Sensory Perception Flashcards
Visual Cues Sensory Adaptation Weber's Law; Thresholds
What is depth perception?
Depth perception is the visual ability to perceive the world in 3D AND distance of an object.
Visual cues allow us to perceptually organize by taking into account which cues?
Visual cues allow us to perceptually organize by taking into account the following cues: depth, form, motion, constancy.
________ cues provide humans with a sense of depth.
Binocular cues provide humans with a sense of depth.
Created by retinal disparity (space b/w eyes).
What visual cues are binocular?
Binocular cues include stereopsis (relative disparity) and convergence.
What is retinal disparity (stereopsis)?
Retinal disparity (stereopsis) is a binocular cue in wh the eyes are separated by some distance (~2.5in in humans); creates depth perception.
↑ diff (disparity) b/w image on ea retina → ↓ distance to observer
How does convergence help provide a sense of depth?
Convergence provides a sense of depth based on how much eyeballs are turned.
- Objects far away → muscles of eyes relaxed, i.e. focus straight ahead
- Objects near → muscles of eyes contract; i.e. turn inward (like trying to cross your eyes)
I.e. ↑ angle of convergence (inward strain) → ↓ distance from object.
Which two binocular cues help provide a sense of depth?
Retinal disparity and convergence help provide a sense of depth.
retinal disparity - space b/w eyes.
convergence - eyes contract/turn for objects close by.
What are monocular cues?
Monocular cues are visual cues that dep on info available to either eye alone.
- Imp for judging distances of far objects bc retinal disparity is negligible.
What visual cues are monocular?
Monocular cues include relative size, interposition (overlap), relative height, shading/contour, motion parallax, and constancy.
Describe how relative size provides a sense of depth.
Relative size is a monocular cue in which the closer of two objects is perceived as larger given the assumption that the objects are the same size.
I.e. if we know two ants are about the same size (absolute size), then the ant that appears larger is perceived as being closer to viewer.
E.g. attached image - Tells us that the orange circles are diff sizes bc they are next to diff sized grey circles, even tho orange circles are actually the same size. Although the circles are the same size, our brains think they’re diff sizes bc we compare them to their surroundings.
Describe interposition (occultation).
Interposition (overlap; occultation) is a monocular cue in wh near surfaces overlap far surfaces.
I.e. an object that partially obstructs another is perceived as nearer.
Observer creates “ranking” of relative nearness.
Describe relative height (elevation).
Relative height (elevation) is a monocular cue in wh higher objects are perceived to be farther away than those that are lower.
I.e. when object is visible relative to horizon, objects closer to horizon are perceived as being farther away, and objects farther fr horizon are closer to us.
Also, if object moves fr position near horizon to position higher/lower than horizon, it will appear to move closer to viewer.
Describe light, shadow, and contour as a monocular cues.
Light/shadow - Closer objects reflect more light than distant objects. The dimmer of two identical objects seem farther away.
Contour - light reflection/shadows provide a sense of form, e.g. crater vs mountain.
Describe motion parallax.
Motion parallax (relative motion) is a monocular cue wh stationary objects further away appear to move slower than objects nearer; provide a sense of motion.
E.g. driving in car, mountains in distance appear to move slowly while road signs nearby appear to move quickly.
Provides a sense of how far away objects are based on how much they move as you move.
Describe the monocular cues of relative clarity, texture gradient, and linear perspective.
Relative clarity - monocular cue; hazy objects perceived as more distant.
Texture gradient - monocular cue; coarse/distinct texture perceived as nearer.
Linear perspective - monocular cue; parallel lines appear to converge w distance.
Which monocular cues provide a sense of constancy?
Size/shape/color constancy are monocular cues wh provide a sense of constancy.
Constancy describes how our perception of object doesn’t change even if image cast on retina is diff.
E.g. people don’t actually change size as they walk away.
Describe size/shape/color constancy.
Size/shape/color constancy is a monocular cue in wh our perception of an object’s size/shape/color doesn’t change even though the size/shape/color cast on retina may change.
