 |
Member-Initiated Symposia2012 Symposia Distinguishing perceptual shifts from response biases Human visual cortex: from receptive fields to maps to clusters to perception Neuromodulation of Visual Perception Part-whole relationships in visual cortex Pulvinar and Vision: New insights into circuitry and function What does fMRI tell us about brain homologies?
|
Human visual cortex: from receptive fields to maps to clusters to perceptionFriday, May 11, 3:30 - 5:30 pm Organizer: Serge O. Dumoulin,
Experimental Psychology, Helmholtz Institute, Utrecht University, Utrecht,
Netherlands
Symposium Description
The organization of the visual system can be described at different spatial
scales. At the smallest scale, the receptive field is a property of individual
neurons and summarizes the region of the visual field where visual stimulation
elicits a response. These receptive fields are organized into visual field maps,
where neighboring neurons process neighboring parts of the visual field. Many
visual field maps exist, suggesting that every map contains a unique
representation of the visual field. This notion relates the visual field maps to
the idea of functional specialization, i.e. separate cortical regions are
involved in different processes. However, the computational processes within a
visual field map do not have to coincide with perceptual qualities. Indeed most
perceptual functions are associated with multiple visual field maps and even
multiple cortical regions. Visual field maps are organized in clusters that
share a similar eccentricity organization. This has lead to the proposal that
perceptual specializations correlate with clusters rather than individual maps.
This symposium will highlight current concepts of the organization of visual
cortex and their relation to perception and plasticity. The speakers have used a
variety of neuroimaging techniques with a focus on conventional functional
magnetic resonance imaging (fMRI) approaches, but also including high-resolution
fMRI, electroencephalography (EEG), subdural electrocorticography (ECoG), and
invasive electrophysiology. We will describe data-analysis techniques to
reconstruct receptive field properties of neural populations, and extend them to
visual field maps and clusters within human and macaque visual cortex. We
describe the way these receptive field properties vary within and across
different visual field maps. Next, we extend conventional stimulus-referred
notions of the receptive field to neural-referred properties, i.e.
cortico-cortical receptive fields that capture the information flow between
visual field maps. We also demonstrate techniques to reveal extra-classical
receptive field interactions similar to those seen in classical psychophysical
“surround suppression” in both S-cone and achromatic pathways. Next we will
consider the detailed organization within the foveal confluence, and model the
unique constraints that are associated with this organization. Furthermore, we
will consider how these neural properties change with the state of chronic
visual deprivation due to damage to the visual system, and in subjects with
severely altered visual input due to prism-adaptation.
The link between visual cortex’ organization, perception and plasticity
is a fundamental part of vision science. The symposium highlights these links at
various spatial scales. In addition, the attendees will gain insight into a
broad spectrum of state-of-the-art data-acquisition and data-analyses
neuroimaging techniques. Therefore, we believe that this symposium will be of
interest to a wide range of visual scientists, including students, researchers
and faculty. Presentations
Reconstructing human population receptive field
properties Serge O. Dumoulin,
Experimental Psychology, Helmholtz Institute,
Utrecht University, Utrecht, Netherlands, B.M. Harvey, Experimental Psychology,
Utrecht University, Netherlands We describe a method that
reconstructs population receptive field (pRF) properties in human visual cortex
using fMRI. This data-analysis technique is able to reconstruct several
properties of the underlying neural population, such as quantitative estimates
of the pRF position (maps), size as well as suppressive surrounds. PRF sizes
increase with increasing eccentricity and up the visual hierarchy. In the same
human subject, fMRI pRF measurements are comparable to those derived from
subdural electrocorticography (ECoG).
Furthermore, we describe a close relationship of pRF sizes to the
cortical magnification factor (CMF). Within V1, interhemisphere and subject
variations in CMF, pRF size, and V1 surface area are correlated. This suggests
a constant processing unit shared between humans. PRF sizes increase between
visual areas and with eccentricity, but when expressed in V1 cortical surface
area (i.e., cortico-cortical pRFs), they are constant across eccentricity in V2
and V3. Thus, V2, V3, and to some degree hV4, sample from a constant extent of
V1. This underscores the importance of V1 architecture as a reference frame for
subsequent processing stages and ultimately perception.
