Emotion Regulation and Emotion Perception: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Emotion Regulation and Emotion Perception (NCT03790878), a randomized na study. The primary outcome is The Ability to Perceive Negative and Positive Emotional Expressions, as Measured by a Behavioral Computer-based Assessment.

2. Clinical Question

3. Trial Design

The registry record describes a randomized, double-masked, parallel study with three behavioral conditions: Mindful Breathing, Habituation, and Control. The study status is completed. The recorded start date is 2019-02-04, and the primary completion date is 2019-12-02. The phase is listed as NA. Duke University is the sponsor.

Participants in the mindful breathing condition received training in a breathing skill to regulate emotional distress during a stressor task. Participants in the habituation condition received an exposure or habituation intervention for distress regulation during that task. The control group completed the task without emotion-regulation training. Each group was scheduled to receive mobile-phone reminders for one week, although the control reminders were intended to test for placebo effects.

The registry reports enrollment of 91. The primary outcome summaries include 25 participants in Mindful Breathing, 27 in Habituation, and 27 in Control. The registry notes that baseline data for one participant were not completed appropriately and were omitted from the primary analysis. It does not report the full recruitment flow, the reasons for any other differences between enrollment and outcome sample sizes, or the trial’s original sample-size justification.

The study’s primary outcome was assessed during the laboratory experiment in week 1. The secondary outcome concerned change in emotional distress, measured in units on a scale from the beginning to the end of the week-one laboratory experiment. The registry does not report the precise masking procedures, the allocation-concealment process, or further operational details of the randomization.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 11 participants were considered screen failures

Overall Study

Analysis population for the primary outcome, as defined in the registry: Baseline data for one participant was not completed appropriately and was omitted from the analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as The Ability to Perceive Negative and Positive Emotional Expressions, as Measured by a Behavioral Computer-based Assessment, measured over During the laboratory experiment (week 1).

The primary endpoint measures participants’ ability to perceive negative and positive emotional expressions using a behavioral computer-based assessment during the laboratory experiment in week 1. The reported unit is percent of correct responses, and the registry describes the outcome as a mean with a standard deviation. Its description states that scores for negative emotional expressions range from 0-100%, with higher scores indicating greater emotion-perception accuracy.

This kind of assessment translates performance on a defined task into a quantitative outcome that can be compared across randomized groups. In general, a behavioral measure can capture task performance directly, while remaining specific to the particular assessment and setting used. The result therefore concerns performance on this computer-based task during the laboratory session; the registry does not establish that it measures every aspect of emotion perception outside that setting.

The endpoint title refers to negative and positive emotional expressions, while its description specifically explains the score range and direction for negative expressions. The registry does not provide additional scoring details for positive expressions, nor does it report whether the displayed summary combines positive and negative expression scores or treats them separately. Those details are not reported in the registry record.

The registry notes that baseline data for one participant were not completed appropriately and were omitted from the primary analysis. It does not describe further missing-data handling for this endpoint.

7. Results Posted in the Registry

• Mindful Breathing: mean 77.31 versus 84.85 on control; difference -7.54 (95% CI -15.79 to 0.71), standardized effect d = -0.49; z-test p = 0.073.
• Habituation: mean 81.93 versus 84.85 on control; difference -2.92 (95% CI -11.99 to 6.15), standardized effect d = -0.17; z-test p = 0.528.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in Emotional Distress, as Measured by Subjective Units of Distress (SUDS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Mindful Breathing: mean 1.14 versus 0.56 on control; difference 0.58 (95% CI -0.01 to 1.17), standardized effect d = 0.53; z-test p = 0.053.
  • Habituation: mean 0.61 versus 0.56 on control; difference 0.05 (95% CI -0.44 to 0.54), standardized effect d = 0.05; z-test p = 0.843.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Mindful Breathing and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Habituation and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary endpoint: change in emotional distress measured in units on a scale between the beginning and end of the week-one laboratory experiment. Its group summaries and unadjusted comparisons show different patterns by intervention, while the sponsor reports an overall multilevel modeling result with p = 0.002. The registry does not provide the model estimate or confidence interval, so the size and precision of that overall result cannot be assessed from the reported analysis alone.

