Skip to main content

Interactions between prefrontal cortex and area V4 in attention

Final Report Summary - VISATT (Interactions between prefrontal cortex and area V4 in attention)

A visual scene is typically crowded containing many different objects that cannot all be processed simultaneously by our visual system. Attention is employed to select behaviorally relevant stimuli and facilitate further visual processing. Studies in humans and non human primates have implicated numerous brain areas in visual attention but how activity in each one of those influences activity in other areas of the network remains largely unknown. Current theories on visual attention have suggested that prefrontal and parietal cortical brain areas are part of an attentional control system. Accordingly, it has been suggested that these areas provide “top-down” signals that modulate sensory processing in early visual areas in favor of attended objects and/or locations that require further analysis by our visual system.
In the current project we test the hypothesis that the prefrontal cortex is a source of top-down signals to early visual areas. To this end we conduct extracellular recordings in the prefrontal cortex and area V4 simultaneously from non human primates engaged in a behavioral task in which attention is guided on the basis of object features. The aims of the study are to:
1) Establish the role of prefrontal cortex (including the frontal eye fields FEF) in feature-specific attention and how it influences processing in early visual areas.
2) Establish the functional and anatomical substrates that mediate attention in prefrontal cortex (including the FEF) and temporal lobe areas.
To this end, we trained two monkeys in a covert attention task and in a memory guided saccade task. Following training we recorded simultaneously from the FEF and V4 while the monkeys were executing the two tasks.
We tested the relative contributions of different functional classes of FEF neurons in attentional mechanisms, in order to explore how distinct classes of FEF neurons interact with area V4 during attention. Our results showed that only visual and visuomovement FEF neurons show enhanced responses in the attention task, whereas the responses of movement cells are not modulated by covert attention. Moreover, visual, but not movement or visuomovement cells, show enhanced gamma frequency (35-80 Hz) synchronization with neurons in V4 during attention. Within FEF, beta synchronization (15-35 Hz) is increased for movement cells during attention but is suppressed in a saccade task. These findings demonstrate that oculomotor and attentional processes are mediated by different mechanisms which can be dissociated at the cellular lever. Thus, they refute "motor theories of attention" which attribute a direct causal role of saccadic activity to attentional processes. Moreover, they suggest that highly specific connections between FEF visual neurons and V4 are responsible for the attentional modulation often reported in visual area V4. We also examined the contribution of fast spiking and regular spiking neurons (corresponding to putative interneurons and pyramidal neurons). Our data show a differential involvement of the two types of neurons in visual processing and in interneuronal interactions. These results together with those of current analysis focused on the effect of feature based attention on neuronal responses and synchrony are expected to shed more light on the cellular mechanisms of visual attention.