Critically evaluate the role of neural structures in sub serving a specific psychological function (Face Perception/Recognition)
Exam ID: 8093783
Word Count: 3000
Humans ability to recognise faces has largely been identified as different to any other aspect of normal object recognition. Human faces are altogether more complex and variable forms of stimuli with factors such as emotion, personal knowledge and eye gaze playing roles in the identification of them (Haxby & Gobbini, 2010). Faces are generally recognised on individual cases, as oppose to a general category identification (Ward, 2015). Looking at general models of face recognition aid in further understanding how different neural structures sub-serve face recognition. A general model for face recognition was proposed by (Bruce & Young, 1986) suggested that humans possess an early level of processing that computes a basic view-dependent structural description of a face, which is also present with general object recognition (see figure 1.) Expression-independent descriptions area also generated as this stage. (Bruce & Young, 1986). This structural coding activates face recogntion units which is stronger when there is a match between the face that has been seen and information that is currently stores in a persons face recognition unit. These FRUs are linked to both person identity nodes as well as the cogntive system as shown in the model. Person identity nodes, which is the point at which person recongition occurs, (Pizzamiglio, et al., 2017) access semantic and name information about the indiviudal face that is being recognised (Bruce & Young, 1986). This is supported by other non-facial cues and elements such as emotions (Bruce & Young, 1986). This model was later adjusted with the addition of an affective-response box that reinforces face-recogntion face recongition, combining it with feelings of emotion combined with the feeling of familiarity when seeing someone that is known to the person (Lewis & Ellis, 2001).
Research looking into prosopagnosia, the cognitive disorder which leads to a inability to recognise faces (Bate & Tree, 2017), can provide the basis of evidence for which neural structures in the brain contribute towards face recognition. Bodamer (1947) orginally described the condition as specific to recognising faces as oppose to other within-category distinctions such as object recongition (McNeil & Warrington, 1993). There are generally two forms of prosopagnosia: that which follows trauma or neurological illness as researched by (Damasio, Damasio, & Van Hoesen, 1982), or prosopagnosia that features the same defecits but without a noticeable brain injury which is described as developmental or congenital prosopagnosia (Bate & Tree, 2017). It is research of the neural structures associated with the emergence of prosopagnosia through trauma that can be beneficial in identifying such structures that are needed for face perception. Although prosopagnosia caused this way is rare, it can still provide evidence for specific face percetion areas of the brain (Pitcher, Walsh, & Duchaine, 2011).
Haxby et al’s heirarchical model for face perception provides the basis for this essay in understanding the neural structures that aid in the process of face perception (Haxby, Hoffman, & Gobbini, 2000, 2002). This model suggests that there is a hierarchally-organised network for face perceptiom: a core system that consists of three bilateral regions in the occipital-temporal extrastriate cortex. These areas include the fusiform face area (FFA), posterior temporal sulcus (pSTS) andn occipital face area (OFA) (Haxby, Hoffman, & Gobbini, 2000, 2002). This core system deals with the general perception and analysis of faces and is subserved by exteneded systems that processes and aims to understand the information that has been obtained from the core system about the face (Haxby, Hoffman, & Gobbini, 2002). Sub-systems included in this extended system include: spatial attention/perception which processess information from faces such as the gaze direction and head position and an auditory/visual system that processes speech-related lip movements. A biographical-semantic knowledge system in the anterior temporal lobe enables the retrieval of name and other information related to the percieved face and the emotional processing sub-system allows for reading of emotions (Haxby, Hoffman, & Gobbini, 2002). This model can be used as a basis for research on the neural structures involved in face perception. It provides a hierarchical structure to define what areas of the brain are most important in subserving faceperception (core-system) as well as other neural structures and processes that also aid in the whole face perception process (extended system)
This essay aims to us Haxby et al’s model to discover which areas of the brain are specialised for face perception and critically evaluate the role that they play in subserving this function. Research using funcional magnetic resonance imaging (fMRI), event-related potentials (ERPs), magnetoencepahlography and transcranial magnetic stimulation (TMS) will be used to provide such evidence to describe each neural structures role in face perception.
