Topics in Behavioural Neuroscience
Questions lecture 1: Function and organisation of the visual
system
1) Explain how the response properties of simple and complex cells and
how they can emerge from combining the input from neurons in the
lateral geniculate nucleus.
Simple cells in V1 have distinct excitatory and inhibitory regions and
respond best to bars or edges of a specific orientation and position. These
properties emerge by pooling inputs from multiple LGN neurons whose
center-surround receptive fields are aligned linearly. When a stimulus
aligns with these receptive fields, the simple cell is maximally activated.
Complex cells also respond to specific orientations but are less sensitive to
precise position and lack distinct on/off regions. They integrate input from
multiple simple cells with similar orientation preferences but different
spatial positions. This allows them to detect oriented edges regardless of
exact location and often in motion. Thus, both cell types emerge
hierarchically by combining LGN input through increasingly complex
integration.
2) Explain the concept of multiple hierarchical streams in the visual
system, all the way from retina to temporal/parietal cortex.
Visual processing follows a hierarchical and parallel organization. It begins
in the retina, where photoreceptors transmit signals to retinal ganglion
cells, whose axons form the optic nerve. These project to the LGN in the
thalamus and then to V1. From there, two main processing streams
diverge:
The ventral stream (V1 → V2 → V4 → inferior temporal cortex) processes
the "what" information: object identity, shape, and colour.
The dorsal stream (V1 → V2 → MT → parietal cortex) processes the
"where/how" information: motion, spatial relationships, and visually guided
actions.
Each level builds on the previous, increasing in feature complexity and
receptive field size, forming the basis for high-level visual perception and
behaviour.
3) Describe the hierarchical processing that results in the perception of
shape.
Shape perception builds hierarchically through progressive stages in the
visual cortex. In V1, neurons detect basic visual features like edges and
orientations. V2 begins to integrate these into contours and borders. In V4,
neurons are tuned to more complex combinations of curves and angles,
enabling the perception of shape components. Finally, in the
inferotemporal cortex (IT), neurons respond selectively to whole objects or
complex shapes, often in a position- and size-invariant manner. This
hierarchical structure enables the transformation of simple visual inputs
into coherent shape perception.
4) Explain the configural superiority effect and its neural basis
,The configural superiority effect describes the phenomenon where a group
of elements forming a meaningful configuration is perceived more easily
than the individual elements alone. For example, adding corners to line
segments can create arrows or shapes that are more discriminable than
the same lines alone. Neuroimaging studies show that this effect involves
higher-level visual areas like the lateral occipital complex (LOC), which
show stronger activation to the whole configuration compared to individual
parts. This suggests that configural perception relies on holistic processing
in higher-tier visual areas, beyond early feature detection in V1.
5) Describe the progression of local to global processing in the context
of face, colour, and motion processing
Visual processing shifts from local to global representations as information
flows through the visual hierarchy.
For faces, early areas (V1/V2) detect features like edges or local contrast,
while areas like the fusiform face area (FFA) integrate these into holistic
face representations.
In color processing, V1 and V2 respond to local wavelength contrasts,
while area V4 integrates this into perceptual color constancy and color-
based object features.
For motion, V1 detects local directional movement, while MT (V5)
integrates these into coherent motion patterns, and MST handles global
motion like optic flow.
This progression reflects the increasing complexity and integration of
visual information across cortical levels.
6) Describe and discuss the dimensional framework and how it relates
to category-, colour-, and motion-selective regions
The dimensional framework proposes that the brain encodes perceptual
features (like shape, color, and motion) as dimensions within a
representational space. These dimensions are processed in both
specialized and distributed ways across the cortex.
Color is primarily processed in areas like V4 but also informs object and
category perception.
Motion is processed in areas MT/MST and contributes to action recognition.
Category-selective regions (e.g., FFA for faces, PPA for places) are shaped
not only by semantic content but also by low-level visual dimensions such
as curvature, texture, or spatial frequency.
Thus, the dimensional framework shows how perceptual and conceptual
properties are intertwined across visual regions.
7) Describe and discuss the role of lateral and feedback connections in
visual processing, giving two example studies
, Lateral and feedback connections allow the brain to refine visual
perception through context and top-down influences. Lateral connections
within an area (e.g., V1) help integrate nearby visual information, while
feedback from higher to lower areas modulates processing based on
expectations or attention.
