filter(group == "moebius")
  818. ### 2. Correlation ----
  819. cor_test_Sunny <- cor.test(df$Sunnybrook, df$ofmt, method = "pearson")
  820. print(cor_test_Sunny)
  821. bf_test_Sunny <- correlationBF(x = df$Sunnybrook, y = df$ofmt, paired = TRUE)
  824. ### 3. Group difference ----
  825. anova <- aov_ez(id = "subject", dv = "ofmt", between = "group",data = df_tot)
  826. summary(anova)
  828. ### 4. Plot Figure 3 ----
  829. # Panel A: Scatterplot (no correlation)
  830. pA <- ggplot(df, aes(x = Sunnybrook, y = ofmt)) +
  831. geom_point(fill = "#F0E442", color = "gray" , shape = 21 ) +
  832. geom_smooth(method = "lm", se = TRUE, color = "#D1C436", fill = "#FAF6C2", alpha = 0.3) +
  833. labs(
  834. x = "Sunnybrook Score (Facial Motor Function)",
  835. y = "OFMT Accuracy (%)"
  836. ) +
  837. theme_minimal() +
  838. theme(plot.title = element_text(hjust = 0.5))+
  839. theme_minimal(base_size = 14) +
  840. theme(legend.position = "none",
  841. text = element_text(size = 10,
  842. family = "Helvetica"), # use Helvetica, size 16 for all text
  843. panel.background = element_blank(), # clean white background
  844. axis.line = element_line(colour = "black"), # black axis lines
  845. strip.text.x = element_text(size = 6.6), # facet labels (x): size 20
  846. strip.text.y = element_text(size = 6.6),
  847. panel.grid.major = element_blank(), # Rimuove le griglie principali
  848. panel.grid.minor = element_blank())
  850. # Colori per le condizioni
  851. colors <- c("moebius" = "#F0E442", "control" = "purple")
  853. descritive <- df_tot %>%
  854. filter(group %in% c("moebius", "control")) %>%
  855. group_by( group) %>%
  856. summarise(mean_acc = mean(ofmt, na.rm = TRUE),
  857. sd_acc = sd(ofmt, na.rm = TRUE),
  858. se = sd_acc / sqrt(n()),
  859. .groups = "drop") %>%
  860. mutate(x_pos = case_when(
  861. group == "moebius" ~ 2 - 0.1,
  862. group == "control" ~ 1 + 0.1,
  863. ))
  866. # Panel B: Group boxplot
  867. pB <- ggplot(df_tot,
  868. aes(x = group, y = ofmt, fill = group)) +
  869. geom_rain(alpha = 0.5, rain.side = 'f1x1', id.long.var = "match") +
  870. geom_point(data = descritive,
  871. aes(x = x_pos, y = mean_acc),
  872. color = "black",
  873. size = 2.5, shape = 18, inherit.aes = FALSE) +
  874. geom_errorbar(data = descritive,
  875. aes(x = x_pos, ymin = mean_acc - se, ymax = mean_acc + se),
  876. width = 0, color = "black", linewidth = 0.5, inherit.aes = FALSE) +
  877. scale_fill_manual(values = colors) +
  878. labs(
  879. x = "",
  880. y = "OFMT Accuracy (%)"
  881. ) + theme_minimal(base_size = 14) +
  882. theme(legend.position = "none",
  883. text = element_text(size = 10,
  884. family = "Helvetica"), # use Helvetica, size 16 for all text
  885. panel.background = element_blank(), # clean white background
  886. axis.line = element_line(colour = "black"), # black axis lines
  887. strip.text.x = element_text(size = 6.6), # facet labels (x): size 20
  888. strip.text.y = element_text(size = 6.6),
  889. panel.grid.major = element_blank(), # Rimuove le griglie principali
  890. panel.grid.minor = element_blank())+
  891. geom_text(data = descritive, aes(x = group, y = 0.62,
  892. label = paste0("M = ", round(mean_acc, 3),
  893. "\nSD = ", round(sd_acc, 3),"\n")),
  894. color = "black", size = 2.5)
  896. # Combine both panels
  897. figure3 <- cowplot::plot_grid(pA, pB, labels = "AUTO", nrow = 1)
  899. ggsave("figures/Fig_3_finale.png", # export for draft
  900. plot = figure3,
  901. device = "png",
  902. dpi = 300,
  903. width = 18,
  904. height = 9,
  905. units = "cm")
  907. ggsave("figures/Fig_3.tiff", # export for draft
  908. plot = figure3,
  909. device = "tiff",
  910. dpi = 1200,
  911. width = 18,
  912. height = 9,
  913. units = "cm")
  915. ## END ----