Fusiform Face area (FFA)
Haxby et al.’s model suggests that a core in the brain that is involved in the process of face perception is the fusiform face area (FFA), it is a face-selective region of the brain in the human extrastriate cortex (on the lateral side of the fusiform gyrus). There is a pleathora of research supporting the role of the FFA in face perception, it remains a separate and more complex process than other forms of object recognition. fMRI experiments show that there fusiform gyrus was significantly more active when participants were shown faces as oppose to whent hey were shown a range of everyday objects (Kanwisher & Yovel, 2006). Further to this it has been found that the FFA activated around two times as strong in experients wherby participants view faces as oppose to any other type of objects (McCarthy, Puce, Gore, & Allison, 1997) (Kanwisher, McDermott, & Chun, 1997). Activation in the FFA is also correlated with the perception of a face in binocular rivalry, the face/vase illusion, contrast reversal and face imagery. Further evidence for the function of the FFA in subserving face percetion comes from epilepsy patients with implamented subdural electrodes in discrete portions of the fusiform gyrus. Large N200 potentials have been produced by stimuli featuring faces as opposed to stimuli featuring scrambled faces, cars or butterflies (Ojemann, Ojemann, & Lettich, 1992) (Allison, et al., 1994).
One of the main debates surrounding the role of the FFA in face perception relates to its potential involvenent in within category object identification. Research from McGugin, Newton, Gore, & Gauthier (2014) showed that activity in the FFA correlated with expertise for certain objects. They used fMRI to investigate the activity of the FFA when participants were presented with stimuli of faces, cars and sofas (categorically visually similar to cars). They found that in the low load conditions (single object category shown), the FFA was more active through an expertise effect, with car expertise predicting an increseed response in the FFA for the sofa images. This therefore suggests that globally similar images to that of an expert category (cars and sofas) produce activation in the FFA, as appose to just images of faces alone.
Further to this Gauthier, Skudlarksi, Gore and Anderson (2000) stated that expertise with unfamiliar objects shows activation in both the FFA and OFA, with both areas showing expertise effects. It was argued however that the expertise effect in this study was potentially due to the “faceness” of the stimuli (birds and frontal views of cars resemble faces) (Xu, 2005). Despite changing the stimuli to eliminate this problem, an expertise effect was still found in the righ FFA for both car and bird expers, thus suggesting that the FFA may not subserve face perception specifically.
Challenging this point, there is evidence to suggest that within-category identifcation and expertise acivates other regions of the ventral occipitotemporal corex and not the FFA (Grill-Spector, Knouf, & Kanwisher, 2004). They found that the FFA is specifically involved face perception alone and plays a minimal role in within-category identification.
It is therefore difficult to say that the FFA is solely involved in face perception, with evidence for both its specialisation in face perception as well as evidence to suggest it having expertise effects with non-faces. Furthermore, TMS is currently not accurate enough to reach the FFA specifically meaning there is currently no way of activating this area of the brain to see for the effects. It is clear that not only further research is needed in this area but the technology used to measure the FFA can reliably determine if it plays a role in face perception/recognition.
Posterior superior temporal sulcus (pSTS)
The second neural structure proposed by Haxby et al. (2000) is the posterior superior temporal sulcus (pSTS). The pSTS is the sulcus that separates the superior temporal gyrus and the middle temporal gyrus and is involved in the perception of the changeable features of a face, including facial expressions as gaze (Haxby, Hoffman, & Gobbini, 2000) (Haxby, Hoffman, & Gobbini, 2002) (Baseler, Harris, Young, & Andrews, 2012). In trend with the FFA, there is also a range of evidence to support the pSTS’s role in sub-serving face perception. The pSTS plays a role in distinguishing the eye gaze as part of the overall face perception. Pelphrey, Morris and McCarthy (2005) used fMRI finding that both neurologically normal individuals as well as those with autism spectrum disorder (ASD) had an activated STS region in response to an actor’s eye gaze changing in a given scene. However, only the neurologically normal individuals showed an activated STS region when presented with an incongruent gaze shift (actors gaze shifts to an unexpected location other than the target object). Additional research finds that an area in the pSTS is activated when participants are shown sequences of faces which are varying in gaze and expression in faces that have a constant identity. It also finds that there is a functional connection between the FFA and the pSTS when participants viewed (Baseler, Harris, Young, & Andrews, 2012). This research therefore supports Haxby et al’s (2000) model that suggests that the pSTS deals with the changeable aspects of the face it also shows how the different structures interact to process faces.
