Perception relies on integrating both new and old information


In a nutshell: New experimental techniques separating attention from expectation show how the brain integrates new sensory information with old information about previous events.

View Paper Abstract

Perception relies on integrating both new and old information

We perceive the world around us by combining new information (such as something we see or hear) with knowledge from previous encounters (about similar things we’ve seen or heard before). However, how the brain integrates so-called ‘low-level’ sensory information with the ‘high-level’ information stored in the brain is not clear.

This integration is believed to involve attention and expectation, but it has been difficult to study how these two processes work – independently and together – because existing experimental approaches struggle to distinguish between them and their effects.

To overcome this limitation, an interdisciplinary team of researchers – comprising PhD student Noam Gordon, philosopher Jakob Hohwy (his supervisor), Brain Function CoE investigator Nao Tsuchiya, and another colleague – designed two new experiments to test how expectation and attention are involved in perception.

In both experiments, participants watched a monitor featuring a series of pictures of a house or a face that were flashed on the screen at different frequencies. The pictures were scrambled to varying degrees and presented sequentially, like a movie, with some images clearer – and thus more easily identifiable – than others.

In the first experiment, participants either looked for the house or the face image appearing a certain number of times, as instructed beforehand (the ‘unexpected’ stimuli), or indicated when a previously memorised pattern of images was violated (the ‘expected’ stimuli). In the second experiment, participants were asked to count either the faces or the houses – thus requiring their full attention to the counted image – or to perform a demanding task that drew their attention away from these images.

As similar images were used in both experiments, the researchers could be confident that any resulting differences in perception were strictly due to the participants’ expectations or attention.

During both experiments, the researchers measured the participants’ brain activity non-invasively using electroencephalography (EEG). By showing the house and face images at different frequencies, and applying unique image-scrambling methods, the researchers could use EEG to distinguish brain activity involving new sensory information from activity involving information already stored in the brain.

As a result of their novel approach, the researchers found the first direct evidence demonstrating that expectation and attention increase the integration of low- and high-level information in the brain. How this integration occurs depends strongly on what the participants expected and what they paid attention to: expectation is linked to the use of high-level information, while attention is linked to the amplification of low-level information.

These results will help researchers to better understand not only perception, but also misperception – which can happen in healthy brains when we view illusions, and in the brains of people with conditions such as schizophrenia or autism.

Next steps:
The researchers plan to study where integration happens in the brain, so they can better understand how this process is involved in abnormal perception.

Gordon, N., Tsuchiya, N., Koenig-Robert, R., & Hohwy, J. (2019). Expectation and attention increase the integration of top-down and bottom-up signals in perception through different pathways. PLoS Biology, 17(4), e3000233. doi: 10.1371/journal.pbio.3000233

Republish this article:

We believe in sharing knowledge. We use a Creative Commons Attribution 4.0 International License, which allows unrestricted use of this content, subject only to appropriate attribution. So please use this article as is, or edit it to fit your purposes. Referrals, mentions and links are appreciated.