Investigating Accelerated Learning and Memory in Healthy Subjects Using a Face Name Memory Task: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Investigating Accelerated Learning and Memory in Healthy Subjects Using a Face Name Memory Task (NCT03309072), a randomized na study. The primary outcome is Associative Memory Assessed by a Face Name Associate Memory Experiment.

2. Clinical Question

3. Trial Design

The registry lists this as a completed randomized, double-masked, parallel-group study sponsored by The University of Texas at Dallas. Enrollment was 30: 15 participants in the active tDCS arm and 15 in the sham tDCS arm. The registry lists the phase as NA. The recorded start date is 2017-09-27, and the primary completion date is 2022-01-25. Details about the randomization procedure, concealment, and how masking was implemented or assessed are not reported in the registry record.

Both arms performed the face-name associative memory task. In the encoding phase, participants studied 60 face-name pairs and assessed the gender of the faces to remain focused on the task. The task description states that the face stimuli consisted of 120 grayscale pictures of human faces; the faces had neutral expressions and faced forward, with half male and half female. A 10-minute consolidation phase followed encoding. Participants were instructed to sit still, relax, and think about nothing in particular. During retrieval, participants saw 60 old and 60 new faces and indicated whether each face had appeared during encoding.

The primary outcome was assessed after a 10 minutes break following the study phase. The intervention descriptions specify active or sham tDCS with the face-name associate memory task. Further details of stimulation delivery, task administration, participant flow, deviations from the intended design, and the analysis population are not reported in the registry record. The registry lists one outcome measure and one analysis, with no secondary outcomes posted.

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.

Overall Study

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 Associative Memory Assessed by a Face Name Associate Memory Experiment, measured over Associate Memory is assessed after a 10 minutes break following the study phase.

The primary endpoint is associative memory, measured as the percent of correct name-to-face matches in the face-name associate memory experiment. It is a continuous task score: higher percentages represent a greater share of correct matches on the stated outcome. The score is assessed after a 10 minutes break following the study phase, following encoding, a consolidation interval, and retrieval.

The task uses learned associations between faces and names, so correct matching is a direct measure of performance on the association tested in this experiment. The registry’s task description also explains that participants studied 60 face-name pairs, then encountered old and new faces during retrieval. The reported endpoint, however, is specifically the percent of correct name-to-face matches; the record does not report separate endpoint results for old-face recognition, new-face responses, or other task components.

As a continuous measure, the outcome is summarized by a mean and standard deviation for each arm. The endpoint is useful for comparing average task performance between the randomized groups. It remains a measure of performance under this particular task and assessment timing. The registry does not report evidence that the score measures memory in settings beyond the experiment, nor does it report additional validation or clinically meaningful thresholds for interpreting a given score.

7. Results Posted in the Registry

• Active tDCS: mean 44.22 versus 35.44 on control; difference 8.78 (95% CI 0.34 to 17.22), standardized effect d = 0.74; z-test p = 0.041.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is a continuous outcome, reported as a mean percentage of correct name-to-face matches. The arm summaries are means and standard deviations. A mean describes the central average score in an arm, while a standard deviation describes the spread of individual scores around that arm’s mean. These summaries help show both the difference in average performance and the variation among participants. They do not show the full score distribution, which is not reported in the registry record.

What does the recalculated mean difference represent?

The recalculated comparison subtracts the sham-arm mean from the active-arm mean using the posted summary statistics. It is an unadjusted difference between the two arm averages, with no adjustment for covariates. Its positive direction reflects the higher posted mean in the active arm. The estimate is expressed in percentage points on the task outcome, rather than as a relative percentage change. The calculation is a summary-statistic comparison and should not be confused with the sponsor’s ANOVA analysis.

The recalculation uses the two arm sizes and their standard deviations to estimate uncertainty around the difference. The registry reports 15 participants in each arm. The recalculated estimate is 8.78, with a confidence interval extending from just above zero to 17.22 and an unadjusted p-value of 0.041. The interval and p-value are connected summaries of the data under the assumptions of the unadjusted comparison; neither accounts for analysis choices that are not described in the registry record.

How should the p-value be understood?

