Wayfinding Intervention in High-Fidelity Long-Term Memory: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Wayfinding Intervention in High-Fidelity Long-Term Memory (NCT04253587), a randomized na study. The primary outcome is MDT Change in Mnemonic Discrimination.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-masked, parallel-group trial sponsored by the University of California, San Francisco. Its status is COMPLETED, with enrollment listed as 50. The recorded start date is 2019-10-01 and the primary completion date is 2021-12-15. The phase is listed as NA. The registry record does not identify which people were masked or describe how masking was maintained.

The primary comparison assessed LabyrinthVR Trackers against Placebo Controls. The tracker intervention used a head-mounted virtual reality display and an adaptive spatial wayfinding game; movement was controlled by participant ambulation while wearing leg-position trackers. The placebo description allows for either a handheld tablet or a wireless virtual reality headset presenting commercially available narrative computer games. The record does not specify how these comparator formats were allocated or whether their use differed across participants.

Several other intervention descriptions appear in the registry, including a teleporting version of the virtual reality game, a rhythm-training game, a tablet version of the wayfinding game, and a wireless virtual reality version. Their presence does not make them part of the reported primary comparison. The posted primary-outcome summaries list 21 participants in each of the two compared arms. The registry record does not explain the relationship between those outcome counts and the enrollment of 50, including whether exclusions, missing outcome data, or other analysis-population decisions account for the difference.

The primary outcome was measured immediately before and immediately after completion of the training regimen. The registry lists six outcome measures and names two secondary outcomes: the Remote Cognitive Module and the Test of Visual Attention Change in Top-down Control. The registry record does not report the original sample-size justification, detailed visit schedule, allocation concealment process, or a full statistical analysis plan.

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 MDT Change in Mnemonic Discrimination, measured over baseline immediately before and post-assessment immediately after training regimen is completed.

The primary endpoint is change in mnemonic discrimination, measured by the Lure Discrimination Index on a recognition-memory task involving common objects. The registry describes the score as ranging from 0.00 to 1.00, with higher values indicating better performance. The assessment was scheduled at baseline immediately before training and again immediately after the training regimen was completed. The posted outcome is a continuous measure summarized by its mean and standard error.

Mnemonic discrimination concerns the ability to distinguish similar items from previously encountered items. In this task, the LDI provides a numerical summary of performance on that recognition-memory challenge. Using a defined score makes it possible to compare average change between trial arms, while the pre-training and post-training assessments relate the outcome to the training period. This is a description of what the endpoint measures, not evidence that a score change necessarily corresponds to a broader or lasting change in everyday memory.

The registry record does not report a threshold for a clinically meaningful LDI change, a minimum important difference, or the expected relationship between this task score and daily functioning. The endpoint should therefore be interpreted as task performance on the stated scale. Its meaning is also bounded by the timing of assessment: the posted primary outcome concerns the period immediately around the training regimen, not longer-term persistence.

7. Results Posted in the Registry

• LabyrinthVR Trackers: mean 0.09 versus 0.01 on control; difference 0.08 (95% CI -0.00 to 0.16), standardized effect d = 0.58; z-test p = 0.059.

8. Statistical Interpretation

9. Statistical Methods Explained

The reported and recalculated statistics answer related but not identical questions. The arm summaries describe average change within each group. A mean difference compares those average changes between groups. The recalculation is explicitly an unadjusted comparison based on the posted summary statistics; it should not be mistaken for a reconstruction of the sponsor’s full analysis.

What does a mean difference describe?

A mean difference is the value for one group minus the value for the comparison group, with direction depending on the stated order. In this comparison, the recalculated estimate is 0.08 for LabyrinthVR Trackers relative to Placebo Controls. It expresses the observed contrast on the LDI score scale, not a percentage change and not a probability. The posted arm means, 0.09 and 0.01, are the basis for that unadjusted contrast.

Means are useful for summarizing continuous outcomes, but they do not show the full distribution of participant-level results. Different patterns of individual scores can produce the same group mean. The registry record does not post individual-level LDI values or describe the distribution of changes. The standard errors of 0.03 for each arm describe uncertainty in the estimated arm means; they are not standard deviations of individual participant outcomes.

How should the p-value be read?

In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. The recalculated unadjusted p-value is 0.059. It is not the probability that the null hypothesis is true, nor the probability that the result occurred “by chance.” It does not measure the magnitude of the mean difference or whether the difference matters to participants.

