PSY3108- lec1-11(12) methods1-hearing1(2)

Phenomenology

Qualitative/naturalistic observation, non systematic of a phenomenon (ex: illusion), verbal description of ones observations, the first step of any study in perceptual phenomenon

Example of Phenomenology: Thatcher Illusion

A phenomenological observation where reassignment of features (and displaying the image upside down) does not change the brain's perception of them

Example of Phenomenology: Purkinjie Shift

Noticed in daylight his flower bed looked light red/dark green during the day, but dark red/light green at twilight.
- This led to the hypothesis of 2 visual systems (ultimately to an understanding of rods/short wavelengths and cones/long wavelengths)

Quantitative Methods

Detailed answers/about assigning numbers to things, threshold seeking methods, magnitude estimation

Psychophysics

Branch of perceptual psychology where we try to place physical reality (physics) and subjective perception (psychology) together

What doe psychophysics deal with?

Traditionally low-level simple stimuli that are easily characterized (can be measured in physical terms, ex: measuring surface area of the desk)

Psychophysics example

- Light has a luminance of 5cd/m^2
- Ask someone how they rate the brightness on a scale from 1-10
- Observe the responses and draw a function between the two

Threshold seeking methods

- Psychophysical method that measures a physical quantity representing a limit of perceptual ability (ie: a threshold)

Examples of threshold seeking methods

Visual acuity test, hearing test, tactile acuity test where they continue to present you with a stimulus until you can no longer interact with it

Where are threshold seeking methods used

Medical settings, professional qualifications

How is threshold measured

Ex: decibels of sound, parts-per-million chemical concentration, candelas/m^2 of luminance (brightness)

Absolute Threshold

Smallest quantity you can detect

Difference Threshold

Smallest detectable difference between 2 stimuli (looking to see if you can tell if two pieces of stimuli are different, not being able to tell the difference implies something is BELOW your difference threshold)

50% Threshold (absolute and difference)

50% Absolute Threshold: physical quantity detectable half of the time on average
50% difference threshold: the physical difference detectable half of the time on average
*But thresholds can be defined for any level of accuracy
We must define thresholds for a certain level of response accuracy

Three classic threshold seeking methods

Method of Adjustment: fast, least reliable (can be frustrating for observer as they have to answer and the answer is not always clear, leads to guesswork)
Method of limits: easy on observer, fairly accurate
Method of constant stimuli: very slow, very accurate

Method of Adjustment (fast, least reliable)

Stimulus intensity is adjusted at will by the observer until they say they can just detect it, threshold is point to which the observer adjusts intensity, repeats trials (10) averaged for threshold... fast but not always accurate due to subjective nature of adjustment

Points of Subjective Equality (PSEs)

Can be found by threshold seeking methods
- When two things are objectively different but they appear the same (Ex: Illusions)
- They can be used to quantify the strength of an illusion
- Ex: If the lower fig has to be 1.2 x the size of the upper one to appear equal, we can say is has 20% PSE

Method of Limits (fairly fast/fairly accurate)

Stimuli of different predetermined, equally spaced intensities are presented in order (either a...)
Ascending Sequence: stimuli go from low to high
Descending Sequence: stimuli go from high to low
- Observer simply responds to whether she perceives each stimulus

Crossover point

- Between yes I see it and no I do not is the threshold estimate for a single sequence
- Average of crossover point from several ascending and descending sequences is taken to obtain a final threshold

Method of Constant Stimuli (slow, accurate)

5-9 predetermined levels of stimulus intensity are presented many times each, in random order
- intensities must span the threshold, so must know approx where it is a priori.
- Multiple trials (often 100s) of each intensity are presented
- Threshold is the interpolated intensity that results in 50% trials

Curve-Fitting Techniques

Used to fit a sigmoidal function to the data

Common choices (functions)

inverse weibel, cumuliative normal, logistic

Free Parameter

When you are fitting a function to data, there is a certain number of numbers that you let "float around", we want to minimize them

Adaptive Methods

Classic Method: Staircase
- How it works: We are trying to figure out the softest sound you can hear. Start at a set volume and continue to lower until you cannot hear. We then raise the volume to the one you could hear. All trials except first few at beginning are right near your threshold/very difficult to hear (can be frustrating for participants)