General statistical explanation: a p-value describes how unusual data at least as inconsistent with a specified null hypothesis would be if that null model and its assumptions were correct. It is not the probability that the null hypothesis is true, nor the probability that the finding occurred “by chance.” The recalculated p-value is unadjusted. The registry does not report whether the analysis was prespecified, whether any other outcomes or comparisons were considered, or whether adjustment for multiple testing was planned. Those omissions matter because repeated analyses can change how a nominal p-value should be interpreted.

The sponsor’s posted analysis is an ANOVA, with a reported mean difference (net) of 10.66 and p <.049. ANOVA is a method for comparing means using a model-based estimate of variation. The registry does not report the ANOVA specification, degrees of freedom, standard error, confidence interval, covariates, or other details needed to reproduce that result. It therefore cannot be assumed that this analysis is identical to the simple unadjusted recalculation.

Why does over-dispersion matter in general?

General statistical explanation: over-dispersion occurs when observed variation is greater than a statistical model expects. If a model understates variation, uncertainty estimates may be too narrow and tests may appear more decisive than warranted. This endpoint is reported as a continuous score with standard deviations, rather than as a count model; the registry does not report an over-dispersion assessment or a model that would make such an assessment possible. It is therefore not appropriate to assert that over-dispersion occurred in this trial. The broader lesson is that uncertainty depends on whether the model reflects the actual variability in the data.

Other important features of the analysis cannot be evaluated from the registry record. The method for handling missing outcome values, whether all randomized participants were analyzed, whether exclusions occurred, and whether any outcomes were transformed are not reported. The record does not report an interim analysis plan, stopping rules, or a detailed statistical analysis plan. The absence of these details does not establish that the procedures were absent; it means that they cannot be assessed from the registry record.

10. Confidence Intervals and Precision

The recalculated mean difference is 8.78 percentage points in favor of the higher observed mean in the active tDCS arm. Its confidence interval extends from just above zero to 17.22. A confidence interval gives a range of values compatible with the data under the method’s assumptions and indicates the uncertainty around the point estimate. It is not a guarantee that the true difference lies within those limits, nor does it describe the range of individual participants’ scores.

The interval is wide enough to allow a range of plausible effect sizes, from a difference close to zero to one substantially larger. Its lower bound is close to zero, so the result is less precise about whether the difference is small or more substantial than the point estimate alone might suggest. The interval also does not account for unreported issues such as missing data, model choices, or multiplicity.

The arm standard deviations, 11.54 for active tDCS and 12.04 for sham tDCS, indicate participant-to-participant variation around the reported means. The sample includes 15 participants per arm, so a limited number of observations contribute to each average. In general, a larger sample can provide more precise estimates when other design and measurement features are comparable, but precision also depends on outcome variability and analysis assumptions.

The sponsor’s ANOVA result has no confidence interval reported in the registry record. Its point estimate and p-value therefore do not provide the same direct interval-based view of precision as the recalculation. Neither result should be read as a precise forecast of what would occur in a new study; the interval describes uncertainty for the estimated contrast under its own analysis assumptions.

11. 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 planning figures are participants per arm, not total event counts. They are illustrations based on the observed effect, not the trial’s original design justification. The assumptions listed for these calculations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set as a stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

For planning for 100% of the observed effect, the figures are 29 participants per arm for 80% power and 38 participants per arm for 90% power. For planning for 75% of the observed effect, they are 51 participants per arm for 80% power and 68 participants per arm for 90% power. For planning for 50% of the observed effect, they are 114 participants per arm for 80% power and 152 participants per arm for 90% power.

In plain language, the 80% power column gives the estimated number of participants in each arm for a future study to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect is present. The 90% power column describes the corresponding planning target with a higher chance of detection, which requires more participants under these assumptions. Planning for a smaller effect requires more participants because a smaller difference is harder to distinguish from variation in the outcome.

These figures are planning illustrations, not observed power for this completed trial. They rely on the observed effect as their starting point, and a planning calculation cannot establish what the effect in a future study will be.

12. 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 = 8.780; pooled standard deviation σ = 11.793.
2. n = 2 × (1.96 + 0.842)² × 11.793² / 8.780² = 28.3, rounded up to 29 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 <- 8.78                      # observed difference in means
sigma <- 11.7926502534418                     # 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(3309072)
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 = 29 (80% power), 38 (90% power)
 75% of observed effect: n per arm = 51 (80% power), 68 (90% power)
 50% of observed effect: n per arm = 114 (80% power), 152 (90% power)
power.t.test: 29.3 
Simulated power at n = 29 per arm: 0.799 (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.