Study2_Confirmatory_Analyses_and_Plot.R, no license · at the source

Overview

Authors: Paola Sessa1,2, Arianna Schiano Lomoriello1,2, Thomas Quettier1,2, Antonio Maffei1,2, Sara Costa3, Marta Nichele4, Pier Francesco Ferrari3,5
ORCID iDs: Paola Sessa, Arianna Schiano Lomoriello, Thomas Quettier, Sara Costa, Pier Francesco Ferrari
  1. Department of Developmental Psychology and Socialisation, University of Padova, Padova 35131, Italy
  2. Padova Neuroscience Center, Department of Developmental Psychology and Socialisation, University of Padova, Padova 35129, Italy
  3. Department of Medicine and Surgery, University of Parma, Parma 43126, Italy
  4. Independent Physiotherapist, Padova 35143, Italy
  5. Social Neuroscience and Comparative Development, Social Neuroscience and Comparative Development, Institut des Sciences Cognitives Marc Jeannerod, CNRS/Université Claude Bernard Lyon 1, Bron Cedex 69675, France
Institutions: University of Padua (Italy); University of Parma (Italy); Lyon 1 Université (France); Centre National de la Recherche Scientifique (France); Institut des Sciences Cognitives Marc Jeannerod (France)
Journal: Proceedings of the National Academy of Sciences of the United States of America, volume 123, issue 37, article e2608511123
Dates: received 12 March 2026; accepted 21 July 2026; published online 11 September 2026; in print 15 September 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1073/pnas.2608511123 · PMID 42726659 · PMCID PMC13578938 · OpenAlex W7212220682
Open access: hybrid, a free copy (OpenAlex)
Status: code verified
Categories: EEG (modality), human (organism)
Methods: Statistics, Machine learning, Connectivity, Spectral & time-frequency
Keywords: embodied cognition, facial paralysis, emotion recognition, sensorimotor calibration, Moebius syndrome
MeSH: Emotions*, Facial Expression*, Facial Paralysis*, Adult, Electroencephalography, Female, Humans, Male, Mobius Syndrome, Young Adult (* major topic)
Topic: Face Recognition and Perception (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: Fondazione Cassa di Risparmio di Padova e Rovigo (Foundation Cariparo) ("Ricerca Scientifica di Eccellenza" grant (2021))
Citations: cited by 1 paper (Europe PMC); 43 references in the paper

Abstract

Although recognizing emotions from facial expressions appears effortless, the field is divided between vision-based accounts, which posit matching to learned visual templates, and embodied accounts, which posit recruitment of facial motor circuits. We propose an ambiguity-gated, developmentally calibrated architecture in which the recognition system draws on sensorimotor information primarily when visual evidence is insufficient, with early motor experience proposed to shape the threshold and gain of this contribution. We tested this model across three studies (N = 185) integrating dynamic prototypical and nonprototypical expressions, congenital (Moebius syndrome) and acquired facial palsy, and standardized severity grading (Sunnybrook Facial Grading System; SFGS). Neurotypical adults (N = 117) confirmed a robust prototypicality-dependent cost on recognition. Both congenital (N = 15) and acquired (N = 19) cohorts showed severity–performance coupling for nonprototypical expressions, consistent with shared reliance on sensorimotor information when visual evidence is insufficient; only the Moebius cohort showed this coupling also for prototypical expressions, consistent with undercalibrated visual templates following lifelong atypical facial motor experience. Exploratory electroencephalography (EEG) revealed reduced sensorimotor–face-network functional connectivity in Moebius, consistent with altered cross-route integration. Reanalysis of an independent dataset of ultra-ambiguous static morphs replicated severity–performance coupling and revealed Moebius–control mean differences only under high perceptual demand. Together, the findings provide strong behavioral support for an ambiguity-gated, developmentally calibrated architecture and motivate direct tests of how developmental motor experience shapes when—and how strongly—the recognition system draws on sensorimotor information.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

OSF 3yux4

License: none: the authors keep all their rights
State: the link answers, verified on 26 September 2026
Evidence: files inventoried
Languages: R (4)
Size: 21 files, 4 scripts
Software Heritage: not checked
Found in: the references
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: easystats (4 files), tidyverse (4 files), BayesFactor (3 files), afex (2 files), brms (2 files), broom (2 files), cowplot (2 files), emmeans (2 files), ggplot2 (2 files)
Availability: 1 check, the latest on 26 September 2026: the link answers (HTTP 200)
  • 26 September 2026: the link answers (HTTP 200)
4 files
At the source: osf.io/3yux4/

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 4 scripts, each with its path and the digest of its content;
  • 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data, Materials, and Software Availability

Behavioral datasets, EEG data, R analysis scripts, and experimental protocols have been deposited in the Open Science Framework (OSF) (43) [Facial palsy reveals the sensorimotor contribution to facial-emotion recognition].

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 7 authors, 5 keywords, 10 MeSH terms, 1 funder, 32 references.

