Reproducibility in Paranormal Research: Can Another Team Obtain the Same Results?

Introduction

An investigation produces an unusual recording. The team repeats the session and obtains a similar result. Another group visits the same location but observes nothing comparable. What should researchers conclude?

Science advances when methods and findings can be examined, challenged and, where appropriate, reproduced. A single striking observation may justify further study, but confidence grows when independent investigators can obtain compatible results under documented conditions.

Reproducibility is particularly important in paranormal research, where observations may be rare, conditions difficult to control and interpretations strongly influenced by expectation.

1. Repeatability and Reproducibility

The terms are related but not identical.

  • Repeatability generally concerns obtaining consistent results using the same method, equipment, operator and conditions over a short period.
  • Reproducibility generally concerns obtaining compatible results when some elements change, such as the operator, equipment, laboratory or analysis process.

Definitions vary between disciplines and standards. A research report should state exactly what it means by each term.

A repeated result from the same device and operator is useful, but it does not establish that an independent team will observe the same effect.

2. Why Unusual Reports Are Difficult to Reproduce

Potential challenges include:

  • The event may be rare or intermittent.
  • Environmental conditions may change.
  • Equipment may differ between sessions.
  • Device clocks may not be synchronised.
  • Investigators may use different definitions of an event.
  • Expectations may influence observation and reporting.
  • The original method may not have been documented in sufficient detail.

A failure to reproduce an event does not automatically prove that the original report was false. It does mean the evidence has not yet demonstrated a reliably repeatable effect.

3. Define the Claim Before Testing

A vague claim such as "the room becomes active" cannot be tested consistently.

A stronger research question defines:

  • The observable outcome.
  • The instrument or observation method.
  • The measurement period.
  • The baseline condition.
  • The threshold for identifying an event.
  • The comparison or control condition.
  • The analysis plan.

For example: "During a 30-minute recording period, does the calibrated temperature sensor at position A show a change exceeding a predefined threshold compared with a matched control period?" This is measurable, though it still requires a justified threshold and suitable controls.

4. Standardise the Procedure

A protocol should specify:

  • Site access and room layout.
  • Equipment models and settings.
  • Sensor positions and orientations.
  • Start and end times.
  • Environmental measurements.
  • Participant roles.
  • Event definitions.
  • Data storage and file naming.
  • Analysis software and parameters.
  • Conditions under which a session is stopped.

Standardisation does not mean every investigation must be identical. It means differences are documented so their effects can be considered.

5. Use Controls and Blinding

A control condition provides a comparison against which an observation can be evaluated. Depending on the question, this may be a quiet period, a different room, a period without questions or a known ordinary sound source.

Blinding can reduce expectation effects. For example, an analyst may review anonymised recordings without knowing which clips were collected during a reported event.

Where possible:

  • Collect baseline data.
  • Randomise the order of conditions.
  • Use independent observers.
  • Keep analysts unaware of condition labels until initial scoring is complete.
  • Define exclusions before examining results.

Controls should be appropriate to the hypothesis. A poorly chosen control can create false confidence.

6. Independent Replication

A useful replication attempt should provide the second team with enough information to follow the original method, while avoiding unnecessary cues about the expected result.

Share:

  • The research question and protocol.
  • Instrument specifications.
  • Calibration and setup information.
  • Data definitions.
  • Analysis code or settings.
  • Predefined outcome criteria.
  • Known limitations.

The second team should document deviations and report unsuccessful attempts. Negative or inconclusive results are part of the research record.

7. Preregistration and Transparent Reporting

Preregistration means recording key hypotheses, methods and analysis decisions before collecting or examining the results. It can reduce the temptation to change criteria after seeing the data.

AERM can use a simple protocol register containing:

  • Project title and date.
  • Primary hypothesis.
  • Primary outcome.
  • Planned sample or session count.
  • Control conditions.
  • Exclusion rules.
  • Analysis method.
  • Responsible reviewers.
  • Amendments and reasons.

Preregistration does not guarantee good research, but it makes the decision process more transparent.

8. How to Interpret Replication Outcomes

  • Consistent replication: supports the reliability of the measured effect under the tested conditions.
  • Partial replication: suggests that some conditions may matter and require further testing.
  • Failed replication: indicates that the original result was not reproduced under the new conditions.
  • Inconclusive replication: data quality or sample size is insufficient for a clear conclusion.

Even successful replication does not by itself prove a paranormal explanation. The effect may have an ordinary cause that has not yet been identified.

Conclusion

Reproducibility is not a demand that every investigation produce the same dramatic event. It is a commitment to clear methods, honest reporting and independent scrutiny. By defining outcomes, standardising equipment, using controls, documenting deviations and sharing limitations, AERM can make its research more useful to others.

A result that cannot be reproduced may still be worth documenting, but its uncertainty must remain visible. Scientific credibility comes from the quality of the process, not from how extraordinary the conclusion sounds.

AERM principle: No anomaly is considered paranormal until environmental, psychological, technical, and scientific explanations have been systematically evaluated.

References and Further Reading

  1. National Academies of Sciences, Engineering, and Medicine. (2019). Reproducibility and Replicability in Science. https://doi.org/10.17226/25303
  2. Nosek, B. A., et al. (2015). Promoting an open research culture. Science, 348(6242), 1422–1425. https://doi.org/10.1126/science.aab2374
  3. Munafò, M. R., et al. (2017). A manifesto for reproducible science. Nature Human Behaviour, 1, 0021. https://doi.org/10.1038/s41562-016-0021
  4. Center for Open Science. Preregistration resources. https://www.cos.io/

About AERM

AERM is an independent research and media initiative committed to transparent documentation and systematic evaluation of unusual reports. Paranormal Frame shares these principles through accessible research content.

Educational Disclaimer

This article is educational and does not establish the validity of any particular paranormal claim. Research involving people, private property or sensitive recordings must follow consent, privacy and safety requirements.