What you see is not what you get

Our visual judgments are driven only in part by the raw input that reaches our eyes. The rest comes from how our minds interpret that input in the context of our prior experiences and expectations. 

Visual perception is much more cognitively complex than it feels. We need to be able to identify objects from different angles, under lighting conditions that make them look different colors, and even when they are partially obstructed. Making sense of the visual world is so complicated that our brains have evolved shortcuts to streamline the process.

One key shortcut is to rely heavily on our expectations to make sense of ambiguous input. Have a look at the image below. At first glance, it appears to be a chaotic pattern of black splotches.

But when you are told there’s a Dalmatian sniffing the sun-dappled ground, the scene snaps into focus.

A second shortcut,“confirmation bias,” builds on the first: We tend to notice and remember information that confirms what we already believe. Not only do our expectations shape what we initially see, but they also shape where we focus our attention and what we remember. For example, someone who believes that the full moon leads to erratic behavior will notice and remember examples that are consistent with that, but may discount strange behavior at other phases of the moon or full moons without unusual incidents. We overvalue information that supports our expectations and overlook details that refute them.

Although these shortcuts may seem like flaws in perception, they are actually features.

We are constantly making unconscious, educated guesses about what is most likely to have caused the input our eyes receive. Although these shortcuts usually help us navigate the visual world quickly and efficiently, they can also lead to disagreements between people who bring with them different expectations when viewing the same stimulus.

The idea that we can see things differently from one another isn’t new. In 2015, a photograph of a certain dress brought public awareness to the fact that our unconscious expectations about ambient lighting affect the way we perceive the color of an object. More recently, the New York Times reported that viewers interpreted simulated body cam footage differently depending on their preexisting opinions about the police.

To quote Anais Nin’s 1961 work “Seduction of the Minotaur”, “We don’t see things as they are; we see them as we are.” We carry our biases, expectations, and lived experiences into every new perceptual experience. Consider watching a sporting event in which a referee makes a questionable call against your team—fans on opposite sides will often see the play differently, each convinced the referee got it wrong or right. Similarly, someone who has seen videos of violent protesters may interpret the actions in the Jan. 7 footage from Minneapolis as threatening, whereas someone who has seen videos of aggressive ICE agents may be primed to see excessive force in their actions. Once primed to see threat or excessive force, confirmation bias leads viewers to focus on details that support their initial interpretation.

What does this mean for controversies like the one surrounding Good’s death? First, it means that well-intentioned people can genuinely interpret the same footage differently—they may not be lying about what they see. However, understanding that perception can be biased does not mean that all interpretations are equally valid or that objective facts are impossible to determine. It actually means the opposite: Because individual human observers are prone to bias, rigorous investigation becomes even more critical. We need multiple perspectives, expert analysis, and—perhaps most importantly—as much contextual information as possible to establish what actually happened.

Realizing that our eyes and our brains can deceive us is precisely why we must demand evidence that goes beyond any single person’s interpretation of what they think they saw.