Psych 120b_(week 1~2)sensory and perception

do you think JND is fixed for all stimulus levels or variable? would you guess that the JND is the same or different for 1 Kg and 100 kg weights?

(Weber's law) -JND is not the same for different standards. -It changes as the intensity changes, but it changes in a very interesting way: the proportion remains the same. (=constant k) (difference threshold)/ standard = constant (change in intensity)/ intensity = k

Two of these laws make the same predictions about JND size. Which are they? (i) Weber's Law: ΔI / I = k (or JND / I = k) (ii) Fechner's Law: S = k log R (iii) Steven's Law: S = a I b

(a) (i) and (ii) (b)(i) and (iii) (c) (ii) and (iii) (d)All make the same prediction the answer is a.

Q. You are participating in an experiment of signal detection. Your task is to say whether you see a spot of light or not in every trial. In this experiment, you will be: • rewarded 50 cents for each correct YES • rewarded 5 cents for each correct NO • penalized 5 cents for each wrong YES • penalized 5 cents for each wrong NO Assuming your sensitivity is fixed, what is the best strategy to make money?

* Be conservative / liberal in saying YES * Move response criterion to the left / right * Drawback: Increased false alarm / miss rate the answers will be 'liberal' in saying yes, move response criterion to 'the left', with drawback which is increased 'false alarm'.

main characteristics of neurons

* computation occurs at 2 levels - within neurons (some computational power) - between groups of neurons (lots of computational power) * neurons are connected to thousands of other neurons * even binary signals from individual neurons would support very sophisticated computation.

Psychophysical methods

* constant stimuli * limits * adjustment

Direct Scaling : Magnitude Estimation

* when a stimulus is above threshold * ex) if we have 2 light spots of different intensities, what should a 3rd light's intensity be so that it is perceived subjectively as half way in between ? * physically, it may not be the midpoint of the 2 intensities.

what Stevens found about magnitude estimation

- If light intensity is physically doubled, does it look twice as bright? - less than doubled - If a pencil is physically doubled in length, does it look twice as long? -- yes - If an electric shock is physically doubled in intensity, does it hurt twice as much? - more than doubled

Weber's law

- The Just Noticeable Difference between two stimuli is proportional to the magnitude of the stimuli = ratio of JND to the reference is roughly constant. - can use to predict a person's JND for any discrimination ex) Difference threshold= 5g Standard = 100g k = 20 ex) Difference threshold=10g Standard = 200g k = 20 (the principle describing the relationship between stimulus and resulting sensation that says the just noticeable difference (JND) is a constant fraction of the comparison stimulus.)

What is difference threshold?

- just noticeable difference (JND)= how different two stimuli must be for the subject to notice that they are not the same. ex) height of two people, intensity of lights, weight of cans. ex) if the weight of 105g can be distinguished from that of 100g, the JND is 5g. -the smallest detectable difference between two stimuli, or the minimum change in a stimulus that enables it to be correctly judged as different from a reference stimulus. -change in a systematic way. -smallest change in weight was always close to 1/40 of the standard weight. - 3 methods above apply here as well. (method of limit, method of constant stimuli, method of adjustment)

Describe receiver operating characteristic curve.

- the bottom x-axis : false alarm Pr(S/n) - the left y-axis,: Hit Pr(S/s) - the top x-axis: correct rejection Pr (N/n) - the right y-axis: Miss Pr(N/s) - d'=0 means a vertical line - d'=.75 means a curve line located higher than the vertical line - d'=1.5 means another curve line even higher than d'=.75. - d'=2 means the highest curve line ** if d' below 0, there must be something wrong in your experiment either on the tools or something else. ** - plots hit rate as function of false alarm rate - if hit and false alarm rate are identical, points fall on the diagonal.

what is absolute threshold ?

