Lecture 1: Introduction
→ Statements
1. our senses enable a complete and truthful perception of the world around us
2. Perception is a passive registration of the world around us
3. Perception is innate you do not have to learn it
But:
- Different pictures with artificial illusions
- Even if you know it is an illusion you still see it
- Also illusion in real world e.g. full moon illusion
- Picture with rabbit and duck in one → people that e.g. saw the duck when they saw
the picture for 1 sec. Cannot generate the other part of the picture (rabbit) without
actually seeing the picture (only people with photographic memory)
- Even though there is a passive registration we cannot always rely on it; we
sometimes have to work for it to understand what we are seeing
- We have different strings in our brain and vision for shape and color so we need
some time to attach them to each other in a recognition task (e.g. yellow triangle
and blue circle)
→ bottom-up vs. Top-down processing
- Perception is the result of:
- Stimulus input (bottom-up information) and knowledge and expectations (top-down
information)
→ speech perception and top-down information processing
1. segmentation
- spaces between words in a sentence facilitate reading
- Segmentation of spoken language→we use different frequencies (pauses); but there
are no spaces after every word like in written language
- We need knowledge of the language to understand the segments of spoken
language; if we do not know the language we cannot tell how many words someone
said and where the segments where
2. Phoneme restoration effect
- e.g. it was time to §ave my friend goodbye
- E.g. it was time to §ave my beard
- Because of our knowledge of the language we expect and hear the right word even
tho the phonemes are missing
- People hear wave and shave→restore the phoneme kind of automatically but also
subjectively only if they know the language of course
1
, Lecture 2: Neural basis of visual perception
Rods and Cones
- concentration of cones is highest in the center of the
fovea
- rod peak is the highest 15-20 degrees away from the
fovea
- 5-6 million cones
- 100-120 million rods
- 1 million ganglion cells (transmit info from eye to
brain)
- this means that many photoreceptors must converge into a single ganglion cells
(because there are way more photoreceptors than ganglion cells) - this is called
convergence
Convergence
- left picture: in the center of the fovea cones transmit info to
ganglion cells one-to-one
- right picture: convergence (not in the center of the fovea);
rods give many signals to only one ganglion cell
- Therefore cones give very clear info to the brain; brain an
differentiate from where the light is coming
- Info from rods are not that clear as cones; brain does not from where exactly the
light is coming because many rods are connected to one ganglion cell
- in the periphery of the fovea (rods)=less good with seeing details; Center of
fovea(cones)=good with details
- convergence has consequences for the details that we can see
2
,Light sensitivity (dark adaptation)
- because of convergence we can see better in the dark with rods than with cones
- because one cone only connects to one ganglion cell but many rods connect to one
ganglion cell so more signals at once are transmitted with rods in the dark
3
, Lateral inhibition (sideways suppression)
- if one cell is active it will suppress the activation of other cells next to it
- Cells need to be active in order to suppressor activate other cells
- higher number on picture means lighter; lower number means darker color
- E.g. cell sends 50 to the ganglion cell but gets 2 inhibitions from neighboring cells
(each x5), so in the end the ganglion cell receives 40 (50-5-5=40)
- contrast in the middle cells is larger (see figure 43-13) than cells in the mach band on
the outside
Center surround configuration
Hermann Grid
- has to do with Center surround configuration
- if you look directly at the white spot—these cells sent the signal directly to one
ganglion cell and hen to the brain
4
→ Statements
1. our senses enable a complete and truthful perception of the world around us
2. Perception is a passive registration of the world around us
3. Perception is innate you do not have to learn it
But:
- Different pictures with artificial illusions
- Even if you know it is an illusion you still see it
- Also illusion in real world e.g. full moon illusion
- Picture with rabbit and duck in one → people that e.g. saw the duck when they saw
the picture for 1 sec. Cannot generate the other part of the picture (rabbit) without
actually seeing the picture (only people with photographic memory)
- Even though there is a passive registration we cannot always rely on it; we
sometimes have to work for it to understand what we are seeing
- We have different strings in our brain and vision for shape and color so we need
some time to attach them to each other in a recognition task (e.g. yellow triangle
and blue circle)
→ bottom-up vs. Top-down processing
- Perception is the result of:
- Stimulus input (bottom-up information) and knowledge and expectations (top-down
information)
→ speech perception and top-down information processing
1. segmentation
- spaces between words in a sentence facilitate reading
- Segmentation of spoken language→we use different frequencies (pauses); but there
are no spaces after every word like in written language
- We need knowledge of the language to understand the segments of spoken
language; if we do not know the language we cannot tell how many words someone
said and where the segments where
2. Phoneme restoration effect
- e.g. it was time to §ave my friend goodbye
- E.g. it was time to §ave my beard
- Because of our knowledge of the language we expect and hear the right word even
tho the phonemes are missing
- People hear wave and shave→restore the phoneme kind of automatically but also
subjectively only if they know the language of course
1
, Lecture 2: Neural basis of visual perception
Rods and Cones
- concentration of cones is highest in the center of the
fovea
- rod peak is the highest 15-20 degrees away from the
fovea
- 5-6 million cones
- 100-120 million rods
- 1 million ganglion cells (transmit info from eye to
brain)
- this means that many photoreceptors must converge into a single ganglion cells
(because there are way more photoreceptors than ganglion cells) - this is called
convergence
Convergence
- left picture: in the center of the fovea cones transmit info to
ganglion cells one-to-one
- right picture: convergence (not in the center of the fovea);
rods give many signals to only one ganglion cell
- Therefore cones give very clear info to the brain; brain an
differentiate from where the light is coming
- Info from rods are not that clear as cones; brain does not from where exactly the
light is coming because many rods are connected to one ganglion cell
- in the periphery of the fovea (rods)=less good with seeing details; Center of
fovea(cones)=good with details
- convergence has consequences for the details that we can see
2
,Light sensitivity (dark adaptation)
- because of convergence we can see better in the dark with rods than with cones
- because one cone only connects to one ganglion cell but many rods connect to one
ganglion cell so more signals at once are transmitted with rods in the dark
3
, Lateral inhibition (sideways suppression)
- if one cell is active it will suppress the activation of other cells next to it
- Cells need to be active in order to suppressor activate other cells
- higher number on picture means lighter; lower number means darker color
- E.g. cell sends 50 to the ganglion cell but gets 2 inhibitions from neighboring cells
(each x5), so in the end the ganglion cell receives 40 (50-5-5=40)
- contrast in the middle cells is larger (see figure 43-13) than cells in the mach band on
the outside
Center surround configuration
Hermann Grid
- has to do with Center surround configuration
- if you look directly at the white spot—these cells sent the signal directly to one
ganglion cell and hen to the brain
4
