Methods and Approach of Cognitive Science

1
Methods and Approach of Cognitive Science

• Brandon Beltz
• February 2, 2005

2
The Reductionist Approach

• An attempt to understand complex events by
  breaking them down into their components.

3
Empirical

• Relying on or derived from observation or
  experiment

4
Common Assumptions of Cognitive Science

• Cognitive processes can be partitioned such that
  individual processes can be studied in isolation
  (reductionism)
• (e.g. language can be studied separately from
  memory)
• Focus is on the individual and their natural
  environment.
• De-emphasizes the roles of culture and society.
• Cognitive processes are assumed to be somewhat
  autonomous from noncognitive capacities (e.g.
  affect, motivation, etc.)
• It is useful (and meaningful) to distinguish
  normal from abnormal cognition

From Von Eckart (1993) Braisby and Gellatly (in
press)
5
Common Assumptions of Cognitive Science

• 5. We can talk of typical cognitive processes,
  which generalize across people.
• Adults are sufficiently alike.
• Answers to basic, empirical questions can be
  given in terms of information processing
• Answers to basic, empirical questions should be
  justified on empirical grounds
• 8. Answers to the basic, empirical questions must
  be constrained by the findings of neuroscience.

From Von Eckart (1993) Braisby and Gellatly (in
press)
6
Guiding Principle

• The information processing approach acts as a
  metatheory which influences the way cognitive
  psychologists think about the mind.

Metatheory
theory
theory
7
Information Processing
8
information

• Knowledge derived from experience, study, or
  instruction.
• A collection of facts or data.

9
processing

• A series of actions, changes, or functions
  bringing about a result
• The process of digestion the process of
  obtaining a driver's license.
• Factory example

10
Analogies from Technology

• The channel capacity analogy
• Like a telephone wire, humans are
  limited-capacity information processors.
• Highlights processing capacity
• The computer analogy
• Human information processing may be similar to
  the way computers work.
• Hardware the Brain
• Software thought
• Highlights stages of processing

11
Measuring Information Processing

• Reaction time (RT)
• The time elapsed between some stimulus and the
  persons response to the stimulus.
• Typically measured in milliseconds.
• 1000 milliseconds 1 second.

12
Reaction Time Experiment
13
Experiment Notes

• 18 GMU undergraduates participated.
• Participants pressed a button on a computer
  keyboard every time they saw an X on the screen.
• Computer measured the time between the appearance
  of the Xs and the button presses.

14
In which condition are reaction times longer?
Constant Timing
Random Timing
15
Experiment Results
16
Average Response Times
Average Response Time (ms)
Constant Timing Random Timing
Participants
17
Reaction Time Across Trials for One Participant
Constant Timing
Random Timing
Reaction Time (in milliseconds)
0
1100
0
1100
Trial
Trial
18
Discussion of Results

• When people know what to expect reaction times
  are faster.
• Learning and memory processes help reduce
  reaction times.
• Issue of expertise
• Application of lab results to real world
• Driving

19
Measuring Information Processing

• Accuracy Measures
• How many errors subjects make.
• An example is how many words a subject correctly
  recalled and which were omitted.
• Like recalling someones name
• Class RT example how many used the wrong hand in
  the switch over condition?
• Often measured as proportion or percent correct.

20
Speed-Accuracy Trade Off

• As speed increases accuracy decreases.
• Very common finding in human performance.

Accuracy
Speed
21
The Standard Theory of Human Information
Processing (1970s)
22

• Standard theory nice general framework, but
  researchers needed finer detail on some problems.

23
Process Models

• A hypothesis about the specific mental processes
  that take place when a particular task is
  performed.

24
The Lexical Decision Task

• Yes or no, do the following letter strings form
  valid English words?
• 1) Robin _____
• 2) Manty _____
• 3) Nurse ______
• 4) Trock ______

25
Lexical Decision Your Turn
26
List 1

• girl____
• hoose____
• chair____
• truck____
• spoog____
• hairn____
• nurse___
• whig____
• piano____

• donkey____
• carv____
• tape____
• klen____
• dime____
• hand____
• polt____
• phone____

27
List 2

• red___
• blue____
• kiad____
• olaf____
• apple____
• zioc____
• pear____
• rain____
• snow____

• vrin____
• salt____
• pepper___
• rose____
• sild____
• drib____
• tulip____
• ahef____

28
List 3

• sleep____
• smick____
• bed____
• bibi____
• thief____
• steal____
• durg____
• opic____
• spider____

• flih____
• web____
• foot____
• polt____
• shoe____
• tonf____
• boy____
• girl____

29
A Process Model of the Lexical Decision Task
30
Assumptions of Strict Information Processing

• Sequential stages of processing
• Independent and non-overlapping stages of
  processing

31
Difficulties of Strict Information Processing

• Solid evidence exists for parallel processing
• Context effects
• Not useful for studying complex cognition
• e.g., problem solving tasks
• processes on longer timescales

32
Parallel Processing

• Cognitive processes operate at the same time

33
Context Effects

• Count the number of Fs in the following
  sentence
• Finished files are the result of years of
  scientific study combined with the experience of
  many years.

34
Top-Down Processing

• When existing context has an influence on earlier
  or simpler forms of mental processes
• In the finished files example, reading is so
  automatic that youve learned to ignore function
  words like of-- an example of top down
  processing.

