lecture 14 - perceptual learning and individual differences Flashcards
(25 cards)
THE NATURE OF LEARNING - OVERVIEW
- Learning refers to the processes by which experiences
change our nervous system and hence our behaviour - We refer to these changes as memories
- Experiences are not “stored”, rather they change the way
we perform, perceive, think, and plan by physically
changing the structure of the nervous system - We must be able to learn in order to adapt our
behaviors to our changing environment. Or another
way…change in the environment can produce learning
learning ‘types’ according to Carlson
stimulus —> perceptual system (changes in neural circuit that detects a particular stimulus) perceptual learning — S-R learning –> motor system/learning (changes in neural circuit that controls a particular behaviour) —> response
s-r learning = stimulus response learning where the association between a given sensory partnering and a given response might increase, the more thats done the better you get at it or there might just be learning that only occurs on the motor side to the primary motor cortex on areas associated with it, so in that case its not your ability to process the stimulus at a sensory side is getting better but what’s happening is that you ability to actually move your muscles faster or better or with more accuracy is what’s being changed .
perceptual learning
‘Changes in neural circuits that
detect and discriminate a
particular stimulus leading to a
measurable behavioural change’
McG Version
Perceptual learning is NOT limited to early experience/critical
periods.
* It is a type of procedural learning that occurs continually in
everyday life
- Learning provides us with the ability to perform an
appropriate behavior in an appropriate situation. - The first part of learning involves learning to perceive
particular stimuli, or to appreciate differences between
stimuli. - Perceptual learning involves learning about things, not what
to do when they are present - Involves learning to recognize new stimuli or to recognize
changes in familiar stimuli - LOCATION of change correlates with WHAT is being learned
PERCEPTUAL LEARNING IN THE REAL WORLD… - radiology - Seitz, 2017
xray images
radiologist is able to tell the differences and issues as this ability has been honed over a long time after looking at 100s of X-rays so there is a lot of training in this type of recognition
also find it in expert facial profilers which are used for CCTV for example
a lot of what they go through is akin to a psychology perceptual learning task
a guy made sure people who do finger print recognition were trained better and was applying the same kind of standards and stance and ideas that we would do in the lab to this very real world and important issue and found by applying these aspects of psychological learning theory and treating it more explicitly as a perceptual learning task the group he was training could do a far better job on any novel fingerprint thrown at them
different regions, different questions
diagram of brain in notes
there are very complex networks involved in perceptual learning eg changes within the primary visual cortex - a lot of the perceptual learning world probably because it evolved out of the psychophysical world is based on visual cortex
the primary sensory modality for perceptual learning in visual perceptual learning
V1 has a map of visual space - there is some point to point topography of where neurones are spawned on where that particular photoreceptor is in visual space and there are also feature detectors that are sensitive to orientation within that particular region of space
investigating perceptual learning
Most simple perceptual learning effects are specific to the trained task or
stimulus. - theres not a lot of generalisation of these skills to different kinds of tasks and that makes them kind of useful from a visual cortex point of view because it means if we know already quite a lot about the response properties of V1 we know those maps, the space within there. if things aren’t generalising and you’re only getting better at particular orientations or particular areas of space then it probably tells us the locus of change or at least on involvement is going to be in primary visual cortex because thats the first cortical representation of all those things.
vernier acuity task -
Configuration of two or three vertical
lines (or dots) is presented. The subject is
asked to indicate whether the lines (or
the dots) are aligned. - its gets harder or easier depending on your performance. Can improve
VISUAL ACUITY
performance decreases when they change the orientation of offset from vertical to horizontal or vice versa
INVESTIGATING PERCEPTUAL LEARNING
two Parts: Respond ‘T’ (as shown in the
figure) or an ‘L’ (ensures fixation at the
centre).
Then indicate whether the orientation of
the target (the three elements with
orientation that differs from that of the rest
of the elements) is vertical (as shown in
the figure) or horizontal.
its a way to make sure people are always fixating because its a two part task you fixate and if you fixate it means if stimuli and presented they are always being presented at the same point in the visual field - you need to keep foveating to be able to say if it was a T or an L in the centre and while your doing that the other thing comes up so the experiment has that built in control to make sure that people can’t then saccade or foveate over to the other stimulus
Stimulus moved to new
location in visual field your performance goes back to almost what it was when you started
learning to do the task in a new position does not wipe the neural correlates of what you learned to do in the old position - so you have visual filed specific learning abilities
INVESTIGATING PERCEPTUAL LEARNING
look at graph in notes
between day learning - there is consolidation overnight
within day learning - no consolidation overnight
within and between day learning - go back to zero learning overnight
within day learning no consolidation - best design for best possible paradigm to show the biggest effects in your experiments
- Different kinds of task and stimuli will show different time courses
of learning. - As will the design of the experiment: feedback, reward, sleep etc
all play a part
PERCEPTUAL LEARNING – INDIV. DIFFS
While the overall trend of
learning in a group may be
similar, there is VARIABILITY in
the starting (or baseline) levels
of performance
* If we want to understand
perceptual learning, we need to
appreciate these individual
differences that occur
NATURALLY in our participant
population.