Cortico-cortical receptive field modeling using
functional magnetic resonance imaging (fMRI) Koen V. Haak,
Laboratory for Experimental Ophthalmology,
University Medical Center Groningen, University of Groningen, Groningen,
Netherlands, J. Winawer, Psychology, Stanford University; B.M. Harvey,
Experimental Psychology, Utrecht University; R. Renken, Laboratory for
Experimental Ophthalmology, University Medical Center Groningen, University of
Groningen, Netherlands; S.O. Dumoulin, Experimental Psychology, Utrecht
University, Netherlands; B.A. Wandell, Psychology, Stanford University; F.W.
Cornelissen, Laboratory for Experimental Ophthalmology, University Medical
Center Groningen, University of Groningen, Netherlands The traditional way to
study the properties of cortical visual neurons is to measure responses to
visually presented stimuli (stimulus-referred). A second way to understand
neuronal computations is to characterize responses in terms of the responses in
other parts of the nervous system (neural-referred).
A model that describes the relationship between responses in distinct
cortical locations is essential to clarify the network of cortical signaling
pathways. Just as a stimulus-referred receptive field predicts the neural
response as a function of the stimulus contrast, the neural-referred receptive
field predicts the neural response as a function of responses elsewhere in the
nervous system. When applied to two cortical regions, this function can be
called the population cortico-cortical receptive field (CCRF), and it can be
used to assess the fine-grained topographic connectivity between early visual
areas. Here, we model the CCRF as a Gaussian-weighted region on the cortical
surface and apply the model to fMRI data from both stimulus-driven and
resting-state experimental conditions in visual cortex to demonstrate that 1)
higher order visual areas such as V2, V3, hV4 and the LOC show an increase in
the CCRF size when sampling from the V1 surface, 2) the CCRF size of these
higher order visual areas is constant over the V1 surface, 3) the method traces
inherent properties of the visual cortical organization, 4) it probes the
direction of the flow of information.
Imaging extraclassical receptive fields in early
visual cortex Alex R. Wade,
Department of Psychology University of York,
Heslington, UK, B. Xiao, Department of Brain and Cognitive Sciences, MIT; J.
Rowland, Department of Art Practise, UC Berkeley Psychophysically, apparent
color and contrast can be modulated by long-range contextual effects. In this
talk I will describe a series of neuroimaging experiments that we have
performed to examine the effects of spatial context on color and contrast
signals in early human visual cortex.
Using fMRI we first show that regions of high contrast in the fovea
exert a long-range suppressive effect across visual cortex that is consistent
with a contrast gain control mechanism. This suppression is weaker when using
stimuli that excite the chromatic pathways and may occur relatively early in
the visual processing stream (Wade, Rowland, J Neurosci, 2010).
We then used high-resolution source imaged EEG to examine the effects of
context on V1 signals initiated in different chromatic and achromatic
precortical pathways (Xiao and Wade, J Vision, 2010). We found that contextual
effects similar to those seen in classical psychophysical ‘surround
suppression’ were present in both S-cone and achromatic pathways but that there
was little contextual interaction between these pathways - either in our
behavioral or in our neuroimaging paradigms.
Finally, we used fMRI multivariate pattern analysis techniques to
examine the presence of chromatic tuning in large extraclassical receptive
fields (ECRFs). We found that ECRFs have sufficient chromatic tuning to enable
classification based solely on information in suppressed voxels that are not
directly excited by the stimulus. In many cases, performance using ECRFs was as
accurate as that using voxels driven directly by the stimulus.
The human foveal confluence and high resolution
fMRI Mark M. Schira,
Neuroscience Research Australia (NeuRA), Sydney
& University of New South Wales, Sydney, Australia After remaining terra
incognita for 40 years, the detailed organization of the foveal confluence has
just recently been described in humans. I will present recent high resolution
mapping results in human subjects and introduce current concepts of its
organization in human and other primates (Schira et al., J. Nsci, 2009). I will
then introduce a new algebraic retino-cortical projection function that
accurately models the V1-V3 complex to the level of our knowledge about the
actual organization (Schira et al. PLoS Comp. Biol. 2010). Informed by this
model I will discuss important properties of foveal cortex in primates. These
considerations demonstrate that the observed organization though surprising at
first hand is in fact a good compromise with respect to cortical surface and
local isotropy, proving a potential explanation for this organization. Finally,
I will discuss recent advances such as multi-channel head coils and parallel
imaging which have greatly improved the quality and possibilities of MRI.