35
Updating the Standard Theory
36
Seven Themes of Cognition
37
Themes of Cognition

• How do seven themes of cognition relate to
  schools of thought we have covered in class?
• What are some specific real-life examples related
  to each theme?

38
A Short Lesson in Brain Anatomy
39
Neurons
40
Inter-neural Communication
Axon, terminal button
Synapse
Neurotransmitters
http//www.le.ac.uk/li/sources/subject8/
41
Neural Activation

• All or nothing firing
• Input from dendrites must exceed certain
  threshold in order for nerve to send signal down
  its axon.
• Neural activation is a chemo-electrical process
• Electrical in polarity changes of cell
• Ion exchanges
• Changes of cell chemical ratios
• Sodium (Na), Potassium (K)
• Diffusion of chemicals
• Neurotransmitters

42
Neuron Anatomy

• Soma
• Nucleus
• Dendrite
• Axon Hillock
• Axon

• Myelin Sheath
• Terminal Buttons
• Synapse
• Neurotransmitters

43
Neural network
http//www.alanturing.net/
44
Cerebral Cortex

• Layer of cells that covers the outside of the
  brain.
• Wrinkled appearance
• Approx 3 millimeters thick
• When stretched, covers about 2.5 square feet

45
Lobes of the Cerebral Cortex
Frontal Lobe (Front of head)
Parietal Lobe
Occipital Lobe (O for back of head)
Temporal Lobe (Cross of T)
http//www.ling.mq.edu.au/units/sph307
46
Functional Division of Cerebral Cortex
47
Subcortical Brain Structures
Cerebellum
48
Subcortical Brain Structures(the real thing)
49
Brain Structures Summary

• Subcortical Structures
• Corpus Callosum
• Thalamus
• Hypothalamus
• Hippocampus
• Amygdala
• Medulla
• Cerebellum

• Cerebral Cortex
• Frontal Lobe
• Parietal Lobe
• Occipital Lobe
• Temporal Lobe

50
Examining the Brains Role in Cognition
51

• Possible to separately study the hardware from
  the software?

52
Localization

• If evidence for a cognitive process is supported
  by empirical behavioral studies, that process
  should be represented and localized in the brain.
• Supported by methods that look at brain
• Damage
• Activation

Eureka! This is why I fall asleep in class.
53
Brain Damage

• If brain area x supports cognitive process y,
  damage to area x will affect process y

54
Brain Activation

• If brain area x supports cognitive process y,
  brain area x will be activated during process y

55
Methods of Investigation

• Electrical Activation
• Single cell recordings
• Electroencephalograms (EEG)
• Event-related potentials (ERP)
• Imaging Technology
• Computerized axial tomography (CAT)
• Positron emission topography (PET)
• Functional magnetic resonance imaging (fMRI)
• Lesions
• Examine damage during autopsy
• Documented brain surgery

56
Electrical Activation
EEG, ERP
http//www.bsp.brain.riken.jp/Photos/
http//www.theory.org/fracdyn/neurodyn/eeg-traces.
gif
57
Imaging
http//www.atr.co.jp/html/topics
http//www.mch.com/clinical/radiology/fmri/images/
fMRI
58
Dissociations

• A disruption in one cognitive process but no
  impairment of another.
• Usually seen through memory studies
• Studied through patients with brain lesions
• Simple versus double dissociations

59
v
v
X
X
Association
v
X
Simple Dissociation
v
X
?
v
X
Double Dissociation
v
X
60
Dissociation Example
Simple Dissociation
Double Dissociation
From Gazzaniga, 2004
61
v
v
X
X
Association
v
X
Simple Dissociation
v
X
?
v
X
Double Dissociation
v
X
62
Split Brain Research

• Sperry -- severed corpus callosums of people with
  severe epilepsy.

Right Hemisphere
Left Hemisphere
KEY
RING
sees word
says ring
interacts with key in left hand
performs action
63
Basic Principles of Brain Functioning

• Contraterality
• Hemispheric Specialization

64
Contralaterality

• The receptor and control centers for one side of
  the body are in the opposite hemisphere of the
  brain.

65
Hemispheric Specialization

• Different brain functions tend to rely more
  heavily on one hemisphere or the other.
• Example The left hemisphere controls language
  for most right-handed people.

66
Hemispheric Specialization
67
Lecture Summary- 1st Half

• Cognitive psychologists often use an empirical,
  reductionist approach to studying cognition.
• There are many assumptions made within cognitive
  psychology which affect what conclusions
  scientists can make.
• Information processing is an important metatheory
  which guides thinking about cognition.
• Time, accuracy measures
• Like all metatheories, information processing has
  changed in the past few decades
• Context effects, top-down processing, parallel
  processing (later class)
• There are many common themes of cognition that
  will recur throughout this course

68
Lecture Summary- 2nd Half

• We discussed how neurons fire and covered many
  structures of the brain
• Information about cognitive processes being
  localized in the brain is supported with brain
  damage and activation studies.
• Many methods to this approach.
• Dissociations are valuable in exploring the
  relationships between and limits of cognitive
  processes.
• Contraterality and hemispheric specialization are
  basic features of brain functioning.
• Split-brain research