* What neural mechanisms might
underlie them? - GABA
MECHANISMS UNDERLYING INDIV
DIFFS IN BASELINE PERCEPTION?
- CONTROVERSIAL topic – but many
researchers agree that the neurotransmitter
GABA (gamma-amino-butyric-acid) plays a
role. - GABA is an INHIBITORY neurotransmitter,
primarily found in cortical
INTERNEURONS. - It can ‘shape’ the size – and SELECTIVITY –
of receptive fields.
if you infuse a GABAergic blocker a chemical essentially into the but of cortex you have already marked that you know corresponds to that area then you get an expansion of that region. bits of the cortex that were surrounding that that would not respond to that before, do now respond to stimulation within here
TACTILE FREQUENCY
DISCRIMINATION: INDIVIDUAL
DIFFERENCES - there are frequency specific features within the somatosensory cortex
frequency is an abstract feature of touch eg a frequency with which something would be pressing down onto your fingers
- Neurons in primary
somatosensory cortex (S1)
respond to vibrotactile
stimuli by firing in phase
with each cycle of the
vibration. - GABAergic inhibition
shapes neural response
properties in S1 (Dykes et
al., 1984; Juliano et al., 1989).
an electrode records the group of neurones that are identified as essentially mapping digit two and then a frequency stimulus - a vibro tactile stimulus - is applied to that with different frequencies and what the initial investigators found was that the amount of firing varies with the exact patterning of the frequency of that stimulus contacting the skin
if you are recording from that area within the sensory cortex you are able to read out with great temporal fidelity what actually happening at the finger because you know the coding strategy and it has the ability to have good phase locking
the phase locking means that very time it is high firing the data lookalike a peaks when its at the maximum displacement on the finger itself on the stimulus so the stimulus and firing go up and down and as you increase the speed or make it go slower at least within the flutter period between about two to 40 Hz you’ll get that kind of pattering - graph in notes
if you disrupt GABAergic signalling you lose the temporal locking
MEASURE INDIVIDUAL TACTILE
DISCRIMINATION THRESHOLDS (PUTS ET
AL, 2011) - frequency discrimination paradigm - diagram of experiment in notes
spectroscopy - subjects put their finger on a little fibre tactile stimulator - you get one of two frequencies and you have to press a mouse button and say which interval contained the highest frequency - left for the first one and right for the second one. then the frequency would get nearer and nearer to the comparator until your roughly 75% or so then you take the value you get from that
- 16 subjects (10 male)
- Avg. age 27.4 yrs
- All right handed
- No history of neurological disease
- Piezoelectric stimulator
- Stimulation on LD2
- Controlled via audio output MATLAB
can you measure the GABA within the brains of each of these people and relate it to their performance on the task?
MRS – NEUROIMAGING GABA
GABA MRS (MEGA-PRESS) in two voxels
Sensorimotor voxel (3cm)3 & Occipital voxel (3cm)3
you are able to use this to get a very small signal with is actually the GABA concentration itself - we relate its particular density in the region where the map is compared to control regions
when the ptps were in the scanner they measure the GABAergic concentration within the somatosensory map area and in the visual area - the visual one is the control as it wont be used for a tactile task
RESULTS – THE NEUROCHEMISTRY OF
INDIVIDUAL DIFFERENCES
- More GABA, better
discrimination only in
sensorimotor cortex. - Significant relationship in sensorimotor
cortex
(r = -0.58, p < 0.05) - NS in visual cortex
(r = -0.04; p > 0.5
more GABA in sensory cortex and no relationship between GABA And visual cortex
individual differences at one of these simple tasks can be linked essentially noninvasively to neurochemcial mechanisms
HOW MIGHT GABA CONTROL INDIVIDUAL
DIFFERENCES IN PERCEPTION?