The sponsor reports p = 0.016 for its ANOVA analysis, classified as equivalence. That value belongs to the sponsor’s analysis and should not be substituted for the recalculated p-value. Since the discrepancy is flagged, the sponsor’s own analysis takes precedence. The registry record does not report enough methodological detail to identify why the two analyses differ.

What does an interval contribute?

In general, a confidence interval describes a range of estimates compatible with the data under a specified statistical procedure and its assumptions. It conveys uncertainty about the effect estimate in a way a single p-value cannot. An interval that extends across zero includes both negative and positive differences under the chosen scale and direction. That does not mean all values in the interval are equally plausible, and it does not answer whether the interval lies within an equivalence margin.

The sponsor’s 95% interval runs from -0.06 to 0.06. The recalculated interval spans zero as well. The sponsor’s interval and p-value must be understood within the sponsor’s stated equivalence analysis; the equivalence margin itself is not reported in the registry record. Consequently, the posted information is insufficient to independently assess the equivalence criterion.

Why can analyses differ?

General statistical methods may use different outcome definitions, adjustment variables, variance assumptions, or analysis populations, and those choices can change an estimate and its uncertainty. That general observation does not identify the cause of the difference in this trial. The registry record does not report enough detail to determine whether the sponsor’s ANOVA used adjustments or other specifications not represented in the simple recalculation. No particular explanation should be assumed.

The sponsor’s reported value is a net mean difference of 0.06, while the unadjusted recalculation is 0.08. Their p-values also differ. Since the registry record flags a discrepancy, interpretation should foreground the sponsor’s analysis and identify the recalculation as a separate, simple comparison of the posted summaries. Neither should be silently substituted for the other.

What is over-dispersion, and is it relevant?

In general, over-dispersion means that observed variability is greater than a statistical model expects. It is especially relevant to some count-data models, where underestimating variability can make uncertainty appear too small. The primary outcome in this trial is continuous, so over-dispersion is not the central issue for the posted mean comparison. The registry record does not report a model diagnostic or a variance assessment for the sponsor’s ANOVA. More generally, the adequacy of uncertainty estimates depends on the analysis assumptions and the data, not only on the reported mean.

The results are therefore best read as a comparison of two reported group summaries, alongside a distinct sponsor analysis. The registry record’s lack of detailed methods limits independent assessment of the sponsor’s model, its equivalence specification, and the relationship between its analysis population and the enrolled population.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate the underlying group contrast. The recalculated mean difference is 0.08, but the uncertainty around it is substantial enough that the recalculated interval extends across zero. This means the posted summaries are compatible with a range of possible differences, including values in either direction; the point estimate alone should not be read as a precise estimate of the contrast.

The sponsor reports a 95% confidence interval from -0.06 to 0.06 around its net mean difference of 0.06. In general, a 95% confidence interval is produced by a procedure that, under its assumptions and repeated use, covers the target quantity at the stated rate. It is not a guarantee that the true value lies in a particular interval, nor does it give the probability that a particular hypothesis is correct. Its interpretation also depends on the analysis and target quantity being appropriate.

The two arm summaries each include 21 participants, and each has a standard error of 0.03. Those counts and standard errors help describe why the estimate is not exact, but they do not reveal the full participant-level variability or the effect of any missing outcomes. The registry record does not report the original sample-size justification or the anticipated precision target. It also does not report a clinically meaningful threshold for LDI change. Thus statistical uncertainty can be described, but the registry record does not establish whether the interval is narrow enough for a clinical or practical decision.

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 illustrative participant counts per arm, not totals across both arms and not observed trial results. They use the observed recalculated effect as a reference and consider planning for 100%, 75%, and 50% of that observed effect. The figures are 47 and 63 participants per arm for 80% and 90% power when planning for the full observed effect; 83 and 111 when planning for 75% of it; and 186 and 249 when planning for 50% of it.

The assumptions for these illustrations 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. These are planning calculations based on the observed result, not the trial’s original design assumptions; those original assumptions are not reported in the registry record.

In general, power is the probability that a study will detect an effect of a specified size under the planning assumptions if that effect is present. The 80% column gives the planned number of participants per arm for that power target; the 90% column gives the planned number per arm for the higher target. Smaller effects need more participants to detect under the same general assumptions, which is why planning for a smaller fraction of the observed effect produces larger counts. These figures are not observed power: they are planning illustrations, not a measure of what the completed trial achieved.

The registry’s planning note distinguishes participant counts per arm from time-to-event calculations, which would report total events. The primary outcome is continuous, and the listed planning figures are participants per arm. They should not be mistaken for the actual number enrolled, the number analyzed, or a recommendation independent of the assumptions used.

