The Doubtful Witness: You don’t really know what you saw

How often have you heard someone say “I know what I saw”. Observations and remembrances of events are deeply flawed but we still rely on our memory to give us a true account and we believe reports of eyewitnesses. These accounts are the primary evidence put forward in support of paranormal reality. Those who believe in the reality of UFOs, Bigfoot, ghosts, or anomalous phenomena are heavily influenced by seemingly legitimate and truthful tellings of strange events people say happened to them. They also contend that there are so many accounts that “there must be something to it” and “all these witnesses aren’t lying”. Investigators collect these reports and then derive ideas, theories, and conclusions from them. But if the reports are not accurate, the data is unreliable and misleading – garbage in, garbage out. Here is the second example of why we need to qualify eyewitness accounts as data. (See the first here.)

Everything that can go wrong with reporting an event

I dislike the phrase “I know what I saw” because it is well-established that we often don’t know what we saw. We perceive some but not all of the event, and our brain fills in the rest. No matter how credible or insistent the witness, we all misperceive and misinterpret observations. The problem of eyewitness accounts used as data is especially bad when it is the majority of evidence put forward in support of a claim that would overturn what we already established as true. Anecdotes or even first-hand accounts can lead us to look for more evidence but they are not good enough on their own to conclude much of anything.

Even claims that aren’t all that implausible, like sightings of ball lightning, have remained on the fringes of science because experimentation is difficult and controlled sightings are very rare. Eyewitness reports constitute the evidence researchers cite to argue there is something here to investigate. How do we know what people are calling ball lightning in their reports is actually that and not an error of perception or interpretation (or something else entirely)? Scientific researchers are aware of this problem.

In Ball Lightning: An Unsolved Problem in Atmospheric Physics by Mark Stenhoff (2002), I found a description of the missteps (p 48-49, referencing Wertheimer and Hartmann sources) that can occur in the establishment of an eyewitness account that is then used as evidence. During the process between the physical event happening and the reporting and recording of it, many possibilities for distortion and errors can and almost certainly do occur rendering the description of the event imperfect.

Starting with the event (an observation of an “energy change or source”), that is, an occurrence that produces some effect in the physical environment, the energy or properties of the event are transmitted to the observer.

Problem 1: The energy (sound, light, noise, heat) can be attenuated, blocked, or distorted during transmission to the observer. This is missed information.

Problem 2: The observers receive that energy – what they see, hear, feel, smell – through sense organs. Not everyone’s senses are as acute as another’s. Consider eyesight or hearing. If one has limitations on their senses, such as colorblindness, not all of the information gets received or it is received differently. Also, the senses may not receive enough information because it may be too dark, the observer’s attention focused elsewhere, or other environmental factors interfere. There can be byproducts of the observation that are misinterpreted. For example, if a bright light is seen on a dark night, the afterimage on the retina can be recorded as part of the event even though it’s a byproduct of the sensory input.

Problem 3: The information from the senses is transmitted to the brain where it perceived. If the brain is missing information (say that the outline of an object is seen in extremely low light conditions or it travels behind blocking objects), assumptions are made to fill in the missing information and we don’t even realize it. Here is where the brain takes over the story. It feels insulting to the witness to suggest their perceptions could be mistaken as they accept their own interpretations as fact. The observer’s perception will be influenced by their degree of alertness or their emotional state. Our perceptions of color, size, features, and distance are calculated based on perhaps limited or misleading sensory information. We are often very poor judges of these qualities when information is limited. (I always recommend E. Loftus’ Eyewitness Testimony book as essential reading.)

Problem 4: The observer tries to make sense of the perceptions based on their previous experiences, knowledge, and belief. The potential for distortion from the original event is particularly notable during shocking, confusing or frightening experiences. Here is where media depictions, suggestion, folklore, religious beliefs, and expectations can change the event into something completely different than what it physically is: it can make Venus into a UFO and a shadow into Bigfoot.

Problem 5: Reporting the event also can introduce false facts and distortions if the witness is led in relating the event or if they are trying to press a particular narrative onto the explanation.

Finally, later retelling of the accounts introduces memory distortions. Whenever the witness retrieves the memory, it is reconstructed. Inadvertent or deliberate changes creep in.

All these processes are natural, affect everyone, and are wholly underemphasized when eyewitness accounts are used as evidence. It’s why such accounts cannot be considered overly reliable and must be supported by other evidence.

These factors are why you should not accept witnesses stories at face value no matter how credible or insistent they are. Human beings make mistakes.

For an example of a morphed story via problem 5 and retelling, see the previous post Masefield’s Montrose Ghost Story.