- the minimum amount of stimulation necessary for a person to detect a stimulus 50%of the time. - the level at which a stimulus will be detected a specified percentage (e.g. 50%, 100%) of the time e.g) a sound, a spot of light, a touch, an odor, a taste - (a graph) not a step function, rather it is sigmoid function. (probability of detection that increases with signal intensity)

method of constant stimuli

-a psycho physical method in which many stimuli, ranging from rarely to almost always perceivable are presented one at a time. Participants respond to each presentation: "yes/no" "same/different" and so on. -The experimenter creates many instances of stimuli with different intensities and presents these stimuli randomly ** can be somewhat inefficient: many stimuli are above or below thresholds, but gives you more information. - pros: Instead of focusing one threshold, we can get multiple thresholds. Good for a new area of study or a field of new paradigm. -cons: time insensitive, takes long time for you and for participants who are more likely to be a college student. if you are not careful to motivate participants to get involved in the experiment, you might get a racked data from participants who got tired and felt fatigued. you would always report hearing a tone that was relatively far above threshold, and almost never report hearing a tone that was well below threshold. In between, however, you would hear other, lower intensities on only a few presentations. In general, the intensity at which a stimulus would be detected 50% of the time would be chosen as your threshold. (ex) detecting a light spot a set of predetermined trials - use strong lights --visible - use very dim lights -- invisible - use lights in between - present all above in random order - fit the curve, and find the intensity that corresponds to 50%( or chance) accuracy. - its graph forms a sigmoidal shape.)

Gustav Fechner

-invented psychphysics and is thought by some to be the true founder of experimental psychology. -is best known for his pioneering work relating changes in the physical world to changes in our psychological experiences. -exact quantitative relation between matter and mind.

what is criterion ?

-it is bias where your cutoff between your response option is placed -biased observer fixed d' (sensitivity), changing criterion.

Describe the components of a neuron and their functions as well.

1) Dendrites : Receives information from other neurons 2) Soma: does most of the computation of information received from pre-synaptic neuron 3) Axon: passes neural impulse on to post-synaptic neurons

what are the 2 types of thresholds?

1) absolute threshold (a threshold for detecting a stimulus (stimulus presented or no?) ex) 2 pencil tips on the skin Uneven sensitivity, fingertips vs elbow ** there is no true of stimulus level in truth, you really have to decide what %of threshold you target into. 2) difference threshold: threshold for detecting the difference between two stimuli (difference or no?)

what are the four possible outcomes under signal detection theory?

1) false alarm 2) correct rejection 3) hit 4) miss

example questions regarding Fechner's law.

1) if light intensity is physically doubled, does it look twice as bright? 2) if a pencil is physically doubled in lenght, does it look twice as long? 3) if an electric shock is physcially doubled in intensity, does it hurt twice as much ? according to Fechner's law, the answer should be "less than double". that is, if light intensity is physically doubled, does it look twice as bright? no, less if a pencil is physically double in length, does it look twice as long? nO, LESS If an electric shock is physically doubled in intensity, does it hurt twice as much ? no, less

3 methods of measuring thresholds

1) method of limits 2) method of adjustment 3) method of constant stimuli

the sequence of action potential (starting at -70mV)

1) stimulus 2) Depolarization (-70mV -> +30mV, Sodium gates close) 3) Repolarization - Potassium gates open - Active sodium and potassium pumps 4) Hyper-polarization 5) Rest potential

5 methods to study of the senses

1. Thresholds 2. Scaling - measuring private experience 3. Signal detection theory- measuring difficult decision 4. Sensory Neuroscience 5. Neuroimaging - an image of the mind.

measurement and how do we measure ?

1. sensation : sensory limits - (original emphasis) > what is dimmest light you can detect? > what is smallest difference in pitch you can distinguish? 2. Perception: what is out in world? > which object is farther away? > is one object bigger than another? how do we measure? (1) psychophysics and (2) signal detection theory.

