"There, all is order and beauty / Luxury, calm and pleasure," wrote Charles Baudelaire in An invitation to travel. Science supports this: order, calm, and beauty are likely linked. According to a new study, what we find beautiful is partly related to the low energy consumption required for an image to be processed by the brain. The perception of beauty may reflect an affective signal triggered when visual coding is efficient and metabolically economical.", explains to Science and Future neuroscientist Yikai Tang, first author of this work published in the PNAS Nexus journal.
Uncomfortable images overstimulate the brain
In their search for the brain determinants of beauty, researchers were initially inspired by its antithesis. Because according to previous worksVisually uncomfortable images include those that overstimulate our sight compared to what exists in nature – for example, with a wide variety of patterns and bright colors.
Read alsoDoes universal beauty exist?
When we look at these uncomfortable images, our brain consumes more resources. This can be seen in functional MRI, which tracks blood flow and therefore oxygen consumption in real time. These results indicate a relationship between visual discomfort and increased physiological responses, or greater metabolic demands.", explains Yikai Tang. What if, conversely, the impression of beauty, of visual pleasure, was linked to images requiring less resource consumption?
Visualizing brain energy consumption using imaging
The scientists were right on the money, because that's exactly what they demonstrated. Radiator (not pleasant), canyon (pleasant), abstract scribbles (not pleasant), these are nearly 5,000 photorealistic images that were shown to more than 1,100 participants who had to rate their pleasantness from 1 to 5, then to four people lying in a functional MRI machine.
The researchers were thus able to link the image pleasantness score to oxygen consumption in certain brain areas, visible in the imaging. And to make the experiment more robust, they compared the results to those of an artificial intelligence model trained to model the complexity of our visual nervous system. This neural network provided a controlled way to estimate metabolic cost directly from the computational demands of visual processing. Demonstrating the same relationship in the network and in the human brain reinforced our conclusion.“, explains Yikai Tang.
Pleasant images consume less energy.
Read alsoWhy facial beauty is so subjective
The results are consistent between the computer model and fMRI, but they are clearest in the imaging. The images that people appreciated more required less neural energy to process.", reports the neuroscientist.
This effect is particularly pronounced in high-level visual areas such as the parahippocampal area (PPA), the occipital area (OPA), and the facial fusiform area (FFA). These regions are known to support higher visual functions such as scene processing, spatial arrangement, and faces.“, emphasizes Yikai Tang.
Our brain rewards energy conservation
Although the researchers did not test the correlation between the different characteristics of the images and their metabolic cost, previous studies shed light on this point. Properties such as minimalism, familiarity, and similarity of patterns could reduce processing requirements"The neuroscientist explains." Given the high metabolic cost of the visual system, the selection of stimuli that are easier to process may have conferred adaptive advantages.. »
Read alsoThe superpower of beauty, dissected by science
In short, when faced with images that require little energy, our ever-economical brain rewards us, in a way, with a feeling of pleasure and appreciation that we interpret as "beauty." The same mechanism might be at play with the faces we find beautiful, but Yikai Tang offers a more nuanced perspective. Facial beauty is a multifaceted and context-dependent phenomenon, shaped by influences that go far beyond perceptual coding.. »
English 
French