E.g. two ladies of approx same size standing near and far -> gives cue of distance b/w ladies rather than diff in size.
E.g. door opening toward observer appears to change from rectangle to trapezoid, but still perceived as same, constant shape.
E.g. red cup with unequal lighting is perceived as same, constant color rather than a gradient.
What is sensory (neural) adaptation?
Sensory (neural) adaptation describes a change in responsiveness to a constant stimulus over time; typ experienced as a change in stimulus.
E.g. resting hand on table -> initially feel table -> sense of table ceases as sensory neurons fire less and less often, and sometimes not at all.
Describe the process of hearing (auditory) adaptation.
Hearing (auditory) adaptation occurs via a small muscle in the inner ear. The muscle contracts in response to higher noise, wh dampens vibrations and protects eardrum.
Hearing adaptation (& muscle contraction) is not instantaneous, i.e. doesn’t protect against abrupt, loud noises like a gunshot.
Involves mechanoreceptors on hairs cells (rather than chemoreceptors)
Describe touch (somatosensory) adaptation.
Touch (somatosensory) adaptation describes a change in responsiveness to a constant barometric/temp stimulus over time.
E.g. resting hand on table -> initially feel table -> sense of table ceases as sensory neurons fire less and less often, and sometimes not at all.
E.g. ice bath.
Describe smell/scent (olfactory) adaptation.
Smell adaptation involves desensitization of receptors in your nose, e.g. forgetting that you’re wearing cologne.
Typ involves negative feedback sys via several factors (mostly CaMK/calmodulin) bound to Ca2+.
Also, “olfactory fatigue”.
Describe proprioception adaptation.
Proprioception adaptation involves a sense of balance/self/where you are in space.
E.g. goggles that distort environ -> adapt to new environ over time.
Describe sight (visual) adaptation.
Sight (visual) adaptation involves up/down regulation in response to light intensity.
Down regulation is an adaptation to high light intensity in wh pupils constrict (less light enters eye/hits retina) and rods/cones become desensitized. Also “light adaptation”.
Up regulation is an adaptation to low light intensity in wh pupils dilate and rods/cone become sensitized (synth light-sensitive molecules). Also “dark adaptation”.
Does the perception of pain decrease over time?
No. Somatosensory adaptation involves large mechanoreceptors which display adapation over time, but pain involves small nociceptor neurons wh do not display adaptation.
What is the difference b/w habituation and adaptation?
Habituation is a behavioral phenomenon, neural adaptation is physiological; not entirely sep.
Conscious control:
Habituation: some conscious control over whether one notices something to wh one is becoming habituated.
Neural adapt: no conscious control.
E.g. cannot consciously force oneself to smell perfume after adaptation.
Stimulus intensity:
Habituation: varies; weak stimuli permit almost immediate habituation; strong stimuli make habituation difficult.
E.g. cool breeze vs fire alarm.
Neural: tied v close to stimulus intensity.
Which law describes the relation between the actual change in a physical stimulus and the perceived change?
Weber’s Law describes the relation between the actual change in a physical stimulus and the perceived change
What is meant by the “just noticeable difference”?
The just noticeable difference (JND) is the threshold at wh you’re able to notice a change in stimuli/sensation.
Applies to all senses.
E.g. a 2lb and 2.05 lb weight feel the same, but a 2lb and 2.2lb weight feel diff; thus 0.2lb is the JND relative to the 2lb weight
I.e. the 0.2lb JND is not absolute; a 5lb and 5.2lb weight likely feel the same; JND closer to 0.5lb.
Describe Weber’s law (Weber-Fechner Law).
Weber’s Law describes the relation b/w actual change in a physical stimulus and perceived change.
ΔI / I = k
where ΔI := JND; I := initial intensity; and k := constant (specific to subject).
E.g. 2lb vs 2.2lb weight and 5lb vs 5.5lb weight -> 0.2/2 = 0.5/5 = 0.1; so k for individual is 0.1.
