Wednesday, 11 January 2012

Separating face from fiction with the fusiform gyrus

"You've got a face, I've got a face. It's all gonna be alright." But is it Noel Fielding? Is it really? And how do you know it's a real face anyway? After all you might simply be looking at that rocky outcrop in the picture which bears resemblance to a face. Or maybe you're looking at a piece of toast branded with the face of Jesus or Erik Estrada. Alright so we can probably give Noel the benefit of the doubt when it comes to his ability to visually descriminate human faces from rocky outcrops and toast. After all we can all readily tell the difference between an actual face and something that just resembles a face. But how is it that we are able to do this? How does our brain help us to sort the face from the non-face? 

Facing the facts on facial recognition

I guess to begin with we should probably take a quick look at how our brain helps us recognise faces in the first place. In short it's all thanks to a nifty little region of the temporal lobe known as the fusiform gyrus, or the fusiform face area (FFA). fMRI studies have consistently shown that when people are presented with facial images, major bilateral activation of the FFA occurs. Activation which is a great deal stronger than that produced by almost any other image. And it doesn't stop there. Like with a lot of cognitive / perceptual research a great deal of our knowledge and understanding stems from the investigation of individuals with acquired brain injury. And research into facial perception is no different. Damage to the FFA and surrounding areas results in a condition known as proposagnosia, which roughly translated means 'not knowing faces'. People with damage in the FFA have so much trouble perceiving faces that they are literally unable to pick their own face from a photographic line-up. Instead they rely on other stimuli such as clothing, voice and topics of conversation as clues to who it is they've been talking to for the past half-hour (it was probably that Noel Fielding fellow, they just kept banging on about faces). 

Remarkably this ability to perceive and recognise faces is not limited to us humans, with almost every species of primate ever tested showing some level of facial recognition ability. In fact even sheep have the ability to spot a face in the herd. However the most remarkable finding regarding facial recognition has to go to a recent study involving golden paper wasps (Polistes fuscatus). In this study the authors found  that the golden paper wasps, who themselves have clearly variable facial featureswere better at learning to  discriminate between images of other wasp faces, than they were at discriminating between images of caterpillars, which they prey on. Moreover these avant-garde arthropods were even able to discriminate between faces of wasps from other species. But hold on that doesn't necessarily mean that the wasps were using facial recognition to discriminate between the images. Surely it makes more sense that they were simply relying on basic pattern recognition of some kind. Well here's the kicker. The wasps uncanny ability to discriminate between faces was severely stunted when the images involved faces without antennae or faces which had simply been rearranged, thus suggesting that their ability to discriminate was based on facial recognition after all. But what about the faces in the rocks? Can they see those? Well probably not. After all they're only wasps. In fact there's a fair chance that pareidolia (the perception that a random stimulus is in some way significant) is a wholly human trait. But that's an argument best left for another day. 

Vision or visage

For now let's get back to focussing on how to distinguish face from fiction. In a recent study, published in the Proceedings of the Royal Society B: Biological Sciences, researchers conducted fMRIs whilst presenting 36 participants with images ranging from nothing like faces to actual faces. As expected when the images of real faces were shown to the participants there was distinct bilateral activation of their fusiform area. Just not at the same time. The researchers found that not only did activation of the left fusiform gyrus precede that of it's right-sided counterpart but it also appeared to change only gradually as the images became more facelike. The right fusiform gyru on the other hand showed dramatically stronger activation to real faces when compared to any of the other images, regardless of their inherent facelike attributes. These findings were particularly exciting as the provided researchers with the first known example of the hemispheric separation of roles in high-level visual processing, something which was usually reserved for such cognitive functions as spatial perception and speech. But what exactly does it all mean? Well put simply, these findings suggest that when you see something resembling a face in some way your left fusiform gyrus activates as if to say "hey look a face!" to which your right gyrus responds "nah that's just a rock". And you know what he's probably right!  


  • Cherian T, Singal G, & Sinha P (2012). Lateralization of face processing in the human brain. Proceedings. Biological sciences / The Royal Society PMID: 22217726
  • Katsnelson, A. (2011). Wasps clock faces like humans Nature DOI: 10.1038/nature.2011.9533 Meng M, 
  • Leopold, D., & Rhodes, G. (2010). A comparative view of face perception. Journal of Comparative Psychology, 124 (3), 233-251 DOI: 10.1037/a0019460
  • Science Daily
  • Sheehan, M., & Tibbetts, E. (2011). Specialized Face Learning Is Associated with Individual Recognition in Paper Wasps Science, 334 (6060), 1272-1275 DOI: 10.1126/science.1211334 


This post was written by Andrew Watt for A Hippo on Campus.


Post a Comment