Psych 85 exam 1 material
Terms
undefined, object
copy deck
- Visual Agnosia
- an impairment of object recognition that is not attributable to a loss of general intellectual ability or to an impairment in such elementary visual perceptual processes as brightness, acuity, depth, and color.
- Associative Agnosia
-
Perception is not impaired but the person can not recognize the object.
-person can still draw the object - Prosopagnosia
- The inability to recognize faces after brain damage. -They can identify people other ways.
- LH experiment 1
- Only 62 percent correct for faces while 92 percent correct for objects. -study done by Farah, Klein, and Levinson
- LH experiment 2
- (faces X similar objects; testing if task difficulty theory) Controls: 67-69 eyeglass frames/85-87 faces LH 63 eyeglass frames/64 faces.
- Face-Inversion effect
- Inverting a face makes it much harder to recognize. -So, face recognition mechanism is specialized for processing upright faces.
- LH experiment 3
- Normals: 94% correct for upright vs. 82% correct for upside-down faces. LH 58% correct for upright vs. 72% correct for upside-down faces. -so his face processor is used (involuntarily) even though it is disadvantageous to do so
- CT
- Inability to learn new faces postmorbid. -could recognize premorbid images of Michael Jackson but not more recent ones
- CK
- Performs in the opposite direction of LH—good face recognition & low object recognition. -shows that face and object recog are separate processes and that they work in parallel
- Behaviorism
- Before the 1950s. B.F.Skinner main promoter. S/R associative experiments used to understand behavior with out delving into the mind (a “black box†to them). Eschewed any mental concepts
- Cognitive Revolution
- Several events in the 1950’s led to a rapid shift away from behaviorism and towards the development of cognitive science George Miller’s studies of verbal short-term memory (1956) Newell & Simon’s Logic Theorist (1956) Chomsky’s critique of Skinner’s Verbal Behavior (1959)
- Information theory
- Information can be thought of in a way entirely divorced from specific content or subject matter as simply a single decision between two equally plausible alternatives (bits). Shannon
- Newell and Simon's Logic theorist
- A program that solved logic proofs and show that a computer could think “intelligently†and use best-guess (heuristic) methods as we do.
- Chomsky's Linguistics
- Explained that there needed to be an inborn sense of grammatical structure and a framework that supports it for language to work. It was not just “picked up.â€
- Fundamental features of Cognitive Science
- Focus on mental representation (mind a machine for manipulating symbols, images, and other representations). Focus on computing (as a metaphor and a tool for simulation). Multidisciplinary approach. Grounding in classical philosophical problems (What is the nature of mind?).
- Course themes
- Humans solve hard problems (often with heuristics). The brain constrains the mind. Multiple levels of analysis (no one special level that supplies complete understanding.
- Three kinds of answers to mind/body problem
- Dualism, Materialism, Functionalism
- Dualism – problems
- How do the mind (non-physical) and body (physical) interact? If a computer or animal can do what a human can do does it also have this immaterial mind substance?
- Mind-Brain identity Materialism
- Mind and brain are the same thing. Relies on the model of reduction in the physical sciences. PROBLEMS: Seems to imply that only brains can have mental states. The brain is always changing it can never be in exactly the same state twice
- Eliminative materialism
- There are no such things as “mental phenomena†as we naively think of them. Once neuroscience has progressed sufficiently, we will realize the uselessness of these terms. ERROR: We still need descriptions at a cognitive level.
- Functionalism
- Mental states are defined by their function. Anything, a computer for example, could have a mind if its functions are similar to one.
- Mind levels of analysis
- Cognitive, Behavioral, and Neural
- Marr's levels of analysis
- Computational (What is the goal? What needs to be accomplished?), Algorithmic (representation of input/output and algorithm for transformation), and Implementation (physical representation) levels. (1980)
- Marr's levels evalutation
- Provides a clear and powerful framework for understanding the operations of cognitive systems. Prob: independence of the three levels is not clear. Marr's views too heavily constrained on the computer metaphor. It is not clear what should count as a distinct computation.
- Modularity of mind
- Phrenology linked. The brain is sectioned regionally for different cognitive functions
- Phrenology problems
- Concepts attributed to areas (bad psychology). Size of brain regions does not reflect the strength of psychological faculties. Brain shape does not affect skull shape. Non-objective (no experimentation)
- Cortical columns
-
Fundamental processing unit of the cortex
Extend vertically through the cortex
Vary from 300-600 microns in diameter
Similar across species whose brain volumes differ by factor of 1000
Evolution is associated with increased number of columns, not column size - Receptive Fields
- Sensory neurons represent particular aspects of the sensory environment. The particular aspects of the environment coded by a neuron are known as its receptive field.