13. Analyses Reported by the Sponsor

The registry reports the sponsor’s own analysis as an ANOVA, with a mean difference (net) of 10.66 and p <.049. A confidence interval for that analysis is not reported in the registry record. The result should be distinguished from the simple unadjusted recalculation based on the posted arm summaries, which estimates a mean difference of 8.78 and reports a confidence interval from just above zero to 17.22, with an unadjusted p-value of 0.041.

These estimates are not identical. The registry does not give enough information about the ANOVA specification or analysis details to explain the difference, and it is not possible to reconstruct the sponsor’s result from the reported information. The sponsor’s analysis is the analysis reported by the registry; the recalculation is a separate, unadjusted check from the available group summaries. The discrepancy flag is false, so the record does not identify a discrepancy requiring the sponsor’s result to take precedence over the recalculation.

The two p-values are close in their reported form, but neither should be interpreted without attention to its method and assumptions. The ANOVA p-value is the sponsor-reported result, while the recalculated p-value belongs to the simple summary-statistic comparison. Details about prespecification, adjustment, missing-data handling, and analysis populations are not reported in the registry record, which limits a fuller comparison of the analyses.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and one analysis, and no secondary outcomes are posted. This tells readers what results appear in the record, but it does not fully describe the set of analyses considered by the investigators. Whether other outcomes, time points, subgroup comparisons, or alternative models were planned or examined is not reported in the registry record.

Multiplicity arises when several hypotheses are tested or several analytic choices are explored. In general, the more opportunities there are to obtain a seemingly notable result, the less directly a single unadjusted p-value reflects evidence from the entire analysis process. The recalculated p-value is explicitly unadjusted. The registry does not report a multiplicity adjustment or explain whether one was planned. The sponsor’s ANOVA p-value is also reported without information about any adjustment.

Missing-data procedures and the number of participants contributing outcome data are not reported beyond the arm enrollment counts. The record does not state whether every randomized participant completed the task, whether all enrolled participants were included in the outcome analysis, or how incomplete or unusable measurements were handled. It also does not report an intention-to-treat, per-protocol, or other named analysis population. Therefore, the relationship between the 15 participants assigned to each arm and the population actually analyzed cannot be confirmed from the record.

The original statistical analysis plan, interim-analysis procedures, and stopping rules are not reported in the registry record. This does not establish that no plans or procedures existed; it means the available record does not allow readers to check whether the posted outcome and analysis match a prespecified plan. These limits matter when interpreting nominal p-values and assessing the completeness of the analysis, even though the study is marked completed and a primary result is posted.

15. 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 among 15 participants at risk in the active tDCS arm and zero among 15 at risk in the sham tDCS arm. It also reports zero other adverse events among 15 participants at risk in each arm, and zero deaths in both arms. These are counts recorded for the participants and categories listed in the registry.

A zero count in a small study does not establish that an event cannot occur, and the figures do not provide a precise estimate of the frequency of uncommon events. They describe what was reported for this trial, not what should be expected in a larger or different population. The registry does not list serious adverse event terms, as no serious events are recorded. Further details about how adverse events were defined, collected, monitored, or attributed are not reported in the registry record.

The recorded counts should be read as safety reporting in this study, not as a general conclusion about the safety of either intervention. They also do not alter the interpretation of the memory endpoint: adverse-event counts and task-performance results answer different questions.

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

The registry’s posted outcome shows higher average correct name-to-face matching scores in the active tDCS arm than in the sham arm in this sample. The simple unadjusted recalculation estimates the difference as 8.78 percentage points, with an interval extending from just above zero to 17.22 and an unadjusted p-value of 0.041. The sponsor’s reported ANOVA gives a mean difference (net) of 10.66 and p <.049. Those are the statistical results recorded for the task and comparison.

They do not, by themselves, establish that active stimulation caused a difference. Randomization can support causal comparisons when implemented and analyzed appropriately, but details about the randomization process, analysis population, missing data, and prespecified analysis are not reported in the registry record. The small number of participants and the breadth of the recalculated interval also leave substantial uncertainty about the size of any difference.

