Thomas Z. Ramsøy

Noise, cognitive load, and liking

In my previous post, I discussed elements of Daniel Kahneman’s concept of “noise” in his recent book with the same title. But there’s a deeper way we can appreciate the concept of noise, and how it affects our thoughts and behaviors.

Take the following examples:

  • You listen to music, but every once in a while it has a digital error and skips a beat
  • You want to see a YouTube movie, but it starts buffering every 5-7 seconds or so
  • You’re reading a book, and the print in the book is bad, so some letters are vague
  • You’re listening to radio, but the connection is bad and white noise is coming and going in the program

The effects of noise

So what’s going on here? I’m sure you recognize these examples.

As you might guess, these are also instances of what we can call “noise.” In a more technical definition, we can define noise as “random fluctuations that obscure or do not contain meaningful data or other information.” In other words, noise can be something that disturbs our otherwise fluent experience of something.

This is key. Noise is not merely randomness — noise is something that qualitatively changes and distorts our experience of something.

“Noise is something that disturbs an otherwise fluent experience of something”

But what are the effects of these examples of noise? What happens to music when there’s a scratch in the sound? When there’s a buffering of your YouTube video?

The effect is pretty straightforward: you enjoy the experience less. Noise disturbs your otherwise fluent experience of something.

But why does it happen? Turns out, it’s all about the emotional brain

Delays are horrifying

Some years ago, I was part of a research project for Ericsson and Vodafone. The aim of the study was fairly simple: to understand how we all respond to poor connection and delays when watching videos and going to websites.

We set up a study where we manipulated the delay that people were exposed to. Some participants got a few delays of a second or so. Nothing serious. Another group didn’t really get any delays at all. But one group got the full monty of video buffering and web page load delays.

People hate delays on their phones, and everything connected to the delay goes down the drain”

The results were pretty clear: Even during moderate delays, cognitive demand soared and emotional responses plummeted. Participants reported being extremely annoyed, giving up on the tasks they had been given.

The interesting part was: the brands associated with the bad experiences took a dramatic emotional hit! If they thought the videos were from YouTube, then emotional responses to YouTube dropped dramatically after the test. If they thought that Vodafone was the service provided, emotions soured like lemons.

The conclusion was pretty straightforward: people hate delays on their phones, and everything connected to the delay goes down the drain!

In other words, something that is unpredictable becomes “noisy” and we respond negatively to it.

In the first Ericsson/Vodafone study, we found that mobile delays were as stressful as watching a horror movie.

Ambiguity produces anxiety

Another type of studies comes from understanding how we respond to choices where the chances for the potential outcome are either known or unknown.

When a choice has a certain outcome where you know the possibilities, we call it a risky choice. Flip a coin: that’s a 50/50 choice. Rolling a dice? That’s 1/6 for getting a particular outcome.

In other choices, you don’t know the chances for the outcome. These choices we say have ambiguous outcomes. Would you bet on how the weather will be in your location in 53 days from now? What are the odds that it will rain? Sure, you can make a statistical calculation based on historical weather data, but you really don’t know.

“When choices are ambiguous, we also tend to avoid them more”

When we scan the brains of people making choices in either risky or ambiguous conditions, we see a pretty dramatic effect: ambiguity makes us more fearful. This is both seen in brain scans and in behavior.

For example, in a study from Paul Glimcher’s lab, it was shown that both risky and ambiguous choices engaged the amygdala, a brain structure typically involved in emotional responses. Often we see that the amygdala is involved in fear and avoidance behaviors, but in my own lab, we have also seen that the amygdala can be just as active in positive emotions, and in signaling whether a choice is easy or not.

In Glimcher’s study, they discovered that ambiguity led to a stronger amygdala activity than risky choices. When choices are ambiguous, we also tend to avoid them more.

In short, when a situation does not contain information to make a situation predictable, the brain responds with negative emotions, and we dislike the situation and other aspects connected to it.

The amygdala responds more strongly to decisions where the outcome is uncertain (ambiguity), compared to decisions where the outcome is known (risk).People also tend to avoid ambiguous choices more than risky choices.

When noise becomes aversive

In a similar vein, we have observed that negative emotions can happen when a situation contains a simple element that is unpredictable.

For example, in a study a few years ago, it was found that unpredictable simple sounds could trigger activity in the amygdala. Yes, the very same brain region we just mentioned when studying risky and ambiguous choice.

In a study by Herry and colleagues, it was found that unpredictable simple sounds would make the amygdala respond more strongly. This happened both in humans and mice. In addition, both species showed avoidance behavior when the unpredictable sound was played, as opposed to a predictable sound.

This led my team and I to ask: can unpredictable sounds affect how we perceive brands?

The test we set up was fairly straightforward: expose people to novel brands, and randomly play predictable or unpredictable sounds while they look at them. Then test afterwards what they think about each brand.

“Unpredictable sounds create a negative emotional response”

The results were published in 2012: people that listened to an unpredictable sound sequence

We then looked at the emotional responses. In this study, we measured emotional intensity — or arousal — by measuring changes in pupil dilation.

Here, we found that unpredictable sounds were associated with stronger emotional intensity. This response is highly connected to amygdala activity.

The conclusion was that unpredictable (noisy) sounds led to a negative emotional response, which produced a negative turn on anything presented at the same time as the sound.

Brands were liked less if they were associated with an unpredictable sound (left). Unpredictable sounds also led to an increased pupil dilation, suggesting a higher emotional engagement and possibly more negative emotional response.

The noise of noise

From these three examples, it is pretty clear that noise is more than a single thing. But the universal effect of noise is equally clear: it is bad for emotional responses, and bad for brands.

So in addition to Kahneman’s concept of “noise” we should think of ways that we can qualify noise more. Is noise something that is related to our senses, our emotional responses, cognitive overload, or something else? The more we understand “noise” the less noisy our understanding will be.