Exposure to false cardiac feedback alters pain perception and anticipatory cardiac frequency
- Department of Psychology, Sapienza University of Rome, Italy
- School of Psychology, University of Aberdeen, United Kingdom
- Department of Neuroscience, Imaging and Clinical Sciences, “G. d'Annunzio” University of Chieti-Pescara, Italy
- Institute of Cognitive Sciences and Technologies, National Research Council, Italy
- Institute for Advanced Biomedical Technologies ‑ ITAB, “G. d'Annunzio” University of Chieti-Pescara, Italy
- UdA-TechLab, Research Center, University “G. d’Annunzio” of Chieti-Pescara, Italy
- Department of Psychology, “G. d’Annunzio” University of Chieti-Pescara, Italy
Figures
Figure 1
Trial timeline of the experiment.
Participants are seated with the electrodes of both the electrical stimulation and the ECG attached. Each trial starts with a fixation cross (60 s), after which the noxious stimulation is administered and the pain intensity (i.e. Numerical Rating Scale) and unpleasantness (i.e. Likert scale) ratings are collected. After each trial, a pause screen of 45 s was shown (i.e. ISI, Inter Stimulus Interval), after which participants proceeded to the next trial by pressing the spacebar. The trial timeline was identical for both the no-feedback and feedback phases, with the only exception that in the feedback phase participants were provided with the acoustic feedback (i.e. slower, faster, congruent) about their HR, reproduced for the whole fixation period (60 s) preceding the shock.
Figure 2
Interoceptive cardiac feedback shapes subjective pain ratings and anticipatory heart-rate responses.
(a) Intensity pain ratings (Numeric Pain Scale), (b) Unpleasantness ratings (5-points Likert Scale), (c) Real cardiac frequency. Values consist of the central tendency of change of pain intensity (a), unpleasantness (b), and heart rate (c) associated with the exposure to the interoceptive faster, slower, and congruent cardiac feedback, relative to the no-feedback. Values of zero on the vertical axis would represent no change relative to the exposure to the no-feedback. Positive and negative values would represent an increase and a decrease, respectively. Brackets denote significance: = significant one-way RM-ANOVA; short bracket = planned Faster–Slower contrast. Asterisks indicate two-tailed p values (*<0.05, **<0.01, ***<0.001, “n.s.”=not significant).
Figure 3
Exteroceptive feedback frequency does not induce interoceptive-like modulation of pain perception and anticipatory cardiac activity.
(a) Intensity pain ratings (Numeric Pain Scale), (b) Unpleasantness ratings (5-point Likert Scale), (c) Real cardiac frequency. Values consist of the central tendency of change of pain intensity (a), unpleasantness (b), and heart rate (c) associated with the exposure to the faster, slower, and congruent exteroceptive feedback, relative to the no-feedback. Values of zero on the vertical axis would represent no change relative to the exposure to the no-feedback. Positive and negative values would represent an increase and a decrease, respectively. Brackets denote significance:=significant one-way RM-ANOVA; short bracket = planned Faster–Slower contrast. Asterisks indicate two-tailed p values (*<0.05, **<0.01, ***<0.001, “n.s.”=not significant).
Appendix 1—figure 1
Interoceptive experiment: Model-predicted heart rate (HR) responses as a function of standardized trial progression (Trial) across feedback conditions (Feedback ×Trial was significant, F(3, 711.19) = 6.16, p<0.001).
The Trial variable is modeled as a continuous, mean-centered predictor capturing temporal dynamics within the session (from –0.5=early trials to +0.5 = late trials). Violin plots show the distribution and density of predicted HR values at early, middle, and late trial points.
Appendix 1—figure 2
Interoceptive experiment: Model-predicted Pain Unpleasantness ratings for each stimulation intensity (StimInt 2–4), across feedback conditions (Stimulus Intensity × Feedback was significant, F(3, 672.32) = 3.90, p=0.0088).
Violin plots and overlaid boxplots depict the distribution (violin) and interquartile range with median (box) of the observed data for each condition. Boxes are centered on the x-axis categories because they summarize the data within each stimulation intensity and feedback condition. Large colored dots and error bars show the model-predicted estimated marginal means (EMMs)± standard errors for each condition.
Appendix 1—figure 3
Interoceptive experiment: Model-predicted pain unpleasantness ratings as a function of standardized trial progression (Trial) across feedback conditions (Feedback × Trial, F(3, 675.33)=3.44, p=0.0165).