This outcome addresses emotional distress rather than accuracy in perceiving emotional expressions. It is related to the study’s intervention rationale because the interventions were intended to regulate distress during a stressor task, but it is not a direct replication of the primary emotion-perception measure. The two outcomes therefore describe different aspects of the laboratory study and should not be treated as interchangeable evidence.

The registry reports one secondary outcome among four outcome measures. The number of measures and analyses raises a general multiplicity concern: when several outcomes or comparisons are examined, the chance of at least one apparently notable result can increase. The record does not report a multiplicity adjustment, a hierarchy for interpreting outcomes, or whether the secondary analysis was prespecified. The secondary p-value should therefore be read with less weight than a result interpreted within a clearly documented analysis plan. It is not, by itself, a complete account of the evidence across outcomes.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change in Emotional Distress, as Measured by Subjective Units of Distress (SUDS)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry labels the primary outcome as a mean of a continuous measure and reports standard deviations for each group. A mean describes the group’s average score; a standard deviation describes how dispersed individual scores are around that average. Those summaries allow a basic comparison of groups, but they do not capture every detail of the participant-level data or of an adjusted model.

What does the overall test address?

The reported primary analysis is an ANCOVA across Mindful Breathing, Habituation, and Control, with p = 0.018. In general, analysis of covariance compares group outcomes while adjusting for one or more covariates. The registry does not name the covariates, specify how they entered the model, or report the adjusted group estimates. It therefore is not possible to explain which adjustments produced the reported result or how large the adjusted differences were. The p-value is evidence against the null hypothesis tested by that model, but it is not an effect size and does not tell which pair of groups differs.

The simple recalculations use the posted group means and standard deviations to estimate unadjusted mean differences between each intervention and Control. They do not reproduce the ANCOVA. An adjusted overall test and unadjusted pairwise comparisons can differ because they answer different questions and may use different statistical information. The record flags a discrepancy; accordingly, the sponsor’s own analysis takes precedence over these recalculations.

What does a confidence interval add?

In general, a confidence interval describes a range of values compatible with the observed data under the method used. It communicates uncertainty about an estimated difference, not the range of individual participant scores and not the probability that a particular effect is true. For both posted primary pairwise recalculations, the interval includes zero. That means the data do not precisely distinguish those estimated differences from no mean difference using these unadjusted calculations. The intervals also permit effects in the opposite direction, so the point estimates should not be read in isolation.

Why distinguish adjusted from unadjusted results?

An unadjusted comparison uses the posted group summaries without incorporating covariates. An adjusted analysis can account for measured variables specified in its model. In general, adjustment may change an estimate or its uncertainty; it does not automatically make a result more or less important. The registry does not report enough model detail to assess the ANCOVA specification, its covariates, or whether assumptions were checked. The posted unadjusted comparisons are therefore best treated as transparent recalculations from summary data, rather than substitutes for the reported model.

What does the reported p-value mean?

In general, a p-value measures how unusual data at least as inconsistent with a specified null hypothesis would be if that hypothesis and the model assumptions held. It is not the probability that the null hypothesis is true, nor does it measure the size or practical importance of an effect. The sponsor-reported p = 0.018 belongs to the three-group ANCOVA. The pairwise recalculated p-values are unadjusted and refer to different contrasts. Their interpretation also depends on the analysis plan and any multiplicity handling, which are not reported in the registry record.

For the primary outcome, the group-level sample sizes are 25, 27, and 27. The record explicitly notes one omitted participant because baseline data were not completed appropriately. It does not state whether the ANCOVA used exactly these analysis populations, how missing observations were handled beyond that omission, or whether any other participants were excluded. Such details matter because the analysis population and missing-data procedures can affect both estimates and uncertainty.

