The University of Tennessee at Martin

The University of Tennessee - Martin

Search The University of Tennessee at Martin:


Dr. Ann Gathers

Assistant Professor of Biological Sciences

224 Brehm Hall



Visual Processing Development



Behavioral perceptual literature identifies three types of visual processing used by the normal population for the recognition of faces and objects:  featural or piece-meal processing, first-order relational processing, and second-order relational processing.  Featural processing refers to recognition based on identification of specific parts of an object or features of a face (i.e. a specific handle on a cup or a specific nose on a face).  Relational processing, also known as configural processing, involves first-order recognition via a general order or pattern of parts (i.e. eyes above nose and nose above mouth = face) and second-order recognition via spatial relationships among parts (i.e. distance between the eyes).  In general, it is thought that individuals with normal visual recognition rely primarily on featural processing for object recognition.  However, developmental studies indicate an increased reliance on second-order processing for face recognition with increasing age (Carey & Diamond, 1977; Mondloch et al., 2003). This developmental increase in second-order processing for faces is thought to be demonstrated by the face inversion effect (Yin, 1969). With development, inversion of stimuli results in decreased accuracy and decreased speed for processing faces but not objects.


Mr. BrainIn addition to behavioral studies regarding types of visual processing for faces and objects, some adult brain imaging studies point to modules of cortex specialized for visual categories. For example, Kanwisher et al. (1997) identified a face-preferential area (significantly stronger responses to face stimuli than other non-face stimuli), the fusiform face area (FFA), in a region of the right ventral temporal cortex called the anterior fusiform gyrus. Though not all adult studies support category-specific cortical modules (Joseph).


Thus, behavioral patterns indicate a change in face processing with age and imaging studies indicate possible face-specific cortex in the mature brain. Until recently, however, fewer studies have investigated how these processing changes are reflected in brain activation patterns and how the developing brain is organized for visual processing. Using functional magnetic resonance imaging (fMRI), Dr. Gathers and others at the University of Kentucky researched developmental changes in brain activation as children (5-11 years) and adults performed passive-viewing and matching tasks with face and non-face stimuli.


Overview of Results:


An fMRI study of adults and two groups of school-aged children (5-8 and 9-11 years) passively-viewing faces and non-face objects revealed an anterior shift in face processing regions. Adults and 9-11 year olds showed bilateral face-preferential activiation in the fusiform gyrus. The 5-8 year old children did not produce any face-preferential activation in the fusiform gyrus. However, more posterior regions in the right and left occipital gyri were recruited for face processing. This study demonstrated an anterior shift in face processing with age which may reflect an increased reliance on configural processing of faces with age. The overall findings indicated that visual processing undergoes neural changes in development. (NeuroReport, 2004)


In another fMRI study, the same age groups performed animal matching and face matching tasks. Stimuli were both upright and inverted. In adults and older children, face-matching performance (reaction time and accuracy) was more disrupted by inversion than in younger children. This developmental difference in performance was reflected in brain activation patterns. Adults and older children showed dissociation of activation with inverted face-matching recruiting posterior occipital cortex and upright face-matching recruiting more anterior regions. Younger children used posterior cortex for face stimuli at all orientations. These age-related changes were not evident in the animal-matching task. Thus, face processing undergoes both behavioral and neural changes in development. The age differences in neural recruitment for upright and inverted faces may reflect changes in visual processing with development. (Cognitive, Affective, and Behavioral Neuroscience, 2006)


A third fMRI study extended previous analyses (NeuroReport, 2004) to address large-scale functional neuroanatomical changes in face processing. All age groups activated a core network of occipito-temporal regions. However, bilateral anterior and left posterior fusiform regions exhibited fine-tuning (increased preference) for faces with age, whereas an additional left posterior fusiform region exhibited broad-tuning (decreased stimulus category preference)with age. A network of social brain regions, including limbic, frontal, and right posterior fusiform also emerged as face-preferential with development. These tuning changes in core face regions and the emergence of a social network with development may reflect potentially different perceptual and social face processing at different ages.







Joseph, J.E., Gathers, A.D., & Bhatt, R.S. (submitted).  Progressive and regressive developmental changes in neural substrates for face processing: Testing specific predictions of the Interactive Specialization account.

Joseph, J.E., Clark, J., Dekhtyar, O., Bhatt, R.S., Corbly, C.R., Gathers, A.D., and Zimmerman, R. (submitted). Development of face processing is driven by changes in functional brain connectivity and laterality.


Joseph, J.E., Johnson, N.F., Bhatt, R.S., Cerullo, M.A., and Gathers, A.D. (submitted). Can perceptual differentiation explain the development of expertise for faces?


Joseph, J.E., Gathers, A.D., Liu, X., et al. (2007).  Neural developmental changes in processing inverted faces. Cognitive, Affective, & Behavioral Neuroscience.    


Gathers, A.D. et al. (2004).  Developmental shifts in cortical loci of face and object processing.  NeuroReport, 15(10), 1549-1553.




Joseph, J. E., Dekhtyar, O., Liu, X., R., C. C., Bhatt, R. S., & Gathers, A. D. (November 2007). Functional brain development for face recognition involves a reduction in redundant pathways: Evidence from functional connectivity analyses. Paper presented at the Society for Neuroscience Annual Meeting, San Diego, CA


Gathers, A.D., et al. (2006). Anatomical Narrowing Without Categorical Specialization: Developmental Changes in Brain Activity During Face and Object Processing in 5-11 Year Olds and Adults.  16th Biennial Meeting of the International Society for
Developmental Neuroscience, Banff, Alberta, Canada.


Gathers, A.D., et al. (2004).  A developmental fMRI investigation of the neural correlates of object and face recognition. Organization for Human Brain Mapping, Budapest, Hungary.

Gathers, A.D., Bhatt, R., & Joseph, J.E. (2004). Face and object processing across development:  A structural similarity mechanism.  Cognitive Neuroscience Society, 11, Program No. D24, San Francisco, California.

Gathers, A.D. & Joseph, J.E. (2003). A functional MRI investigation of the neural correlates of object and face recognition in children.  Organization for Human_Brain Mapping, New York, New York.

Gathers, A.D. & Joseph, J.E. (2003). A comparison of visual shape processing in children and adults.  Society for Neuroscience Abstracts, 29, Program No. 890.10, New Orleans, Louisiana.