Difference from the Method of Limits

We adapt the set of things we present to you based on your previous behaviour, makes us efficient/we do not waste our time with stimuli that is too visible or not visible enough to you

Reversal Points

Points where the participant "goes back" on their previous statement of the amount of stimulation needed in order to detect the stimuli.
Calculation: Add sum of reversal points and divide by # of reversal points to calculate threshold
* reversal can be going from yes to no or no to yes

Two types of response bias

Address problems with potential bias (response bias)
- Lax criterion/lax bias: participants tend to say "yes I see it" a lot
- Strict criterion/strict bias: participants tend to say "no I do not see it" if they are not completely sure

Forced choice method

You are given x (often 2-8) number of options and must choose between them. Ex: 2 boxes on the screen with light (in 1 box) on each trial, person must choose 1 on each trial, continue until the person begins answering incorrectly, helps determine the person's threshold and limits lax and strict biases.

Performance floor with forced choice

The performance floor can be affected by the number of forced choice options you give (ex: if there are 2 (2AFC) boxes on each trial, you have a 50% performance floor, 25% for 4 (4AFC) boxes, etc.)

Threshold Finding with other senses

Same method can be used with a wide variety of sensory qualities. Different physical units can be used depending on the modality:
- Sound amplitude, touch pressure, smell/taste intensity

Difference threshold

Answering the question of whether you can detect the difference between two stimuli
Ex: if you can detect the difference between 10 and 11 candelas, your difference threshold is 1 candela
- This produces a "Just Noticeable Difference" (JND) subjectively
- Magnitude of stimulus increases=difference threshold increases (delta I)

Weber's Law

Describes the relationship between stimulus intensity and difference threshold. Holds true for many senses and across a moderately wide range of intensity levels. Shows that our sensory systems tend to work in proportional terms (percent changes)

Fechner's Law

Determined a formula that proved the idea that subjective sensation is related to stimulus intensity
ln=log to the base e
k= constant/a thing that does not change within a particular set of experimental circumstances
I- stimulus intensity
I0=stimulus intensity at absolute threshold

Fechner's Assumption

Assumed that each time you went up by one difference threshold (10%/subjectively 1 JND) that is related to an equal jump in subjective intensity

Does fechner's law work?

It does for some modalities (loudness, weight), but not others (electric shock), it more closely monitors the response of individual neurons than people

Magnitiude Estimation

Technique pioneered by steven to examine the relationship between subjective perception and objective stimulus intensity
- All stimuli are well above threshold
- Observer needs to be given a standard stimulus and value for its intensity
- Observers assigns numbers to the test stimulus relative to the standard

Magnitude estimation responses

Response expansion: as intensity increases, perceived magnitude increases more slowly than the intensity
Response compression: as intensity increases, perceived magnitude increases more quickly than the intensity

Sensitivity

Symbolized by (d')

How is sensitivity measured in a stimulus detection test?

Present stimulus 100 times and note the percentage of times the participant says they detect it.

What is a problem with measuring sensitivity by asking participants to say 'yes'?

A person who always says 'yes' will perform perfectly, regardless of actual sensitivity.

What is a good way to measure sensitivity in a stimulus detection test?

Only present the stimulus on 50% of the trials in random order.

What are test trials in a stimulus detection test?

Trials where the stimulus is presented.

What are catch trials in a stimulus detection test?

Trials where no stimulus is presented.

Are absolute thresholds related to actual sensitivity

ex: an individual doing method of limits could just say "yes" to all stimuli (lax criterion) and look like he is incredibly sensitive

What are the 4 possible results on each trial?

