perception and cognition - applied cognition (module 3)

what does applied cognitive psychology do

- uses theories of cognition to inform solutions to real world problems
- allows measurement of cognitive processes that produce behaviour
- finds novel solutions to problems that modify behaviour
- furthers our understanding of cognition in the wild

features of applied cognitive psychology

- theory driven
- experimental
- problem focused
- representative

feature integration theory

focused attention is not required to detect individual features of a stimulus, but is required to bind those individual features together

marr (1982) three levels of explanation for cognitive science

- computational problem being solved
- algorithm/model used (software)
- how algorithm/model is implemented in the brain (hardware)

implication rates of eyewitness misidentification

60%

stimulus representativeness

task should use faces and individuals representative of real world targets and foils that might appear in lineups

task representativeness

task should use a format and have parameters that are as much as possible similar to those experiences by real witnesses (brief exposure duration, long retention intervals, incidental encoding)

traditional metaphors of memory

- typical laboratory paradigm
- storehouse metaphor

typical laboratory paradigm

- list learning experiment
- memory performance measured % of items recalled/recognised

storehouse metaphor (quantity-oriented)

- memory as a storage place
- focus on accuracy of reports
- memory assessed in output bound manner
- forgetting is loss of correspondence

correspondence metaphors of memory

- everyday memory research
- correspondence metaphor (accuracy-oriented)

everyday memory research

- e.g. eyewitness report of a crime
- memory performance measured by faithfulness to past events

correspondence metaphor (accuracy-oriented)

- memory is about past events
- focus on accuracy of reports
- memory assessed in output bound manner
- forgetting is loss of correspondence

recall--recognition paradox

recognition better in the lab, recall better in eyewitness studies

koriat and goldsmith 1994 experiment 1

assessed memory by quantity and accuracy through:
- forced recall
- free recall
- forced recognition
- free recognition

the PEACE model in investigative writing

- preparation and planning
- engage and explain
- account clarification and challenge
- closure
- evaluation

cognitive interview pros

- build rapport, foster control over report
- context reinstatement
- open ended report
- followed by focussed report
- reexamine report from different perspectives

cognitive interview evaluation

- meta-analysis suggests 34% increase in accuracy
- suggests criterion change
- some research shows interview no more effective than standard interview
- need to understand which ingredients drive increase in accuracy and at what cost

monitoring witness (memory content/analytic cues)

highly detailed memories more likely to be correct

monitoring witness (perceptual fluency/inferential cue)

ease of perception of an item at encoding/test

monitoring witness (retrieval fluency/inferential cue)

ease of remembering an item from memory

control over report option

volunteering info or saying dont know

control over grain size

fine detail v general information

cognitive interview

give control to witnesses via transfer of control instructions

fine witness info

i.e. specific hair colour (black, brown, blonde, grey)

coarse witness info

i.e. shade (light or dark)

current best practice in interviewing

- self administered interview
- uses techniques but no interviewer required
- validated across contexts

why apply cognition to space

- humans required for space missions
- artemis program
- testing cognitive performance in extreme environments helps adapt to future earth
- safety of habituation is dependent on supporting cognition

threats to cognitive abilities in space

- sleep deprivation
- radiation bursts
- spikes in C02
- changes in microbiome/immune system
- noise
- extreme changes in temp and humidity
- microgravity

benefits of space research to earth life

- better understanding of human cognition and augmentation with technology
- translation of learning to support people with loss in functional capacity
- design of spaces and habitats to support human cognition (biophilic, light, physical dimensions)

inner circle cognitive distractions

low demand i.e. listening to radio

triangle sources of distraction

moderate demand i.e. talking to a passenger

high demand sources of distraction

manipulating info on secondary device i.e. phone/tablet

features of attention

- input stimuli
- attenuating filter
- selective filter
- response process

inattentional blindness

inability to see objects directly in field of view because attention directed to another object

measuring cognitive workload and distraction

- recognition memory
- DRT paradigm
- ERPs and fNRIS

pros and cons of recognition memory

pros
- easy to administer
- distinguish between cognitive and visual distraction with eye tracking
cons
- cannot distinguish failure to encode from failure to retain info

pros and cons of DRT

pros
- no self report, real time measures, minimal addition of load from measurement
cons
- simple measures that does not specify source of load

fNIRS pros and cons

pros
- relatively non-invasive and temporally accurate measure
cons
- correlational only (cannot tell why there is an increase)

adaptation to vehicles to reduce distraction

- barring use of touch screens
- break assist and alert
- restrictions on passengers with inexperienced drivers (p1/p2)

are all vulnerable to cognitive distraction

- some evidence that some are better multitaskers
- experience level does matter
- working memory capacity decreases impact of cognitive distraction on RT

where is the cognition in sport

- attention and spatial awareness
- speeded info processing
- memory and expertise
- emotion regulation
- judgement and decision making
- motor imagery
- reasoning and problem solving

speeded info processing

- carrying out quick decisions under time pressure
- athletes have been shown to have superior processing speed

motor imagery

- defined as the mental representation of action without concomitant body movement

mechanisms in motor imagery

- areas of the brain involved in actual movement and imagined movement are largely overlapping
- implications on performance and rehabilitation during injury

emotion regulation

- regulation of stress response so it is not overwhelming

mind-body cognition

- embodied cognition
- many features of cognition shaped by aspects of organism's entire body

neurovisceral integration model (thayer and lane 2000)

- how to explain relationship between physiological experience of anxiety and cognitive impact it may have
- physiological adaptation to stress should also allow up in personality measures

heart rate variability (HRV)

- consists of changes in time intervals between consecutive heartbeats called interbeat intervals (IBI)
- lower HRV associated with various forms of mental and physical health problems

psychological flexibility in sport

- ability to regulate stress and emotion
- measure of acceptance and avoidance
- individual athletes more psychologically flexible than team sport athletes (d=.44)

types of machine learning (ML) models

supervised and unsupervised

types of supervised ML

classification or regression

psychology tasks conquered by ML

- classification of mental health conditions
- identification of suicidality risk
- chat bots for counselling
- risk assessment of offenders
- facial recognition
- radiology (diagnostics)

humans v machines thinking differences

- humans satisfice machines optimise
- humans take a bet on most salient cues
- machines take in all data and combine it with a model to minimise a loss

XAI

explainable AI

what is XAI

- global explanation of the model in terms of the most important rules from the training data that translate to parameters and output

problems with xai

- only approximate the underlying method used to produce the output
- impractical to implement
- simply not possible
- illusion of explanatory depth (IOED)

when is it unethical to develop ai

- no understanding of scope of problem being solved
- no understanding of consequences of solution in broader society
- no real problem to begin with
- no understanding of human cognition in cotext

bounded optimality

theory for designing optimal programs for agents with performance- limited hardware that must interact with environments in real time

limitation of notion of boundedness

stands in contrast to idea of embodied cognition and extended mind thesis