Thinking 1 Flashcards
what is thinking?
the flexible organisation and manipulation of internal representations
what do rationalists emphasise?
- the constructivist nature of the human mind
- emphasise the importance of a priori existing concepts in the understanding of the sensory inputs we receive in order to perceive the world
what does empiricism emphasise?
- the importance of sensory observations to build this
- emphasise the sensory data
how does the brain think?
uses ‘algorithms’ and assumptions to actively construct an image of the world
- e.g. perceptual illusions, Gestalt laws - laws of continuity, law of closure etc
- brain tries to make meaningful objects from the sensory input
why does the brain make meaningful objects from sensory input?
- as we do not have unambiguous information coming through our senses
- incomplete “projection” into the brain, e.g. 3D-2D projections
- even colour can be ambiguous - blue/black, white/gold dress
- shadows looking like different animals
give examples of how the brain integrates specific observations (e.g. for visual object recognition: contour lines/ shape, colour) with all kinds of contextual information?
- the probability of the occurrence of a particular ‘object’ - depending on the narrative context, observers either see the young (‘wife’) or the old woman (‘mother in law’)
- other available information, e.g. the sound the object makes
what is the ventriloquist effect?
The perceived location of a sound is shifted in space by simultaneously occurring visual stimulus at incongruent location
what is the McGurk effect?
The perceived sound of a spoken syllable is altered by incongruent visual input of lip movement
who discovered perception as an unconscious inference?
Hermann von Helmholtz and Wilhelm Wundt
what is Bayesian cognition?
contradicting camps
- human cognition is based on Bayesian algorithms and this is what brains have evolved to do
- humans fail in taking prior probabilities (=base rate) into account
what are the 3 factors of probability?
- probabilities are non negative (real) numbers between 0 and 1 (0<p<1)
- The probability of the certain event is 1
- The probabilities of all separate events that comprise a set add up already and they add up to 1
what is conditional probability?
p(A I B) = the probability of B given A
the probability if a particular (hypothetical or real) ‘event’t, e.g. A=2 with in the set of another event
what is the Bayes Theorem equation?
p(H1 I O) = p(H1) X p (O I H1) / sum of p(Hi)p(O/Hi)
p(O I H1) = the probability of a certain observation O given that hypothesis H1 was true, also called the likelihood of observation O given H1
p (H1) = the prior probability of hypothesis H1 being true
p (O) = the prior probability of making observation O
p (H1 I O) = the posterior probability of hypothesis H1 being true given observation O being made
what is a bayesian inference?
to infer the observations to the (probability of the) hypothesis
3 key facts of Bayes theorem
- inference of ‘ground truth’ (the state of the world) on the basis of (limited) data always comes with uncertainty, since the likelihood, e.g. p (A I B) are virtually never o or 1. The data alone rarely ever tells us with 100% certainty whether Hypothesis 1 or 2 is correct
- in order to judge the best judgement / inference under uncertainty, the Bayes theorem is crucial
- For judging which hypothesis is the most likely, consideration of the prior probability is often crucial. individual observations are rarely enough, we need contextual information
what is Bayesian perceptual influence?
calculate the ‘posterior
- probability’ of perceptual hypotheses given the sensory evidence and prior probability of the causes
- sensory and prior information weighted according to their ‘precision’ to compute posterior belief (subjective percept)
how do we use prior/contextual information?
help influence the perceived level of threat from one and the same person
how do we increase the certainty about objects in the real world?
combine signals from different sensory modalities
- The Bayes theorem allows to derive how these ideally combine to come to a common judgement, in the same way as individual observations and prior probability are “combined”
what is a probability density function?
something that has a normal distribution also known as Gaussian distribution
- defines the probability function representing the density of a continuous random variable lying between a specific range of value
- SD is the width, wider pdf distribution
what is a cumulative density function?
- gives the probability that the random variable X is less than or equal to x
- never reaches one
- SD is the slope, flatter slope in cdf
what did Ernst & Banks 2002 do?
- visual and haptic 3D - virtual reality
- task is to compare the height of two ‘bars’ - presented in VR visually and haptically
- can measure the accuracy of judgment in visual domain (presented separately)
can estimate a person’s accuracy in each modality (separately) for vision also under various levels of noise - in combined presentations, judge how much their judgement depends on visual, how much on haptic perception
The psychometric function Ernst & Banks 2002
- 2 AFC: judge whether the second stimulus was larger than the first (“standard” stimulus)
- The psychometric function is often modelled as a ‘cumulative Gaussian function’
- slope of the psychometric function indicates the ‘performance’ of an observer:
- The better, the steeper
- This is captured by the SD (variance)
haptic and visual stimuli Ernst & Banks 2002
2 AFC (serial comparison)
1. “standard stimulus” S (55mm)
2. comparison stimulus ↑↓
measure the psychometric function:
- under haptic stimulation only
- under visual stimulation only
- at different noise levels
get different psychometric functions with different slopes
what happens if use combined stimuli
- present both visual and haptic
- when visual stimuli was different from haptic stimulus (kind of a visuo-haptic McGurk effect), allows to estimate the empirical weight with which each sensory modality contributed to the joint estimate
RESULTS: followed pretty much exactly the optimal weight, as derived from the Bayes Theorem