Weber’s law predicts a linear relationship b/w incremental threshold and background intensity; ΔI = Ik, plot of ΔI vs I is linear (constant).
T/F: sensation is proportional to the logarithm of stimulus intensity.
True
Sensation is proportional to the logarithm of stimulus intensity.
(Relates Weber’s law to sensory perception/nervous sys)
what are the types of somatosensation?
The four types of somatosensation are temp (thermoception), pressure (mechanoception), pain (nociception) and position (proprioception).
How are the intensities of somatosensations encoded in the body?
Intensities of somatosensations are encoded in the body by how quickly neurons fire.
E.g. thermoception: if it’s really cold outside, then neurons might not less rapidly.
E.g.
What are the three types of timing encoded by somatosensory neurons?
The three types of timing encoded by somatosensory neurons are:
-
Non-adapting: consistently fire action potentials
- Equal spacing b/w successive APs, i.e. constant firing rate.
-
Slow-adapting: starts firing APs v quickly then slows down over time.
- Space b/w successive APs increases.
- Fast-adapting: starts firing APs v quickly, stops, then fires again when stimulus stops.
Describe how somatosensation provides info about the location of a stimulus.
The location of a stimulus is encoded by dermatomes.
Ea part of the body is innervated by a partic nerve, and that nerve goes up to the brain (sends signals).
T/F: non-adapting receptors always fire action potentials even if there is no stimulation. They produce action potentials constantly.
True
Non-adapting receptors always fire action potentials, even if there is no stimulation. They produce action potentials constantly.
Recall: equal spacing b/w successive APs.
The body is always checking homeostasis, balance, etc. using non-adapting receptors, so these neurons act as a “control group” to differentiate what’s normal and what’s a new stimulus.
T/F: Slow and fast-adapting receptors don’t produce APs if they are not stimulated by associated stimulus modality (temp, pressure, pain, etc.).
False
Slow and fast-adapting receptors ALMOST don’t produce APs if they are not stimulated by associated stimulus modality (temp, pressure, pain, etc.).
Describe how fast-adapting receptors function.
Fast-adapting receptors almost don’t produce APs if they are not stimulated by associated stimulus modality (temp, pressure, pain, etc.).
As long as stimulus intensity is changing, fast-adapting recpetors produce APs (and v quickly).
When stimulus intensity is constant, fast-adapting receptors stop producing APs (actually, prod v v spaced APs).
What are the four possible outcomes when signal detection theory is applied to a data set where stimuli were either present or absent.?
When stimuli either present/absent, the four possible outcomes of SDT are:
- Hit - stimulus/signal is present, and you recognize it as present.
- Miss - stimulus/signal is present, but you don’t recognize.
- False Alarm - stimulus/signal is absent, but you recognize it as present.
- Correct Rejection - stimulus/signal is absent, and you recognize it as absent.
Describe how signal detection theory might be applied to memory experiments.
SDT & memory experiments:
E.g. items are presented on study list for later testing. A test list is created by combining these ‘old’ items w novel, ‘new’ items that did not appear on the study list.
- On ea test trial the subject will respond ‘yes, this was on the study list’ or ‘no, this was not on the study list’.
- Items presented on the study list are called Targets, and new items are called Distractors.
- Saying ‘Yes’ to a target constitutes a Hit, while saying ‘Yes’ to a distractor constitutes a False Alarm.
In signal detection theory, what does the variable d’ (d-prime) represent?
In SDT, the variable d’ (d-prime) represents discriminatability, i.e. the distance b/w the means (peaks) of the signal/stimulus and noise distribution curves; it is a “sensitiity index”.
E.g. large d’ w a v bright green light; small d’ w a dull green light.
If hits > miss → signal is strong; if miss > hits → signal is weak.
In assessing strategy using SDT, what does ‘B’ represent?
‘B’ in SDT:
-
‘B’ represents a threshold intensity specific to the subject.
- B can be modified w experience (or simulation).