One study (e.g., Gilbert & Wiesel) showed that collinear facilitation in V1
depends on lateral connections, helping in contour integration.
Another study using fMRI and masking (e.g., Lamme et al.) showed that
conscious perception is associated with feedback from higher areas to V1,
as masked stimuli that bypass feedback remain unconscious.
These connections are essential for integrating features, predicting
sensory input, and enabling conscious awareness.
Questions lecture 2: Plasticity
1) In visual cortex we find ocular dominance columns and orientation
columns. Explain one line of research (methodology & results) for
each of these columnar structures that shows that their properties
are strongly altered by experience.
The plasticity of ocular dominance columns has been demonstrated
through studies of monocular deprivation in kittens by Hubel and Wiesel.
Using single-cell recordings, they found that closing one eye during a
critical developmental period led neurons in the primary visual cortex to
respond predominantly to the open eye, significantly reducing cortical
representation of the deprived eye. This showed that ocular dominance is
shaped by early visual experience. Similarly, the plasticity of orientation
columns was revealed in an experiment by Blakemore and Cooper, where
kittens were raised in environments with only vertical or horizontal lines.
Neurons in their visual cortex became selectively tuned to the exposed
orientations, and responses to non-experienced orientations were
diminished. These findings demonstrate that both ocular dominance and
orientation selectivity in V1 are highly experience-dependent during
development.
2) Describe the evidence in monkeys that exposure to faces is a
necessary condition for face-specific behavior and neural responses.
Compare these effects to the long-term visual deficits in face
perception seen in patients that were born with congenital cataract
and for who vision was restored after a few months.
Evidence from monkeys shows that early visual exposure is necessary for
normal face processing. In a study by Sugita, macaques were raised
without seeing faces for several months. When finally exposed, they
initially lacked face-selective behavior and neural responses in face-
processing areas, though some abilities recovered with later experience,
indicating a sensitive period. In humans, individuals born with congenital
cataracts and treated after a few months show persistent deficits in face
perception, including poor recognition and reduced activation in face-
selective brain regions like the fusiform face area. Unlike monkeys, these
deficits often do not fully recover, suggesting that the critical period for
Questions lecture 1: Function and organisation of the visual
system
1) Explain how the response properties of simple and complex cells and
how they can emerge from combining the input from neurons in the
lateral geniculate nucleus.
Simple cells in V1 have distinct excitatory and inhibitory regions and
respond best to bars or edges of a specific orientation and position. These
properties emerge by pooling inputs from multiple LGN neurons whose
center-surround receptive fields are aligned linearly. When a stimulus
aligns with these receptive fields, the simple cell is maximally activated.
Complex cells also respond to specific orientations but are less sensitive to
precise position and lack distinct on/off regions. They integrate input from
multiple simple cells with similar orientation preferences but different
spatial positions. This allows them to detect oriented edges regardless of
exact location and often in motion. Thus, both cell types emerge
hierarchically by combining LGN input through increasingly complex
integration.
2) Explain the concept of multiple hierarchical streams in the visual
system, all the way from retina to temporal/parietal cortex.
Visual processing follows a hierarchical and parallel organization. It begins
in the retina, where photoreceptors transmit signals to retinal ganglion
cells, whose axons form the optic nerve. These project to the LGN in the
thalamus and then to V1. From there, two main processing streams
diverge:
The ventral stream (V1 → V2 → V4 → inferior temporal cortex) processes
the "what" information: object identity, shape, and colour.
The dorsal stream (V1 → V2 → MT → parietal cortex) processes the
"where/how" information: motion, spatial relationships, and visually guided
actions.
Each level builds on the previous, increasing in feature complexity and
receptive field size, forming the basis for high-level visual perception and
behaviour.
3) Describe the hierarchical processing that results in the perception of
shape.
Shape perception builds hierarchically through progressive stages in the
visual cortex. In V1, neurons detect basic visual features like edges and
orientations. V2 begins to integrate these into contours and borders. In V4,
neurons are tuned to more complex combinations of curves and angles,
enabling the perception of shape components. Finally, in the
inferotemporal cortex (IT), neurons respond selectively to whole objects or
complex shapes, often in a position- and size-invariant manner. This
hierarchical structure enables the transformation of simple visual inputs
into coherent shape perception.