Traditional remedy (catch trial)

= the design of a clinical trial in which part of the experiment would be to include blank items as a variable (the catch). For example, there is an interval in an auditory test (e.g. audiometry) wherein no actual signals are given but the participant's responses are recorded anyway. See blank trial. CATCH TRIAL: "A catch trial helps correct the possibility of guesswork, especially in tests which require a simple yes-or-no response." - if feedback is provided, then this may make the subject overly conservative (being too strict on themselves, saying less yes) - if feedback is not provided, then the experimenter can use performance on catch trials as a measure of response bias, but it remains unclear how much the threshold is influenced by the responsebias.

Signal Detection Theory

= the detection of a stimulus depends on both the intensity of the stimulus and physical/psychological state of the individual. - signal detection helps us understand how we make decisions under conditions of uncertainty. - Quantifies the ability to discern information = bring patterns (signal) and random patterns that distract from the information (notice, consisting of background and random acitivity in the nervous system) - measures obtained from a series of presenations - d' (sensitivity index) - Criterion of observer (response bias) - separate sensitivity from response tendencies - criterion (bias) vs sensitivity - characterizes sensitivity independently of response bias. ** sensitivity (internal perceptual effect) , but there is always a noise distribution ** ex) internal : close eyes in the dark room external : reading mammogram and misread the picture ** Near threshold, it is difficult to tell a real stimulus from a particularly vigorous surge of internal noise **

method of adjustment

=a method of limits in which the participant controls the change in the stimulus. ex) detecting a light spot =subject in control of a knob - turning a knob up until just invisible - turning the knob down until just invisible - back and forth - take the average pros: relatively quick and easy to run cons: if your participants are unreliable, your data could be bad (therefore, you need to make sure your participant understood the process and whether he/she was motivated to give you false results)

Weber fraction

He said that " ...in observing the disparity between things that are compared, we perceive not the difference between the things, but the ratio of this difference to the magnitude of things compared." -ΔI / I (standard intensity)= k (or JND / I = k) -the constant of proportionality in Weber's law - ex) the length of two lines : the detectable change ratio was 1:100.

is there an absolute level that is just barely detectable?

No, a probability of detection that increases with signal intensity.

Fechner's law

P = k log S P = Perceptual or sensory magnitude k = constant S = physical intensity (stimulus) - focused on the relationship between the physical magnitude of a stimulus and its perceived intensity = if JNDs are the just noticeable difference, each new perceived intensities is +1 JND) (weber focused on the relationship between the JND and the standard) a principle describing the relationship between stimulus and resulting sensation that says the magnitude of subjective sensation increases proportionally to the logarithm of the stimulus intensity. our psychological experience of the intensity of light, sound, smell, taste or touch increases less quickly than the actual physical stimulus increases. ** Both Weber's and Fechner's laws state that we become less sensitive to difference as stimulus intensity increases ** **the graph is not linear; it is a graph of log. Fechner's law graph represents a line of log graph, indicating the magnitude (the differrence between two standard stimulus on the x-axis) becomes larger as the intensity of stimulus gets higher. (on the y-axis) ** Fechner's law is basically the same as Weber's law in that they both talk about JND being increminally different and the ratio between JND or difference stimulus and standard stimulus are the same which forms the constant k.**

Steven's Law

P= k S^n P= perceptual or sensory magnitude S= physical intensity (stimulus) k = constant n = exponent, specific to each sensory dimension. (1) n<1 ; the graph becomes compressive (ex) brightness (2) n=1 ; the graph is vertical (ex. apparent length) (3) n>1 : the graph is exaggerating. (ex. electric shock) log(p)= nlog(s)+ log(k) this is a linear function - independent variable: log (S) - Dependent variable: log (P) - k is a constant takes into account the type of sensory modalities (ex. audio, visual, etc) - the magnitude of subjective sensation is proportional to the stimulus magnitude raised to an exponent - takes the specific stimulus into account - take log of both sides to get linear function

Compare those psychophysical laws.

Weber & Fechner's laws: : Can only explain cases where people become less sensitive to difference as stimulus intensity increases. Steven's law : - more general; can explain situations where people are more, less, or equally sensitive to differences as stimulus intensity changes.

