Biology Reference
In-Depth Information
topographically specific. In areas that preferentially process a particular stimulus feature (e.g.,
color or motion), increases in baseline activity were shown to be stronger during the
expectation of a preferred compared to a nonpreferred stimulus feature (Chawla et al., 2000).
Interestingly, patterns of neocortical activity evoked by real visual stimulus and its voluntary
imagination are similar (Mechelli et al., 2004). Even in early visual cortical areas, visual
mental imagery could evoke activity with precise visual field topography (retinotopy)
(Slotnick et al., 2005). Therefore, C-BG-Th-C loops that participate in involuntary and
voluntary attentional modulation of visual processing are overlapped. This is consistent with
known experimental data and general theories of attention that assume involuntary and
voluntary attentional processes converge on a common neural architecture (Hunt and
Kingstone, 2003; Kincade et al., 2005).
After the appearance of visual stimulus neural pattern representing this stimulus is
superimposed with the neuronal representation of imagined stimulus. Then contrasted
selection of total pattern is performed by C-BG-Th-C loops based on dopamine release in
response to real stimulus. If real and imagined objects have similar properties, initial cortical
representation of real stimulus becomes stronger, and its subsequent contrasted selection
requires smaller number of cycles of circulation in the C-BG-Th-C loops. Thus the perception
of the voluntary attended stimulus, which is similar to expected one, can be faster, in
comparison with its perception without attention. Since processing of visual information
occurs in the same neural networks, irrespective of a pathway of dopaminergic cell excitation,
dopamine-dependent effects caused by top-down activation of dopaminergic cells can
maintain and develop effects cased by their bottom-up excitation. Remarkably, the analysis of
experimental data also led to assumption that top-down processes could modulate involuntary
attention (Arnott et al., 2001).
In our model, mechanism of visual attention is built into the mechanism of visual
processing. It becomes apparent in selection of a stimulus (its attribute) for the best
processing and contrasted amplification of neuronal cortical representation of this stimulus
(attribute). The output BG signal acting on the thalamus performs the role of “attentional
filter” (Fig. 3). This signal depends on both the real stimulus, and traces of previous
processing of similar stimuli in diverse C-BG-Th-C loops.
Earlier it was proposed that the interaction between cortical and dopaminergic inputs to
striatal neurons and disinhibition of the SC via the striatum and SNr (i.e. via the direct
pathway through the BG) may underlie purposeful saccades (Hikosaka et al., 2000). Saccades
could be inhibited via the striatum, globus pallidus and SNr (i.e. via the indirect pathway
through the BG) (Hikosaka et al., 2000). As distinct from mentioned model, we assume that
in presence of dopamine, both direct and indirect pathways through the BG synergistically
disinhibit SC, whereas in the absence of dopamine, SC are synergistically inhibited via direct
and indirect pathways through the BG. Therefore, voluntary or involuntary evoked dopamine
release and subsequent disinhibition (through the BG) of SC projected onto oculomotor
structures, can promote focusing of eyes on attended stimulus (Fig. 3), and thus additionally
strengthen responses of thalamic and neocortical neurones.
According to known experimental data, visual brain areas separately and asynchronously
process different features of the same object (Zeki, 2001). However, vision produces unified
perceptual experience indicating the solution of the “binding problem”. Several lines of data
support and evolve the idea that attention enhances the binding (Neri, 2004; Paul and Schyns,
2003; Saenz et al., 1990). Without attention, binding is less effective (Reeves et al., 2005).