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 = 0.080; pooled standard deviation σ = 0.137.
2. n = 2 × (1.96 + 0.842)² × 0.137² / 0.080² = 46.4, rounded up to 47 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 <- 0.08                      # observed difference in means
sigma <- 0.137477270848675                     # 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(4253587)
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 = 47 (80% power), 63 (90% power)
 75% of observed effect: n per arm = 83 (80% power), 111 (90% power)
 50% of observed effect: n per arm = 186 (80% power), 249 (90% power)
power.t.test: 47.3 
Simulated power at n = 47 per arm: 0.797 (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 sponsor’s posted result is an ANOVA analysis of net mean difference, classified as equivalence. It reports an estimate of 0.06, a 95% confidence interval from -0.06 to 0.06, and p = 0.016. The simple unadjusted recalculation from the posted arm summaries gives a mean difference of 0.08 and p = 0.059. These are different estimates from different analyses, and the registry record does not explain their discrepancy.

Because the discrepancy is flagged, the sponsor’s own analysis takes precedence. The recalculation remains useful as a transparent, unadjusted comparison of the displayed means, but it should not be presented as the sponsor’s result or used to override it. The registry record does not describe the sponsor’s full model specification, the analysis population, or the reason for its equivalence classification. It also does not report the equivalence margin. As a result, the label and statistics can be relayed as reported, but an independent assessment of the equivalence conclusion cannot be made from the registry information alone.

The difference between the sponsor’s p-value and the recalculated p-value is not, by itself, proof that either calculation is erroneous. It does signal that the underlying estimands, assumptions, or methods may not be aligned; which explanation applies is not reported in the registry record. Interpret the sponsor’s estimate and interval as the principal reported analysis and keep the recalculated contrast clearly identified as unadjusted.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures, names two secondary outcomes, and reports two analyses. The primary endpoint is the change in LDI score. The named secondary outcomes are the Remote Cognitive Module and the Test of Visual Attention Change in Top-down Control. The registry record does not describe the full set of outcome definitions or the analyses associated with every listed measure.

When a study evaluates multiple outcomes, time points, or comparisons, the chance of obtaining at least one apparently notable result can increase if each is assessed separately. General approaches to multiplicity include adjusting statistical thresholds or defining a hierarchy of outcomes in advance. The registry record does not report whether multiplicity was addressed, which outcomes were considered confirmatory, or how any adjustment was conducted. The two posted analyses are not enough to establish the full analysis family.

Information about missing data and analysis populations is also limited. The overall enrollment is listed as 50, while the primary outcome summaries include 21 participants in each compared arm. The registry record does not explain this difference, indicate whether all randomized participants were included in the reported analysis, or describe missing-outcome procedures. It does not report whether participants were analyzed according to their randomized assignment or whether other analysis populations were used.

The registry record also does not report how baseline values were handled in the sponsor’s model, whether any covariates were included, or how deviations from the training regimen were treated. The original sample-size justification, interim analyses, and detailed statistical plan are not reported in the registry record. These omissions limit assessment of selective analysis choices, missing-data sensitivity, and how closely the posted estimates follow the trial’s planned methods. They do not establish that any particular method was or was not used.

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 posted serious adverse event figures list 25 participants at risk in the LabyrinthVR Trackers group and 24 at risk in the Placebo Controls group. The reported number with a serious adverse event is zero in each group, and deaths are also listed as zero in each group. The registry lists no serious-event terms.

These figures describe the serious adverse events recorded for the populations and periods represented in the registry entry. They do not establish that no event of any kind occurred: the record separately lists other adverse events as zero for both groups, but the meaning and collection procedures are not detailed in the information reported. The at-risk counts differ from the primary-outcome counts and from the overall enrollment, and the registry record does not explain those population differences.

A zero count in these groups is not proof that the interventions carry no risk. Serious events can be uncommon, and the registry entry does not provide enough detail to assess ascertainment, follow-up duration, event definitions, or completeness of reporting. These are descriptive safety figures, not medical advice or a basis for an individual treatment decision.

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

The posted primary-outcome summaries show a mean LDI change of 0.09 in the LabyrinthVR Trackers arm and 0.01 in the Placebo Controls arm. The simple unadjusted recalculation gives a difference of 0.08. This is a numerical contrast in the measured recognition-memory task over the period from immediately before to immediately after training. It is not, by itself, evidence about memory performance beyond that task or beyond that assessment period.