The outcome is bounded by its stated 0-100% score range for negative expressions. In general, statistical models for continuous outcomes can be sensitive to distributional features such as skew, unequal variability, or boundary effects. The posted summaries do not establish whether those features were present or whether they were evaluated. The registry does not report diagnostic checks or alternative model analyses, so no conclusion about the appropriateness of particular assumptions can be drawn from the record alone.

The registry also lists a secondary emotional-distress outcome and reports a multilevel modeling analysis for it. In general, multilevel models can account for correlated observations or nested data, but the record does not describe the model structure, the repeated measurements included, or the exact estimand. It reports an overall p-value but no model estimate or confidence interval. The statistical meaning of that result is therefore less fully documented than would be needed to interpret its magnitude.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate a group difference. The primary pairwise confidence intervals are wide enough to include zero and values on either side of zero. For Mindful Breathing versus Control, the estimated unadjusted mean difference is -7.54; its interval extends from -15.79 to just above zero. For Habituation versus Control, the estimate is -2.92 and the interval ranges from -11.99 to 6.147. These intervals indicate that the point estimates alone do not convey the uncertainty in the comparisons.

In general, a confidence interval is more informative than a point estimate alone because it shows the range of estimates that remain reasonably compatible with the data and calculation method. A broad interval signals limited precision; it does not prove that the true difference is zero. An interval containing zero also does not establish equivalence between groups. Equivalence would require a prespecified margin and an analysis designed to assess whether plausible differences fall within that margin; such a margin is not reported in the registry record.

The sponsor’s ANCOVA result is reported as p = 0.018, but the registry gives no corresponding adjusted effect estimate or confidence interval. That omission limits interpretation of the size and precision of the overall adjusted result. A p-value by itself cannot show whether an estimated difference is small or large, or how narrowly it has been estimated.

Precision is also affected by the number of participants contributing to an analysis and the variability of their scores. The posted primary group sizes are 25, 27, and 27, and the corresponding standard deviations are 14.36, 17.94, and 16.0. In general, smaller samples and greater variability tend to leave more uncertainty in estimated differences. The available summaries do not support an inference about what additional participants would have observed outcomes.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the original trial design. For these continuous outcomes, the figures are participants per arm, not total events. The planning assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect under the listed assumptions, if that effect and the planning model were appropriate. The 90% power column gives the corresponding planning number for a 90% chance. These are design targets, not descriptions of what happened in the completed study. The figures are given separately for each pairwise intervention-versus-control comparison.

The rows for 100%, 75%, and 50% of the observed effect illustrate how the planned effect size changes the required enrollment. A smaller effect is harder to distinguish from random variation, so detecting a smaller effect generally requires more participants. For example, the primary Mindful Breathing comparison lists 65 participants per arm for 80% power at 100% of the observed effect and 257 per arm at 50% of that effect. The corresponding Habituation figures are 532 and 2,128 participants per arm. These examples are planning illustrations, not proposed changes to the completed study.

The sample-size curve should be read as a visual representation of this relationship between the effect size assumed and the participants needed under the same planning framework. Neither the curve nor the listed planning figures is observed power. Observed power is not an appropriate substitute for examining the effect estimates and their uncertainty after data collection.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -7.540; pooled standard deviation σ = 15.235.
2. n = 2 × (1.96 + 0.842)² × 15.235² / -7.540² = 64.1, rounded up to 65 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -7.53999999999999                      # observed difference in means
sigma <- 15.2348484731552                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3790878)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 65 (80% power), 86 (90% power)
 75% of observed effect: n per arm = 114 (80% power), 153 (90% power)
 50% of observed effect: n per arm = 257 (80% power), 344 (90% power)
power.t.test: 65.1 
Simulated power at n = 65 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s reported primary analysis is an ANCOVA comparing the three groups, with p = 0.018. The registry does not provide an adjusted effect estimate, confidence interval, covariates, or further model details for that analysis. The simple recalculations use posted group means and standard deviations to compare each intervention separately with Control, without adjustment. They give different p-values and confidence intervals because they are not the same analysis as the ANCOVA.