Juliano looked at the amount of phase locking you get when the GABA system is in tact. when you disrupt it you get the same amount of firing but that exquisite locking between when the stimulus is occurring and when youre getting the increase a firing completely goes and its that pattering that is essentially a representation off the stimulus
intact - could hear a tone beautifully
disrupted - would sound like static and you have completely messed up the fundamental frequency of the audio tone and the stumps being applied
- Improvement in vibrotactile
frequency discrimination
through training leads to
improvements in phase-locking
in SI neurons in monkeys. - Between-subject mechanisms
similar to within-subject
improvement?
BUT WE ARE ONLY AT THE
BEGINNING OF LINKING
NEUROSCIENCE AND INDIVIDUAL
DIFFERENCES…
- Brain Derived Neurotrophic
Factor (BDNF) - Protein, acts on neurons in the
central and peripheral nervous
system - In development, acts to
encourage growth and
differentiaton (maturation) of
neurons - In adults…important for
LEARNING and
MEMORY…and found in high
concentration in the BASAL
FOREBRAIN…
BDNF is involved in motor plasticity in adults
mike Myers did his auditory experiment again and used something to block BDNF - if you do the experiment with BDNF present you get an overrepresentation of the kinds of tones that you stimulate with. when a BDNF blocker us replied its not able to do its job you dont get that representation and that map itself is not able to hold its usual structure based on the new stimulation
EVEN THE CRITICAL PERIOD
MAY BE ‘REVERSABLE’…
recent experiments in the visual cortex have suggested this
* Back to interneurons (GABAergic,
control receptive field size of sensory
cortex)
- they still have a job when your an adult but their influence in the critical period is way higher
* After the critical period in visual cortex
they undergo a change that surrounds
them with a ‘PERINEURONAL NET’ –
essentially scaffolding to ‘hold’ the
plasticity in place.
* Dissolve the perineuronal net with a
PROTEASE – visual cortex RECOVERS
aspects of critical period plasticity - may give a second chance to treat conditions like amblyopia
critical period
CRITICALLY PLASTIC
GABERGIC INTERNEURONS
CONTROLLING CORTICAL RF
PERINEURONAL NETS REDUCE
EFFECT
CRITICALLY PLASTIC…
AGAIN
APPLY PROTEASE –
BACK TO CRITICAL
PERIOD?
in diagram PV is a particular marker for types of interneurons - they can control the size of the cortical receptive field - in the critical period the perineuronal nets come up and the influence of them decreases - theres still a functional network but their influence has decreased - if you then burn all that away with the protease you are almost back again to a critical period state
PERCEPTUAL LEARNING
- It can occur without the participant paying attention
to the stimuli - It is often initially rapid but then slow and
incremental - It can be retained across a span of years even if the
task is no longer practiced - GABAergic interneuronal function is important for
individual performance on simple perceptual tasks
Summary: Learning and Memory
Definition of Learning:
Learning is the process where experiences alter the nervous system, leading to changes in behavior. These changes are known as memories. Rather than being “stored,” experiences reshape neural circuits, affecting how we perceive, act, think, and plan.
Forms of Learning:
Perceptual Learning:
Recognizing previously perceived stimuli.
Involves categorizing and identifying objects, people, or situations.
Changes occur in sensory association cortices (e.g., visual, auditory).
Stimulus–Response Learning:
Learning to perform a specific behavior in response to a stimulus.
Connects sensory perception to motor action.
Includes:
Classical Conditioning: A neutral stimulus (e.g., tone) becomes associated with an important one (e.g., puff of air), triggering a response (e.g., blink).
US (Unconditional Stimulus): Puff of air
UR (Unconditional Response): Blink
CS (Conditional Stimulus): Tone
CR (Conditional Response): Blink to tone
Instrumental Conditioning: (mentioned, not detailed)
Hebb Rule (Donald Hebb, 1949):
A synapse strengthens when it is repeatedly active while the postsynaptic neuron fires.
In the example: repeated tone-air pairings cause the synapse related to the tone to strengthen, eventually triggering a blink on its own.
Key Concepts:
Learning alters brain structure, particularly synaptic strength.
Classical conditioning illustrates how associations are formed neurologically.
Modern techniques now allow us to measure synaptic changes, validating Hebb’s theory.
perceptual learning
Definition:
Perceptual learning is the process by which we learn to recognize stimuli that we have previously experienced. It allows us to identify and categorize objects, people, and situations based on sensory input.
Purpose:
The main function is to interpret the world accurately and efficiently.
