psych 30 midterm 1 FINAL

psychophysics

the relationship between external stimuli and perception

What is the + vs - when looking at receptive fields?

+ is excitatory field

- is inhibitory field

There are __ times as many rods than there are cones

20

two-point touch threshold

smallest distance btwn 2 points required for a person to feel touch on 2 points of the skin instead of one

just noticeable difference (JND) or difference threshold

smallest change in a stimulus that can be detected

Weber’s Law (or Weber’s fraction)

.. JND = constant proportion of original stimulus intensity

Weber’s fraction = JND / intensity_{ORIGINALstimulus}

—> the set fraction to help quantify change in stimulus required for noticeable difference

unconscious inference

when you brain tries to fill in the unknown by interpreting ambiguous visual data

Fechner’s law

S = k log R

—> explains that the perception of a stimulus intensity increasing is slower than the actual stimulus increasing

(less general than Stevens’s Power Law)

absolute/detectable threshold

minimum intensity of a stimulus that can be detected

what are the three methods to measure the absolute threshold?

method of constant stimuli, method of limits, method of adjustment

method of constant stimuli

stimuli of different intensities are shown randomly and multiple times to help estimate sensory thresholds

method of limits

gradually increase and decrease the intensity of a stimulus and measure when person either starts or stops hearing the stimulus - since ppl tend hear the stimulus start later when increasing the intensity and hear it stop earlier when decreasing it, balance this noise out to get an average threshold

method of adjustment

technique where the subject themselves controls the intensity of stimulus to identify thresholds

magnitude estimation

method tracking perception of different stimulus intensities used to develop Stevens’s Power Law

subject assigns a numerical value to the intensity of a stimulus to compare perception of different intensities (ex - if sensation is 2x as strong for something, the number they assign is likely to be double that too)

Stevens’s power law

the sensation/perception of a stimulus scales as an exponent (which varies depending on the type of sensation being measured) of the intensity

S = kI^a

sensation = constant (intensity) ^{exponent (depend on sensation/stimulus)}

exponent < 1 = perceptual/response compression

exponent > 1 = perceptual/response expansion

cross-modality matching

matching the same intensity across different types of stimuli (like finding a light that seems to be the same amount of brightness as a piece of music is loud)

signal detection theory

framework to measure people’s choices when they are uncertain about a stimulus being present or absent

4 possible outcomes: hit, miss, false alarm, correct rejection

key terms: sensitivity, criterion

criterion (in signal detection theory) vs absolute threshold

criterion: subjective threshold that determines whether a person chooses to decide if a signal is present or absent (can adjust criterion based on person’s objectives - ex) can lower criterion if listening for important phone call))

absolute threshold: (more objective) minimum intensity of a stimulus for person to detect it 50% of time

sensitivity/discriminability (d’ in signal detection theory) — what is it, how calculate, what #s mean (higher vs lower)

value shows the ease in which a person can differentiate the presence and absence of signals (higher number = larger difference / more clearly can differentiate when signal is present and when it is absent)

d’ = Z (hit rate) - Z (false alarm rate)

receiver operating characteristic (ROC) curve

a graph showing how well a model/subject identifies or denies a stimulus, plotted as false alarm rate (x) by hit rate (y)

perceptual judgements depend on 2 things…

1. sensory information

2. criterion

what are catch/null trials and why might you add them to an experiment?

catch trials normally = diff stimulus/trial to check for attention lapses/random guessing

BUT in sensation and psychophysics experiments… catch = null

catch/null trials have no stimulus — get measurement of how likely one is to hit regardless of if stimulus is there

what is the hit rate (or miss rate) = to? (fraction)

hit rate = # hits / # trials that stimulus is present

miss rate = # miss / # trials that stimulus is present

what are correct rejection or false alarm rates? (fractions)

CR = # CR / # trials stimulus absent

FA = # FA / # trials stimulus absent

which rates when summed together equal 1?

hit rate + miss rate = 1

correct rejection rate + false alarm rate = 1

why do people make errors in perceptual judgements?

neuronal responses are noisy (standard distribution)

scaling

relationship between stimulus intensity and perceived magnitude

ex) Stevens’s Power Law, Weber’s Law

the only way to increase the discriminability is to…

increase the stimulus intensity

does each point on the ROC curve have a different or the same d’?

the same d’ because discriminability is independent from criterion

What does the pupil control?

The amount of light entering the eye.

Which eye structure acts like a camera's adjustable aperture?

The pupil BUT the IRIS is what actually controls the size of the pupil

Why does the pupil dilate in dim lighting?