The Trial variable is modeled as a continuous, mean-centered predictor reflecting the temporal progression of the task (from –0.5=early trials to +0.5 = late trials). Violin plots display the distribution and density of predicted ratings at early, middle, and late stages of the session.
Appendix 1—figure 4
Interoceptive experiment: Model-predicted Numeric Pain Scale of intensity ratings (NPS) for each stimulation intensity (StimInt 2–4), across feedback conditions (Feedback × Stimulus Intensity, F(3, 667.07) = 3.4631, p=0.016).
Violin plots and overlaid boxplots depict the distribution (violin) and interquartile range with median (box) of the observed data for each condition. Boxes are centered on the x-axis categories because they summarize the data within each stimulation intensity and feedback condition. Large colored dots and error bars show the model-predicted estimated marginal means (EMMs)± standard errors for each condition.
Appendix 1—figure 5
Interoceptive experiment: Model-predicted Numeric Pain Scale of intensity ratings (NPS) as a function of standardized trial progression (Trial) across feedback conditions (Feedback × Trial, F(3, 669.15)=9.86, p<0.001).
The Trial variable is modeled as a continuous, mean-centered predictor reflecting the temporal progression of the task (from –0.5=early trials to +0.5 = late trials). Violin plots display the distribution and density of predicted ratings at early, middle, and late stages of the session.
Appendix 1—figure 6
Exteroceptive experiment: Model-predicted heart rate (HR) responses as a function of standardized trial progression (Trial) across feedback conditions in the Exteroceptive condition.
The Trial variable is modeled as a continuous, mean-centered predictor capturing temporal dynamics within the session (from –0.5=early trials to +0.5 = late trials). Violin plots show the distribution and density of predicted HR values at early, middle, and late trial points. Feedback x Trial was not significant; values are shown for consistency with the interoceptive experiment.
Appendix 1—figure 7
Exteroceptive experiment: Model-predicted pain unpleasantness ratings for each stimulation intensity (StimInt 2–4), across feedback conditions (Stimulus Intensity ×Feedback, F(3, 569.16) = 8.21, p<0.0001).
Violin plots and overlaid boxplots depict the distribution (violin) and interquartile range with median (box) of the observed data for each condition. Boxes are centered on the x-axis categories because they summarize the data within each stimulation intensity and feedback condition. Large colored dots and error bars show the model-predicted estimated marginal means (EMMs)± standard errors for each condition.
Appendix 1—figure 8
Exteroceptive experiment: Model-predicted pain unpleasantness ratings as a function of standardized trial progression (Trial) across feedback conditions (Feedback ×Trial, F(3, 568.85)=11.47, p<0.0001).
The Trial variable is modeled as a continuous, mean-centered predictor reflecting the temporal progression of the task (from –0.5=early trials to +0.5 = late trials). Violin plots display the distribution and density of predicted ratings at early, middle, and late stages of the session.
Appendix 1—figure 9
Exteroceptive experiment: Model-predicted Numeric Pain Scale of intensity ratings (NPS) for each stimulation intensity (StimInt 2–4), across feedback conditions (Feedback × Stimulus Intensity, F(3, 570.65) = 3.00, p=0.0301).
Violin plots and overlaid boxplots depict the distribution (violin) and interquartile range with median (box) of the observed data for each condition. Boxes are centered on the x-axis categories because they summarize the data within each stimulation intensity and feedback condition. Large colored dots and error bars show the model-predicted estimated marginal means (EMMs)± standard errors for each condition.
Appendix 1—figure 10
Exteroceptive experiment: Model-predicted Numeric Pain Scale of intensity ratings (NPS) as a function of standardized trial progression (Trial) across feedback conditions (Feedback ×Trial, F(3, 569.80)=12.13, p<0.001).
The Trial variable is modeled as a continuous, mean-centered predictor reflecting the temporal progression of the task (from –0.5=early trials to +0.5 = late trials). Violin plots display the distribution and density of predicted ratings at early, middle, and late stages of the session.
Appendix 1—figure 11
Individually calibrated stimulation intensities (in mA) associated with five subjectively defined pain levels (NPS 10, 30, 50, 70, and 90), shown separately for the Exteroceptive and Interoceptive experiments.
Each violin plot includes the distribution of individual values, boxplots with interquartile range and median, and the mean ± standard error.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Exposure to false cardiac feedback alters pain perception and anticipatory cardiac frequency
eLife 12:RP90013.
https://doi.org/10.7554/eLife.90013.3