Further to this, research shows that the pSTS is activated in participants who were shown stimuli of faces, animals and faceless animals when compared to stimuli of houses (Chao, Martin, & Haxby, 1999). A similar response to faces has also been found in macaques, with a face-selective region being found in the temporal cortex providing cross-species evidence for neural structures sub-serving face recognition (Desimone, 1991). The pSTS has also been shown to play a role in interpreting other dynamic information, such as biological motion (Grossman, Battelli, & Pascual-Leone, 2005). This study used TMS to provide a temporary disruption to the activity of the pSTS whilst measuring participant’s sensitivity to biological motion. It was found that partiticapnts were less sensitive to biological motion when the pSTS was stimulated, therefore suggesting that it plays a role in this process. Interpreting dynamic information plays an important role in face perception, with a similar inversion effect found with biological motion and face perception.
Occipital Face area (OFA)
The third neural structure proposed by Haxby et al. (2000) is the occipital face area (OFA). The OFA is generally involved in the construction of low-level representations of the physical features in a face, including the eyes, nose and mouth at a early stage of visual perception (Pitcher, Walsh, & Duchaine, 2011). However, recent research suggests that the OFA is also involved in higher level processing of the face, not just the representation of features (Ambrus, Windel, Burton, & Kovács, 2017). In this study TMS was used in conjunction with a sequential sorting face matching paradigm. When the OFA was stimulated, there was no differences found between trained and novel face identities when compared to a control group. This suggests that the OFA is not limited to low level processing of facial features but also has a role in identity and encoding of faces, therefore challenging Haxby et al’s model.
The OFA’s importance in face categorisation has been questioned however. There is research on patients with acquired prosopagnosia, documented with lesions to the OFA but still showing activity in the FFA. This shows how their face categorising ability still remains intact, suggesting the FFA does not work alone and Haxby et al’s hierarchical model is potentially organised in reverse. (Solomon-Harris, Mullin, & Steeves, 2013). TMS to the OFA in ‘normal’ individuals found no change in ability to categorise intact faces but an impairment in face identity discrimination. This research therefore suggests that the OFA is not necessarily needed for face categorisation (no lower level processing required) meaning that other areas of the brain can take over. It is important to mention however that face recognition still requires the OFA’s lower level processing, supporting Haxby et al’s model and showing that the OFA is infact important for face recognition.
It is also worth mentioning some of the neural structures in the extended system of Haxby et al’s (2000) model that sub-serve face perception. The Amygdala, a set of neurons that are located in the medial temporal lobe, is an area that is activated when responding to emotions. This is a key component of face recognition and the perception of facial expressions, it however responds the strongest to fear (Whalen, 1998) but is also responsive to other emotions, including positive ones (Todorov, 2012). Linking to this further, the amygdala also responds consistently to emotionally neutral faces (Todorov, 2012), thus showing that it may also play a role in other aspects of face recognition other than merely perceiving faces. Further to this the amygdala is also activated in response to observing bizarre faces in addition to novel faces which are atypical or unexpected (Todorov, 2012). The amygdala has also shown a greater response when compared to the pSTS a test where faces morph from one emotional expression to another (e.g. anger to happiness) as oppose to when the emotional expression of the face remains the same (Harris, Young, & Andrews, 2014). It has also been found that the amygdala is involved in the encoding of the identity of neutral-expression faces, with no evidence for its involvement in processing the direction of gaze (Mormann, et al., 2015). This research provides some support for the role of the amygdala in sub serving face perception, however, in line with Haxby et al’s model it remains a less important structure. Evidence supports the model suggesting that the FFA, OFA and pSTS are the main neural structures involved in face perception.
In summary, through fMRI and TMS research, it is clear that there are three main neural structures that sub-serve face perception. Haxby et al’s (2000) model demonstrates a hierarchical model suggesting that there are three core neural structures that deal with the main processes in face perception. There is evidence to support the involvement of theses three structures and their roles in sub-serving face perception. However, the role of the FFA still is open to debate, with research suggesting its involvement in within-category identification and expertise effects providing a counter argument to its specificity.