The sponsor’s own analysis reports a net mean difference of 0.06, with a 95% confidence interval from -0.06 to 0.06 and p = 0.016, and classifies the analysis as equivalence. The recalculation differs, and the sponsor’s analysis takes precedence because the discrepancy is flagged. The registry record does not report the equivalence margin, so the meaning of the sponsor’s equivalence classification cannot be independently evaluated from the posted details alone.

Neither a p-value nor a mean difference establishes whether an observed score change is important to participants. The registry record does not provide an LDI threshold for meaningful change, or evidence linking the score change to everyday memory, independence, or quality of life. It also does not report longer-term follow-up for this primary outcome. Conclusions should therefore remain tied to the stated task and immediate assessment window.

The outcome summaries include 21 participants per arm, whereas enrollment is listed as 50. The registry record does not explain the difference or describe how missing data were handled. That limits confidence about how the analyzed participants relate to all enrolled participants. The posted safety figures record zero serious adverse events in each group, but those counts do not establish absence of risk. Overall, the registry record describes a randomized comparison and an immediate task-based result while leaving important questions about analysis details, precision, and broader interpretation unresolved.

17. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation requires more than reading a group mean or a single p-value. The posted arm means are 0.09 and 0.01, and their simple unadjusted contrast is 0.08. The sponsor’s reported net mean difference is 0.06, with a different p-value and an equivalence classification. Keeping these analyses distinct prevents an unadjusted reconstruction from being mistaken for the trial’s own analysis.

It also illustrates the role of precision. An estimate is not a complete description of evidence: uncertainty intervals show which range of contrasts remains compatible with the analysis. Here, the sponsor’s 95% interval extends across zero, and the recalculated interval does too. The registry record does not give the equivalence margin, so the reported equivalence label cannot be assessed fully from the available statistical details.

The outcome itself matters to interpretation. LDI is a score on a recognition-memory task, with higher values indicating better performance. The posted result concerns immediate change around training, not a direct measure of broader function or lasting effects. Without a reported meaningful-change threshold, statistical differences cannot automatically be translated into practical importance.

Finally, the enrollment and outcome counts, the unreported missing-data procedures, and the absence of a detailed statistical plan show why transparency about populations and analysis decisions matters. Planning illustrations can help readers understand how assumptions about effect size and power influence participant needs, but they do not reconstruct the trial’s original design. These are general statistical lessons drawn from the features of this registry record, not claims about outcomes beyond those posted.

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.

Several trial-specific limits affect interpretation. The primary outcome summaries include 21 participants in each of the two compared arms, while overall enrollment is listed as 50. The registry record does not explain this difference or report the analysis population and missing-data handling. It is therefore not possible to determine from the posted information how fully the outcome summaries represent the enrolled trial population.

The reported sponsor analysis and the unadjusted recalculation differ, and the registry record does not explain why. The sponsor’s analysis takes precedence, but its detailed model specification is not reported. Although the analysis is classified as equivalence, the margin and its rationale are not reported in the registry record. This prevents independent evaluation of what degree of difference was treated as equivalent.

The primary endpoint is a task-based score assessed immediately after training. The registry record does not report a clinically meaningful threshold for LDI change or longer-term primary-outcome follow-up. The findings therefore cannot be extended from the posted task score to everyday functioning or persistence without additional evidence.

The comparator description includes different presentation formats, while the registry record does not specify their distribution. It also describes multiple other intervention variants, but the primary comparison concerns only LabyrinthVR Trackers and Placebo Controls. The record does not provide enough information to assess how design details, participant characteristics, or adherence may have influenced the observed contrast. These are limits of what the registry record reports, not evidence that any particular problem occurred.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Wayfinding Intervention in High-Fidelity Long-Term Memory — NCT04253587.
• Registry record last updated 2023-02-27; status last verified 2023-02.
• Results first posted 2023-02-27.
• 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

The registry record identifies NCT04253587 as a completed, randomized, double-masked, parallel-group trial sponsored by the University of California, San Francisco, with enrollment of 50. It compares LabyrinthVR Trackers with Placebo Controls for the primary outcome, change in LDI score on a mnemonic discrimination task. The posted arm means are 0.09 and 0.01, respectively, with 21 participants in each arm’s outcome summary.

The sponsor reports an ANOVA net mean difference of 0.06, a 95% confidence interval from -0.06 to 0.06, and p = 0.016, classified as equivalence. The unadjusted recalculation differs, and the sponsor’s own analysis takes precedence. The equivalence margin, detailed analysis plan, and explanation of the outcome-count difference are not reported in the registry record.