- Area MT
-
Region in the middle temporal area
Sensitive to visual motion - Cortical maps
- The sensory cortices of the brain are laid out in a topographical manner
- Brain Plasticity
-
During development
experience determines how the brain gets wired up
During adulthood
experience results in changes in the brain’s wiring - How does learning happen?
-
The brain has molecular processes that change its processing based on experience
The strength of individual synapses changes due to experience
D. O. Hebb: neurons that fire together, wire together
Learning results in the growth of new synapses
These processes are the basis of learning and memory - Studying the human brain
-
How can we study the human brain?
Study patients with lesions or brain diseases
Brain imaging techniques - Virtual lesions
-
Transcranial magnetic stimulation (TMS)
Induces momentary electrical current in cortex - Recording from the human brain
-
Penfield stimulated the surface of the cortex during surgery
The brain doesn’t have pain receptors!
Stimulating different areas caused different sensations
Activity can be recorded using implanted electrodes
Patients undergoing surgery for epilepsy often have electrodes implanted for up to a week before surgery
Can record from these electrodes while performing a task - Brain imaging
-
Electroencephalogram (EEG):
measures electrical activity at the scalp. (good speed and cheap; bad spatial resolution)
Functional MRI:
Measures blood flow
in the brain. (good spatial and decent time resolution; bad price, size, and magnetic conflicts) - Multimodal imaging
-
New techniques allow the integration of fMRI and MEG/EEG results
Combines the advantages of each technique - Retinotopic organization
-
The visual cortex is laid out in an orderly topographic fashion
central vision gets more space than peripheral vision - Visual illusions
-
(Backus & Oruç, 2005):
the visual system constructs percepts that represent that which would most likely have evoked the same pattern of sensory neural activity
This can lead to visual illusions - The Generic View Principle
-
The visual system assumes that the current image is one that is likely to have arisen
It’s not an extremely rare accident of viewpoint
This heuristic allows many other assumptions to be made
However, it will sometimes fail
If it did not make this assumption, it would be stuck deciding between many different possible interpretations of the retinal image - monocular cues to depth
- Interposition, Relative size, Linear perspective.
- Binocular cues to depth
- Binocular disparity (information coming to each eye is different)
- View-based theory
-
Object is represented in terms of low-level features
Memory stores representations of each object from many different viewpoints
A new object is recognized by matching it to all of the existing views
If no match, then a new view is created in memory - Evidence for view-based theory
-
Object recognition is viewpoint-dependent
People are better at recognizing objects from particular viewpoints
Neural evidence
Some neurons fire for specific objects in a viewpoint-specific manner - Structural description theory
-
Memory stores representations of the 3-D structure of the object
Objects are represented in terms of a set of atomic geometric features, called geons
The stimulus is represented in terms of geons, and the list of geons is matched against memory - Evidence for structural description theory
-
Priming is invariant to rotation only when the same parts are visible
the same geons are represented
Part versus line priming results
Biederman & Cooper (1992) - Levels of object description
-
“animal†- superordinate level
“dog†- basic level
“Jack Russell Terrier†-
subordinate level
Experts in a domain tend
to use the subordinate level - Evidence for expert recognition view of face recognition
-
The fusiform “face area†is also activated by other categories of expert stimuli
Inversion effects for other expert stimuli, AND the inversion effect may be due to expertise. - Spatial attention
-
Humans can selectively attend to specific areas of space
like a zoom-able spotlight
Stimuli that appear within the attentional spotlight are processed faster than those outside the focus of attention.