The registry record flags a discrepancy between the sponsor’s analysis and the recalculation. In that situation, the sponsor’s own analysis takes precedence. The recalculated contrasts remain useful as unadjusted descriptive comparisons, but they should not be presented as a reproduction of the sponsor’s model or used to overturn its reported result. The overall ANCOVA result also does not, by itself, establish which particular pairwise comparison accounts for the overall finding.

For the secondary emotional-distress outcome, the registry reports a multilevel modeling analysis with p = 0.002, without an estimate or confidence interval. The record does not provide model details sufficient to compare that analysis directly with the simple recalculations. In both outcomes, a fuller account of the model and its estimated effects would be needed to understand the relationship between the reported analysis and the summary-statistic calculations.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a three-group ANCOVA for the primary endpoint and a multilevel modeling analysis for the secondary endpoint. It also lists four outcome measures and three analyses. The record does not describe a prespecified hierarchy among outcomes, a multiplicity adjustment, or how the reported p-values should be interpreted across the several outcomes and group comparisons. Without those details, the possibility of chance findings from examining multiple results cannot be evaluated fully.

The primary record notes that baseline data for one participant were not completed appropriately and were omitted from the primary analysis. It does not describe the handling of other missing data, whether the analysis followed an intention-to-treat principle, or whether participants were analyzed according to their randomized groups. These details are not reported in the registry record. The primary group sizes are 25 for Mindful Breathing, 27 for Habituation, and 27 for Control, while the safety denominators differ for Mindful Breathing; the registry does not explain that difference.

For the secondary endpoint, the group summaries include 26 participants in Mindful Breathing, 27 in Habituation, and 27 in Control. The registry does not explain whether missing measurements occurred or how repeated measures were handled in the multilevel model. The model’s structure and assumptions are also not reported. These omissions limit assessment of whether the analyses used consistent populations and how missing observations may have affected the results.

The registry does not report interim analyses, stopping rules, or the original sample-size justification. Those design details should not be inferred from the retrospective planning illustrations. Nor does the record explain whether the reported models were selected before examining the data. These uncertainties do not invalidate the posted results, but they limit how comprehensively the statistical analysis can be evaluated.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports zero serious adverse events and zero other adverse events in each of the three groups. The safety denominators were 26 for Mindful Breathing, 27 for Habituation, and 27 for Control. It also reports zero deaths in each group. No serious adverse-event terms are listed.

These figures describe what was recorded in the registry for the participants counted in each safety group. They do not establish that no adverse events could occur, and they do not provide a precise estimate of the frequency of uncommon events. The registry does not give further information about how events were collected, the observation period for safety reporting, or whether any events were excluded from these summaries. The Mindful Breathing safety denominator is not the same as its primary-outcome sample size; the record does not explain the difference.

17. What the Result Does, and Does Not, Mean

The results describe a randomized, double-masked, parallel comparison of two emotion-regulation interventions and a control condition on a laboratory computer-based assessment. The primary group means were 77.31 for Mindful Breathing, 81.93 for Habituation, and 84.85 for Control. The unadjusted recalculated comparisons have confidence intervals that include zero, while the sponsor reports an overall three-group ANCOVA with p = 0.018. The record does not provide the adjusted effect estimate or interval needed to describe the magnitude and precision of that overall result.

These results do not establish which specific group differences account for the overall ANCOVA finding. The unadjusted pairwise recalculations and sponsor-reported adjusted analysis answer different questions, and the registry flags a discrepancy; the sponsor’s own analysis takes precedence. The available results also do not establish that the groups are equivalent, because no equivalence margin or equivalence analysis is reported.

The outcome is performance on a behavioral assessment during a laboratory experiment in week 1. The record does not establish how the findings relate to emotion perception in other settings or to outcomes beyond that assessment. Likewise, the secondary emotional-distress measure concerns a separate endpoint and cannot be used as a substitute for the primary measure.

