Viewing headless bodies causes face adaptation

VIEWING a stimulus for a prolonged period of time results in a bias in the perception of a stimulus viewed afterwards. For example, after looking at a moving stimulus for some time, a stationary stimulus that is viewed subsequently appears to drift in the opposite direction. These after-effects reveal to us the properties of our perceptual system. They occur because the neurons which are sensitive to the initial stimulus re-calibrate their responses; they adapt to compensate for the earlier enduring stimulus, and so can continue to encode current stimuli efficiently.

It was long thought that the properties of the initial (or adapting) stimulus have to be similar to those of the subsequent (or adapted) stimulus for any after-effect to occur. But a new study published this week in the journal Current Biology contradicts this assumption about perceptual adaptations, by showing that viewing photographs of headless bodies causes the brain to adapt to faces that are viewed afterwards.  

Another long-held assumption about perceptual adaptation is that such after-effects only occur in response to simple stimuli such as motion and colour. Recently, however, this too has been contradicted, by a number of studies which show that adaptation can occur in response to high level stimuli. One study, published in 1998, showed that a briefly flashed circle, when presented after a line, appears as an ellipse elongated perpendicular to the line. Another showed that prolonged viewing of a male face made subsequently presented faces appear more feminine than they otherwise would.

The new study, conducted by Avniel Singh Ghuman and his colleagues of the Laboratory for Brain and Cognition at the National Institute of Mental Health in Bethesda, Maryland, follows on from these earlier findings, and provides the first demonstration that adaptation after-effects can occur across different categories visual of stimuli.

In one experiment, participants were shown photographs of two individuals (panel A, below) and trained to identify each one. They were then presented with a continuum generated by morphing the two faces in the photographs (panel B), and asked to identify them. Before each of the identification trials, they briefly saw a photograph of only the body of one of the individuals in the photos. This was found to bias the subsequent identification of the morphed faces- viewing the body of one individual made the participants more likely to identify the morphed faces as depicting the other. The initial stimulus had caused adaptation to the faces presented afterwards, even though it did not contain a face itself.


One explanation for this effect is that the participants had formed a strong association between the faces and bodies, so that the bodies when viewed alone evoked a strong mental image of the face associated with it. To address this, the researchers performed a second experiment which did not require the participants to learn any associations between particular bodies and faces. Instead, this oen was designed to investigate whether the gender of a headless body would influence the perceived gender of a face presented after it. The participants were therefore shown photographs of headless male or female bodies (panel D, above), followed by one of five faces from a morphed male-to-female continuum (panel E).

Again, the same effect was observed. Face perception was significantly biased away from the gender of the viewed body; that is, viewing a photo of a male body made the participants much more likely to perceive morphed face as female, and vice versa. The effect was also observed when this experiment was repeated using back view photos of males and females, ruling out the possibility that it had something to do with the way the faces were cropped from the photographs. By contrast, there was no adaptation when the participants were first presented with scrambled male/ female photos.

The researchers then used the same photos of male and female bodies, but presented them to the participants for 1, 5, 10 and 20 second durations. This time, the magnitude of the after-effect was found to increase with longer exposure to the adapting stimulus. This is a defining characteristic of perceptual adaptation, so the finding that the magnitude of the effect increased with longer exposure shows that it was not due to some other process, such as a bias in decision-making while viewing the faces. Finally, the researchers found that photographs which have gender connotations, such as a football player or a high heeled shoe, failed to produce any adaptation, showing that the images of bodies do not evoke a generalized, abstract representation of identity or gender which then biases th eperception of faces. 

Thus, viewing photographs of headless bodies alters perception of subsequently viewed faces. In other words, the adaptation after-effect can cross from one category of  visual stimulus (bodies) to another (faces), whereby features of the later stimulus are inferred rather than directly perceived in the earlier stimulus. Apparently, this occurs because the body photos cause neurons which encode face-specific visual features to adapt their responses.

But what does this body-to-face adaptation tell us about the neural mechanisms underlying the processing of bodies and faces? Two mechanisms could account for the effect. Viewing a photo of a body may indirectly activate the neural representation of a face, so that its configuaration is automatically inferred from the body. Alternatively, the body and face are represented in the brain together, and this representation is adapted with prolonged viewing of the body alone.

One area of the brain area likely to be involved is the fusiform gyrus on the medial, or inner, surface of the temporal lobe. The fusiform gyrus contains an area known to be critical for processing faces (the fusirom face area) and another which processes bodies (the fusiform body area). The body and face sub-regions of the fusiform gyrus are adjacent to each other, so their proximity could facilitate the interactions necessary for the body-to-face adaptation effect.


Ghuman, A., et al. (2010). Face Adaptation without a Face Curr. Biol. 20: 32-36. DOI: 10.1016/j.cub.2009.10.077.

Suzuki, S., & Cavanagh, P. (1998). A shape-contrast effect for briefly presented stimuli. J. Exp. Psychol. Hum. Percept. Perform. 24: 1315-1341. [PDF]

More like this

If you look at a waterfall for about 30 seconds, and then shift your gaze to a nearby stationary object, such as a rock or a tree, that object will seem to drift slowly upwards. This well known optical illusion demonstrates a phenomenon called the motion after-effect, which is thought to occur as a…
Music can be thought of as a form of emotional communication, with which the performer conveys an emotional state to the listener. This "language" is remarkably powerful - it can evoke strong emotions, and make your heart race or send tingles down your spine. And it is universal - the emotional…
I really do love illusions of all sorts, in large part because they fit nicely into my narrative about the fallibility of human thought, but illusions are also great as windows into the ordinary working of our brains. For example, color afterimages provide direct evidence for opponent-processing…
"WHEN a man sits with a pretty girl for an hour," said Albert Einstein, "it seems like a minute. But let him sit on a hot stove for a minute, and it's longer than any hour." Einstein was describing one of the most profound implications of his Theory of General Relativity - that the perception of…

Fascinating! This seems to be almost the opposite of priming. I imagine there may be some interesting illusions based on the tension between priming and adaptation.

I wonder whether various words and phrases would have this effect? I'd expect that "high heeled shoe" would not, but what about "high cheekbones"?

Typo: fusirom for fusiform in the last paragraph.


By Chris Phoenix (not verified) on 13 Jan 2010 #permalink

Looks simple. Neurons take time to regenerate after firing.