- Any signal/stimulus intensity < B → respond ‘NO’; > B = ‘YES’
- Probability of a HIT is the area > B beneath the signal curve and excluding the area beneath the noise curve.
- Probability of a FALSE ALARM is the area > B beneath the noise curve.
In assessing strategy using SDT, what does ‘D’ represent?
‘D’ in SDT:
- D is threshold intensity relative to the index.
- D = d’ - B
- Recall: B is a threshold specific to the subject.
- Any signal/stimulus intensity D → ‘YES’
- Probability of a HIT is the area > D beneath the signal curve and excluding the area beneath the noise curve.
- Probability of a FALSE ALARM is the area > D beneath the noise curve.
In assessing strategy using SDT, what does the variable ‘C’ represent?
‘C’ in SDT:
-
C is criterion: B (randomly chosen) is either ideal, conservative, or liberal dep on how easily noise and actual data (signals/stimuli) are discriminated.
- C = B - d’/2
- If d’ goes to infinity → B - d’/2 <<< 1 → C <<< 1 → data (signal/stimulus) is always discriminated/detected → strategy is considered v liberal.
- Ideal observer → minimizes possibility of miss and false alarm
- Any signal/stimulus intensity C → ‘YES’
- Probability of a HIT is the area > C beneath the signal curve and excluding the area beneath the noise curve.
- Probability of a FALSE ALARM is the area > C beneath the noise curve.
- Strategy is conservative when C > 0.
- Strategy is liberal when C < 0.
In assessing strategy using SDT, what does ‘β’ represent?
‘β’ in SDT:
E.g. friend calls your name twice. Even though your criterion (C) is to respond, you fail to do so the first time your name was called. The noisier the room , the more likely you’ll be affected by your bias (β). Assuming your name is called out w same loudness, a noisy room is one that has a small discriminatability (d’). Thus, bias deps on the d’ of the data and your criterion.
I.e. if data was absolutely non-discriminate (d’=0, room is just as loud as your name being called) → signal and noise distribution curves overlap → odds of differentiating signal from noise is up to chance (50%) → does not matter where you choose your threshold (B) bc β = e^(d’ x C) = e^(0) = 1 → β = 1 means your criterion is solely dep on your bias, as long as B false somewhere on overlapping noise/signal curve).
I.e. if data is absolutely discriminate (d’ = infinity, room is perfectly quiet and name called infinitely loudly) → signal/noise curves do not overlap whatsoever → odds of differentiating signal from noise is nearly certain (~100%) → your criterion is v large negative # (C= B - d’/2) → β = e^(d’ x C) = e^(∞ x -∞) = e^(-∞), wh converges to zero (β = 0) → β = 0 means bias doesn’t matter and your signal detecting ability deps solely on choice of threshold (B).
-
‘β’ is bias, i.e. how often subject is able to detect and carry out criterion (C; C = B - d’/2).
- ln(β) = d’ x C → β = e^(d’ x C)
- Threshold equal to the ratio of the height of signal curve to height of noise curve
- “Bias is extent to wh one response is more probable than another”
- β is logarithmic bc its affect ↑ exponentially as d’ ↓.
- β also ↑ as B ↑ (Weber’s Law; for any stimuli, assuming d’ constant)
In SDT, what value of C is considered the ideal threshold?
In SDT, C = B - d’/2. The strategy is considered ideal when C = 0, i.e. highest ratio of hit : false alarms.
-
C > 0 → conservative, i.e. subject responds ‘NO’ more often and ‘YES’ less often than ideal observer.
- I.e. rati of hits : false alarms increases (fewer false alarms, but total # of hits also decreases)
-
C < 0 → liberal, i.e. subject responds ‘NO’ less often and ‘YES’ more often than ideal observer.
- I.e. Ratio of hits : false alarms decreases (more and more false alarms)
Describe bottom-up processing.
Bottom-up processing begins w the stimulus; stimuli influence what we perceive (our perception)
- No preconceived cognitive constructs of the stimulus (never seen before).
- Data driven - stimulus/perception directs cognitive awareness.