Cite

This paper

Sessa, P., Schiano Lomoriello, A., Quettier, T., Maffei, A., Costa, S., Nichele, M., & Ferrari, P. F. (2026). Facial palsy reveals the sensorimotor contribution to facial-emotion recognition. Proceedings of the National Academy of Sciences of the United States of America, 123(37), e2608511123. https://doi.org/10.1073/pnas.2608511123

BibTeX

@article{sessa2026facial,
author = {Sessa, Paola and Schiano Lomoriello, Arianna and Quettier, Thomas and Maffei, Antonio and Costa, Sara and Nichele, Marta and Ferrari, Pier Francesco},
title = {{Facial palsy reveals the sensorimotor contribution to facial-emotion recognition}},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
year = {2026},
month = sep,
volume = {123},
number = {37},
pages = {e2608511123},
publisher = {National Academy of Sciences},
issn = {0027-8424},
doi = {10.1073/pnas.2608511123},
url = {https://doi.org/10.1073/pnas.2608511123},
pmid = {42726659},
pmcid = {PMC13578938}
}

RIS

TY - JOUR
AU - Sessa, Paola
AU - Schiano Lomoriello, Arianna
AU - Quettier, Thomas
AU - Maffei, Antonio
AU - Costa, Sara
AU - Nichele, Marta
AU - Ferrari, Pier Francesco
TI - Facial palsy reveals the sensorimotor contribution to facial-emotion recognition
T2 - Proceedings of the National Academy of Sciences of the United States of America
J2 - Proc Natl Acad Sci U S A
PY - 2026
DA - 2026/09/11
VL - 123
IS - 37
SP - e2608511123
SN - 0027-8424
PB - National Academy of Sciences
DO - 10.1073/pnas.2608511123
UR - https://doi.org/10.1073/pnas.2608511123
LA - en
ER -

CSL-JSON

{
"id": "10.1073/pnas.2608511123",
"type": "article-journal",
"title": "Facial palsy reveals the sensorimotor contribution to facial-emotion recognition",
"container-title": "Proceedings of the National Academy of Sciences of the United States of America",
"author": [
{
"family": "Sessa",
"given": "Paola"
},
{
"family": "Schiano Lomoriello",
"given": "Arianna"
},
{
"family": "Quettier",
"given": "Thomas"
},
{
"family": "Maffei",
"given": "Antonio"
},
{
"family": "Costa",
"given": "Sara"
},
{
"family": "Nichele",
"given": "Marta"
},
{
"family": "Ferrari",
"given": "Pier Francesco"
}
],
"container-title-short": "Proc Natl Acad Sci U S A",
"volume": "123",
"issue": "37",
"page": "e2608511123",
"DOI": "10.1073/pnas.2608511123",
"PMID": "42726659",
"PMCID": "PMC13578938",
"ISSN": "0027-8424",
"publisher": "National Academy of Sciences",
"URL": "https://doi.org/10.1073/pnas.2608511123",
"language": "en",
"issued": {
"date-parts": [
[
2026,
9,
11
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1162/imag.a.1258 [code]
Non-specific increase in alpha power during a neurofeedback session targeting its downregulation.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: BayesFactor, afex, brms, 5 other tools, EEG
[2] doi:10.1038/s41467-026-73865-9 [code]
Histamine shapes the neurocomputational dynamics of human learning.
Journal: Nature communications
In common: BayesFactor, brms, easystats, 5 other tools
[3] doi:10.1093/nc/niag029 [code]
A data-driven approach to identifying and evaluating connectivity-based neural correlates of conscious visual perception.
Journal: Neuroscience of consciousness
In common: BayesFactor, afex, easystats, 5 other tools
[4] doi:10.1162/opmi.a.358 [code]
Tracking Visual Statistical Learning with Steady-State Visual Evoked Potentials: Effects of Exemplar and Category Information.
Journal: Open mind : discoveries in cognitive science
In common: BayesFactor, afex, brms, 4 other tools, EEG
[5] doi:10.1038/s41467-026-74565-0 [code]
The functional neurobiology of dispositions towards negative emotions.
Journal: Nature communications
In common: BayesFactor, afex, brms, 4 other tools
[6] doi:10.1371/journal.pone.0355165 [code]
Pupillary dynamics during hands-off L2 driving and transitions of control under high cognitive load.
Journal: PloS one
In common: BayesFactor, brms, easystats, 4 other tools
[7] doi:10.1016/j.isci.2026.116747 [code]
Age and loneliness relate to reduced trust learning and alterations in amygdala function.
Journal: iScience
In common: brms, easystats, broom, 4 other tools
[8] doi:10.1038/s41597-026-07350-9 [code]
An open multi-center MEG-EEG dataset for studying conscious visual perception.
Journal: Scientific data
In common: BayesFactor, afex, easystats, 3 other tools, EEG
[9] doi:10.1038/s42003-026-10229-5 [code]
The neurocomputational mechanisms underlying the impact of social comparison on effort investment.
Journal: Communications biology
In common: afex, brms, easystats, 3 other tools, EEG
[10] doi:10.1523/eneuro.0076-26.2026 [code]
Exogenously Driven Neural Reactivation of Spatially Matching Visual Working-Memory Contents.
Journal: eNeuro
In common: BayesFactor, brms, emmeans, 3 other tools, EEG

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.