We all know that the appearance of objects often changes. For example, we cannot distinguish objects in dim evenings or fog. It is because the contrast of the objects decreases.
Still, if we are familiar with objects, the brain can identify them even if they become indistinct. The exact mechanism contributing to the perception of low-contrast familiar objects remains unknown.
Researchers from the National Institute for Physiological Sciences (NIPS) in Japan reveal a new flexible mechanism of information representation, changing in correlation with visual contrasts.
In the primary visual cortex (V1), the area of the cerebral cortex dedicated to processing basic visual information, the visual responses have been considered to directly reflect the strength of external inputs. Thus, high-contrast visual stimuli elicit strong responses and vice versa.
In this study, Rie Kimura and Yumiko Yoshimura found that in rats, the number of V1 neurons preferentially responding to low-contrast stimuli increases after repeated experiences.
In these neurons, low-contrast visual stimuli evoke stronger responses, and high-contrast stimuli evoke weaker responses. These low contrast–preferring neurons show a more evident activity when rats correctly perceive a low-contrast familiar object.
In Science Advances, it was first reported that low-contrast preference in V1 is strengthened in an experience-dependent manner to represent low-contrast visual information well. This mechanism may contribute to the perception of familiar objects, even when they are indistinct.
“This flexible information representation may enable a consistent perception of familiar objects with any contrast,” Kimura says. “The flexibility of our brain makes our sensation effective, although you may not be aware of it. An artificial neural network model may reproduce the human sensation by incorporating not only high contrast–preferring neurons, generally considered until now, but also low contrast–preferring neurons, the main focus of this research.”
Journal Reference
- Rie Kimura, The contribution of low contrast–preferring neurons to information representation in the primary visual cortex after learning, Science Advances, Vol 7, Issue 48 -DOI: 10.1126/sciadv.abj9976.