Connecting Brain, Purpose

1
Connecting Brain, Purpose LanguageUnderstandin
g Language Behavioras the Control of Perception

• Gary A. Cziko
• University of Illinois at Urbana-Champaign
• g-cziko_at_uiuc.edu garycziko.net

2
(start recording)
3
Apologies (5)

• Not a brain or neuroscientist
• Will not summarize or synthesize findings from
  neural science
• Theory and computer simulations I will present
  are not my own (Wm. T. Powers)
• Relativity little research has been done to
  support theory and arguments I will make
• Dont have fancy PowerPoint slides

4
Excuses (3)

• Am interested in biological bases of complex
  human behavior education language
• Without Miracles (1995) Darwinian
• The Things We Do (2000) Bernardian
• Will share with you a fundamental building block
  for understanding how mind might be created from
  neurons
• Will do some research, simulations and data
  analysis with you to demonstrate some behavioral
  phenomena and their possible underlying neural
  basis

5
Goals of presentation (4)

• Explain how we can get human language from
  neurons (in well under an hour!)
• Provide a different perspective on behavior
• Neither behavioristic (S-R)nor cognitive (S-O-R)
  (lineal A -gt B causality)
• Behavior as purposeful (circular causality)
• Offer preliminary implications for understanding
  language and its underlying neural basis
• How we might get mind from neurons
• From the simplest unit of perception, thought
  behavior to the highest form of human perception,
  thought behavior

6
Presence of mentality(read-along quote)

• The pursuance of future ends and the
• choice of means for their attainment are
• the mark and criterion of the presence of
• mentality in a phenomenon. We impute no
• mentality to sticks and stones, because they
• never seem to move for the sake of anything,
• but always when pushed, and then indifferently
• and with no sign of choice. So we unhesitatingly
• call them senseless.
• --Wm. James, 1890 (American psychologist)

7
Behavior demonstratingpresence of mind

• The rubber-band demo
• What is the subject doing?

8
Computerized Rubber-Band Demo

• Powers computer demo of compensatory tracking
  (Demo 1 Step F)
• Correlation between (visible) Cursor and Handle
  (S-R)
• Correlation between (invisible) Disturbance and
  Handle

9
Towards a model(read-along quote)

• What we have is a circuit, not an arc or
• broken segment of a circle. This circuit is more
• truly termed organic than reflex, because the
• motor response determines the stimulus, just as
• truly as sensory stimulus determines movement.
• Indeed, the movement is only for the sake of
• determining the stimulus, of fixing what kind of
• a stimulus it is . . .
• --John Dewey, 1986 (American psychologist
  educator)

10
A model of compensatory tracking

• Demo 2 Step G closing the loop
• Explain components
• Behavior with open loop
• Close loop
• Vary reference level (purpose)
• Step
• Continuously

11
(No Transcript)
12
Characteristics ofPerceptual Control

• Provides a model of purposeful (intentional)
  behavior
• Internal reference level (goal, purpose,
  objective, intention) essential
• Perception (stimulus) affects behavior
  (response) AND behavior affects perception
• Circular rather than lineal (one-way) causation
• Perceptual control shown by LOW correlation
  between perception (input) and behavior
  (output)
• Test of the controlled variable
• Behavior understood as a means of controlling
  perception
• Not vice versa as in both behaviorism(S -gt R)
  cognitive psychology (S -gt O -gt R)
• No clear independent or dependent variable
• Circular not straight line (lineal) causality

13
A Hierarchy of Perceptual Control (3)

• More complex behavior seen as control of more
  complex perceptions
• Inputs from lower-level perceptions combined to
  form higher-level perceptions
• not new, e.g, complex cells, neural nets
• Outputs from higher levels sent to levels as
  reference levels (goals, purpose)
• new

14
(No Transcript)
15
Demonstration of Hierarchy

• Volunteer maintains arm parallel to ground
• Told to drop arm to side when hand pushed from
  above
• Predictions?
• Change in reference level from horizontal to
  vertical
• Lower level will perceive push before upper level
  perceives it and is able to change the reference
  level for the lower system
• Watch what happens!

16
The Whys and Hows of Behavior (2)

• Answers to why questions about behavior found
  by going up to higher-level perceptual control
  systems
• Why am I saying these sentences right now?
• Answers to how questions about behavior found
  by going down to lower-level perceptual control
  systems
• How am I saying these sentences right now?

17
Applications of PCT to Language (4)

• Language behavior is purposeful, intentional,
  functional, goal-directed
• Provides a social means of controlling ones
  perception
• Language is hierarchical
• auditory intensities gt phonetic features gt
  phonemes gt morphemes gt lexemes gt phrases gt
  clauses gt discourse
• Hierarchy of control from articulation to
  pragmatics
• Language behavior is usually successful despite
  many disturbances (e.g., environmental noise
  objects in mouth)

18
PCT provides a particular form of a connectionist
model (3)

1. Circular perceptual control systems arranged
   hierarchically
2. Neurons can compute (perceptrons) and transmit
   information (senory and motor systems)
3. Neurons can act as a comparators (using
   inhibitory and excitatory synapses)

19
PCT provides an architecture for research
discovery (4)

• Sensory (afferent) systems combining signals from
  lower levels
• Motor (efferent) systems sending reference levels
  to lower-level comparators (desired perceptions,
  not motor commands)
• Comparators (comparing upper-level outputs with
  lower-level inputs, with difference sent as
  reference level to lower level)
• Evidence of control behavioral and neural
  (controlled variable)

20
Conclusions (3)

• Advances in neurosciences are providing details
  at the micro level
• An understanding of behavior as the control of
  perception provides a macro framework for
  understanding the micro-level details
• Combining the two provides new important insights
  into the connections among brain, behavior,
  language and human purposes
• i.e., how mind (and mindful behavior) might be
  constructed from neurons

21
Resources

• Powers, W. T. (1973). Behavior The control of
  perception. Chicago Aldine de Gruyter.
• Gary Czikos books (full text online via
  garycziko.net)
• Without Miracles (1995)
• The Things We Do (2000)
• Control Systems Group website
• www.ed.uiuc.edu/csg (these other computer
  demos)
• Joel Walters (English, Bar-Ilan U)
• Application of PCT to understanding bilingualism

22
Connecting Purpose, Brain, and LanguageUnderstan
ding Language Behavioras the Control of
Perception

• Gary A. Cziko
• University of Illinois at Urbana-Champaign
• g-cziko_at_uiuc.edu garycziko.net

23
(No Transcript)
24

• What we have is a circuit, not an arc or broken
  segment of a circle. This circuit is more truly
  termed organic than reflex, because the motor
  response determines the stimulus, just as truly
  as sensory stimulus determines movement. Indeed,
  the movement is only for the sake of determining
  the stimulus, of fixing what kind of a stimulus
  it is, of interpreting it.(J. Dewey 1986, p.
  353)
• It is possible to step back and treat the mind
  as one big monster response function from the
  total environment over the total past of the
  organism to future actions.--Allen Newell
  (1998).

25
(Overview)

• Introduction
• Demonstration model of purposeful behavior
• Hierarchies of perceptual control
• Applications to language
• Implications for neural science
• Resources for further exploration

26
Demonstration Model of Perceptual Control
(overview)

• Rubber-band demo
• Powers computer demo of compensatory tracking
  (Demo 1 Step F)
• Dewey quote
• Powers computer model of compensatory tracking
  (Demo 2 Step G closing the loop)