In general, trial results are interpreted in light of the study population, outcome definitions, analysis plan, uncertainty, and missing-data handling. Several of those details are not reported in the registry record. The posted numbers therefore support a bounded statistical interpretation of the recorded study, rather than a broad conclusion about the effects of these interventions in other contexts.

18. Why This Trial Matters Statistically

This study presents a useful statistical distinction between an overall adjusted test and pairwise unadjusted comparisons. The reported ANCOVA addresses the three groups collectively, whereas the recalculations compare each intervention with Control using posted summary statistics. Those analyses can yield different results without being direct contradictions, because they may involve different model structures and questions. The registry’s discrepancy flag makes it especially important to identify which result comes from which analysis.

The example also shows why effect estimates and confidence intervals are needed alongside p-values. A p-value can indicate how compatible the data are with a null hypothesis under a particular model, but it does not convey the size or precision of a group difference. Here, the sponsor-reported overall result is not accompanied by an adjusted estimate or interval, limiting assessment of its magnitude. The unadjusted pairwise estimates have intervals that include zero, emphasizing the uncertainty in those specific contrasts.

The study also raises practical questions about analysis populations and multiple outcomes. A participant was omitted from the primary analysis because baseline data were not completed appropriately, but broader missing-data procedures are not described. The registry lists multiple outcome measures and analyses but does not document multiplicity handling. These reporting details are essential to understanding how results were produced and how much confidence to place in any one finding.

Finally, the planning illustrations show how target sample sizes depend on the assumed effect and desired power. They are not evidence of achieved power or a record of the trial’s original design rationale. Keeping planning calculations distinct from observed results is central to sound interpretation.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves important details about the primary analysis unspecified. It reports an ANCOVA and p = 0.018, but not the covariates, adjusted group means, effect estimate, confidence interval, or model diagnostics. As a result, the magnitude and precision of the sponsor’s overall adjusted finding cannot be assessed from the record alone. The pairwise recalculations are unadjusted and should not be treated as reproducing that model.

The participant population is not described in detail beyond the study procedures. The registry does not provide information about participant characteristics that would help assess how broadly the findings may apply. It also does not report the precise masking procedures or allocation-concealment process. These omissions limit assessment of trial conduct and generalizability.

For the primary endpoint, one participant’s baseline data were omitted because they were not completed appropriately. Other missing-data procedures and the analysis population are not described. The primary Mindful Breathing sample size is 25, while its safety denominator is 26; no explanation is given. The registry also does not report the original sample-size justification, interim analyses, or stopping rules.

The endpoint description gives a score range and direction for negative expressions but provides fewer scoring details about positive expressions. The statistical model assumptions and diagnostics are not reported. The record does not describe a multiplicity strategy despite listing multiple outcome measures and analyses. Finally, the secondary multilevel model is reported without an effect estimate, confidence interval, or model specification. These reporting gaps constrain interpretation without allowing the missing details to be inferred.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Emotion Regulation and Emotion Perception — NCT03790878.
• Registry record last updated 2021-02-24; status last verified 2018-12.
• Results first posted 2021-02-24.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT03790878, titled Emotion Regulation and Emotion Perception, is a completed randomized, double-masked, parallel behavioral study sponsored by Duke University. It compared Mindful Breathing and Habituation interventions with a Control condition among 91 enrolled participants. The primary endpoint was percent correct on a computer-based emotion-perception assessment during the week-one laboratory experiment. The registry reports group summaries and an overall ANCOVA result of p = 0.018, but does not provide the adjusted effect estimate or confidence interval.

The record also reports change in emotional distress as a secondary outcome and a multilevel modeling result of p = 0.002. One participant’s baseline data were omitted from the primary analysis. The record does not fully describe the analysis populations, missing-data procedures, multiplicity handling, or original sample-size rationale. The recalculated pairwise comparisons are unadjusted; because the registry flags a discrepancy, the sponsor’s own analysis takes precedence.