Hit, miss, false alarm, correct rejection

Participant correct vs incorrect responding

Criterion distinguishes the two (one has a higher likelihood to say yes, the other has a higher likelihood to say no)... but they are equally sensitive to the light

Calculating Ph/Pfa

Out of 100, B got 80, so her Ph = 0.8
Out of 100, B got 10, so her Pfa = 0.1

How do sensitivity experiments differ from threshold seeking methods

We vary the intensity of stimulus, and look for the intensity that will get us that

Calculating Basic Sensitivity

We want to reward hits and punish false alarms
Therefore: Basic Sensitivity/ BS = Ph - Pfa
... But this is not the most effective way to do this

Issue with BS method

Is is easy to receive a small increase in performance but takes time to make large increases that are proportionally the same (test score example)

d' (d-prime) = Stronger calculation than BS

Converting proportion of hits and FAs
- Find corresponding z-score for Ph and Pfa
- Place z scores in the formula (do NOT multiply)
- Subtract z-scores from one another and calculate
- Then you have the individuals d' (d-prime)

Zero and One

When you get z-scores of 0 or 1, you get z-scores of - or + infinity
- If you receive a proportion of 0.00 or 1.00, will will treat them as 0.01 or 0.99

What is the relationship between criterion and sensitivity?

The flip side of sensitivity is criterion or response bias.

What does a lower c (criterion) value indicate about a person's response tendency?

The person has a higher likelihood to say yes (lax criterion).

What does a positive c value indicate about a person's response tendency?

The person has a tendency to say no when they are unsure (strict criterion).

Isosensitivity Curves (ROC curves)

A range of Ph and Pfa values can produce the same d'
- Sensitivity remains the same but criterion shifts

Why does criterion shift

Payoff matrix
- If the cost of a false alarm or the reward for a rejection is great, one will tend to make ones criterion stricter (and vice versa)
- Ex: winning money or betting on money
- Ex of payoff matrix leading to lax criterion: radiologist

Alternatives to d'

We use them because Ph or proportion correction fail to take into account differences between performance and linearity

Signal Detection theory

- Arises the concepts of sensitivity and criterion
- SDT suggests that any attempt at detecting a signal (stimulus) has to contend with competing noise
- Noise in this sense is random variations from the environment

Stimuli of touch

Somatosensory neurons respond to many physical changes in skin: pressure, vibration, temperature variations

Pressure

- Most basic touch stimulus
- Measured in pascals M/m^2

Vibration

- Rhythmic changes in pressure across time
- Frequency of vibration (amount of times x is moving from side to side per second, measured in hertz = cycles/second)
- Detected by mechanoreceptors
- Amplitude on vibration (distance of movement from end to end) measured in meters
- This is the basis for the sense of movement across the skin (important in tool use, for example)

Temperature

- Measured in degrees celcuius
- You detect cold/hot when it changes from your physiological zero (often 32 degrees)
- Cold/warmth is detected by cold/warmth thermoreceptors
- Detection threshold

Pain

- Not a stimulus, but an excessive intensity of any stimulus produces painful sensations
- pain thresholds vary according to many factors
- Pain is a complex and subtle topic

Anatomy of skin

- Heaviest/largest sense organ by far
- Three main layers
Epidermis: outer protective layer, composed of dead keratinized cells
Dermis: Below the epidermis, contains many types of somatosensory receptors
Hypodermis: layer of fatty tissue below the skin, contains receptors
- Two types of skin: Hairy and hairless (palms and soles = hairless skin), with different elastic properties

Tactile receptors in the skin

Learn sketch and label diagram

Mechanoreceptors

- Sensory neurons that signal pressure on skin
- Respond to mechanical deformation of the cell itself or an associated structure
3 main anatomical categories
- Encapsulated
- Accessory-structure-associated/ASA
- Free nerve endings

Encapsulated Mechanoreceptors

- Sensory neurons with specialized endings growing out of their peripheral axon terminal
Three types
- Meissner: terminal surrounded by oval capsule full of layered cells
- Ruffini: terminal surrounded by spindle shaped structure
- Pacinian: terminal surrounded by large onion like structure

ASA Mechanoreceptors

Merkel Disc: nerve fiber surrounds a separate skin cell called merkel cells
Root Hair Plexus: Nerve fibre surrounds a hair follicle Mechanoreceptors

Free nerve endings

Simply neurons with peripheral axon terminals that terminate close to skin's surface

What is a receptive field?

The location that a stimulus must be placed in order to affect the firing rate of a given neuron.

What is the receptive field for somatosensory neurons?

A patch of skin.

What is the receptive field for vision?

