Recognizing Faces

Recognizing Faces

Introduction

The ability of humans to recognize faces is truly remarkable and extraordinary. Throughout the length of a day, human beings rely constantly on their brain's ability to identify faces. One cannot possibly imagine how confusing life would become if the brain failed to distinguish these faces. The cognitive process involved in recognizing faces has enthralled neurobiologists for a long time.

Discussion

A vast majority of neurobiologists and researchers believe that the brain processes and perceives human faces in a different manner as compared to other visual objects. For instance, researches have established that rotating the image of a human face by 180 degrees impedes the brain's ability to process the face much more than, say, inverting simple visual objects.

Moreover, recently conducted studies provide sufficient evidence to support the theory that the brain may utilize particular neurons that are specifically tuned to the identity of a particular person. The studies suggest that these neurons, which are present in the “fusiform” area of the face, become particularly active every time an individual encounters a face.

However, in an issue of 'Neuron' published on April 6, 2006, Maximilian Riesenhuber and his colleagues of Georgetown University Medical Center provided evidence to support the theory that the fusiform area of the face contains strongly inter-connected circuitry that aids the brain in the recognition of faces based mainly on the selective processing of the features and shapes of a particular face (Cell Press, 2006).

In this particular study, the researchers started by constructing a computational framework that represented the method through which their assumed neuronal circuitry might function. This model served to predict how the circuitry would facilitate the brain in processing and perceiving distinguishable features and shapes of the human face such as the nose, eyes, mouth etc.

The researchers discovered that their model was able to capture aspects of facial perception even when the circuitry that was integrated into their model did not code them explicitly. In order to illustrate that their model could explain how the neuronal circuitry could be programmed to other objects, the researchers tested its behavior when it came to processing images of cars.

They were able to gather that the model functioned as efficiently as in the case of faces and produced similar recognition characteristics. The research team experimented with their “shape-related” model by making volunteers identify images of faces whose characteristics could be reconstructed with the help of a computer program.

At the same time, the brains of the volunteers were scanned for specific patterns of activity in the fusiform area of the face with the help of functional magnetic resonance imaging (fMRI). The fMRI technique uses radio waves and magnetic fields to assess the flow of blood in various regions of the human brain due to activity. The researchers discovered that the findings from the fMRI agreed with those of the computational model.

As a result, the researchers reached the conclusion that they had successfully demonstrated how a computational execution of a physiologically possible neural model of face recognition and processing can quantitatively explain ...