- Inductive reasoning - perception always matches reality/environ.
E.g. plane cockpit → never seen before, so visual stimuli completely drive our preception.
Describe top-down processing.
Top-down processing uses background knowledge to influence perception, e.g. where’s waldo.
- Theory driven - perception influenced by expectation.
- Uses deductive reasoning.
- Not always correct, i.e. expectation may not fit reality.
E.g. attached image w incomplete image of a cube. Use expectations and deductive reasoning to “fill in the gaps”.
Describe Gestalt’s Principle of Similarity.
Gestalt’s Principle of Similarity:
Similar objects are grouped t/g, e.g. columns of squares and circles are grouped separately rather than grouping t/g rows of alternating squares/circles.
Describe Gestalt’s Principle of Pragnanz (Good Form; Good Gestalt).
Gestalt’s Principle of Pragnanz (Good Form; Good Gestalt):
Tendency to group/organize forms of similar shape, pattern, color, etc.; i.e. into simplest form possible.
E.g. olympic rings → organized into 5 discrete circles rather than more complicated shapes.
Describe Gestalt’s Principle of Proximity.
Gestalt’s Principle of Proximity:
Objects that are close t/g are grouped t/g.
Describe Gestalt’s Principle of Continuity (Good Continuation).
Gestalt’s Principle of Continuity (Good Continuation):
Objects are grouped/perceived as following the smoothest path. I.e. if two objects intersect, objects are pereceived as individual, uninterupted objects.
Describe Gestalt’s Principle of Closure (reification).
Gestalt’s Principle of Closure (Reification):
Objects grouped t/g are seen as a whole; missing info is filled in (using top-down processing).
Describe Gestalt’s Principle of Symmetry.
Gestalt’s Principle of Symmetry:
Objects are perceived as symmetrical; grouped around center point.
What is the Law of Common Fate?
The Law of Common Fate is one of Gestalt’s Laws of Grouping in wh objects (stimulus elements) are likely to be perceived as a unit if they move t/g. E.g. birds in flying-V or a swarm of bees.
Allows moving objects to be discerned despite other details missing (color, outline, etc.)
A diagnostic test is developed for a potentially fatal disease. In order to effectively treat the disease, those diagnosed with it must immediately be placed on a costly and poorly tolerated medication. Which threshold should be used for this test?
- A – a liberal strategy which would make sure everyone with the disease was treated
- B – an ideal observer strategy to try to minimize both false positives and false negatives.
- C – an arbitrary threshold to get the majority of disease cases while minimizing false positives
- D – a conservative strategy which would eliminate false positives
A diagnostic test is developed for a potentially fatal disease. In order to effectively treat the disease, those diagnosed with it must immediately be placed on a costly and poorly tolerated medication. Which threshold should be used for this test?
-
A – a liberal strategy which would make sure everyone with the disease was treated.
- In the case of a fatal disease w an effective treatment, it is best to err on the side of caution and adopt a liberal detection strategy → ensure treatment for every affected patient.
- B – an ideal observer strategy to try to minimize both false positives and false negatives.
- C – an arbitrary threshold to get the majority of disease cases while minimizing false positives
- D – a conservative strategy which would eliminate false positives
This image best demonstrates what idea about perception?
- Gestalt law of pragnanz
- Size constancy monocular cue for depth
- Gestalt law of proximity
- Relative size monocular cue for depth
This image best demonstrates what idea about perception?
- Gestalt law of pragnanz
- Reality is reduced to its simplest form in our perceptions.
- Size constancy monocular cue for depth
- Relies on knowledge that objects do not change size as they travel through space.
- Gestalt law of proximity
- Tend to group things by spatial relatedness, e.g. may see two groups of circles before you process the components of those groups. H/e, that is not the principle demonstrated here.
-
Relative size monocular cue for depth.
- Tells us that the orange circles are diff sizes bc they are next to diff sized grey circles, even tho orange circles are actually the same size. Although the circles are the same size, our brains think they’re diff sizes bc we compare them to their surroundings.