A patch of retina at the back of the eye.

What is the receptive field for hearing?

A patch of a membrane in the inner ear.

What is the receptive field for smell?

A patch of olfactory mucosa in the nasal cavity.

What is the receptive field for taste?

A patch of the tongue.

How are receptive fields mapped out

Mapped out by measuring neuron firing rates while moving a stimulus across the sensory surface
- Can have excitatory and inhibitory zones, such as "centre/surround" organization
- Applies to neurons at any stage in the sensory process, from the sense organ up to the cortex
- A given patch of the sense organ has many RFs that overlap one another

Center surround Receptive Fields

- Vary in size and have either an: excitatory centre/inhibitory surround (ONcentre)
- Inhibitory centre/excitatory surround (OFFcentre)

From skin to spine

- When stimulated, mechanoreceptors produce action potentials
- Transmitted to the spine by dorsal root ganglion neurons (DRG neurons), a type of bipolar neuron

DRG Adaptation speeds

- Another way to classify dorsal root ganglion neurons
- Slow adapting: respond continuously to stimulus
- Fast adapting: respond at the onset and offset of stimulus

Neuron skin depth

Fast Adapting (FA) and Slow Adapting(SA) further break down based on depth in skin:
- Type I (FA-1 & SA-1) are found near surface. Have small receptive fields
- Type II (FA-II & SA-II) are found deeper. Have large receptive fields

Spatial Resolution

- Means: how good is your sensory systems at determining exactly where a stimulus is
- Determined mainly by size and density of receptive fields

Temporal resolution

- Degree to which a sensory system can determine exactly when a stimulus occurs
- Determined mainly by adaptation speed of individual receptors
- Faster adaptation yields better temporal resolution

Dorsal Column-Medial Lemniscus Pathway

- Large fibers (Aa and AB)
- Cross over the medial lemniscus of brainstem

Anterolateral Pathway

- Small fibers (Ao and C)
- Carries info about temperature and pain

Serial Processing

Very rarely does a single neuron that takes care of everything, it's often a chain; neurons connect to one another in a sequence

Parallell Processing

The chains are not all carrying the same types of information; they bring different information to the brain simultaneously

Contralateral processing

Sensation from left side of body cross over to right side of brain and vice versa
- However, signals are then communicated to the ipsilateral side so that the whole brain is involved (not in the case of missing corpus callosum)

Where are DRG neuron cell bodies located?

Roots of the spine, specifically in the dorsal root ganglion (lump in the root that goes to the spine)

Mechanoreceptor signals are primarily carried by what class of DRG neurons?

A-beta

Somatotopic Organization

Adjacent neurons carry signals from adjacent body parts
- Size of brain area is not related to the size of body parts; body parts with high acuity have larger amounts of associated brain tissue

Subcortex and Somatosensation

- Spinal neurons from the DCML pathway (touch) terminate in the ventral posterior nucleus (VPN) of the thalamus

Cross Modal Integration

Different types of sensory input are initially processed in different streams but rejoin eventually

Organization of neurons

More brain tissue for parts of the body that have more receptors (ex: finger tips, lips)
- Neural plasticity can lead to changes in how cortical cells are organized

Topographic Organization

Means the brain area are organized according to some feature of the sense organ, outside world, or something else
- Roughly a point to point correspondence between locations on the sensory organs and location in cortex (not in accordance with size)

Receptive Fields of Cortical Neurons

Each cortical neuron receives inputs from a limited patch of skin- this is its receptive field. The size of RFs vary according to body area (ex: very large RFs for back and tiny ones for fingertips)

Columnar Organization of SI

Going down through 6 layers of SI in one spot, one finds neurons with RFs in the same location on the body

Neural Plasticity

- Neurons can change/be plastic, but can occur in a bad way (brains way of changing in response to experiences, injury, etc)

Musicians Cramp/Focal Dystonia

- You abruptly lose the ability to move individual fingers with skill, people find that the parts of the cortex that previously processed information from individual digits have grown over each other/fused
- Seen in musicians, caused from "overdoing it" with practice and can lead to an inability to play properly.

What are the two pathways that carry touch info up the spine to the brain?

DCML, Anterolateral