For years and years there has been a lot of speculation as to whether face processing is a special process. Some of the issues involved in face recognition are the same as for recognising any other possible object, other issues seem to be unique to just face recognition. The question remaining, is if the brain’s most efficient solution is to have special mechanisms for face recognition, or to simply extend the abilities of existing object recognition mechanisms.
Faces are surely special in a lot of phenomena, you can think of the inversion sensitivity and face pareidolia. Faces are extremely hard to recognise and put ogether when they’re upside down, this is the inversion sensitivity. Face pareidolia is the phenomena in which an otherwise random arrangement of stimuli is perceived as a face. Other objects do not seem to have the same amount of sensitivity for these kinds of phenomena as face recognition does. This makes us wonder about how this ‘specialness’ of face recognition came to be.
Is there a specific set of cognitive processes for dealing with face identity information compared to other forms of objects or is it just the result of domain- general brain processes which become more efficient because of all the practice we have. To try to find an answer to these questions, we’re going to have a look at the research that has been done and the evidence that has been found over the last couple of years. There are two sides to this story. The domain- generality and domain-specificity theories are the two opposites.
To see if we can find an answer to our questions I wil explain the two theories a bit and see what evidence there is for both theories. (PROSOPAGNOSIA) We will start off with findings and evidence for the domain-specificity theory. For face processing to be domain-specific, we need systems in the brain to be different. Farah (1996) had three criteria; systems should be functionally independent, physically distinct and the two systems should process information in a different way. In her research in 1996, she found evidence for the domain-specificity theory by testing on prosopagnosic patients.
Prosopagnosia is defined as the inability to recognize individual faces following brain damage, but visual processing and intellectual functioning remains intact (Busigny, 2010). The subjects in Farah’s research had an impairment in face recognition, but not in object recognition. By testing the subjects, it was found that the system in which the atients had an impairment was not necessary for recognizing objects other than human faces. This is complementary to the ‘functionally independent systems’ criteria Farah talked about.
In other research in which a prosopagnosic patient was tested, it was found that the patient had poor recognition of faces. This was linked to a perceptual impairment in face processing. Her object recognition was still intact and she still showed a normal ability to learn to recognize non facial objects. Along with this was still an impaired learning of facial exemplars used in the study (Riddoch, 2008). This case therefore also provides evidence for special perceptual processes for faces. (PROSOPAGNOSIA OTHER WAY) As you could see in the previous paragraph there has been a lot of research done on patients with prosopagnosia.
These patient did not experience the loss of their abilities to recognise objects. The evidence we gained through the prosopagnosic patients are of course very helpful, but no developmental cases with the opposite impairment had been reported yet, until Germine (2008) started researching a patient with normal face recognition, but with impaired object recognition. As Germine stated in an article: The existence of a case of non-face developmental visual agnosia would indicate that the development of normal face recognition mechanisms does not rely on the development of normal object recognition mechanisms. Which in other words, would mean that face recognition is special, because it is functionally independent. (INVERSION)
As I stated earlier in this essay, there are quite a few difficulties regarding face recognition. One of them is the face inversion effect. Faces are extremely hard to recognise when they are upside down. The recognition of faces tends to be best when presented upright (Busigny, 2010). The inversion effect has way more effect on faces than it does on any other type of stimulus. Farah did some research on a prosopagnosic subject and found that the subject was better at matching inverted faces than upright faces (1995).
This is the opposite of the usual face inversion effect. This finding implicates that there might be a domain-specific face processing system, because where there used to be an inversion effect in humans during face recognition, there is an opposite effect present in humans who are no longer able to recognise faces. Here you can see that even when there is a place in the brain that’s malfunctioning, he specialized systems still control the behaviour. To research this finding a bit more, we are going to have a look at the findings of Busigny and Rossion (2010).
Busigny and Rossion found something else than Farah did in 2009. They found that there is an absence of the face inversion effect in prosopagnosia patients, by testing patients themselves, but also by summarizing other research articles that tested the same finding. The finding by Busigny and Rossion is also evidence for the domain-specificity effect, because we see that once there is no more face recognition, the face inversion effect disappears, hile everything else in the brain stays intact. This is evidence for the hypothesis that face processing systems are functionally independent and physically distinct. INFANTS)
Another piece of evidence regarding the domain- specific theory is the research finding that new born babies have a visual preference for human faces over almost any other category of stimulus (Pascalis, 2009). New born babies also seem to be very good at detecting information from the face. You can think of identity, emotions and eye gaze directions (de Haan, 2001). De Haan also found that new born babies begin to form face prototypes at the age of 3 months already. All of this could be evidence for a specialized face processing system that is operating inside our brains from very early on.
To test this theory a bit further Sugita (2007), did some research on new born monkeys. The infant monkeys were separated from their mother immediately after birth and reared separately by human caregivers who had facemasks on. The monkeys were prevented from seeing faces or face-like stimuli for either 6 months, 12 months or 24 months. What Sugita et al. found was the same as was found in research with new born human babies; all infant monkeys showed a preference for faces. They referred looking at both human and monkey faces, even without having seen face-like stimuli since birth.
This could be further evidence of the domain-specificity theory, because even without seeing faces, the brain was still wired to have a preference for looking at faces more than it did for any other object. On the other side, we have the domain-generality theory. This theory states that faces are not necessarily special for the visual system, it just appears to be special. We are as good as we are at face processing, because we have so much training. We spend most of our time looking at faces, which makes us an xpert at face processing and recognising faces.
This theory is called the expertise hypothesis. Face processing requires a different level of processing than other objects need to be processed. Subordinate-level identification is the default for faces. Because of this and all the practice human beings have, the brain might have developed a specialist processing skill for faces. There has also been a lot of research regarding this hypothesis. To make sure we have an idea of both the domain- specificity and domain-generality hypotheses, I’ll list a couple of findings regarding the domain-generality hypothesis here.