2. Information Theory

Question 1

information theory

Answer 1

• foundation of cognitive psychology
• thoery acts as metaphor, as many assumptions are not always met
• serve as channel
• refer to diagram and explanations on page 6 of notebook

Question 2

noise

Answer 2

anything not relevant to the task at hand

Question 3

early evidence for information theory

Answer 3

refer to slide 11 of powerpoint for early evidence

Question 4

what approach to psychology could be linked to information theory ?

Answer 4

behaviourist appearance (SR relationship)
- S=transmitter message
- R=recipient’s message channel, blackbox

Question 5

historical origins: Hartley

Answer 5

• in 1920s, telecommunication companies sought balance between rate of transmission and intelligibility
• refer to slide 8 for example
• Hartley defined amount of info in a message as:
  1. nb of available symbols (binary, alphabetical, etc)
  2. nb of symbols in a particular message (sentences can be short, long depending on the message

Question 6

binary input vs alphabetical input: what is the difference?

Answer 6

binary input has less input compared to alphabetical input, therefore, the message is less likely to be lost

Question 7

historical origins: Shannon (1940s)

Answer 7

• focus on structure and organization, linking it with physical laws
• terms developed from cryptography - info encoded, then decoded
• applied to are of transmission, channel capacity, signal/noise discrimination (works at social level too)
• considers probability and uncertainty
• high probability events carry less info than low probability events (bird vs falcon spotting)
• info characterized by decrease of uncertainty

Question 8

misinformation vs disinformation

Answer 8

misinformation
- unintentional
disinformation
- intentional, trying to deceive
these work because they act as they are real information, they decrease uncertainty in the wrong way

Question 9

issues with information theory

Answer 9

• initial conception focused on structure, not content - excludes content
• does not account for existing info
• requires many conditions that must be satisfied
  1. estimates probability of events (impossible in real world)
  2. amount of info needs to be critical factor

Question 10

Information theory fills the black box with information, similar to behaiourism… What happens between SR in the channel?

Answer 10

idk

Question 11

applications of information theory

Answer 11

models of decision making
- probability and simple decision making
- accumulation of evidence (source of RTs and its variability)
- helps us understand what makes decision making difficult
information availability and RTs
- increase in RT will result from nb of response alternatives (Hick’s law)
- the more options we need to consider, the longer it will take

Question 12

sequential sampling models

Answer 12

• consider accumulation of evidence over time, after an find. obtains a critical amount of evidence (or time runs out!), they make a decision

Question 13

sampling evidence from sequence: sequence

Answer 13

• we assume neurons fire and go into refractory period
• nervous system shoots out info in packets, in a sequence
• consequently, we sample the environment
• nature of process will determine info we can tolerate (RT)

Question 14

simple (?) decision-making examples

Answer 14

• refer to Tinder exmaple, slides 15-16-17
• refer to drink example, slides 18-45

Question 15

evidence accrual

Answer 15

recording of evidence

Question 16

evidence accrual is determined by

Answer 16

1. innate bias
   - LTM; experience and preference
   - endogenous attention (e.g. “It’s loud! it smells!”; activates our sense towards things we prefer)
   - in decision-making research, we find that people look for confirming evidence
   - source of info, source credibility
   - influences our perception
2. limited nb of items can only be retained in memory for a limited time
   - working memory, decay
3. evidence accumulated on previous trials
   - attention
4. evidence accumulated on most recent trial
   - working memory; retention

Question 17

confirmation bias

Answer 17

if something is below our threshold of awareness, we go with confirmation bias

Question 18

channel capacity

Answer 18

• limited nb of items can only be retained in memory for a limited amount of time
• retaining all info would be burdensome for our cognitive and perceptual systems
• 7+/- = chunkable
• 4 is not chunkable

Question 19

accumulated evidence: most recent trial

Answer 19

• last evidence may be most salient
• e.g. car insurance - usually recent evidence = weighed more than older evidence
• primacy affected by long term memory
• recency affected by working memory (as long as it doesn’t exceed capacity)

Question 20

accumulated evidence: previous trial

Answer 20

• more of something in a previous trial might displace memory for something else in current trial
• often we don’t think of a reason to disconfirm erroneous-memory

Question 21

determinants of RT

Answer 21

• the more evidence, the more events must be counted, the loner it takes
• more alternatives we have = higher RT
• smaller the difference between alternative, the longer the RT - sorting through basic categorization process
• both interact to determine response time
• difficulty with only thinking about the nb of internal events
• doesn’t,t consider what happens after the evidence = accumulated (i.e. a response must be selected)

Question 22

Hick’s law

Answer 22

• RT increases with nb of response of alternatives but reaches an asymptote
• refer to slide 54 of powerpoint for graph

Question 23

why do assumptions matter

Answer 23

• extent to which attention is allocated to a target stimulus will determine the amount of information encoded
• when encoding an item in either short/long term memory, evidence must be accumulated for it
• complex cognition (e.g. language, problem-solving, intelligence) are reliant on more fundamental processes