top of page

How to decide between competing explanations and theories

Updated: Jul 4

Imagine you need to choose between believing…

  • Whether the 2020 U.S. general election was 'rigged'/'stolen' or not

  • Whether some particular historical account of events occurred as someone claims or whether there are credible alternative accounts

  • Whether homeopathy works through a placebo effect or some mechanism unrecognized by modern science

  • Whether UFO sightings are evidence of extraterrestrial life or misidentified terrestrial phenomena

  • And so on, and so on.

How do you pick an explanation or theory when options conflict?

Last week, we told you about our replication study, which found evidence of a real Visual Mandela Effect, a phenomenon whereby large numbers of people appear to have the same memory that does not match reality. For instance, many people remember the Monopoly Man as having a monocle. The effect got its name because many people (incorrectly) thought they remembered that Nelson Mandela had died in prison.

In that newsletter, we also told you about two competing explanations for this effect:

  1.  The parallel worlds explanation:  the theory that there is an alternate universe in which Nelson Mandela died in prison and that we, somehow, have memories from that universe - perhaps because we have somehow crossed over from that one into this one, where he survived and became President of South Africa. Some people even go so far as to say that the mechanism for this change in the universe is the Large Hadron Collider (particle accelerator) at CERN in Switzerland; they say that when the Large Hadron Collider is turned on, we jump from one universe or timeline into another and our reality changes.

  2. The psychological explanation: the theory that our brains use our existing knowledge and expectations to interpret new experiences and encode memories. This can lead to memories being unintentionally altered or distorted to fit our pre-existing beliefs and assumptions about how things are supposed to be. People may have the same false memories simply because our brains work in similar ways when compressing and recalling information.

This week, we're going to show you how you can choose between these theories, and (more importantly) an approach to choose between any competing theories or explanations more generally. Even if you think that one of these two theories sounds ridiculous, it's an interesting question to ask: how should a person decide which of two theories to believe, in general? We're going to demonstrate an approach that philosophers call ‘principles of theory selection’. These can be used in all domains whenever you have to choose between two or more potential explanations.

Applying Theory Selection Principles to the Mandela Effect

In this case, we have two theories (parallel worlds vs. cognitive biases and memory mechanisms), which are both attempting to explain the Visual Mandela Effect. Even if we looked at all of the available evidence (which we can’t do here), we couldn’t know with 100% certainty which explanation is correct. But that doesn’t mean the theories are equally good! We’re now going to list a few of the most relevant principles of theory selection and see how the two theories compare. This will help us determine which theory it is more rational to believe in, and show how you can use this approach any time you encounter competing explanations in your own life.

1. Empirical Adequacy A theory should be able to explain and predict observable phenomena accurately. It should fit the available empirical evidence.

When it comes to this principle, here’s how it shakes out for the two theories we’re considering:

Parallel Worlds

✅ There are some reasons to believe in the existence of parallel worlds (for instance, there is the many-worlds interpretation of quantum mechanics - which is not the most popular interpretation among scientists, but it does have adherents). In particular, the many-worlds interpretation of quantum mechanics does fit the empirical data that's available - though other theories (that don't involve parallel worlds) also fit the data just as well.

❌ There is no empirical evidence supporting the notion that travel between parallel worlds is possible or has happened.

❌ There is empirical evidence against the theory that CERN has anything to do with changing timelines or opening portals to parallel worlds. For example, there are higher energy collisions of particles happening in our atmosphere all the time, than the ones happening in the Large Hadron Collider at CERN.

Cognitive Biases and Memory Mechanisms

✅ There is a vast body of empirical evidence from the discipline of psychology that supports the effects and mechanisms employed by this theory. Some of these were discussed in our last article, where the theory was outlined in more detail. One such mechanism is that we don't store our memories like a tape recording - they are highly compressed. When we recall memories, evidence suggests we are actually reconstructing elements of the memory automatically based on guesses the brain makes. The reason many people (falsely) remember, for example, that the Monopoly Man has a monocle may simply be that our memories don't encode that information, but our brains predict that the sort of character that the Monopoly Man is likely has a monocle, so when we recall the memory a monocle is added without us being aware of it. 

2. Falsifiability The best theories (a) make testable predictions that (b) could lead to observations that prove the theory false. For example, Einstein's theory of general relativity predicts that light will bend around massive objects. During a solar eclipse, astronomers observed starlight bending as it passed near the sun, confirming the prediction. If the light had not bent, this would have provided strong evidence against the theory. 

Parallel Worlds

❌ The theory that the Mandela Effect is due to movement between parallel worlds is not obviously falsifiable. It’s not clear what predictions it makes, or what we could observe that would prove it false. It might even be impossible to prove false.

✅ The theory that CERN is the mechanism by which travel between parallel worlds occurs does make a prediction: it predicts that we should see more reports of Mandela Effects after CERN undertakes experiments. There is not currently any robust evidence that this is the case, but the fact that this theory makes a prediction that can be tested does count in its favor.

