Why Do Some People Have Difficulty in Recognising Faces?

Prosopagnosia is the term used to describe an inability to recognise faces. This is a very specific impairment and there has been debate about whether such a specific impairment really exists. The alternative suggestion is that there is a more generic impairment in visual processing which partly manifests as an inability to recognise faces. In a PLOS One paper titled ‘Deficits in long-term recognition memory reveal dissociated subtypes in congential prosopagnosia’ (freely available here) the authors Stollhoff and colleagues investigate this further. They sample a group of people with Congenital Prosopagnosia (CP). In other words they have suffered with prosopagnosia since birth*(1). In the introduction the authors summarise previous research into prosopagnosia when we learn that researchers have tried to understand the phenomenon by comparing facial recognition with other important functions including object recognition, the processing of facial expressions and face detection. In this experiment the research team recruited 16 people with CP and 36 controls *(2).  Prosopagnosia was confirmed using a semi-structured interview examining difficulties with facial recognition as well as specialised assessments of visual processing. The researchers also wanted to exclude neurological and psychiatric illness *(3). Without going into too much detail the researchers used a number of tests to investigate visual processing. They investigated the ability of subjects to recognise famous faces, to access knowledge about visually presented information, to perceive visually presented information and to remember faces that had been presented one year previously.

Why Don’t Certain People Recognise Faces?

Comparing the subjects with CP to the control group the researchers found many differences. These differences may have been nothing more than associations. However the functions are similar in some ways to the function of recognising faces and so it can be reasonably assumed that there was some meaningful association between the phenomenon. Nevertheless even with these findings the researchers were unable to characterise the CP group as having a unique set of functional impairments. Instead they found that the subjects with CP had many different patterns of impairment. Some people had difficulty linking visual information to stored knowledge. Others had difficulty remembering previously presented faces. Yet other people had difficulty in processing the visual information. They concluded by suggesting that there were three types of congenital prosopagnosia (although these categories had been proposed prior to this study)

(1) Apperceptive Prosopagnosia: In this form, the person has difficulty in perceiving the face. The suggestion is that visual perception of faces is a two-stage process. In the first stage the person will retrieve visual information from the face. In this model, the face is properly considered as a3-dimensional object. In the second stage, this 3-dimensional information is then integrated into the visual perception of the face. Although the finer details of this model are not clear, the researchers note that in some cases of prosopagnosia, the person may inspect the face for a longer period of time and there may also cases in which it takes longer to integrate this information. Thus we can see that more general mechanisms of visual perception can potentially interfere with the visual perception of faces and it is important to be aware of these before concluding that the difficulty is more specifically related to faces.

(2) Associative Prosopagnosia: In this form of prosopagnosia, the person will be able to perceive the face correctly but will have difficulty in associating the face with stored knowledge.  The authors reference work by A Young on face recognition and presumably this is the same researcher who has proposed a model of facial recognition in the Delusional Misidentification Syndromes showing the potential relevance of this subject area to schizophrenia and other forms of psychosis. The researchers note that these two forms of prosopagnosia are unlikely to occur in an absolute form but rather that there should be an overlap of impairment in both stages of processing and this is what they find in the subjects in this study.

(3) Amnesic Prosopagnosia: In this form of prosopagnosiathe person has difficulty in retaining the association between the face and the associated semantic memory. In this study the subjects were tested at two points a year apart. They were asked to recall faces presented a year previously. The researchers found a significant group difference between the people with ‘congenital prosopagnosia’ and the control group. The people with prosopagnosia were significantly more likely to miss faces that were presented previously. Interestingly this didn’t hold for images of shoes that had been presented and so this was more convincing evidence for an impairment in visual processing specific to faces.

However even with these three categories there is still something of a black-box approach. We can see this by asking a few simple questions. What for instance is the process by which the spatial information about a face is integrated in the brain? Does it happen in the visual association cortices? How long does this take to occur? How many neurons are involved? Are certain types of face easier to remember? With such questions it is easy to see that the models are necessarily very simple and even with relatively sophisticated methodology it will likely be difficult to add more detail to these models.