You need to determine a subject's absolute threshold for detecting light. You don't trust the subject to turn the light dial themselves, and you also don't have time to pre-select a bunch of light values and present them randomly to the subject. Which of the following methods should you use?

a) Method of limits b) Method of adjustment c) Method of constant stimuli d) None of the above. These can only be used for difference thresholds. answer is a.

method of limits

a psychophysical method in which the particular dimension of a stimulus, or the difference between two stimuli, is varied incrementally until the participant responds differently. - experimenter begins with a set of stimuli that vary in intensity but then increases or decreases the intensity of these stimuli until the stimuli are able to reflect thresholds. ex) detecting a light spot In each trial, present a light spot, ask for a response, and then incrementally - Increase intensity until just visible - Decrease intensity until just invisible - Back and forth - Take the average, this is the 50% threshold -experimenter manipulates the volume of intensity -participants respond with yes or no -trial number alternate going up & down in intensity pros: you are not wasting time for using different thresholds that would not reflect a participant's threshold (Single threshold) cons: subject bias; you need to determine whether your participants are motivated to give you false results, which can rack your results

On different trials you are presented with sounds of varying loudness. The task is to report when the sound is audible. Does the threshold in this task reflect one's hearing sensitivity? q1) would those with good hearing do well in this task? and those with bad hearing do worse? q2) how about the other way around? Does a low threshold in this task indicate good hearing?

a1) yes, a person with good hearing may have a low threshold, meaning they might have a high sensitivity. a2) no, they might lie to you in order to meet your expectation. in other words, having a low threshold does not indicate one's having good hearing. (Response problems)

we typically make a distinction between units of physical entities and measures of people's perception

ex) we measure the physical intensity of a sound in decibels, but we refer to our sensation as loudness. ex) frequency is a measure of a physical phenomenon, while pitch describes a psychophysical response to that physical phenomenon.

what is sensitivity (d')?

how good are you at picking up the signal.

analysis of a signal detection theory

if you want to increase the likelihood of winning $50 by getting a right answer, you may want to shift your criterion to the left to maximize the 'hit rate while you are also increasing the 'false alarm' rate (you are being liberal) if you want to be in a safe area and do not want to risk of losing any, then you will shift your criterion to the right on the graph to maximize 'correct rejection' rate and to minimize miss rate. let's say that the reward for the hit is $5 compared to -$50 of penalty for the miss, whereas the money for each false alarm and correct rejection are each -$50 and $5. =you want to try to shift your criterion as far right as you can to avoid the risk of the miss and false alarm.

according to Weber's law, if the relationship between stimulus intensity and psychological intensity were linear, what would the JND looks like?

it would keep increasing. (45degree vertical line)

What is response problem?

response tendencies or strategies influence threshold. example) descending limits --> subject keep saying yes on every trial even though after a certain level they no longer hear it. Subject trying to look good.

Form a 2x2 table using two IVs: response and stimulus

stimulus (on) response (on) = hit stimulus (on) response (off) = miss stimulus (off) respone (on) = false alarm stimulus (off) response (off) = correct rejection

Q. What stimulus intensity lies at the next psychological step?

stimulus intensity :8,12,18,27,40.5 change in stimulus intensity :4,6,9,13.5 the answer is 40.5+ (0.5x40.5)= 60.25

Q. What's the JND for the next psychological step ?

stimulus intensity 8,12,18,27,40.5 change in stimulus intensity 4,6,9,13.5 the answer is 0.5x40.5= 20.25

Sensation

the ability to detect the pressure of a finger and, perhaps, to turn that detection into a private experience your own sensory experience is directly accessible only to you.

Perception

the act of giving meaning and/or purpose to those detected sensations.

two-point touch threshold

the minimum distance a which two stimuli are just perceptible as seperate.

psychophysics

the science of defining quantitative relationships between physical and psychological(subjective) events. study of sensations evoked by physical stimuli (e.g. light intensity, mechanical pressure) dependent variables: (1) accuracy and (2) reaction time