To let more light enter the eye.

Why can dilating the pupil make the image blurrier?

A wider opening admits more unfocused rays of light, reducing image sharpness.

convergence vs pooling

convergence = structural wiring where many photoreceptors connect to a lesser # of bipolar cells and ganglion cells/neurons

pooling = functional aggregation/combination of spatial information

What is accommodation?

Changing the eye's lens shape to focus objects at different distances.

Which muscles control accommodation?

The ciliary muscles.

Why is an adjustable lens needed in addition to the cornea?

Because objects at different distances require different focusing power to form a sharp image on the retina.

When focusing on a close object, should the eye increase or decrease its focusing power?

Increase it.

what is myopia and how do corrective lenses help with it?

myopia = eye’s focusing power is too strong (image focused in front of the retina)

CONCAVE LENSES (think curve inwards so makes eyeball less round) decrease the eye's overall focusing power so the image focuses on the retina

If nearby objects are focused behind the retina (hyperopia), what change is needed?

CONVEX LENSES (think bulge outwards more)

Increase focusing power.

what is presbyopia and why does it happen?

the ability of accommodation declines with age (inevitable)

—> bc lens becomes harder and the capsule encircling the lens lose elasticity

Where is the image formed inside the eye?

On the retina, at the back of the eye.

What is visual transduction?

Conversion of light energy into an electrical signal in the retina

Which retinal cells carry out light transduction?

Rods and cones (photoreceptors).

What does a photopigment do when it absorbs light?

It changes shape, initiating the photoreceptor's electrical response — bleached and now must regenerate

What photopigment is found in rods?

Rhodopsin.

How does photoreceptor output vary with light intensity in the lecture's simplified model?

It is graded and varies with light intensity, rather than being all-or-none.

Why can't a retina with only one kind of rod pigment provide normal color vision?

A single pigment type cannot distinguish wavelength differences from intensity differences.

Which cone type preferentially absorbs short wavelengths?

S cones (short, blue region).

Which cone type preferentially absorbs medium wavelengths?

M cones (medium, green region).

Which cone type preferentially absorbs long wavelengths?

L cones (long, red region).

Why does having three kinds of cones enable color vision?

Their different wavelength sensitivities allow the visual system to compare their responses.

Which photoreceptors are better suited to low-light vision?

Rods.

Why does seeing in the dark improve after spending time in darkness?

Dark adaptation — biphasic process (takes time bc photopigments must regenerate)!

1. Cone adaptation (faster photoopsin regeneration but higher threshold for light detection than rods)

—> rod-cone break (rods take over as the more light-sensitive and dominant photoreceptor)

2. Rod adaptation (slower rhodopsin regeneration but lower light detecting threshold (greater sensitivity) than cones)

OVERALL light-detection threshold decreases!

describe the fovea (photoreceptor types, features)

fovea = densely packed CONES (no rods) + relatively little convergence (preserves fine spatial detail) and good for detailed tasks

Does light or the neural signal pass through ganglion-cell layers before reaching photoreceptors?

Light. The neural signal originates when photoreceptors transduce it.

What is the main vertical neural pathway through the retina?

Photoreceptors → bipolar cells → ganglion cells.

Which cells send their axons out of the eye through the optic nerve?

Retinal ganglion cells.

What is the optic disc?

The location where ganglion-cell axons exit the retina to form the optic nerve.

Why does the optic disc produce a blind spot?

It has no photoreceptors to detect light.

Why do smaller photoreceptor sampling units (receptive fields) support higher spatial resolution?

They allow the image to be sampled at more frequent intervals (can differentiate between closer points rather than averaging them)

Where is spatial resolution generally higher: the fovea or the periphery?

The fovea.

Why does the peripheral retina have lower spatial resolution?

1. MORE photoreceptor information pooling / convergence

2. LARGER ganglion-cell receptive fields

3. more RODS (less cones)

What changes about ganglion-cell dendritic trees farther from the fovea?

They become larger.

Why must the retina compress visual information?

There are far fewer ganglion cells than photoreceptors, limiting how much can be sent to the brain.

Why is pooling useful despite reducing spatial detail?

It compresses information so fewer ganglion-cell outputs can represent input from many photoreceptors.

What is a retinal ganglion cell's receptive field?

The region of the visual field where stimulation influences that cell's response.

A neuron responds only when light falls within a certain visual region. What is that region called?

Its receptive field.

How is a ganglion cell's firing rate measured?

Action potentials (spikes) per second.

What does an excitatory input do to ganglion-cell firing?