The endpoint concerns performance on a specific face-name associative memory experiment, assessed after a 10 minutes break following the study phase. It does not establish how participants would perform on other memory tests, whether any difference lasts beyond the stated assessment, or whether findings apply to populations other than healthy adults. The task score is not accompanied by a threshold indicating what magnitude would be meaningful outside the experiment.

The registry’s stated purpose concerns accelerating learning and improving associative memory performance, but the posted outcome is a task score from this study. The results should therefore be interpreted at that level. They do not establish a general effect on learning, memory, or everyday functioning. Nor do the reported adverse-event counts establish the absence of risk. The record supports describing the observed group results and their statistical uncertainty, while leaving broader interpretations unresolved.

17. Why This Trial Matters Statistically

This trial illustrates how an outcome can be directly tied to a carefully described task while still having limits as a measure of a broader concept. Correct name-to-face matching provides a concrete continuous score for associative memory in the experiment. Its interpretation remains bounded by the task, the assessment timing, and the healthy-adult population listed in the registry.

It also highlights the difference between an observed contrast and the precision with which that contrast is estimated. The active arm has a higher posted mean, but the recalculated confidence interval spans values from close to zero to a larger positive difference. A point estimate alone would conceal that uncertainty. Reporting the interval alongside the means and standard deviations helps readers consider both the direction and plausible magnitude of the contrast.

The comparison between the sponsor’s ANOVA and the unadjusted recalculation is another useful statistical distinction. Both are reported as analyses of the arms, but they have different estimates, and the registry does not contain enough methodological detail to account for the difference. This shows why a reported result needs an analysis description that lets readers understand what was modeled and how uncertainty was calculated.

Finally, the registry record leaves important aspects of statistical planning and data handling unclear. Original sample-size assumptions, multiplicity procedures, missing-data handling, analysis populations, and interim analyses are not reported. These gaps do not erase the observed results, but they limit how fully readers can evaluate their robustness. The example emphasizes that statistical interpretation depends not only on a p-value, but also on design, endpoint, precision, and transparent reporting.

18. 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 describes a small randomized comparison, with 15 participants in each arm. The limited number of participants contributes to uncertainty in the arm means and in the difference estimate. The recalculated interval is broad and begins close to zero, so the data do not pin down the size of the difference precisely.

The posted endpoint is one task-based measure of correct name-to-face matches, assessed after a 10 minutes break following the study phase. The record does not report later assessments or results on other memory measures. The inference is therefore limited to the recorded task score and timing; the registry does not establish persistence of a difference or generalization to other forms of memory performance.

Participant characteristics beyond the description of healthy adults are not reported in the registry record. This limits assessment of how representative the sample might be of other healthy populations. The record also does not provide detailed eligibility criteria, participant flow, adherence, or implementation information for the stimulation and sham conditions. Although the study is listed as randomized and double-masked, the procedures used to randomize participants and maintain masking are not described.

The original sample-size justification and detailed statistical analysis plan are not reported. Nor are missing-data methods, analysis populations, multiplicity procedures, or interim analyses. The sponsor’s ANOVA is reported without enough detail to reproduce it or explain why its estimate differs from the unadjusted recalculation. Finally, the registry’s adverse-event counts are zero for the listed categories in both arms, but event definitions and collection procedures are not described. These limitations constrain interpretation of both the efficacy-related task result and the safety reporting.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Investigating Accelerated Learning and Memory in Healthy Subjects Using a Face Name Memory Task — NCT03309072.
• Registry record last updated 2022-02-03; status last verified 2022-01.
• Results first posted 2020-09-11.
• 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.

22. Record Summary

This completed, randomized, double-masked, parallel-group study enrolled 30 healthy adults, with 15 assigned to active tDCS and 15 to sham tDCS during a face-name memory task. The primary outcome was the percent of correct name-to-face matches, assessed after a 10 minutes break following the study phase.

The posted means were 44.22 in the active arm and 35.44 in the sham arm. The simple unadjusted recalculation estimates a mean difference of 8.78, with a confidence interval from just above zero to 17.22 and p = 0.041. The sponsor-reported ANOVA gives a mean difference (net) of 10.66 and p <.049. No adverse events or deaths are recorded in either arm. Important details about the original design assumptions, missing data, analysis populations, and statistical plan are not reported in the registry record.