Unfortunately, most fMRI research is still essentially performed in the same
old 3 by 3 by 3 mm style - which was adequate when using a 1.5T scanner and a
birdcage head coil. I will introduce simple high resolution techniques that
allow fairly accurate estimates of the foveal organization in research subjects
within a reasonable timeframe of approximately 20 minutes, providing a powerful
tool for research of foveal vision.
Population receptive field measurements in
macaque visual cortex Stelios M. Smirnakis,
Departments of Neurosci. and Neurol., Baylor
Col. of Med., Houston, TX, G.A. Keliris, Max Planck Inst. For Biol.
Cybernetics, Tuebingen, Germany; Y. Shao, A. Papanikolaou, Max Planck Inst. For
Biol. Cybernetics, Tuebingen, Germany;
N.K. Logothetis, Max Planck Inst. For Biol. Cybernetics, Tuebingen,
Germany, Div. of Imaging Sci. and Biomed. Engin., Univ. of Manchester, United
Kingdom Visual receptive fields
have dynamic properties that may change with the conditions of visual
stimulation or with the state of chronic visual deprivation. We used 4.7 Tesla
functional magnetic resonance imaging (fMRI) to study the visual cortex of two
normal adult macaque monkeys and one macaque with binocular central retinal
lesions due to a form of juvenile macular degeneration (MD). FMRI experiments
were performed under light remifentanyl induced anesthesia (Logothetis et al.
Nat. Neurosci. 1999). Standard moving horizontal/vertical bar stimuli were
presented to the subjects and the population receptive field (pRF) method
(Dumoulin and Wandell, Neuroimage 2008) was used to measure retinotopic maps
and pRF sizes in early visual areas.
FMRI measurements of normal monkeys agree with published
electrophysiological results, with pRF sizes and electrophysiology measurements
showing similar trends. For the MD monkey, the size and location of the lesion
projection zone (LPZ) was consistent with the retinotopic projection of the
retinal lesion in early visual areas. No significant BOLD activity was seen
within the V1 LPZ, and the retinotopic organization of the non-deafferented V1
periphery was regular without distortion. Interestingly, area V5/MT of the MD
monkey showed more extensive activation than area V5/MT of control monkeys
which had part of their visual field obscured (artificial scotoma) to match the
scotoma of the MD monkey. V5/MT PRF sizes of the MD monkey were on average
smaller than controls. PRF estimation methods allow us to measure and follow in
vivo how the properties of visual areas change as a function of cortical
reorganization. Finally, if there is time, we will discuss a different method
of pRF estimation that yields additional information.
Functional plasticity in human parietal visual
field map clusters: Adapting to reversed visual input Alyssa A. Brewer,
Department of Cognitive Sciences University of
California, Irvine Irvine, CA, B. Barton, Department of Cognitive Sciences
University of California, Irvine; L. Lin, AcuFocus, Inc., Irvine Knowledge of the normal organization of visual field map clusters allows us to study potential reorganization within visual cortex under conditions that lead to a disruption of the normal visual inputs. Here we exploit the dynamic nature of visuomotor regions in posterior parietal cortex to examine cortical functional plasticity induced by a complete reversal of visual input in normal adult humans. We also investigate whether there is a difference in the timing or degree of a second adaptation to the left-right visual field reversal in adult humans after long-term recovery from the initial adaptation period. Subjects wore left-right reversing prism spectacles continuously for 14 days and then returned for a 4-day re-adaptation to the reversed visual field 1-9 months later. For each subject, we used population receptive field modeling fMRI methods to track the receptive field alterations within the occipital and parietal visual field map clusters across time points. The results from the first 14-day experimental period highlight a systematic and gradual shift of visual field coverage from contralateral space into ipsilateral space in parietal cortex throughout the prism adaptation period. After the second, 4-day experimental period, the data demonstrate a faster time course for both behavioral and cortical re-adaptation. These measurements in subjects with severely altered visual input allow us to identify the cortical regions subserving the dynamic remapping of cortical representations in response to altered visual perception and demonstrate that the changes in the maps produced by the initial long prism adaptation period persist over an extended time.
|