Increases its firing rate

A ganglion cell increases firing when light hits its center and decreases firing when light hits its surround. What type is it?

ON-center/OFF-surround.

If light increases on both center and surround, must ON-center firing increase?

No. The outcome depends on the balance of center excitation and surround inhibition.

What are the three factors affecting the net response of a center-surround receptive field?

Area of overlap, light intensity, and the weighted balance of excitation and inhibition.

Why doesn't every excitatory or inhibitory input contribute equally?

Synaptic inputs can have different weights (strengths).

Why do center-surround receptive fields help with contrast detection?

They compare illumination in a region with illumination around it, emphasizing local differences.

Would a uniform bright surface always activate an ON-center cell more strongly than an illuminated center alone?

No. Uniform illumination can also recruit surround inhibition.

Which researcher studied retinal ganglion-cell receptive fields in the 1950s?

Stephen Kuffler.

How can a researcher identify a ganglion cell's receptive field?

Present light at different visual-field locations while recording changes in firing rate.

Why must a subject fixate when mapping a visual receptive field?

To keep stimulus locations stable relative to the retina.

What surprising observation challenged the idea that a bigger light always gives a bigger response?

Kuffler found that larger illumination could reduce firing compared with a small center light.

What is the receptive field response (formula, what # means - higher vs lower)

response = perceived brightness (higher # = brighter perception)

receptive field response = (W_exc * A_exc * I_exc) + (W_inh * A_inh * I_inh)

• excitatory vs inhibitory

  • W = weight (inh = -#)

  • A = area (portion out of 100%)

  • I = intensity

Why does a nearby object require more focusing power than a distant object?

Nearby objects send more diverging light rays into the eye, requiring stronger convergence to focus on the retina.

Why can peripheral vision detect dim objects well despite having lower spatial resolution?

the periphery has more rods (which provide high light sensitivity due to pooling signals from multiple photoreceptors) as opposed to the cones-rich foveal center

A retinal region pools many photoreceptors into one output neuron. What benefit does this provide?

Increased sensitivity through spatial summation of weak signals.

Why does high photoreceptor convergence reduce spatial acuity?

Signals from different spatial locations are combined, making it harder to distinguish exactly where stimulation occurred.

An engineer wants a retinal sensor optimized for detecting faint light rather than fine detail. Should it use high or low convergence?

High convergence, because summing weak signals increases sensitivity at the expense of spatial resolution.

What is the fundamental sensitivity-acuity tradeoff in retinal processing?

Greater pooling improves detection of weak signals but reduces the ability to distinguish fine spatial details.

A ganglion cell responds at its baseline firing rate when light appears outside a certain region. What does this reveal?

That location is outside the cell's measured receptive field.

An ON-center ganglion cell fires more strongly to a small central light than to a large light covering its entire receptive field. Why?

The large stimulus activates both excitatory center and inhibitory surround, reducing the net response.

Why can white intersections in a Hermann grid appear darker than neighboring white line segments?

More bright area falls within the inhibitory surrounds at intersections, reducing the modeled net response.

Two receptive fields have responses of $92.5$ and $87.5$ under the same model. Which location is predicted to appear darker?

The location with response $87.5$, because its modeled output is lower.

If center excitation remains constant but the proportion of illuminated inhibitory surround increases, how should net output change?

It should decrease because the inhibitory contribution becomes more negative (slower ganglion cell firing)

If increasing a stimulus from $50$ to $65\,\text{units}$ is just noticeable, what is the Weber fraction?

$0.30$, because $\frac{\Delta I}{I} = \frac{15}{50} = 0.30$.

Why does Weber's Law predict that larger starting stimuli require larger changes to notice a difference?

Because the just-noticeable difference is proportional to the initial stimulus intensity.

If a stimulus doubles in intensity and Weber's Law holds, what happens to the absolute just-noticeable difference?

It doubles, because the Weber fraction stays constant.

If a stimulus doubles and Stevens' exponent is $1$, what happens to perceived magnitude?

It doubles.

If a stimulus doubles and Stevens' exponent is $0.5$, does perceived magnitude double?

No. It increases by a factor of $\sqrt{2}$, approximately $1.41$.

If a stimulus doubles and Stevens' exponent is $2$, what happens to perceived magnitude?

It quadruples, because $2^2 = 4$.

Why can't a high hit rate alone establish that a signal detector is highly sensitive?

A liberal decision criterion can produce many hits while also producing many false alarms.

A detector becomes more conservative. What happens to its hit rate and false-alarm rate?

Both generally decrease because stronger evidence is required before reporting a signal.