Cognitive Biases and Memory Mechanisms

✅ The foundations of this theory are experiments that have been conducted and published in academic journals. For these experiments, falsifiable hypotheses are typically formulated and tested as part of the work.

3. Simplicity/Parsimony

Other things being equal, simpler theories are generally preferred over more complex ones (this is sometimes known as the principle of ‘Occam's Razor’). Simplicity comes in a variety of forms, such as:

  • Having fewer assumptions

  • Positing fewer kinds of entities

  • Being able to be described using fewer parameters or with a lower "complexity" model

  • Requiring fewer logical or mathematical tools to express the theory.

Of course, simplicity does not guarantee truth (just as none of the other principles for theory selection guarantee truth), but it is a heuristic that there are reasons to think help lead to correct conclusions more often, all else equal. 

Parallel Worlds

❌Assumptions: This theory requires several substantial assumptions, including the existence of parallel worlds, the possibility of movement between these worlds, and mechanisms (such as the Large Hadron Collider) that could facilitate such movement. None of these has been empirically established (though there may be independent reasons to believe in the existence of parallel worlds from quantum mechanical considerations, as mentioned above).

✅/❌ Kinds of Entities: It posits the existence of multiple, possibly infinite, universes and a way for consciousness or memory to traverse between them. There are some proposals in physics that could explain the former (which are compatible with empirical evidence). On the other hand, there are neither well-established theories of physics nor empirical evidence for the latter. 

Cognitive Biases and Memory Mechanisms

Assumptions: This theory operates with fewer assumptions, primarily relying on well-documented cognitive processes and psychological phenomena that have empirical evidence behind them.

✅/❌ Kinds of Entities: It involves familiar entities such as social influence and known cognitive biases. The existence of these things is not required by the Parallel Worlds theory, so it looks like both theories require different kinds of entities. It’s not clear which one is simpler in this regard.

4. Explanatory Power

A good theory should provide deeper explanations and unify diverse phenomena under a coherent framework rather than just describing observations.

Parallel Worlds

❌ The theory that the Mandela Effect is the result of movement between parallel worlds doesn't seem to explain anything other than the Mandela Effect.  So, it appears to have weak explanatory power.

Cognitive Biases and Memory Mechanisms

✅ The same cognitive biases and memory mechanisms that provide an explanation for the Mandela effect are also useful for understanding other phenomena, such as why eyewitness testimony is pretty often not reliable, why people have been found to be overconfident in their memories of tragic events, and so on. So, this set of explanations has good explanatory power for a variety of phenomena.  

5. Fertility/Fruitfulness

A fertile theory should open up new areas of research and suggest novel predictions and experiments.

Parallel Worlds

✅/❌ The theory of parallel worlds that humans could somehow be switched between could be a fascinating area of new research in theory - if anyone had any idea of how to do such switching. However, the idea of many worlds in quantum mechanics (the most rigorous proposed idea of how parallel worlds could work) provides no such known mechanism. 

✅ The CERN mechanism does suggest predictions and experiments that could be conducted. These would revolve around the research question of whether there are more reports of Mandela Effects after CERN activity or not.

Cognitive Biases and Memory Mechanisms

✅ The cognitive biases theory opens up numerous avenues for research in psychology, neuroscience, and social science. It encourages studies such as manipulating variables in controlled settings to observe their effects on memory accuracy. It also supports longitudinal studies on how memories change over time and the factors influencing these changes.

✅ This theory also leads to specific, testable predictions and experiments. Researchers can design studies to investigate how false memories are created, how compression works in memory, the impact of social reinforcement on memory, and the ways in which cognitive biases, stereotypes, and schemas affect recall.

What’s the verdict?

If we zoom out and look at how these two different explanations of the Mandela Effect fared, when we applied the principles of theory selection, we can see that the cognitive biases and memory mechanism  explanation had many more points in its favor and those points tended to be much stronger. When the parallel worlds explanation had a point in its favor, it tended to come with a caveat such as “There is not currently any robust evidence that this is the case”.

For this reason, the methodology of applying principles of theory selection pushes us to adopt the cognitive biases and distortions explanation of the Mandela Effect, over the Parallel Worlds explanation.

This process of applying principles of theory selection can be used across nearly all domains. So, the next time you’re deciding between two explanations / theories, consider taking a systematic approach by applying the principles of:

  • Empirical adequacy

  • Falsifiability

  • Parsimony / simplicity

  • Explanatory power 

  • Fertility / fruitfulness

Ultimately, this can help you tcome tomore informed conclusions based on the strength of the evidence and the quality of the reasoning supporting each explanation.

If you found this article helpful, please forward this content to your friends.  Word-of-mouth is a powerful way to help us gain visibility!


Gerald B
Gerald B
Jun 20

As Bertrand Russell said, when choosing between competing theories, if the experts disagree, it's best to withhold judgment. (paraphrased)

Jun 20
Replying to

What if nine out of ten experts agree? Does not expert consensus have a legitimate place in the process of reason? Then dissent means carrying the onus of providing new evidence or thinking that can convince the majority. If it still fails to do this after repeatedly trying, then either it still lacks sufficient data or it is just wrong.

bottom of page