Attention speeds processing. - The neural basis of spatial attention
-
The parietal lobe is necessary for spatial attention
Damage to the right parietal lobe in humans leads to neglect of the left side of space
An attentional problem rather than a visual problem per se
Neurons in the parietal cortex code for different parts of space. - “What†versus “Where†pathways
-
Ventral (Temporal) pathway:
Object identification
Dorsal (Parietal)pathway:
Space/Action - Darwin’s dangerous idea
-
Natural selection always occurs if you have:
1. Variation in traits
2. That lead to variation in fitness (number
of offspring left in next generation)
3. Correlation in traits between parents and
offspring - Adaptationism
-
Structure reflects
function
⬢ natural selection is a
feedback process
that "chooses"
among alternative
designs on the basis
of how well they
function. It is a hillclimbing
process - Evolutionary psychology
-
ultimate +
proximate:
– use ultimate fitness reasons (adaptationist
reasoning) to explain the design of proximate
mechanisms - EEA
-
Environment of
Evolutionary
Adaptedness - Evolutionary disequilibrium
-
Mismatch between
current environment
and EEA
⬢ Examples:
⬢ fears and phobias
⬢ Fat and salt - Human kin detection
-
Biologist William Hamilton showed that there is
an evolutionary advantage to helping kin
⬢ For similar reasons (probability of gene
sharing), mating with close kin is
disadvantageous (recessive deleterious alleles) - The Westermarck mechanism
-
Edward Westermarck, a Finnish
anthropologist, speculated that
children raised together may
experience diminished sexual
attraction towards each other
The “Westermarck mechanism†is
now seen as a mechanism for
identifying kin
It uses cohabitation as a cue,
because shared genes cannot be
directly detected - Transduction of sound
-
Sound is transduced into neural signals by hair cells in the cochlea
Bending in one direction excites the cell
Bending in the other direction inhibits the cell - Hair cells and transduction
-
The hair cells fire in response to vibrations
100 picometers in a hair cell is equivalent to a motion of 10 mm at the top of the Eiffel Tower
Most signal comes from the ~3500 inner hair cells
Outer hair cells (~ 12,000) alter the vibration of the basilar membrane
enhances selectivity - Fourier analysis in the cochlea
-
The cochlea decomposes the auditory signal by frequency
Different areas along the cochlea are sensitive to different frequencies
Activity of the receptors in each region codes for the amount of sound at that particular frequency
Complex sounds are made up of a number of different frequencies
The lowest is known as the fundamental frequency
Other frequencies are known as harmonics - The auditory pathway
-
From the cochlea, the signal travels via the auditory nerve to the brainstem
Most of the signals cross over to the contralateral hemisphere in the brainstem
After traveling to the inferior colliculus and MGN of the thalamus, the signal then arrives at the auditory cortex. - Place coding of pitch
-
The basilar membrane is laid out in a tonotopic manner
Vibrations in different parts of the membrane reflect different pitches
Hair cells at each point on the membrane code for specific frequencies based on firing to "best liked" frequencies. - Timing code for pitch
-
Humans can hear sounds up to ~20,000 Hz
Naïve proposal:
A neuron fires for every cycle
Problem:
A neuron can only fire about 500 times a second, due to its refractory period
Solution
The volley principle - The volley principle
-
Many neurons work together to code each cycle of the sound
Each particular neuron fires at the peak of a cycle, but only does so every few cycles
Evidence for volley principle
Neurons will phase-lock to a stimulus. - Localizing sounds
-
Two sources of information:
Interaural time/phase differences
Interaural intensity differences - Interaural time differences
-
Eccentric sounds take slightly longer to arrive at the far ear (< 1 millisecond)
The phase of the sound wave will be slightly different between the two ears
More important at low frequencies (< 1500 Hz) - Interaural level differences
-
The head casts an auditory shadow on the far ear
More important for high frequencies
To cast a shadow, an object must be large in comparison to the wavelength of the signal. - Neural mechanisms for sound localization
-
There are neurons in the midbrain and auditory cortex that respond to timing differences
Other neurons have receptive fields for different regions of space
Integrate timing and level information - Localizing sounds in elevation
-
How do we localize the elevation of a sound source?
The pinna (outer ear) acts like an acoustic antenna
It resonates differently depending upon the elevation of the sound
It differently reflects sounds depending upon the elevation
Pinnae differ greatly between individuals
We learn to use our own ears to localize in elevation - Cross-modal perception
-
Visual and auditory signals can affect perception in the other modality
Vision affects audition
McGurk effect
Audition affects vision
Shams et al. illusion - Synesthesia
-
“joined sensationsâ€
the ability to experience sensation in one sense as a perception in a different sense
hearing colors
tasting shapes
or across different aspects of a single modality
seeing letters as colored
involuntary and unsuppressible
occurs in ~ 1/25,000 people
more often in females (> 3:1)
runs in families - Visual search
-
Searching for a simple feature (e.g., blue vs. red) is “pre-attentiveâ€
Difficulty does not increase as the number of distractors increases
Searching for a complex feature (e.g., 2 vs 5) requires attention
Difficulty increases as the number of distractors goes up - Modularity and cross-modal perception
-
The strong modularity of mind thesis proposed that modules are “informationally encapsulatedâ€
Cross-modal perception shows that modalities can affect each other at very low levels
Suggests that modularity is false in its strongest form.