Basic Assumptions (Caramazza, 1984) - PowerPoint PPT Presentation

Loading...

PPT – Basic Assumptions (Caramazza, 1984) PowerPoint presentation | free to download - id: 72aeae-Y2U5N



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Basic Assumptions (Caramazza, 1984)

Description:

Basic Assumptions (Caramazza, 1984) Neurological specificity: organization of the mind parallels organization of the brain (monist philosophy) modularity of mind ... – PowerPoint PPT presentation

Number of Views:26
Avg rating:3.0/5.0
Slides: 21
Provided by: Psychol92
Learn more at: http://www.columbia.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Basic Assumptions (Caramazza, 1984)


1
Basic Assumptions (Caramazza, 1984)
  • Neurological specificity organization of the
    mind parallels organization of the brain (monist
    philosophy)
  • modularity of mind (functional) modularity of
    brain (anatomical)
  • Transparency the pattern of spared and impaired
    performance of the patient can lead to valid
    conclusions about the nature and function of the
    impaired processing components
  • brain damage allows one to spot more easily the
    workings of processes that are opaque in
    normals flawless performance
  • Subtractivity performance of the brain damaged
    patient is just like the normal, intact cognitive
    system minus those components that were damaged
    by the injury.
  • does not mean that behavioral compensation does
    not occur (using functions of intact modules to
    compensate for damaged ones)
  • but, implies that the mature brain does not
    develop new cognitive modules following injury
    (substantial reorganization)
  • Modularityfunctional and anatomical

2
A hierarchy of modules
  • Basic Modules
  • carry out a single type of computation
  • no interaction with other modules until finished
    with its process

3
Modularity Definitions
  • Fodor, Modularity of Mind, 1983
  • Informational encapsulation module functions in
    isolation from processes going on elsewhere
    makes them more efficient
  • Autonomous does not share processes with other
    modules
  • Domain specificity each module can only process
    one particular type of input (not a gatekeeper)
  • Mandatory (Automatic) and Fast
  • Innate determined by genetics (nature)
  • Hardwired not affected by nurture
  • Modularity, as adopted by cognitive
    neuropsychology
  • Domain Specificity
  • Informational encapsulation
  • can be selectively impaired (single/double
    dissociations)
  • remains intact in the face of gross intellectual
    decline (dementia)
  • not affected by beliefs, desires, or expectations
    (Coltheart)

4
Task Dissociation LogicAssociation Patient X
shows a deficits on task A and B
  • Patient Xs damage could have impaired a single
    functional module shared by task A and B
  • Patient Xs damage could
  • have impaired two
  • separate functional modules
  • Modules could be anatomically adjacent (e.g.,
    Gerstmanns syndrome), or not, if damage is
    widespread

5
Task Dissociation Logic
  • Single dissociation Patient X (case or group) is
    significantly more impaired on Task A than Task B
    relative to a control group.
  • Dissociations are considered to be stronger
    if
  • Task A and B are of equal difficulty (avoid
    resource artifact)
  • Patient(s) is/are normal (as good as controls)
    on one task.

6
Task Dissociation Logic
  • Double dissociation Patient X (case or group) is
    more impaired on Task A than B (relative to
    controls) Patient Y (case or group) is more
    impaired on Task B than A.
  • Considered gold standard of cognitive
    neuropsychology research

7
Alternatives to modularity
black box module
vs.
  • Connectionist architectures
  • Processing is graded, distributed and interactive
  • Conceptually more similar to neural populations
  • When distributed systems are lesioned they can
    sometimes simulate neuropsychological impairment
  • (Farah McCelland, 1991 Farah, 1994)

8
Alternatives to modularity
  • Compromise
  • Distributed systems inside of modular boxes
  • Cascading processes

9
Methods Case vs. Group Studies
  • Case Studies assess behavior of an individual
    subject who demonstrates a highly specific
    pattern of deficits
  • Vs.
  • Group Studies assess average performance of a
    group of patients similar with respect to area of
    brain damage, etiology or disorder (i.e.,
    amnesia, anomia).

10
Case Studies
  • Advantages (argued by advocates)
  • often provides evidence of highly specific
    cognitive modules
  • fine-grained analysis (conjunctions of results
    from multiple tasks) can be used to develop
    highly sophisticated models of cognitive systems
  • less susceptible to Type II error
  • Disadvantages (argued by detractors)
  • diffuse etiologies make localization difficult
    (where is the neuro in this type of
    neuropsychology?)
  • localization of cognitive function to particular
    brain regions is secondary (radical view), or
    even unimportant (ultra view)
  • why should brain damage fractionate along
    cognitive lines?
  • questionable generalizability and replicability
  • selection bias select patients on the basis
    of whether they demonstrate a particular deficit
    as proof of a theory that the deficit represents
    a modules
  • problems with Type I error
  • how do you know whether a single individual is
    truly representative of the normal population
    (premorbidly atypical subjects)

11
Group Studies
  • a priori grouping small groups (N gt 15)
    pre-selected to have overlapping/ homogeneous
    lesion or disorder
  • If system is modular it is only important there
    is a region of overlapping damage, not that there
    is heterogeneity outside this region

12
A priori grouping
  • Advantages
  • good localization
  • good generalizability/predictability
  • some heterogeneity in patients means that when
    if effect is found it must be a large and robust
    enough effect to overcome the uncontrolled
    variance.
  • Disadvantages
  • inter-subject variability could lead to Type II
    error
  • even if homogeneous with regard to lesion or
    clinical disorder, may not be homogeneous with
    regard to experimental tests
  • dont give enough behavioral tests to get at the
    heart of the problem

13
Group Studies
  • a posteriori grouping large groups (N lt 15) with
    more heterogeneous lesions, grouped by behavior

Left frontal damage
a) With apraxia of speech
b) Without apraxia of speech
14
A posteriori grouping
  • Advantages
  • reduces between-subject variability by using it
    as blocking variable
  • Disadvantages
  • like case-studies, this advantage depends on
    use of multiple tests and specificity of those
    tests (how good they are at tapping into
    dissociable behaviors)

15
Sources of variance
  • Neurological
  • Age
  • No two brains are exactly alike (like
    snowflakes?)
  • Occult brain conditions (undetected brain damage)
  • Differences in lesion size
  • Differences in pre-morbid organization
  • Differences in post-morbid plasticity
  • High-resolution structural and functional imaging
    can provide some insight into these sources

16
Sources of variance
  • Behavioral
  • Different approaches to the assessment task
  • distraction, strategy, motivation

Memory Performance ( Correct)
Memory Performance ( Correct)
17
Sources of variance
  • Behavioral
  • Different approaches to the assessment task
  • personality differences, cultural differences

18
Etiologies (causes) Diffuse
  • Head Trauma
  • Open, penetrating (rare focal lesion, though may
    see some diffuse signs)
  • Closed, coup-contrecoup (common diffuse lesions
    white matter disconnection)

19
Etiology (causes) Focal
  • Vascular Disorders(best for group studies)
  • Cerebro-vascular accident (CVA) hemorrhagic
    (aneurysm) or obstructive
  • Oxygen deprivation (anoxia/ischemia) and
    apoptosis (cellular suicide)
  • Surgical Resection
  • Tumors
  • Epilepsy
  • AVM

20
(No Transcript)
About PowerShow.com