Flexible Processing in Procedural-Based Categorization

1
Flexible Processing in Procedural-Based
Categorization
W. Todd Maddox1, J. Vincent Filoteo23 Grant C.
Baldwin1 1University of Texas 2Veterans Medical
Research Foundation, and 3UCSD
Perceptual Shift Exp.

• Participants Healthy young adults.
• Task Perceptual categorization task 5.
• Single line stimulus. Length and orientation
  vary across trials.
• Trial procedure
• Session Procedure
• 3 100-trial pre-change blocks
• 3 100-trial post-change blocks

• Much evidence supports the existence of
  neurobiologically distinct category learning
  systems e.g., 1 4.
• Procedural-based system mediated by the tail of
  the caudate nucleus and best suited for
  (nonverbalizable) information-integration
  category learning.

Stimulus Response Feedback

• Pre-change performance equated.

"Correct, that was an A" or "Wrong, that was
an A"
A
B

• Different stages of procedural-based
  categorization are differentially impacted by
  perceptual shifts and stimulus-label switches.
• Breaking cortico-caudate connectivity (through
  perceptual shifts) leads to a large initial
  performance cost, and slowed re-learning.
• Breaking the caudatepallidal connectivity
  (through stimulus-label switches) leads to a
  small initial cost, and slowed relearning.
• Breaking the cortico-caudate-pallidal
  connectivity leads to a large initial performance
  cost, but fast relearning.

Pre-Change Information-Integration Categories

• Perceptual shift leads to bigger cost than
  stimulus-label switch.
• Is pre-3 accuracy predictive of cost?
• Yes, if a perceptual shift (rPS .55, rSS
  .41)
• No, if stimulus-label switch (rSLS -.14)

• (A see above) Category representation
  component Many-to-one mapping from visual areas
  to the tail of the caudate.
• (B see above) Stimulus-label compatibility
  component System associates a response label
  with a region of perceptual space
• Caudate to globus pallidus

Post-Change Information-Integration Categories

• To determine whether there are different stages
  of procedural-based category learning.
• To determine whether these stages are
  differentially impacted by
• Stimulus shifts in the perceptual space
• Stimulus-label switches
• Both
• To determine the flexibility of re-learning
  following these experimental manipulations.

• More post changes lead to better re-learning.

• 1 Ashby, F. G., Alfonso-Reese, L. A., Turken,
  A. U., -Waldron, E. M., (1998). A
  neuropsychological theory of multiple systems in
  category learning. Psychological Review, 105,
  442-481.
• 2 Ashby, F. G., Maddox, W. T. (2005). Human
  Category Learning. Annual Review of Psychology,
  56, 149-178.
• 3 Maddox, W. T., Ashby, F. G. (2004).
  Dissociating explicit and procedural-learning
  based systems of perceptual category learning.
  Behavioural Processes, 66(3), 309-332.
• 4 Smith, E.E., Patalano, A.L., Jonides, J.
  (1998). Alternative strategies of
  categorization. Cognition, 65, 167-196.
• 5 Ashby, F. G., Gott, R. E. (1988). Decision
  rules in the perception and categorization of
  multidimensional stimuli. J Exp Psychol Learn Mem
  Cogn, 14(1), 33-53.
• http//homepage.psy.utexas.edu/homepage/group/Madd
  oxLAB/

• Perceptual shift cost larger than stimulus-label
  switch cost.
• The more locked in the larger the cost for
  perceptual shift, but not stimulus-label switch.
• Despite larger initial costs, the more post
  changes, the better post-change learning.