Without recognizing a stimulus, we cannot properly respond or learn from interactions with it.
Perceptual learning forms the foundation for more complex behaviors, such as decision-making or planning.
Sensory Modalities Involved:
Perceptual learning occurs in all sensory systems:
Visual: Recognizing faces, shapes, objects, or movements.
Auditory: Recognizing voices, speech sounds, or specific tones.
Tactile: Identifying objects by touch (e.g., textures).
Olfactory: Recognizing smells (e.g., food, danger signals).
Gustatory (taste): Recognizing and categorizing tastes.
Examples:
Recognizing someone by their face, walk, or voice.
Understanding spoken words in your native language.
Telling the difference between similar musical notes or accents.
Learning to identify a familiar perfume or the scent of smoke.
Neural Basis:
Perceptual learning is mediated primarily by changes in the sensory association cortex.
Visual perceptual learning → Visual association cortex
Auditory perceptual learning → Auditory association cortex
And similarly for other senses.
These regions adapt through experience-dependent plasticity, meaning repeated exposure strengthens the brain’s ability to detect and interpret subtle differences.
Key Features:
Experience-dependent: Requires repeated exposure or attention to the stimulus.
Passive or active: Can occur simply through repeated exposure (passive) or through active training or attention.
Long-lasting: Once developed, these recognitions are often stable over time.
Improves discrimination: Learners become better at telling apart very similar stimuli (e.g., shades of color, sounds of different instruments).
Role in Learning Hierarchy:
Perceptual learning is often the first step before other types of learning:
Before we can associate a stimulus with a response (as in stimulus–response learning), we must first recognize the stimulus
types of learning
Instrumental (Operant) Conditioning
Involves learning the association between a behavior (response) and its consequence (stimulus).
Unlike classical conditioning (automatic responses), it deals with voluntary, learned behaviors.
Reinforcement increases the likelihood of a behavior (e.g., food after pressing a lever).
Punishment decreases it (e.g., pain following an action).
Stimuli become cues that trigger the learned response (e.g., sight of a lever triggers pressing it).
Reinforcement strengthens connections between sensory and motor circuits in the brain.
Motor Learning
A subset of stimulus–response learning focused on acquiring new movements.
Involves modification of motor system circuits based on sensory feedback (e.g., from muscles, eyes, joints).
Needed for complex skills like dancing, using tools, or playing sports.
The more novel the skill, the more neural adaptation is required
Instrumental Conditioning
Learns association between voluntary behavior and consequence.
Reinforcement increases behavior; punishment decreases it.
Stimuli act as cues (e.g., lever → pressing).
Strengthens links between sensory and motor brain circuits.
Motor Learning
Learning new movements through sensory feedback.
Involves changes in motor circuits.
Essential for complex, novel skills (e.g., dancing, tool use).
Regionally Specific Human GABA Concentration Correlates with Tactile Discrimination Thresholds
(Puts et al., Journal of Neuroscience, 2011)
Key Aims:
Investigate whether GABA levels in the brain correlate with tactile perception.
Focus on frequency discrimination of touch using vibrotactile stimuli.
Test if this relationship is region-specific (sensorimotor vs. occipital cortex).
Methods:
Participants: 16 healthy adults.
Task: Vibrotactile discrimination — identify frequency differences around a 25 Hz stimulus on the index finger.
Technique: Magnetic Resonance Spectroscopy (MRS) to measure GABA concentration in:
Sensorimotor cortex (touch-related)
Occipital cortex (visual control region)
Results:
Higher GABA in the sensorimotor cortex correlated with better tactile discrimination (r = –0.58, p < 0.05).
No correlation was found in the occipital region.
No link between performance and brain structure (e.g., cortical thickness or gray matter volume).
Suggests GABA levels, not anatomy, account for perceptual differences.
Interpretation:
GABA may enhance tactile discrimination by:
Increasing inhibitory control and sharpening neural responses.
Improving signal-to-noise ratio in somatosensory processing.
This is consistent with earlier findings in visual discrimination tasks.
Relevance:
May inform understanding of sensory processing deficits in conditions like:
Autism spectrum disorder
Schizophrenia
Highlights GABA as a potential target for intervention.
Limitations:
MRS voxel size is large (3×3×3 cm³), includes both sensory and motor regions.
GABA signals also contain macromolecular contributions (~40%).
Conclusion:
First direct evidence linking individual GABA levels in the sensorimotor cortex to touch discrimination ability.
Supports the idea that GABAergic inhibition plays a general role in sensory perception.