I would interpret this as follows. There is utility in providing a label to investigate a phenomenon further. However in the case of prosopagnosia, given the complexity of the visual apparatus there are likely to be many different ways of interfering with the visual ‘machinery’ to effect the outcome of prosopagnosia. Once the process of face recognition is fully elucidated it will be possible to predict how pathological occurrences might occur within the system as well as relating actual clinical cases to this model. Until that stage is reached however the further classification offers an approximation to the ideal model. The ‘resolution’ of this model is determined by sophistication of the diagnostic apparatus becoming in effect a function of the resolution of the imaging equipment, reliability of the psychometric assessments, a function of the ‘models’ underlying the psychometric assessments as well as the clinical and imaging paradigms, the history taking characteristics, the limits of language (in the history taking) in describing visual phenomenon (in effect the ability of the mind to communicate phenomenological experiences in the perceptual domain) as just a few examples. This means that while the characterisation of the heterogeneity within this small population may represent an iterative process in the refinement of the model of CP (part of the normal science described by Thomas Kuhn (see here)), there is sufficient lax within the system to enable a different experiment using alternative assessment tools and a different population to entirely reframe the characteristics of the heterogenous CP population and offer a different model validated by that same CP population.

An Alternative Question

We can replace the question of ‘Why Don’t Certain People Recognise Faces?’ with the question of ‘What type of answer to this question would satisfy us?’. For instance at one level we might say that we want an answer that is simple to understand and remember. Someone in a research setting might want an answer that they can use to formulate an experimental paradigm to build upon this answer whilst another person might want an answer that they can challenge. For me the most satisfying answers would be those that offered for instance the clinician a pragmatic tool for use in a clinical assessment (interesting and similar is this recent BMJ paper) or for more abstract academic purposes a model which can be described with a specific language or framework which researchers in many fields of neuroscience could contribute to and which would enable the researchers to see where that specific model fits into the ‘bigger picture’. However this bigger picture itself is difficult to define but is needed ultimately to make sense of the huge global surge in clinical neuroscience and associated research that we are seeing.


*(1)  This is perhaps something of a misnomer as visual perception takes some time to develop. Phenomenon such as ocular dominance columns present just one example of why it takes sufficient training for the visual cortex to process images and transform them into complex perceptions. Thus a newborn baby would not be expected to recognise faces. Thus there should be a critical point in development when face recognition ‘develops’ and it is at that point that prosopagnosia can be meaningfully discussed.

*(2) I wasn’t clear on how the people with CP were recruited but I assume that as part of the research took place at the Institute of Human Genetics there was some process which led to people with prosopagnosia being assessed at that centre and then  recruited for this study. The only alternative I can think of offhand is that the study was marketed to the general population although I couldn’t see any such details in the methodology section at the end of the paper.

*(3) The terms however are sufficiently broad to be difficult to interpret. For instance many people would reach the threshold for adjustment reactions during the course of a year as just one example. Possibly what the researchers are referring to are the more common illnesses including schizophrenia, bipolar, unipolar depression and so on. However this can be argued to be semantics as we might expect with sufficient technological resources to be able to identify the cause of the prosopagnosia and to couch it in diagnostic terms meaning that if it isn’t currently a neurological or psychiatric disorder (e.g neurodevelopmental disorder) then with this additional information and due process it soon would be.

Index: An index of the site can be found here. The page contains links to all of the articles in the blog in chronological order. Twitter: You can follow ‘The Amazing World of Psychiatry’ Twitter by clicking on this link. Podcast: You can listen to this post on Odiogo by clicking on this link (there may be a small delay between publishing of the blog article and the availability of the podcast). It is available for a limited period. TAWOP Channel: You can follow the TAWOP Channel on YouTube by clicking on this link. Responses: If you have any comments, you can leave them below or alternatively e-mail justinmarley17@yahoo.co.uk. Disclaimer: The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s