4) Explain the configural superiority effect and its neural basis
,The configural superiority effect describes the phenomenon where a group
of elements forming a meaningful configuration is perceived more easily
than the individual elements alone. For example, adding corners to line
segments can create arrows or shapes that are more discriminable than
the same lines alone. Neuroimaging studies show that this effect involves
higher-level visual areas like the lateral occipital complex (LOC), which
show stronger activation to the whole configuration compared to individual
parts. This suggests that configural perception relies on holistic processing
in higher-tier visual areas, beyond early feature detection in V1.
5) Describe the progression of local to global processing in the context
of face, colour, and motion processing
Visual processing shifts from local to global representations as information
flows through the visual hierarchy.
For faces, early areas (V1/V2) detect features like edges or local contrast,
while areas like the fusiform face area (FFA) integrate these into holistic
face representations.
In color processing, V1 and V2 respond to local wavelength contrasts,
while area V4 integrates this into perceptual color constancy and color-
based object features.
For motion, V1 detects local directional movement, while MT (V5)
integrates these into coherent motion patterns, and MST handles global
motion like optic flow.
This progression reflects the increasing complexity and integration of
visual information across cortical levels.
6) Describe and discuss the dimensional framework and how it relates
to category-, colour-, and motion-selective regions
The dimensional framework proposes that the brain encodes perceptual
features (like shape, color, and motion) as dimensions within a
representational space. These dimensions are processed in both
specialized and distributed ways across the cortex.
Color is primarily processed in areas like V4 but also informs object and
category perception.
Motion is processed in areas MT/MST and contributes to action recognition.
Category-selective regions (e.g., FFA for faces, PPA for places) are shaped
not only by semantic content but also by low-level visual dimensions such
as curvature, texture, or spatial frequency.
Thus, the dimensional framework shows how perceptual and conceptual
properties are intertwined across visual regions.
7) Describe and discuss the role of lateral and feedback connections in
visual processing, giving two example studies
, Lateral and feedback connections allow the brain to refine visual
perception through context and top-down influences. Lateral connections
within an area (e.g., V1) help integrate nearby visual information, while
feedback from higher to lower areas modulates processing based on
expectations or attention.
One study (e.g., Gilbert & Wiesel) showed that collinear facilitation in V1
depends on lateral connections, helping in contour integration.
Another study using fMRI and masking (e.g., Lamme et al.) showed that
conscious perception is associated with feedback from higher areas to V1,
as masked stimuli that bypass feedback remain unconscious.
These connections are essential for integrating features, predicting
sensory input, and enabling conscious awareness.
Questions lecture 2: Plasticity
1) In visual cortex we find ocular dominance columns and orientation
columns. Explain one line of research (methodology & results) for
each of these columnar structures that shows that their properties
are strongly altered by experience.
The plasticity of ocular dominance columns has been demonstrated
through studies of monocular deprivation in kittens by Hubel and Wiesel.
Using single-cell recordings, they found that closing one eye during a
critical developmental period led neurons in the primary visual cortex to
respond predominantly to the open eye, significantly reducing cortical
representation of the deprived eye. This showed that ocular dominance is
shaped by early visual experience. Similarly, the plasticity of orientation
columns was revealed in an experiment by Blakemore and Cooper, where
kittens were raised in environments with only vertical or horizontal lines.
Neurons in their visual cortex became selectively tuned to the exposed
orientations, and responses to non-experienced orientations were
diminished. These findings demonstrate that both ocular dominance and
orientation selectivity in V1 are highly experience-dependent during
development.
2) Describe the evidence in monkeys that exposure to faces is a
necessary condition for face-specific behavior and neural responses.
Compare these effects to the long-term visual deficits in face
perception seen in patients that were born with congenital cataract
and for who vision was restored after a few months.
Evidence from monkeys shows that early visual exposure is necessary for
normal face processing. In a study by Sugita, macaques were raised
without seeing faces for several months. When finally exposed, they
initially lacked face-selective behavior and neural responses in face-
processing areas, though some abilities recovered with later experience,
indicating a sensitive period. In humans, individuals born with congenital
cataracts and treated after a few months show persistent deficits in face
perception, including poor recognition and reduced activation in face-
selective brain regions like the fusiform face area. Unlike monkeys, these
deficits often do not fully recover, suggesting that the critical period for
